| Original Full Text | This electronic thesis or dissertation has beendownloaded from Explore Bristol Research,http://research-information.bristol.ac.ukAuthor:Floredin, IsabellaTitle:Care at the end of lifemeasuring outcomes for young people and their families for use in economic evaluationGeneral rightsAccess to the thesis is subject to the Creative Commons Attribution - NonCommercial-No Derivatives 4.0 International Public License. Acopy of this may be found at https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode This license sets out your rights and therestrictions that apply to your access to the thesis so it is important you read this before proceeding.Take down policySome pages of this thesis may have been removed for copyright restrictions prior to having it been deposited in Explore Bristol Research.However, if you have discovered material within the thesis that you consider to be unlawful e.g. breaches of copyright (either yours or that ofa third party) or any other law, including but not limited to those relating to patent, trademark, confidentiality, data protection, obscenity,defamation, libel, then please contact collections-metadata@bristol.ac.uk and include the following information in your message:•Your contact details•Bibliographic details for the item, including a URL•An outline nature of the complaintYour claim will be investigated and, where appropriate, the item in question will be removed from public view as soon as possible. This electronic thesis or dissertation has beendownloaded from Explore Bristol Research,http://research-information.bristol.ac.ukAuthor:Floredin, IsabellaTitle:CARE AT THE END OF LIFE: MEASURING OUTCOMES FOR YOUNG PEOPLE ANDTHEIR FAMILIES FOR USE IN ECONOMIC EVALUATIONGeneral rightsAccess to the thesis is subject to the Creative Commons Attribution - NonCommercial-No Derivatives 4.0 International Public License. Acopy of this may be found at https://creativecommons.org/licenses/by-nc-nd/4.0/legalcode This license sets out your rights and therestrictions that apply to your access to the thesis so it is important you read this before proceeding.Take down policySome pages of this thesis may have been removed for copyright restrictions prior to having it been deposited in Explore Bristol Research.However, if you have discovered material within the thesis that you consider to be unlawful e.g. breaches of copyright (either yours or that ofa third party) or any other law, including but not limited to those relating to patent, trademark, confidentiality, data protection, obscenity,defamation, libel, then please contact collections-metadata@bristol.ac.uk and include the following information in your message:•Your contact details•Bibliographic details for the item, including a URL•An outline nature of the complaintYour claim will be investigated and, where appropriate, the item in question will be removed from public view as soon as possible. CARE AT THE END OF LIFE: MEASURING OUTCOMES FOR YOUNG PEOPLE AND THEIR FAMILIES FOR USE IN ECONOMIC EVALUATION ISABELLA FLOREDIN A dissertation submitted to the University of Bristol in accordance with the requirements for award of the degree of DOCTOR OF PHILOSOPHY in the Faculty of Population Health Sciences Bristol Medical School Submitted in July 2023 2 3 ABSTRACT This thesis reports work to assess outcomes at the end of life (EOL) for young people. The work was conducted in response to a need to capture wider outcomes, for young people at the EOL and close persons, that is their family and friends, for the economic evaluation of interventions at the EOL. The first part of the work focused on outcomes for young people at the EOL where there is lack of appropriate measures for use in economic evaluation. Twenty-one in depth interviews with young people, parents and bereaved family members were undertaken. Constant comparison was used to identify attributes for a measure. The attributes are: Experience and enjoy, Independence, Freedom from physical and emotional suffering, Formal care and support, Control and Identity. The second part focused on outcomes for close persons and explored use of a close person measure in this context. Nine cognitive interviews with bereaved family members were undertaken for an initial assessment of the measure. Although completion errors were identified, the measure appears to be relevant for use in this context. The final part focused on the integration of outcomes for young people and close persons within an economic evaluation. A survey was used to explore relative weights given by the general population (n=566) to outcomes between close persons and young people of different age groups and at different time points in the last year of life. Mean weights were similar across tasks, at around 45% of weight given to close persons. Weights given to close persons increased as the trajectory got closer to death. Higher mean weights elicited for close persons were associated with socio-demographic and care related characteristics of the sample. This work significantly enhances the assessment of outcomes for young people and close persons in economic evaluation of interventions at the EOL. 4 5 COVID-19 STATEMENT This statement reports on the impact of COVID-19 disruptions on the empirical work undertaken for this thesis and steps taken to adjust the research. While adjusting empirical work undertaken is a common part of any research study, COVID-19 had a significant impact on the design and conduct of this study. This was particularly due to the research requiring recruitment of young people with life-limiting and life-threatening conditions, and their families, for qualitative work. Two funded extensions were received from Wellcome (three-month extension) and the University of Bristol (three-month extension) to mitigate the impact of COVID-19. Yet, these extensions were not sufficient on their own to enable the initial empirical work planned to be undertaken and therefore the initial focus of the study has shifted. Pre COVID-19, the planned focus of the study was to develop a young person end-of-life measure for use in economic evaluation by undertaking qualitative work with young people at the EOL. Work to develop the measure was planned to involve two phases. The first planned phase was to develop conceptual attributes for a measure through in-depth interviews with young people at the EOL. The second planned phase was to generate meaningful wording for these attributes through semi-structured interviews with young people at the EOL. A third planned objective of the study was to undertake initial assessment of the construct validity of the descriptive system of the newly developed measure. 6 The process of discussing alternative options for the early empirical work began in March 2020 with the most likely approach to begin with a meta-ethnography, rather than primary empirical research. This approach would have enabled conceptual attributes to be generated for the measure using existing literature and provide a draft measure followed by primary data collection around the choice of attributes and wording for the measure down the line if restrictions were lifted by that point. A scoping search to look at existing qualitative studies on young people at the end-of-life was undertaken. The aim of the search was to identify existing studies in which the views of young people near the end of their life are explored. Relevant study findings would then have been synthesized to conduct a meta-ethnography on what is important to young people near the end of life. However, the availability of qualitative studies undertaken directly with young people near the end-of-life was extremely limited and therefore it was decided to proceed with primary qualitative work, but using online interviews, to develop the measure. Recruitment of study participants (young people with life-limiting and life-threatening condition and parents/guardians for young people not able to take part themselves) for the in-depth interviews began in January 2021, with plans focused on recruitment through charitable organisations and hospices. Although some charitable organisations and hospices were willing to circulate recruitment materials to patients, as a result of the pandemic, it was extremely difficult to establish meaningful relationships with hospices and charitable organisations or to recruit through them via 7 staff. There were severe delays in recruitment with hospice settings not being open to the idea of researchers entering the settings given the vulnerable young people and families using the services and staff were subject to considerable over-work. In addition, the cancellation of the ‘Together for Short Lives’ paediatric palliative care conference which the researcher intended to attend in March 2020 further reduced opportunities to make valuable initial contacts in this area. Even following societal lockdowns, the situation had not eased, as the initial difficulties of the pandemic were then added to by the extreme workload pressures that health providers were under. As a result, recruitment strategies had to be adapted early on, and had to rely largely on social media. This recruitment strategy did produce interviews but at a very slow rate, given that this is an extremely vulnerable group, the interviews were on a sensitive topic and access was not enabled through an individual trusted by the informant. The six-month funded extension received enabled recruitment for in-depth interviews to continue until September 2022, increasing the sample size so that the qualitative findings of the study could be robust. However, due to the severe delays in recruitment it was not feasible in the remaining time of the study to continue with a second phase of semi-structured interviews to finalise the measure. In response to the recruitment difficulties, alternative elements to the research were added. These were to test the validity of an existing measure with bereaved family members of young people and to undertake an online public survey relating to citizen weights for the 8 allocation of resources between young people at the EOL and their family and friends. Although these additions shifted the focus of the study significantly, they have added to the scope of knowledge and understanding that would be contributed by the study. 9 ACKNOWLEDGEMENTS First, I would like to thank my four supervisors: Joanna Coast, Paul Mitchell, Samantha Husbands and Susan Neilson. Thank you for your time, input, guidance, and support throughout this process. It has been invaluable, and I have learned a lot from you. I would also like to thank Tim Peters whose input and knowledge have been of great value to my learning. Acknowledgements are also due to the Wellcome Trust for funding the PhD studentship and making this research possible. The completion of this research would not have been possible without the participants who generously gave up their time to take part in this study. Thank you for your willingness to share your personal experiences. I would also like to thank all the charitable organisations and hospices for their help in recruiting participants. Parts of this work have been presented at conferences and seminars. These include the Human Development and Capabilities Association summer school and conference, the International Health Economics Association conference, the South-West Palliative and End of Life Research Group and the LifeCourseCap Advisory Group. Thank you to all those who offered feedback on this work. I would also like to thank my parents for their support throughout my education, and my friends for their encouragement. Alex, thank you for the delicious food breaks and your words of advice. Nikos, Terry, Margaritis, Vana and Popi thank you for the 10 entertainment, encouragement, and warnings of the difficulties that this journey may entail. Mark, Louise, Barny, Josie and Yuri, thank you for your help and comments on different parts of this work. Last but not least, Lee, thank you for giving up your time to read a full draft of this thesis, for putting up with me when things were difficult, and always offering to help in any way that you could. Your support has been invaluable in completing this study. 11 AUTHOR’S DECLARATION I declare that the work in this dissertation was carried out in accordance with the requirements of the University's Regulations and Code of Practice for Research Degree Programmes and that it has not been submitted for any other academic award. Except where indicated by specific reference in the text, the work is the candidate's own work. Work done in collaboration with, or with the assistance of, others, is indicated as such. Any views expressed in the dissertation are those of the author. SIGNED: .......Isabella Floredin .... DATE:........12/04/2024....... 12 13 TABLE OF CONTENTS INTRODUCTION ........................................................................................................... 31 Chapter 1. Care at the end-of-life for children and young people ................................. 35 1.1. Children and young people .............................................................................. 35 1.1.1. Definitions: child; children; children and young people and young people 35 1.2. End-of-life and palliative care provision ........................................................... 36 1.2.1. Definitions: end-of-life, end-of-life care, palliative care ........................... 36 1.3. Key differences between adult and paediatric palliative care ......................... 38 1.3.1. Life-limiting conditions and life-threatening conditions in children and young people ........................................................................................................... 38 1.4. End-of-life and palliative care provision for children and young people ......... 44 1.4.1. Child development and paediatric palliative care .................................... 44 1.4.2. Care planning at the End-of-life in children and young people ................ 48 1.4.3. Current challenges and unmet needs in PPC and EOLC in the UK ............ 52 14 1.5. Conclusion ........................................................................................................ 66 Chapter 2. Theoretical frameworks and Practical aspects of economic evaluation of health care interventions. ............................................................................................... 69 2.1. Normative and positive approaches ................................................................ 70 2.2. Health economics and economic evaluation ................................................... 71 2.3. Welfarism (1900s-1940s) ................................................................................. 72 2.3.1. Utility ......................................................................................................... 73 2.3.2. Pareto efficiency ....................................................................................... 74 2.3.3. Kaldor-Hicks efficiency .............................................................................. 75 2.3.4. Cost-benefit analysis (1950s-70s) ............................................................. 76 2.3.5. Cost-effectiveness analysis – late 60s ....................................................... 79 2.3.6. Cost Utility analysis (CUA) and QALY ........................................................ 80 2.3.7. Valuation methods - QALYs ...................................................................... 82 2.4. Extra-welfarism - mid to late 80s ..................................................................... 84 2.4.1. Outcome focus in economic evaluation ................................................... 84 2.4.2. Source of valuation of relevant outcomes ................................................ 86 15 2.4.3. Weighting of relevant outcomes .............................................................. 87 2.4.4. Interpersonal comparability of relevant outcomes .................................. 87 2.4.5. Cost-Consequence analysis ....................................................................... 88 2.5. Capability approach- beyond health and maximisation .................................. 89 2.5.1. Functionings and capabilities .................................................................... 90 2.5.2. Freedom and agency ................................................................................. 91 2.5.3. Capabilities and EOLC for CYP ................................................................... 94 2.5.4. Critique of the Capability Approach .......................................................... 95 2.6. Shift of focus in health economics ................................................................... 98 2.7. Economic evaluation of child health .............................................................. 100 2.7.1. Shift of focus - from child's human capital to child indicators approach100 2.7.2. Child's own perspective and capability approach ................................... 101 2.7.3. Concepts of wellbeing and well-becoming ............................................. 102 2.7.4. Family as a unit of analysis ...................................................................... 103 2.8. Conclusion ...................................................................................................... 104 Chapter 3. Measuring outcomes in health and care for use in economic evaluation .. 107 16 3.1. Measuring outcomes in health and social care.............................................. 108 3.1.1. Definition of outcome measurement ..................................................... 108 3.1.2. Purpose of measuring outcomes in health and social care .................... 109 3.1.3. Types of outcome measures used in health and social care in the UK .. 111 3.1.4. Use of generic or specific measures in economic evaluation of health care 111 3.2. Methodological and practical considerations in measuring and valuing outcomes for use in economic evaluation of health and social care ....................... 113 3.2.1. Psychometric properties of measures .................................................... 113 3.2.2. Issues with valuing outcomes in economic evaluation........................... 116 3.3. Challenges in measuring EOLC outcomes for use in economic evaluation ... 117 3.3.1. Important outcomes of interventions may not be limited to HRQOL outcomes at EOL stage. ......................................................................................... 117 3.3.2. QALYs and time ....................................................................................... 119 3.3.3. Important outcomes may not be limited to patient outcomes in EOLC. 121 3.3.4. Valuation methods for EOLC ................................................................... 126 17 3.3.5. Practical challenges with patient and proxy measure responses at the EOL stage 127 3.4. Challenges in measuring outcomes in child health for use in economic evaluation .................................................................................................................. 130 3.4.1. Challenges in measuring outcomes in different developmental stages . 130 3.4.2. QALYs in children .................................................................................... 136 3.4.3. Important outcomes may not be limited to child outcomes .................. 138 3.5. Measuring outcomes for use in economic evaluation of EOLC in CYP .......... 139 3.5.1. Challenges specific to measuring outcomes in EOLC in CYP ................... 139 3.5.2. Measures typically used in PPC and EOLC in CYP ................................... 141 3.6. Methodological challenges in measuring capabilities in economic evaluation 143 3.7. Measuring outcomes for use in economic evaluation of care at the end-of-life in young people and those close to them ................................................................. 145 3.7.1. Measuring outcomes at the end-of-life for young people ..................... 145 3.7.2. Measuring outcomes at the end-of-life for those close to young people 164 18 3.8. Conclusion ...................................................................................................... 167 Chapter 4. Qualitative methodology and methods ...................................................... 171 4.1. Introduction .................................................................................................... 171 4.2. Methods for identifying attributes in measures for use in economic evaluation.................................................................................................................. 172 4.2.1. Identifying conceptual attributes ........................................................... 172 4.2.2. Guidance on the use of qualitative methods to develop conceptual attributes for a measure. ...................................................................................... 174 4.3. Methods to assess content validity and feasibility of use of measures. ....... 184 4.3.1. Use of cognitive interviewing to assess content validity and feasibility of use. 185 4.4. Guidance on conducting qualitative research with a vulnerable population to explore sensitive topics. ............................................................................................ 188 4.4.1. Research participants and recruitment .................................................. 188 4.4.2. Data collection method ........................................................................... 193 4.4.3. Conducting qualitative research with a vulnerable population in the context of COVID-19 ............................................................................................. 198 19 4.5. Conduct of in-depth interviews to develop the conceptual attributes for an EOL measure for young people. ................................................................................ 202 4.5.1. Research Design ...................................................................................... 202 4.6. Conduct of cognitive interviewing ................................................................. 218 4.6.1. Sampling and recruitment ...................................................................... 219 4.6.2. Research Design ...................................................................................... 219 4.7. Ethical issues and conduct of qualitative interviews ..................................... 224 4.7.1. Process to obtain consent from participants. ......................................... 224 4.7.2. Data Storage and Sharing ........................................................................ 225 4.7.3. Assessment and management of potential risks of the research .......... 226 4.8. Summary ......................................................................................................... 230 Chapter 5. Young people at the end of life - Capabilities and conversion factors ....... 231 5.1 Recruitment and participant characteristics .................................................. 232 5.2 Presenting interview findings ......................................................................... 237 5.3 Capabilities and conversion factor themes .................................................... 237 5.3.1 Experience and enjoy .............................................................................. 240 20 5.3.2 Independence ......................................................................................... 248 5.3.3 Free from physical suffering ................................................................... 251 5.3.4 Free from emotional suffering ................................................................ 257 5.3.5 Formal care and support ......................................................................... 262 5.3.6 Control..................................................................................................... 266 5.3.7 Identity .................................................................................................... 274 5.4 Summary......................................................................................................... 281 Chapter 6. Feasibility of use and content validity of ICECAP-CPM ............................... 283 6.1 Participant characteristics .............................................................................. 283 6.2 Findings........................................................................................................... 286 6.2.1 Measure completion rating findings: errors and struggles. ........................ 286 6.2.2 Views of respondents on measure ......................................................... 292 6.3 Summary......................................................................................................... 293 Chapter 7. Quantitative methodology and methods.................................................... 295 7.1 Introduction .................................................................................................... 295 7.2 Methods for eliciting relative weights for different outcomes ...................... 296 21 7.3 Conduct of survey ........................................................................................... 300 7.3.1 Study design ............................................................................................ 300 7.3.1.2 Development of the survey ................................................................. 301 7.3.2 Sampling and recruitment for panel survey ........................................... 310 7.3.3 Data collection ........................................................................................ 314 7.3.4 Analysis .................................................................................................... 316 7.4 Summary ......................................................................................................... 320 Chapter 8. Survey results .............................................................................................. 321 8.1 Introduction .................................................................................................... 321 8.2 Survey results ................................................................................................. 321 8.2.1 Survey data and cleaning ........................................................................ 321 8.2.2 Characteristics of the respondents ......................................................... 322 8.2.3 Estimation of weights .............................................................................. 326 8.2.4 Associations between pie task responses and characteristics of the sample 337 8.3 Summary ......................................................................................................... 347 22 Chapter 9. DISCUSSION ................................................................................................. 349 9.1. Introduction .................................................................................................... 349 9.2. Summary of findings ....................................................................................... 350 9.2.1. Outcomes for young people at the EOL .................................................. 350 9.2.2. Outcomes for those close to young people at the EOL .......................... 351 9.2.3. Relative weights of outcomes ................................................................. 352 9.3. Discussion of findings ..................................................................................... 353 9.3.1. Outcomes for young people at the EOL .................................................. 353 9.3.2. Outcomes for those close to young people at the EOL .......................... 365 9.3.3. Relative weights of outcomes at the EOL ............................................... 368 9.4. Reflections on strengths and limitations of the work .................................... 372 9.4.1. Reflections on qualitative work .............................................................. 372 9.4.2. Reflections on quantitative work ............................................................ 385 9.5. Implications for policy .................................................................................... 389 9.5.1. Including capabilities as broader outcomes in economic evaluation of EOL 389 23 9.5.2. Including capability outcomes for those close to young people at the EOL in economic evaluation ......................................................................................... 392 9.6. Implications for service planning and clinical decision making ..................... 393 9.7. Implications for research ................................................................................ 395 9.8. Future research to develop the capability measure ...................................... 396 9.8.1. Development of a young person EOL measure ...................................... 396 9.8.2. Outcomes for those close to young people ............................................ 398 9.8.3. Relative weights between outcomes ...................................................... 399 9.8.4. Outcomes for other CYP age groups at the EOL ..................................... 400 9.9. Next steps ....................................................................................................... 401 9.10. Conclusion ................................................................................................... 401 24 LIST OF TABLES Table 1 - Reporting ability of health status by age ....................................................... 132 Table 2 - Relationships between studies included in the review. ................................ 155 Table 3 - Main themes identified in included studies ................................................... 159 Table 4 – Young people characteristics ........................................................................ 234 Table 5 - Descriptive characteristics of participants (n=21) ......................................... 235 Table 6 - Participant characteristics .............................................................................. 285 Table 7 - Errors and struggles identified by type across ICECAP-CPM dimensions (n=9) ....................................................................................................................................... 288 Table 8 – Budget pie chart tasks in survey:................................................................... 308 Table 9 - Sample size by MOE value.............................................................................. 313 Table 10 - Conceptually similar groups for explanatory variables ................................ 318 Table 11 - Characteristics of the survey sample ........................................................... 324 Table 12 - Mean weights elicited by pie chart task (n=566) ......................................... 329 Table 13 - Reasoning responses for Task 1 ................................................................... 332 25 Table 14 - Reasoning responses for Tasks 2a and Task 2b ........................................... 334 Table 15 - Reasoning responses for Task 3a and Task 3b ............................................. 336 Table 16 - Initial analysis (Characteristics unadjusted) – Parametric tests .................. 339 Table 17 - Analysis within groups (p-values for characteristics adjusted for others in the same group) .................................................................................................................. 341 Table 18 - Analysis across thematic groups (p-values for characteristics adjusted for all other variables in the relevant final model) ................................................................. 344 Table 19 - Mean weights elicited for family and friends group by subgroups ............. 345 Table 20 - Comparing dimensions of CYP measures with capability themes identified. ....................................................................................................................................... 361 Table 21 - Dimensions of adult PC and EOL measures ................................................. 363 Table 22 - Search terms used in Medline database. ..................................................... 409 Table 23 - Characteristics of included publications in systematic review. .................... 414 Table 24 - Quality assessment summary against CASP checklist criteria ..................... 457 Table 25 - Mean weights elicited by pie chart task (n=587) ......................................... 563 Table 26 - Pie task Median (IQR) weights elicited – non-parametric descriptive statistics ....................................................................................................................................... 564 26 Table 27 - Initial analysis – p-values for characteristics unadjusted (non-parametric tests) .............................................................................................................................. 565 27 LIST OF FIGURES Figure 1 - systematic review flow diagram ................................................................... 150 Figure 2 - Capabilities and conversion factors identified.............................................. 282 Figure 3 - Task 1 (task used in the survey study) .......................................................... 303 Figure 4 - Task A (task used in the deliberative study by Coast et al. 2021) ................ 304 Figure 5 - Task 2 (task used in the survey study) .......................................................... 307 Figure 6 - Histograms showing weights elicited by task. .............................................. 327 Figure 7 - Deliberative work introductory instructions for task completion by Coast et al. 2021. ......................................................................................................................... 559 28 LIST OF ABBREVIATIONS Activities of daily living – ADLs Advanced care plan - ACP Best-worst scaling – BWS Capability approach – CA Cardiopulmonary resuscitation – CPR Children and young people – CYP Cost-benefit analysis – CBA Cost-consequence analysis – CCA Cost-effectiveness analysis – CEA Cost-utility analysis – CUA Do not resuscitate – DNR Economic and Social Research Council – ESRC End of life – EOL End of life care – EOLC Health care professional - HCP Health related quality of life – HRQOL ICECAP-A – ICEpop CAPability measure for adults ICECAP-O – ICEpop CAPability measure for older people ICECAP-SCM – ICEpop CAPability Supportive Care Measure International children’s palliative care network – ICPCN International classification of diseases version 10 – ICD-10 29 Incremental cost-effectiveness ratio - ICER Life-limiting conditions – LLC Life-threatening conditions – LTC Medical Research Council – MRC National Children's Bureau - NCB National health service - NHS National Institute for Health and Care Excellence – NICE Paediatric palliative care - PPC Paediatric Palliative Medicine – PPM Palliative care – PC Patient reported outcome measures – PROMS Person trade off – PTO Quality adjusted life year – QALY Quality of life - QOL Royal College of Paediatrics and Child Health – RCPCH Standard gamble – SG Time trade off – TTO Together for short lives (TfSL) United nations – UN United Nations International Children's Emergency Fund – UNICEF United Nation's convention on the rights of the child – UNCRC Willingness to accept – WTA Willingness to pay – WTP World health organisation – WHO 30 31 INTRODUCTION This thesis explores the assessment of outcomes for young people at the end of life (EOL) for use in economic evaluation of health and care interventions. This assessment involved identifying important outcomes for young people at the EOL and assessing use of an existing outcome measure with those close to young people at the EOL (close persons). To enable outcomes for close persons to be included alongside outcomes for young people at the EOL, an integrative economic evaluation framework is also explored to generate information about how the public weights between outcomes for young people and outcomes for close persons. Chapter one sets the context for the thesis. This chapter introduces terminology that is used within the thesis and provides a broad review of the literature on aspects of paediatric palliative care (PPC) and EOL care provision for children and young people (CYP) relevant to the UK context. Key differences between adult and paediatric palliative care are highlighted. The rising prevalence in the number of CYP who may benefit from care interventions at the EOL and the increasing demand for, and costs of, interventions at the EOL suggest that there is need to properly assess outcomes of these interventions within economic evaluation for the best use of the resources available. Chapter two reviews the theoretical frameworks underpinning economic evaluation of health and care interventions. The shift from welfarism to extra-welfarism and the 32 growing interest in the application of the capability approach (which allows for a broader evaluative space to assess outcomes beyond health and beyond the patient), is outlined. Terms used within the different theoretical approaches are introduced, including for the capability approach, the meanings of capabilities and conversion factors and how these terms are used. Chapter three has two parts. The first part focuses on the practical aspects of economic evaluation and outcome measurement. The particular challenges in measuring outcomes at the EOL and for child health are outlined followed by the challenges in using the capability approach as an alternative evaluative space to measure outcomes for use in economic evaluation. In this part of the chapter, it is suggested that important outcomes at the EOL for CYP may not be limited to health and therefore the broader evaluative space that the capability approach can offer could be useful to draw on in this context. Arguments are also made to include close person outcomes of interventions at the EOL within economic evaluation. Resulting challenges in relation to incorporating outcomes beyond the CYP, within economic evaluation are presented. The second part of chapter three reviews existing work to identify gaps in research on the assessment of EOL outcomes for young people and their close persons. A lack of appropriate measures for young people at the EOL is found and empirical work with this population is limited. There is, however, an existing close person measure which aims to capture broader outcomes for those close to the dying. 33 Chapter four outlines the qualitative methodology and methods used in this research. This chapter has two key parts. The first part describes the methodology and methods used to identify what is important to young people at the EOL. This work aimed to inform the conceptual attributes for a young person EOL measure. A ‘bottom-up’ approach using qualitative methods with the relevant population to develop and validate measures is justified. The second part outlines the qualitative methodology and methods used to assess an existing close person measure with bereaved close persons of young people at the EOL. Chapter five reports the results of twenty-one in-depth interviews with young people, parents and bereaved family members. The findings focus on what is important to young people at the EOL. Seven capabilities and conversion factors were identified as important to young people. Chapter six reports on the results of nine cognitive interviews with bereaved family members of young people to assess feasibility of use and content validity of an existing close person measure. Although completion errors were identified, the close person measure appeared to be relevant for use in this context. Chapter seven outlines the quantitative methodology and methods used in this research to explore how broader outcomes for young people and outcomes for close persons could be integrated within economic evaluation and decision-making for resource allocation. The quantitative methods used to obtain societal values for the relative weights between outcomes for young people and outcomes for close persons 34 are justified. The chapter then describes the development of the survey, sampling, data collection and analysis of quantitative data. Chapter eight reports on the quantitative findings of this research undertaken with a representative sample from the UK population (n=566). This part of the work estimated relative weights between outcomes for young people at the EOL and outcomes for close persons for different young person age groups and at different time points in the last year of life. Weights elicited were similar across tasks, at around 45% of weights given to close persons. Weights given to close persons increased as the trajectory got closer to death. Independent associations between weights obtained and the characteristics of the sample were then examined. Higher relative weights elicited for close persons were associated with socio-demographic and care related characteristics of the sample. The final chapter of this thesis provides a summary and discussion of the key findings reported in Chapters five, six and eight. The work is discussed in light of existing evidence. It then reflects on the appropriateness of the qualitative and quantitative methodology and methods used in this thesis and the implications of the findings for policy making. Future research across a number of areas where this research can be taken forward is then suggested. The chapter concludes by outlining the key contributions of this research for the assessment of outcomes at the EOL for young people and those close to them. 35 CHAPTER 1. CARE AT THE END-OF-LIFE FOR CHILDREN AND YOUNG PEOPLE 1.1. Children and young people 1.1.1. Definitions: child; children; children and young people and young people In this thesis a number of terms are used for children and young people: ‘child/children’; ‘children and young people’ (CYP) and ‘young people’. The terms ‘child/children’ are used in the thesis to refer to children below the age of 18 yearsa.1 The term ‘CYP’ is used to refer to CYP below the age of 25 yearsb.2 In the empirical chapters of the thesis in which the findings are reported, that is Chapter 5, 6 and 8; and in Chapter 9, where the thesis findings are discussed, the term ‘young people’, a As defined by United Nations Children's Fund (UNICEF) and United Nations Convention on the Rights of the Child (UNCRC). b As used in the latest (2019) NHS England long term plan report. 36 rather than the terms ‘child/children’ or ‘CYP’, is used to refer to young people, aged 14-25 years, which is the young people age group that the empirical work of the thesis is focused on. 1.2. End-of-life and palliative care provision 1.2.1. Definitions: end-of-life, end-of-life care, palliative care The terminology around End-of-Life (EOL), End-of-Life Care (EOLC) and Palliative Care (PC), is confusing and terms are often used with different meanings within clinical and academic circles.3 The term ‘supportive care’ is often used to denote palliative care and the term ‘terminal care’ refers to care provided in the last days or hours of life.3, 4 The terms PC and EOLC are often used interchangeably by the public.5, 6 The World Health Organisation (WHO) defines PC, for adults and children, as an approach that: “improves the quality of life of patients and their families facing the problem associated with life-threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychosocial and spiritual.”(para.1).7 37 There is thus a multidisciplinary approach to PC that involves care being provided by different health care professionals (HCPs) including doctors, nurses, physiotherapists, occupational therapists, social workers and/or mental health professionals. Furthermore, PC uses a broad approach to care that involves support provided to the family of the patient. PC is not limited to, but includes, EOLC.8 Whilst EOLC involves care for people at the last phase of life, people may receive PC at an earlier stage of their illness when they are undergoing treatment for their condition4, 9 There is uncertainty in relation to when EOL begins as it can be difficult to predict the time left at the end stage of the illness.3 There have been attempts to use prognostic tools as a way to predict survival time, although it has been argued that the majority of these tools are not superior to the clinician's predictions.10 In the UK, current National Health Service (NHS) information on what EOLC involves suggests that EOLC “may last a few days, or for months or years” (para. 12 )9 and people are perceived to be approaching EOL “when they are likely to die within the next 12 months, although this is not always possible to predict” (para. 14).9 As with adults, there is uncertainty around the timing of EOLC for children and young people (CYP).11 The International Children’s Palliative Care Network (ICPCN) states that the definition of the beginning of EOLC varies according to the individual patient and the judgement of the HCP involved.12 The ICPCN definition further suggests that in some cases it is the patient or the family that first recognises the beginning of EOL.12 38 1.3. Key differences between adult and paediatric palliative care Although adult PC and paediatric palliative care (PPC) share similarities in terms of the multi-disciplinary approach to care and the uncertainty around when the EOL begins, there are important differences between them.11, 13, 14 PPC is a different specialty from adult PC.15 PPC begins “when illness is diagnosed, and continues regardless of whether or not a child receives treatment directed at the disease”.7 Some CYP need PPC from birth, for a long period of time and may then transition to adult PC, whilst adults usually acquire a need for PC near the EOL and for a shorter period of time.8, 13, 16 1.3.1. Life-limiting conditions and life-threatening conditions in children and young people Compared with adults, the conditions that CYP have are very diverse, and diagnoses are often specific to childhood and young adulthood.17 The terms ‘life-threatening condition’ (LTC) and ‘life-limiting condition’ (LLC) are used to describe the population of CYP who could benefit from receiving PPC.18, 19 A LTC is understood as a condition where a cure is possible and a medical intervention may be successful but there is possibility of premature death.20 A LLC is a condition in which there is no reasonable hope of cure and from which the patient will die.20 There is lack of consensus among 39 experts on when these two terms should be used due to uncertainty about which conditions have no hopes of cure.18 1.3.1.1. Life-limiting and life-threatening conditions and care provision in the UK There have been calls for an increased need for specialist PPC and EOLC following studies reporting a rising prevalence of LLCs/LTCs in CYP in the UK15, 17, 19, 21 and internationally.22-24 A 2011 estimate of the number of CYP (0-19 years old) with LLCs/LTCs who would benefit from PPC globally is around seven million.24, 25 In the UK, prevalence of LLCs/LTCs in children (0-19 years) has risen from 32,975 in 2001/2002 to 86,625 in 2017/2018 and future prevalence is estimated to be between 67.0 and 84.2 per 10,000 children (0-19 years), i.e. between 96,275 and 121,023, by 2030.19 Rising prevalence of LLCs/LTCs in CYP is in part attributed to increased survival following improved care and advances in treatment and medical technology, including use of aggressive treatment of complications, that contribute to higher life expectancy and quality of life.8, 19, 26 As a result, many children diagnosed with LLCs/LTCs who would have previously died at a younger age are now living into adolescence and sometimes into early adulthood.15, 19, 27 The number of young people with a LLC/LTC reaching adulthood is increasing with a significant number of deaths in young people in their early 20s.19 40 1.3.1.2. Directory of life-limiting conditions in children and young people in the UK An extensive list of LLCs in CYP - including both malignant and non-malignant conditions – has been compiled (2011) in a directory using data from death certificates, children's hospices and PC teams across the UK.28, 29 The directory includes almost 400 classifications of diseases (ICD-10c).28, 29 However, it is suggested that the list is not exhaustive as methods used to collate data could not capture conditions that are too rare; for the directory to remain current there also needs to be a method of adding new conditions when they become apparent.28, 29 It is noted that the list of conditions that are currently considered to be life-limiting could change in the future with advances in medicine.28, 29 1.3.1.3. Distribution of life-limiting diagnoses in the UK According to the most recent study on prevalence of LLCs/LTCs in CYP aged 0-19 years in the UK (2020), an increase in prevalence of LLCs/LTCs was observed for the period between 2001/2002 to 2017/2018 in all diagnostic groups: Circulatory, Congenital, Gastrointestinal, Genitourinary, Haematology, Metabolic, Neurology, Oncology, Perinatal, Respiratory, Other.19 In 2017/2018 the highest prevalence was in the c International classification of diseases version 10 (ICD-10) 41 congenital abnormalities diagnostic group (27.2 per 10,000) and the second most prevalent diagnostic group was neurological disorders (10.8 per 10,000).19 Lowest prevalence was in the circulatory diagnostic group (2.4 per 10,000) and ‘other’ group (1.7 per 10,000).19 It is important to note that although this study found congenital abnormalities as the category with the highest prevalence in ages 0-19 years, the most recent NHS England long term plan (2019) notes that cancer remains the biggest cause of premature death among CYP of 5-14 years of age.2 1.3.1.4. UK demographics of life-limiting conditions in children and young people Recent studies (2012, 2014, 2020) reporting demographic information on prevalence of LLCs in CYP (0-19 years old) in the UK show an increase in prevalence over all demographic and deprivation categories although not to an equal extent.15, 19, 21, 30 Prevalence is higher in the more deprived locations and locations where more ethnic minority groups live, whilst more rural areas have lower prevalence.15, 19, 21, 30 For the time period 2017/2018, higher prevalence of LLCs/LTCs in CYP (0- 19 years old), is reported: - in children under 1-year old (226.5 per 10,000).19 - among boys (72.5 per 10,000) rather than girls (60.0 per 10,000).19 - among children of Pakistani origin (103.9 per 10,000).19 42 - among children living in areas of higher deprivation (13% in the most deprived areas, 8% in the least deprived areas).19 1.3.1.5. Illness trajectories Defining different illness trajectories is a useful aid for clinicians when offering information to patients and their family on probable needs in the near future, which could enable them to feel in greater control of their situation.18, 31 Trajectories are also useful for informing advance plans for appropriate care, choosing a preferred place of care and death and taking into consideration the multiple needs - physical, psychological, social and spiritual - that the patient and family may have at different points in time.31, 32 Most adults at the EOL have a progressive illness, such as cancer, cardiovascular disease or a respiratory disorder.31 There are three different illness trajectories commonly observed in a progressive chronic illness in adults31: - steady progression with a clear terminal phase (example: cancer).31 - gradual decline interrupted by acute deterioration episodes and recovery, followed by a more sudden death (example: organ failure).31 - prolonged gradual decline (example: dementia or frail elderly).31 It has been noted that PPC differs in terms of the conditions, and their trajectories, that CYP have. These are more diverse and often rare.33, 34 The variety of illness trajectories in CYP includes17, 33: 43 - Conditions that could be potentially curable. These conditions include cancer, organ failure and children on long-term ventilation.17 - Conditions for which therapy prolongs life before premature death. These conditions include cystic fibrosis and Duchenne muscular dystrophy.17 - Conditions which are progressive and there is currently no cure available. These conditions include severe metabolic conditions such as batten disease.17 - Conditions which are not progressive, but the child will likely die before adulthood as a result of complications. These conditions include severe cerebral palsy and complex disabilities following for example a spinal cord injury.17 Apart from the above classification of LLCs/LTCs, which was developed by a UK charity called Together for Short Lives (TfSL)d, another recent study developed a framework for the above categories called ‘The Spectrum’e.35 Although the framework uses the categories of conditions as developed by TfSL, it also considers the distinct care needs for each prognostic category.35 The different categories on the Spectrum are colour coded36: - Green for CYP who are diagnosed with a LTC.36 d Together for Short Lives is a leading charity in the UK for CYP with LLCs/LTCs. e Also known as The Rainbow or Spectrum of Children’s palliative care needs. 44 - Yellow for CYP who may live for many years but death before 18 years of age is not unexpected.36 - Amber for CYP with progressive deterioration and death is not unexpected in months or years.36 - Red for CYP who are critically ill, and death is expected within next the few weeks.36 The Spectrum can be used by a clinician or a care team, who may choose for the same child a different colour category at different times, as their situation and care needs change.37 1.4. End-of-life and palliative care provision for children and young people 1.4.1. Child development and paediatric palliative care Throughout childhood, cognitive, emotional and physiological development is ongoing, and so PPC providers need to have a good understanding of, and respond to, those developmental changes when providing care to a child.16, 17 It is important to note that the terms 'child' and 'childhood' have a number of different definitions depending on the discipline, culture and historical periods; thus they are both very ambiguous terms.38 In the legal sciences and according to the English legal 45 system, a child is a person under the age of 14, 16 or 18 depending on the case and context.39 In the Children Act 198940 and 200441 and the Family Law Act 199642 a child is defined as a person under 18 years of age. In the social sciences a child is considered to be a person who hasn't yet reached an age that has full rights and obligations, including economic, in society.38 It is useful to consider all these different definitions in this thesis as each one focuses on different aspects of childhood. Literature on PPC and EOLC suggests that children process information differently depending on their developmental stage.17, 33, 43 Developmental stage affects how they may assess pain and other symptoms. It also affects how able a child is to take part in decision-making, how they understand death and how they experience their illness overall.33, 43 For example, infants may have a limited understanding of their illness that is restricted to experiencing physical symptoms such as pain.33 An adolescent will have a more developed understanding of how their illness impacts their life, for example in terms of changes in their body image and an increased dependency on the person with parental responsibility.33 A child may also be able to understand their situation better than their verbal ability allows them to express to others.43 According to theories within the psychology discipline, a child goes through different developmental stages within childhood.44 Developmental theories include biological, social, emotional and cognitive development changes observed across childhood and adulthood.44 The developmental stages of childhood consist of different stages from early, middle and late childhood to adolescence, during all of which the knowledge and 46 cognition of the child develops.45 Piaget (1924, 1932, 1936) argued that a child's biological maturity is a prerequisite for the child to be able to engage with their social environment.44, 46-48 Erikson (1950) argued that people go through eight different developmental stages across their life course.49 The stages for CYP are: infancy (0-1 years old); toddler (1-2 years old); early childhood (2-6 years old); early school years (6-12 years old); adolescence or youth (12-18 years old) and young adults (18 years old to late 20s).49 In each of these stages the child develops different aspects of their identity.44, 49 Infants, being solely dependent on adults, develop a basic trust in others, whilst toddlers, who begin to become less dependent, behave with more autonomy.49 In early childhood, children begin to socially engage and interact with other children mainly through playing, from which they develop feelings of guilt, shame and learn to take responsibility for their actions.49 Then through their early school years children continue to socially interact with more people outside their family circle through which they further develop their intellectual skills.49 According to Erikson, during adolescence, children are predominantly concerned with their identity and sexuality and then as young adults their concerns are around developing and establishing relationships which are intimate and secure.49 Erikson recognised that transition from one stage to the next one, depends on the social context and the opportunities available to the child from the person(s) with parental responsibility.38, 49 A need to look beyond the developmental model for an explanation of CYP's understanding of death and major illness has been highlighted in the PPC literature.43 47 It has been argued that there are no clear-cut age-specific groups in CYP for each developmental level and at times CYP who experience illness may develop a 'mature' understanding of their illness beyond their years.33, 43, 50, 51 In early childhood (up to 6 years old), young children are usually not considered to have the emotional maturity to separate physical from emotional suffering.51 Yet, it has been suggested that young children may realise that they are at the EOL stage not only because of the progression of the disease but also because of how others around them react and feel.51 Bluebond-Langer (1978) suggested that CYP with major illnesses are aware of what is going on and adapt to their environment.50 For example, if a child senses that others around them feel uncomfortable when discussing death and dying then the child may avoid discussing this with them in an attempt to protect their parents from having conversations that might upset them.43, 50 Perhaps this uncertainty about the cognitive ability of a child is one of the reasons why the age range and age categories used for CYP in different PPC related studies varies.15, 23, 34, 52 As this study is focused on care provided in the UK, the thesis uses the term CYP as used in the latest (2019) NHS England long term plan report in relation to service provision for CYP.2 This term includes children up to and including the age of 17 and young adults between 18 to 25 years old.2 48 1.4.2. Care planning at the End-of-life in children and young people Care planning, also called care goals, is considered to be an important part of EOLC discussions for both adults and CYP at the EOL.53 Goals agreed vary for each person and could refer for instance to resuscitation preferences, treatment decisions, future care needs, physical and emotional comfort, informing and involving the patient, goals related to cultural and spiritual needs, support for family and preparing for death.54 Goals of care are discussed and agreed between the patient and HCPs as well as family.53 When discussing goals with the family, the family may wish to actively participate in the discussions and decision-making or may wish to defer to HCPs to develop those care goals.53 Discussions around goals and regular reassessment of goals, as goals and preferences may change as illness progresses, can help the patient and family feel more in control over their condition.54, 55 Yet, views on the goals of care may be divergent between patients, family and HCPs.55, 56 For instance, in a study (2000) on care goals for children receiving EOLC, parents' primary goal at the time of the diagnosis focused on cure for the condition while HCPs considered additional goals related to PPC.55 49 1.4.2.1. Relationship and transition between curative treatment, palliative, and end-of-life care in the UK Transition from curative treatment – focused on extending life - to PC and EOLC – focused on pain and symptom relief – is challenging for both adults and CYP.11, 57 Although PPC is often viewed by families as their last option when curative treatment has failed or is not thought to be appropriate, CYP may benefit from PPC alongside curative treatment (parallel planning).16, 17, 58, 59 Parallel planning can be challenging due to the unpredictable nature of LLCs in children.60 There are often no clear boundaries between the two types of care, and care varies for the different conditions and situation.17, 58, 61 Different ways that PPC and curative treatment are offered include: - A gradual shift from curative treatment to PPC.17 - Curative treatment alongside PPC, each one at different stages and at different degrees.17 - Only PPC when curative treatment is not thought to be appropriate.17 - Curative treatment followed by only PPC.17 Clinicians often find it difficult to recognise, consider and communicate transition from curative treatment to PPC and EOLC for CYP.59, 62 This is due to heterogeneity and infrequency of LLCs in childhood which as a result hinders the development of a sufficient scientific evidence-base to guide practitioners in providing prognostic 50 information.60, 62 In addition, the use of newly developed treatments makes it even more difficult for practitioners to provide prognostic information due to the uncertainty involved.62 Finally, some practitioners do not feel confident enough to communicate bad news to the family and patient and may postpone EOL discussions.11, 59, 62 The challenge in early recognition and communication of the need for transition to EOLC can hinder opportunities for appropriate EOLC that focuses on treatment to reduce suffering for the child as well as the family.59, 62, 63 1.4.2.2. Advance care planning The use of an 'Advance Care Plan' (ACP) in PPC, is endorsed by the National Institute for Health and Care Excellence (NICE).16, 64, 65 ACPs include records of child and family wishes, decisions related to future and current EOLC and plans agreed with all the people involved - professionals, family and child - for potential intermittent deteriorations.17, 66 The ACP provides a guide for professionals involved, as well as the family, to make decisions about the child's care particularly when the child becomes too unwell to be able to communicate with them.66, 67 ACPs are specific to the needs and wishes of both the child and the family 66 and may include preferences in relation to social care needs, spirituality and religion, funeral arrangements, place of death and needs of parents and siblings.68 51 1.4.2.3. A ‘good death’ The concept of a 'good death' is often discussed within EOLC 69-71 and considered to be one of the most important outcomes in EOLC for both the patient and their family.72, 73 The concept was initially formulated for adult care, but it is now acknowledged to also be an important part of care provided to CYP.18 There is little agreement on what constitutes a ‘good death’, and it is suggested that it is not possible to have a single adequate definition of the concept as differences in socio-economic and cultural contexts influence perceptions of a ‘good death’.70-72, 74 Even though there may be similarities between what constitutes a ‘good death’ for adults and for CYP, such as with regards to managing physical suffering, fundamental differences in the clinical and social context between adults and CYP means that this concept is different for adults and CYP.73, 75, 76 Studies reviewing literature on what constitutes a good death for adults and CYP have attempted to identify themes.70-72, 77 Common themes for adults included: being in control and having choice about the dying process and treatment, being comfortable and pain-free, having dignity, emotional wellbeing, trusting relationships with care providers, having values and beliefs honoured and minimising burden to family.70-72, 77 Themes of a good death for CYP shared the aspects of practical support and being in control that are evident in the adult themes but also included aspirational goals such as doing as much as possible.71 52 1.4.3. Current challenges and unmet needs in PPC and EOLC in the UK It is suggested that growing demand for PPC and EOLC services in the UK has not been met with an adequate increase in provision of health and social care resources.21, 23, 78, 79 Although the focus of this thesis is not on PPC and EOLC service provision issues it is important to have an awareness of the particular challenges and unmet needs of families highlighted in recent UK reports. A brief understanding of the current issues families experience is important to provide important context for the empirical work in this thesis. 1.4.3.1. Funding of PPC and EOLC in the UK PPC and EOLC in the UK is provided by both statutory and voluntary sectors in a number of settings, including the CYP’s home, in hospitals and in children's hospices.8, 80 The NHS and local authorities usually show preference for care being provided at home as costs involved are significantly less than care provision costs in a hospital or hospice.81-83 The majority of PPC is funded by the NHS, local government and the voluntary sector.8, 80 It is suggested that it is difficult to know how funding is distributed as there is no record of what is being funded across England.84 The contribution of the voluntary sector is crucial and helps ease the pressure on the NHS as it offers a significant proportion of PPC, including EOL care.8, 80 At present, 53 there are 53 children’s hospices in the UK which offer a range of services.79, 85, 86 It is estimated that in 2018-2019, children's hospices in England provided active clinical support to 8160 CYPf (based on an average of 240 CYP per hospice charity) and that 518 CYP have received EOLC (based on an average of 15 CYP per hospice charity).79 NHS England has acknowledged a gap in funding for children's hospices and that the rise of clinical care costs has been greater than increases in local NHS funding for this area of care.2, 80 On average, funding contributed to children’s hospices by NHS England in 2021/2022 was £17 million annually, which represented only around 15% of what children’s hospices spent on care they provided.80 The 2019 NHS long-term plan stated that contribution to hospices by NHS England will increase up to £25 million annually in 2023/2024.2, 80 Caseloads in children’s hospices, including EOLC, have also increased by 11% from 2020/2021 to 2022/2023 and there are growing costs of care provision due to inflation and increased demand for PPC.80 The funding that hospices receive from NHS England also varies significantly across local areas. For example, in 2019, 15% of children's hospice charities did not receive any funding from their local clinical commissioning groupg (CCG).79, 80 Attention has also been drawn to gaps in funding and capacity issues in children's hospices across the UK. These particularly f Of an estimate total number of 62,000 CYP with LLCs in the UK g Up to 2022, Clinical Commissioning groups (CCGs) were responsible for the commissioning -planning, funding and monitoring - of PPC. CCGs were accountable to NHS England which allocated funding to local CCGs. In July 2022, Integrated Care Systems (ICSs) were introduced and Integrated Care Boards (ICBs) replaced CCGs with their functions transferred into ICBs. 87 54 affect community care and 24-hour support provision, as well as recruitment of specialist and qualified staff.88 Local authority funding is another important source of funding for social care provision in PPC and local authorities are responsible for funding services such as respite and bereavement support for families.79, 80 However, as with NHS funding, TfSL reports that funding provided by local authorities is inadequate and does not reflect the amount of care provided.79, 80 For instance, in 2018-2019 local authorities contributed an average of only 1.7%79 and in 2021/2022 an average of 3.4 % of children’s hospice charity costs.80 Due to funding gaps, it is also suggested that the extent to which families can have 24/7 EOLC at home by HCPs with access to advice from specialists varies significantly depending on where the family lives.89 In 2022, it was reported that such access to 24/7 EOLC was not met fully in 79% of the Integrated care systems (ICSs) areas in England.89 1.4.3.2. Provision of advance care planning According to current NICE guidelines, HCPs should develop an ACP for each CYP with a LLC that includes both current and future care.8 However, provision of ACPs in PPC services in the UK varies significantly.8, 17 It is suggested that this is partly due to specialised staff shortages across the UK8 and shortages of PPC consultants and paediatric nurses in particular.8, 90 Another issue reported is a concern around liability 55 when discussing ACPs in the eventuality that preferences discussed with family or the CYP were not eventually met.8, 90 1.4.3.3. Access to specialist PPC A number of reports and studies on PPC highlight the need for specialist care as CYP with LLCs/LTCs often have complex and diverse needs due to the great number of different conditions.34, 52 The nature of conditions that CYP have, can often be unpredictable with changing levels of health – remissions and medical crises – which makes them more difficult to manage.34, 52 NICE guidelines suggest that care should be provided by a specialist multidisciplinary team which includes different professionals and teams depending on the child’s needs.16 According to guidelines, the team should include at a minimum a PPC consultant, a nurse and a pharmacist with expertise in PPC and experts in family and child support in EOLC who can provide social, psychological, practical and spiritual support.16 Although NHS England states that specialist PPC teams should have a PPC medical consultant available to lead the service, it is suggested that this is rarely the case.8, 78 According to the Royal College of Paediatrics and Child Health’s (RCPCH), in 2018, there were only 14 PPC consultants in the UK91 whilst the estimated number of PPC consultants needed in the UK was between 40 and 60.8 The TfSL report (2018) suggests that access to appropriate services that offer specialist PPC differs in areas across England.8 Access depends on funding decisions made by the 56 different local authorities and NHS CCGsh.8 The geographical inconsistency in ability to access specialised services represents, as described by the report, a ‘postcode lottery’ 8, 34 where only about a third (35%) of CCGs across England were implementing in 2018 the UK government’s guidelines on EOLC and one in five (17%) did not offer bereavement care for family members.8, 17 1.4.3.4. Family-oriented care PC for adults is suggested to be patient-centred, with involvement from family acting as proxies in decision making, when necessary, while PPC is considered to be child-and-family-centred.11, 59, 92 One of the central principles of PPC is that support should not only be provided to the child but also to the family involved.93 NICE EOLC guidelines and PPC general principles support family involvement in decision making and recognise the importance of family-oriented care.16, 18 A family-oriented approach to care, includes assessment of family needs and wishes by the care team involved and supports the use of a holistic multi-agency and multi-disciplinary approach to avoid multiple assessments with the child and parent or carer, as well as siblings and extended family.17 CYP with LLCs/LTCs usually have a high level of dependency on the person(s) with parental responsibility who often provide high levels of care.17 It is commonly h CCGs have now been replaced with Integrated care boards (ICBs) in NHS England in July 2022. 57 accepted that caring for a child with a LLC/LTC brings major emotional and practical pressure to the family.52, 94-96 It is also recognised that other people in the community who are close to the child – friends and teachers – can also be significantly affected by the child's illness and death.95 Apart from the significant impact on the mental health of and relationships between family members, caring for the child often affects the financial situation in the family and the physical health of the primary caregiver(s).52, 94, 96, 97 In a recent UK study (2022), it is suggested that family and in particular mothers often feel that they have to ‘fight’ with services so that the necessary support, care and equipment can be provided to their child and themselves.96 Having to ‘fight’ for necessary services and especially for extended periods of time, rather than spending time giving care directly, was reported as a major stressor for families.96 At least one member of the family might need to reduce the time spent on other responsibilities, such as work or taking care of other siblings.94 Research studies on the impact of caring for a child in the UK and Israel show that women in the family are more likely to contribute more time than male members of the family, to caring responsibilities.52, 94, 96-99 Studies of family perspectives also report that often the experience of a LLC/LTC in a family has the greatest impact on families with the lowest income, as caring for a sick child especially for extended periods of time leads to extra costs and fewer opportunities to earn.52 Although there is financial support available for families in the UK, this does not always match the necessary associated costs.52, 96 58 Provision of support to the family can be beneficial across the illness trajectory, from the point of diagnosis to the time of death and in bereavement.16, 17, 93 Support for families might include offering short breaks (respite care), social care supporti, psychological support, condition specific information, training on how to care for their child and advice on finances.16, 17, 34, 52, 95 However, it is reported that many of these family needs are often left unmet.34, 52, 96, 97 This is partly due to the focus being on caring the affected child by the family and services.52, 96, 97 The need for adequate provision of short breaks for parents or guardians34 and the need to receive care on a 24 hour basis during the week8 have been highlighted in UK reports. Whilst home-based care is becoming more common and a less expensive option than hospital based care, it often requires looking after the child for many hours a day.82 Short breaks give the chance for the parent to spend some time away from caring responsibilities and give the opportunity to the child to socially interact with other people.17, 100 Short breaks can be provided either at home, or in a hospice or other residential setting provided by the voluntary or statutory sector.17, 100 The breaks can last for a few hours or for several days.100 Age restrictions found in referral criteria for short breaks have been reported by parents as one issue with accessing short i for example, provision of aids and home adaptations to enhance child's independence. 59 breaks.34 In addition, general availability of respite services is dependent on the geographical areas that care is received.34 Other unmet needs reported by families are around communication with, and coordination of, services providing care and support.18, 34, 52, 101-103 There is a wide range of care settings providing care and support to CYP with LLCs/LTCs, and their family, both in the statutory and voluntary sector.17, 88 Thus, effective communication between all the different services and professionals, is essential to assist with continuity of care, avoid duplication of information in care plans and to clarify the responsibilities of each service.17, 104, 105 To coordinate multi-disciplinary care, there is often a named person from the care team involved who will take a lead role in coordinating care, especially if there are multiple admissions and change of care settings.17, 106 However, it is suggested that families are having to repeat their story to multiple professionals as communication between service providers is often not consistent. This is especially an issue when young people transition to adult services.52 1.4.3.5. Transition from children to adult services The need for developmentally appropriate services had been particularly highlighted to better deal with PPC needs specific to the child’s developmental stage.8, 15, 19, 27, 107 One of the current challenges in PPC in the UK is the transition from children's to adult services.15, 108 The number of adolescents in the UK with LLCs/LTCs is rising and medical advances increase the likelihood that these adolescents will reach adulthood.15, 20, 107 60 The increasing number of adolescents with LLCs/LTCs has drawn attention to a need for adequate transition to adult services when they reach adulthood so that health care engagement can be maintained throughout the life course.15, 20, 62, 107, 109 In the UK, prior to reaching adulthood, young people with LLCs/LTCs will prepare for the transition of care to adult services, with initial discussions on transition starting from as early as the age of 14 with transition at ages 18+.107, 110 Preparing for transition may be challenging for young people and their family, involving a period of change and uncertainty in terms of future care provision and needs being met following transition.110 This is partly due to the nature of adult PC provision, which differs significantly from PPC. Whilst PPC offers longer term care, including respite care and family support, adult PC services tend to focus on short term care, usually around EOLC and symptom control.15, 110 A significant number of children's hospices in the UK retain patients up to their early 20s as a younger adult may still be in need of longer-term care.15 The need for more gradual, planned, and well-coordinated transfer to adult services has been highlighted.107, 111-113 A systematic review (2021) looking at the differences between health care use pre- and post-transition to adult care for young people with LLCs suggests that, post-transition to adult services, there is strong evidence for an increase in the number of visits to emergency departments and a reduction in the numbers receiving physiotherapy.109 Another systematic review (2017) shows that parents have raised concerns in relation to adequate service provision in transition to 61 adult services.107 In paediatric services, care was viewed by parents as being person-centred to the child's complex and unique needs.107 However, during and following transition to adult services, care lacked adequate resources and parents felt that they were left to act as their child's transition co-ordinator, with responsibility for communication between services which placed a significant burden on them.107 Other systematic reviews on the experience of young people with congenital heart disease112 and long-term conditions111 suggest that transition experiences are similar across different conditions. The studies included experiences of transition to unfamiliar relationships with professionals and surroundings (e.g., to older adult wards), feelings of insecurity and lack of preparation for what was ahead, and potential limitations in the adult provision available.111, 112 The NHS England long term plan report (2019) states that part of the new plan for providing services for CYP includes a selective move to a '0-25 years' model to improve continuity of care.2 It further notes the importance of a safe transition to adult services which is person-centred and age-appropriate, in achieving better health outcomes and better patient engagement.2 The plan notes that a successful transition could be achieved if it is based on need rather than age.2 Yet, the most recent NICE guideline (2017) on EOLC for infants, children and young people only includes young people up to the age of 17 years old whilst the adult quality standards and guidelines apply to those receiving EOLC aged 18 years old and over.114, 115 There is no justification provided for this age choice in the NICE guidelines.114, 115 62 1.4.3.6. Ethical and legal challenges in PPC One of the goals of PPC is to minimise conflict between poor clinical care, poor communication, and cultural misunderstandings.55 Such conflict may raise ethical or legal concerns in relation to decision-making.55 Decision-making includes making decisions about who decides and who is involved, making decisions about life-sustaining treatment and whether to stop or start a treatment.55 In paediatrics, HCPs follow age limits in law to assess a child's competency to consent to medical interventions.116 In the UK, young people over the age of 16 years are presumed in law to be competent to give their consent to medical treatment unless they lack mental capacityj at that point in time.116, 117 Assessing mental capacity to make decisions for interventions is a challenging topic for HCPs in both PPC and adult PC practice.118, 119 In adults, for example, assessing capacity is common in conditions such as dementia.120 It should not be assumed that under the age of 16 years CYP are not competent to consent to a medical intervention; a skilled HCP, who is involved in the child’s care needs to assess competency to consent for each individual case and in relation to a specific decision.116 If a child or young person lacks competency to make a decision j to make a decision because of an impairment of, or a disturbance in, the functioning of the brain. The Mental Capacity Act (2005) applies for a person aged 16 and over (Mental Capacity Act 2005). 63 about a medical intervention, then a person with parental responsibility, a court or an appointed proxy can make that decision given that it is in the 'best interests'k of the child.116 If agreement for an intervention cannot be reached between the child, HCPs and parents then legal advice or action is sought.116 The legal principle that overrides others is that decisions in relation to treatment must be chosen in the child's 'best interests'.17 An 'objective' 'best interests' judgement is challenging to make and sometimes the involvement of the court system is necessary to make that decision.121 Recent court cases in the UK in which agreement could not be reached between the child, HCPs and those with parental responsibility include Charlie Gard (2016-2017),121, 122 Alfie Evans (2018)123 and Tafida Raqeeb (2019).124 Disagreement in all three cases was around the decision to withdraw life-sustaining care.121, 123, 124 Whilst the hospital and medical staff thought it would be in the best interest of the child to withdraw care, the parents did not agree with this.121, 123, 124 In all three cases the court ruling was in favour of the medical staff's view.121, 123, 124 There have also been cases in which the family may show a preference for not informing and sharing with the child all the information and details of their condition.125 There are different reasons for this. For example, some families may feel k A 'best interests' judgement involves testing as objectively as possible and considering all factors, risks and benefits, on what would be in the child's best interest (Children and young people ethics toolkit, 2019). 64 that this is a way to protect their child from difficult conversations in relation to their illness.125 Some may feel that the child may not be able to understand and process the information about their condition and treatment options.125 The issue is often not whether the family chooses to tell the child about their condition but rather who should do that, how and when to tell the child.126 Just as parents want to be able to choose what to tell their child in order to protect and advocate for them, so too medical professionals as ethical practitioners are required to respect patient autonomy and advocate what is best for their patient.125 The NICE guidelines for CYP at the EOLC highlight the central role children should have in decision making.16, 17 The guidelines also recommend that HCPs should use different methods to best communicate information to the child, such as through play activities, taking into account their level of understanding, age, family situation, beliefs and values.16 It is important to note that the principle followed in all the NICE guidelines is for the HCP to involve both the child and the family in decision making as much as possible but it is also acknowledged that professionals should be aware that CYP may feel differently to their family or HCPs about how their care is or should be influenced by their beliefs and values.16 The ethical and legal guidelines and principles that HCPs in the UK need to follow when working with CYP correspond to principles in broader United Nations (UN) legislation on the importance of children's voices being heard.127 Article 12, in the UN convention on the rights of the child (2009), assures to children who are capable of forming their 65 own views in matters affecting them, the right to express those views and to be given the opportunity to be included in discussions on those matters.127 Similar to the guidelines for HCPs, it recognises the need to assess those views expressed by the child in relation to their age and maturity.127 1.4.3.6.1. Cultural and spiritual needs Ethical challenge concerns include how best to integrate cultural and spiritual needs and practices of different population groups in PPC.17, 128 The child’s or family’s beliefs and values can influence care, treatment, dying and bereavement decisions.16, 129 Beliefs and values may significantly differ depending on location and family background. 17 For example, Jehovah’s witnesses refuse blood transfusions.130 NICE guidelines recommend asking both the child and the family about their beliefs and integrating their beliefs into care and decision-making.16 It is also acknowledged that professionals may also be influenced by their own personal beliefs and values which can be different to those of the patient, and this affects the process of decision making for care and treatment.129, 131 It is suggested that when there is disagreement between the beliefs or values of the patient, family or professionals, it may be helpful to involve other services, such as a chaplaincy service, to develop a mutually acceptable care plan.17 66 1.5. Conclusion Chapter 1 has introduced concepts and terminology in relation to PPC and EOLC that will be used throughout the thesis. The chapter outlined similarities and differences between adult PC and PPC. Similarities between adult PC and PPC included a multi-disciplinary approach to care and support to family members. Differences between adult PC and PPC were related to the developmental needs of CYP, health conditions and illness trajectories. The chapter then outlined the particular challenges and unmet needs in PPC and EOLC provision in the UK, including needs in relation to transition to adult services. Given the rising prevalence in the number of CYP with LLCs/LTCs, the demand and cost of PPC services increasing over time and the challenging funding context, it is important to consider the best use of resources allocated to health and social care services for CYP with LLCs/LTCs and their families.19, 79, 80 Given the particular challenges evident for young people going through transition to adult services, with parents often being left to act as the young person’s transition co-ordinator, there is a particular need to understand the best use of resources for interventions for young people going through transition to adult services and support for their families. One of the key areas of PPC research priorities in the UK is finding the best ways to deliver care to CYP and their families within the available resources.17 To achieve this, there is a need to look at the cost-effectiveness of different interventions to facilitate 67 decision making in health and social care.3, 17 Chapter 2 outlines the different economic approaches used to inform decision making in health and social care. 68 69 CHAPTER 2. THEORETICAL FRAMEWORKS AND PRACTICAL ASPECTS OF ECONOMIC EVALUATION OF HEALTH CARE INTERVENTIONS. With the rising prevalence in the number of CYP with LLCs/LTCs there is increasing demand for PPC and EOLC provision. Given the limited resources in health care, decisions have to be made in relation to the best use of the resources available. Within the health economics literature, there are different approaches used to inform how resource allocation decisions are made in practice. The normative theoretical and methodological frameworks used in health economics for economic evaluation of health care are outlined in this chapter. These include the shift towards extra-welfarism and the growing application of the capability approach. The final section of this chapter is focused on economic evaluation of child wellbeing. Given the focus on CYP with LLCs/LTCs for the thesis as a whole, throughout this chapter, examples are given of economic evaluations that use the different types of methods discussed. 70 The normative frameworks are presented in order of chronological developments. This is so the development of, and arguments associated with, different frameworks and techniques can be situated within a historical context and can be used to show how focus has shifted towards new and current ways of assessing outcomes of health care interventions. In particular, by presenting it in the order of time, the reader can make two interesting observations. Firstly, that methodological approaches in economic evaluation at times precede theory132, 133 and secondly, that economists may use the same terms but with a different meaning or focus attached to them in different theoretical frameworks.134 Both observations add to a better understanding of how theories and techniques emerged or why they became more prominent than others at different historical periods. Finally, as Forget points out, a historical perspective in health economics assists theoretical innovation (p.618).135 2.1. Normative and positive approaches Both normative and positive frameworks have been used within the discipline of health economics throughout time.136 Whilst positive economics focuses on description, prediction and facts, a normative analysis in economic evaluation is concerned with what should be done with available resources to achieve a desirable outcome, such as one in which social welfare is maximised.137, 138 Normative analysis involves making value judgements about alternative ways of delivering health and care which are socially preferable rather than merely describing or predicting a 71 phenomenon.137 Positive economics does not involve making value judgments but rather describes and/or analyses an economic phenomenon, e.g. the impact of inflation rate on economic growth.138 Since this thesis relates to normative economics – how limited resources should be used in health and care – this chapter will focus on normative literature. It is important to note that although the chapter introduces concepts used in the normative literature, some concepts, such as ‘utility’, have been interpreted from both normative and positive perspectives.136 2.2. Health economics and economic evaluation Whilst the discipline of economics has a long history, health economics, as a formal sub-discipline of economics, is a younger field that has developed in more recent times.139 The discipline of health economics is concerned with the efficient allocation of limited healthcare resources to maximise benefits and aims to promote efficiency in decision-making in the health care sector.140 The term “health economics” started appearing in books around the 1940s141-145 and important growth of the health economics field began in the 1960s.146 Around the same time the National Health Service (NHS) in the UK was founded (1948), to provide free, universal and comprehensive healthcare, but within a fixed overall budget.147, 148 In terms of the context for hospice care, with medical advancement in the 1950s, there 72 was also a shift from adult hospice care being a nursing initiative towards care that was being led by a medical specialty. 74 One of the first modern hospices, St Christopher’s Hospice, was founded in 1967 for adults 149, 150 and the world’s first children’s hospice, Helen House, in Oxfordshire was founded in 1982. 149 Economic evaluations which involve a comparative analysis of costs and outcomes of alternative interventions 151 are commonly used in health economics and are applied extensively in the NHS context. Applied economic evaluation on PC and EOLC services, however, remains very limited and especially in relation to EOL care services for CYP. 152, 153 For instance, a relatively recent systematic review (2020) of economic evaluation of PC models included only five relevant studies in their review and all were in adults. 154 The review suggested that the most common approach in economic evaluations in a PC and EOLC setting was to look at only the cost part of interventions 154 which is the type of analysis (costing analysis) also used in the economic evaluation model in the NICE guidelines for EOLC for infants and CYP. 16 Such analysis, where only costs of relevant interventions are described, does not help to inform decision making in terms of the relative benefits of interventions across interventions. 2.3. Welfarism (1900s-1940s) A large part of the health economics literature is based on welfare economics.136 Welfare economics is “the systematic analysis of the social desirability of any set of arrangements, for example a state of the world or allocation of resources, solely in 73 terms of the utility obtained by individuals”(p.210).137 Brouwer et al.134 argue that there is no consensus on the content of welfarism in the literature, and Culyer 136 notes that some academics set fewer limits than others on its properties and what may count as a social welfare function. The neo-classical welfarist economics framework is built on the following four normative principles134: I. Utility: Individuals can choose their preferred option in a rational way in order to maximise their welfare.134 II. Individualism: Individuals are the best, and for some economists the only, judges of their welfare and utility function.134 III. Consequentialism: Considers only the outcomes for individuals in terms of their consumption and consequent utility obtained rather than the process that led to the outcomes.134 IV. Welfarism: Suggests that the value of a situation depends solely on the utility levels achieved by individuals in that state or situation.134 2.3.1. Utility ‘Utility’, i.e. “the level of satisfaction that consumers obtain through having their desires met” (p.22),137 has been used in different theoretical frameworks and historical periods, but the term did not always have the same meaning.134, 155, 156 As Praag notes, although utility as a concept was necessary to develop normative and positive 74 economic theories, economists have been uncomfortable with its use because science is based on measurement but it is questionable whether measuring utility is possible (p.363).156 In welfarist Paretian economics, utility is determined by commodities and the consumption of goods and services which is the only means of producing welfare.134 In the classical welfarist tradition, utility was usually measured cardinallyl. These individual utilities could then be summed numerically, such that the highest total utility would mean a social optimum was achieved.134 By 1900, economists acknowledged that cardinal utility was not essential for price theory and that ordinalm utility was sufficient.135 In the neo-classical welfarist approach, utility was usually measured ordinally and value judgements were achieved by using the Pareto principle.134 2.3.2. Pareto efficiency The Pareto principle (1906) is a value judgement concerned with achieving an optimal state of affairs by reference to individual utility gained at each state.137 An improvement in social welfare would involve an increase in utility for one individual without a utility loss for another.137 An optimum state (Pareto optimality) exists if any l Numerically; measured in terms of units m Use of rankings instead of values/numbers 75 resource reallocation would mean that utility for at least one individual would be reduced.137 One of the criticisms of the Pareto principle is that the value judgement is only concerned with not having any losers and thus fails to look at social gains in relative terms and to compare trade-offs between people’s welfare.137, 157 For instance, if in a status quo there is unequal distribution of welfare (or utilities) a redistribution would be inefficient – as it involves losers – even though the current state of affairs is not socially equitable.158 2.3.3. Kaldor-Hicks efficiency In the late 1930s, Kaldor159 and Hicks160 suggested an alternative way to maximise utility within Paretian welfarist economics that would not be as restrictive in practice as the Pareto principle and could have broader application. The ‘Kaldor-Hicks compensation’ principle suggests that what would constitute an improvement would be for those that are better off, in a state of unequal distribution of welfare (or utilities), to be able to compensate those that are worse off.161 In this way a Pareto improvement could be achieved. The principle does not require actual compensation but instead offers the possibility of hypothetical compensation.162 Thus, the principle answers the question of whether an economic decision would maximise utility whilst questions of redistribution are left to others to deal with.155 It is also quite difficult to see how the concept of compensation could be applied in a health care setting. How would a sick individual or population be compensated by a healthy individual or population? Even if there was a way for compensation to be feasible in practice in 76 monetary terms, rather than health terms, it doesn't necessarily mean this would be thought of as acceptable or welcomed.133 2.3.4. Cost-benefit analysis (1950s-70s) The underlying rationale of the ‘Kaldor-Hicks compensation’ principle is used in cost – benefit analysis (CBA) which is a ‘compensation test’ within Paretian welfarist economics.163 In CBA the value of costs and benefits are defined by means of monetary compensation and if the benefits of a particular intervention exceed the costs then the intervention is justified.137 This includes situations where an intervention is justifiable if the benefits for the whole society exceed the costs, but some members of society are being left worse off.164 The aim of CBA is to achieve economic efficiency by achieving the optimal use of resources at the least cost.137 A type of cost-benefit methodology called Return on Investment (ROI) analysis 165 has been recently used in evaluation of CYP hospices in Scotland. 166 In ROI analysis, all of the costs and benefits of an intervention are calculated in monetary units. However, it is suggested that this type of analysis may be inappropriate for the EOLC context for CYP as its application can oversimplify complex cost and outcome dynamics of interventions and therefore miss out on important benefits of interventions. 153 For example, the evaluation of CYP hospices in Scotland only accounted for health and social care resource use avoided and societal benefits for adult caregivers who were able to work.166 77 If a cost-benefit analysis is applied in economic evaluation of health care interventions, health services would be considered as another good produced which, if consumed, can produce utility.164 Such analysis fixes a monetary value on life which is controversial and has been criticised by practitioners and policy makers.135 As Broome points out in his critique of CBA, “no finite amount of money could compensate a person for the loss of his life, simply because money is no good to him when he is dead.”(p.92).163 This approach views health as just a means to an end in which individuals ‘consume’ health to produce utility and if they cannot receive health, it would be equally acceptable to be compensated in other ways to produce utility.155, 167 One of the burning topics in the UK around the late 1950s and early 1960s was around the way forward with dealing with the rising NHS costs.167 It was suggested that health care should be based on consumer choice, even if consumers have to rely on experts (e.g. GP) to consume it as we would do with lawyers or architects.168 This argument supported the idea that health is a good that is no different to other goods bought – such as private insurance – and if the aim is to maximise consumer satisfaction then it should be supplied through the market.167 Using the same principle - that health and health care is a compensatable and consumable good - ‘Willingness to Pay’ (WTP) and ‘Willingness to Accept’ (WTA) measurements have been used in health economic evaluation. WTP measures the maximum price which a consumer is willing to pay to buy a product whilst WTA is the minimum price an individual is willing to accept as a compensation for being left worse 78 off in a situation.137 An important difference between the two measures is that WTA is not constrained by one’s ability to financially afford an intervention.164 Yet, in a WTA measure, responses will still be affected by the respondent’s wealth background. Furthermore, both measures put a direct price on life which, at least in countries where health care is public and universal, is a highly contentious view. Even if people are willing to provide a measure of their hypothetical WTP/WTA, it is still something that people could find offensive or uncomfortable to answer as they are required to think about health and care in money terms.135, 169 In the context of a debate regarding the ‘consumption’ of health and marketisation of health care, attention to the concept of ‘externalities’ in health care economic evaluation arose in the early 1960s.162 Externalities, or otherwise ‘spill-over effects’, are the indirect costs and benefits involved in the production and consumption of a good or service which are external to the direct costs and benefits involved in the production and consumption of that good or service.137 The external effects can be either positive – for example immunisation has a positive external benefit to others who have not ‘consumed’ it by reducing their chances of ill health – or negative – such as antimicrobial resistance.137, 161 The early work on ‘externalities’ by economists was intended to answer questions on the efficiency of a public health care service such as the NHS.147, 162, 170, 171 Culyer (1976) developed the concept of a ‘caring externality’172 as an external utility effect that has an impact on “the subjective values of others”(p.194).161 For instance, 79 when a parent is distressed when their child is sick or is relieved to see their child recover. Culyer argued (2010) that, in the case of beneficial externalities, a Pareto efficiency – where benefits and costs to society are equal 137 – will not be achieved in a competitive market if the marginal benefit involved, for instance the parent’s relief is larger than the net marginal cost of the service to the child.161 2.3.5. Cost-effectiveness analysis – late 60s It has been suggested that CBA has been ‘eclipsed’ in the literature since the early 1980s by cost-effectiveness analysis (CEA).173 The first CEA study was published in 1968 on renal failure,146 and throughout the 1970s and 1980s the number of studies using cost-effectiveness analysis to evaluate the efficiency of health care interventions increased rapidly.173 In CEA, whilst costs are measured in monetary terms (as in CBA), the outcomes of a health care intervention are measured in non-monetary terms and the aim is to maximise health outcomes.137, 174 Whilst CBA has been used to answer whether a particular intervention is worth the cost, CEA has been used in comparing the relative efficiency of alternative interventions dealing with a given problem to decide which of the alternative interventions can provide the most benefit for the least cost.137, 174 80 2.3.6. Cost Utility analysis (CUA) and QALY Cost-utility analysis (CUA) is another way to incorporate preferences into economic evaluation in health care.175 The term ‘cost-utility analysis’ was first used in 1981.135, 176 Whilst CEA measures the outcome of particular types of health care, CUA measures the outcome of health care as a whole.137 Yet, Culyer notes that it is often difficult to distinguish between CEA and CUA and there are no major differences between the two.161 Also, in more recent contributions, some authors consider CUA as a type of CEA and thus only distinguish between CBA and CEA.177, 178 It is important to note that ‘Utility’ in CUA does not have the same meaning as in Paretian economics that relates to commodities measured in monetary terms.136 Utility under a welfarist approach represents preferences numerically and it is considered to be the final outcome of an economic evaluation in which people consume health care for the purposes of producing utility.155 In CUA the focus shifts towards a 'health-related utility' in which health is valued in itself rather than for the utility it produces.155 In CUA health outcomes are measured in terms of ‘Quality adjusted life years’ (QALYs). QALYs provide an important measurement tool currently used in healthcare decision-making in the UK, by NICE,179, 180 as well as in a number of other countries, such as the Netherlands.181 As defined by NICE, the QALY is a “measure of the state of health of a person or group in which the benefits, in terms of length of life, are adjusted to reflect 81 the health”.182 A year of life in perfect health corresponds to the value of one QALY.182 If a person’s health is compromised, then the QALYs would be adjusted to reflect this.182 In this way the QALY measure captures both morbidity (quality of life) and mortality (length of life) into a single value for the purpose of evaluating health care interventions.179 Although the QALY framework has been extensively adopted in economic evaluations in a number of countries, it is suggested that its methods may not have the same relevance for interventions where outcomes of care in terms of only QALYs may not be sufficient. 183 PC and EOLC interventions are one example, where care is not focused on curative treatment or on extending life but on improving quality of life in a holistic sense for the remaining time.3 Interventions may be intended to deliver a broader range of outcomes such as comfort and spirituality which are not included in the health focused framework of QALYs. 184, 185 Furthermore the impact of interventions on those close to the dying person is not usually taken into account within the QALY framework, which tends to be patient centric. 152 Capturing the impact of EOLC interventions for those close to the dying CYP in economic evaluation may be particularly important because CYP are substantially dependent on adults 49 and because intense grief may be experienced by those close to a CYP who has died. 186 For these reasons the economic evaluation examples in NICE guidance for EOLC for infants and CYP only use a costing analysis approach and the QALY framework was not used as it was considered unsuitable as an outcome measure for this context. 16 82 Health-related quality-of-life (HRQOL) measurements were used in clinical practice prior to the development of health economics as a discipline 187 but the term QALY was first used in 1972.188, 189 The development and use of QALY by policy makers in Britain in the mid-80s was linked to the rising costs of running the NHS,190, 191 and thus the need for a rationing agenda in health care.191-193 2.3.7. Valuation methods - QALYs Different methods have been proposed to answer questions of how to weigh preferences in measuring trade-offs between mortality and morbidity. Methods includes 'time-trade-off' (TTO) and 'standard gamble' (SG).194 The TTO method involves asking an individual to choose and to state their preference between two alternative scenarios: living in an impaired health state for their rest of their life or living in full health for a shorter period of time.194 The principle is similar to WTA and WTP, in that both involve willingness to sacrifice, but what is measured in TTO is the participant's willingness to sacrifice time rather than money in order to avoid an impaired health state.164 In the SG method, the choice is between remaining in the current health state or taking a gamble to either be in full health or face immediate death.194 Using these methods can make it difficult to measure preferences in health care as – similar to WTP and WTA methods – they pose questions that could be uncomfortable for participants to answer, if the hypothetical scenario asks them to take risks in relation to their health.164, 194 83 Another issue with the methods mentioned above is the problem of who decides the trade-off between mortality and morbidity.135 Mooney (1994) argued that it is not for doctors or economists to define the trade-off between mortality and morbidity but rather for patients and the community; and that the issue of valuation is not a medical or technical matter but rather a social matter.195 Moreover, within a community individual preferences related to trade-offs between morbidity and mortality may vary and be influenced by a variety of factors such as age, gender or social class.137 Thus, there is also the question of whose values within a community should count.136 Mooney (1998) argued that often within QALY measures it is assumed that QALY gains count equally regardless of who gets them while individuals may have differences between them in the utility associated with the health gains.196 He suggested that if these differences are not taken into account, this may mean that "even if health were the only argument in the patients' utility functions, we would not maximise utility even if we did maximise health gains"(p.104).196 Under this approach, measuring social welfare does not focus on the individuals' judgement of their utility but instead on only maximising health gains. NICE guidelines currently suggest that a QALY is of equal weight regardless of who receives it. 197 Therefore, there is no distinction made by age nor any other individual characteristics. Yet, it has been suggested that societal views may differ and there is evidence that some people may place a higher or lower weight on health gains accrued by CYP compared to older adults. 198, 199 It is likely that weighting QALYs for CYP at the 84 EOL would be even more controversial and challenging for policy and decision makers as the additional characteristic of EOL needs to be considered in terms of how weights are distributed and in what order. 14 Stevens et al. suggest estimating joint distributions of preferences across a range of individual characteristics (equity considerations) to elicit a series of equity profiles (patient groups). 14 However, further research is required into these joint distributions so that the framework can be compatible with individual weights, and for work in EOLC and CYP, these joint distributions would need to include characteristics relevant to this group. 2.4. Extra-welfarism - mid to late 80s There is little consensus on the definition and distinctive characteristics of the extra-welfarist approach.134, 169, 200 Brouwer et al. (2008) conducted a systematic comparison between the welfarist and extra-welfarist approach and concluded that the two approaches have some substantive differences in terms of outcome focus in an economic evaluation, valuation sources and weighing of outcomes and interpersonal comparisons.134 These differences are outlined below. 2.4.1. Outcome focus in economic evaluation In the mid to late 1980s, Culyer (1989) attempted to develop a theoretical basis to support non-welfarist approaches within health economics.155, 169 Culyer (1989) argued that most of the cost-effectiveness literature in health economics is 'extra-welfarist' as 85 the focus is on identifying the least cost method to maximise health as opposed to utility.136, 201 Whilst within a welfarist theoretical perspective, utility relates to the consumption of commodities (goods and services) and social welfare is a function of individual utilities,136, 202 within an extra-welfarist perspective social welfare is not determined exclusively by individual utilities and the consumption of commodities. Instead, Culyer argues that what determines social welfare becomes less restrictive and includes an 'intervening category'(p.10)201 called 'characteristics of people'(p.10)201 which relates to 'non-utility information'(p12)201 such as levels of nourishment or fellowship. Culyer (1990) distinguishes between the 'commodities and characteristics of commodities' group and 'characteristics of people' group. Whilst the former group relates to what is needed to eliminate deprivation or ill health, such as the provision of health care, the latter relates to a description of deprivation, social welfare or desired states.201 Yet, Birch et al. (2003) argue that although under the extra-welfarist approach, as defined by Culyer (1989), individual utilities are not considered as a sufficient way to measure social welfare, Culyer does not specify what determines social welfare and what 'characteristics of people' are most appropriate and sufficient in defining and measuring social welfare.202 The theoretical shift of focus has been recognised by academics as offering something 'extra' to welfarism.203 Yet, it has been argued that, although in theory the extra-welfarist perspective advocates for a broader evaluative space than that of utility 86 under the welfarist perspective, in practice (i.e. using QALYs) the focus is narrowed down, almost always, to health only.137, 155, 202 Furthermore, the maximisation rule used under the welfarist perspective is adopted under the extra-welfarist perspective without much debate as to why the focus continues to be on maximisation.155, 169 In contrast to Paretian welfarist approaches, under the extra-welfarist perspective health is an end in itself, hence in practice it is not possible to redistribute or compensate for once produced.155 As Coast argues, the separation of efficiency and equity found under the Paretian welfarist approach, where the compensation rule applies is no longer possible under an extra-welfarist approach as health once produced, can no longer be redistributed to achieve equity.155, 204 Thus the argument used for maximisation under the welfarist approach does not justify its application under the extra-welfarist approach.155 2.4.2. Source of valuation of relevant outcomes Culyer argues that another distinctive characteristic within extra-welfarism is that valuation is not necessarily or exclusively based on people's preferences as in the welfarist approach.161 Apart from the affected individual, a 'decision-maker' could be a source of valuation.134 This approach has been criticised as being paternalistic as the affected individual, i.e. a patient, is not the source of valuation but instead an 'agent' acts on their behalf and makes decisions on how the patient would or should act.134 87 2.4.3. Weighting of relevant outcomes Apart from whose values should be considered in decision-making under an extra-welfarist perspective, there is also the ethical issue of how to weight health related benefits. Some weighting methods used under the extra-welfarist approach were outlined earlier on – e.g., TTO. When weighting health related benefits, different characteristics of people receiving the health care – e.g., their age – could be considered in economic evaluation.134 Brouwer et al. argue that, whilst under welfarism it is not clear whether weighting is of importance to utility information, in extra-welfarism weighting is often considered in relation to dealing with equity and distribution issues in decision making.134 Whilst Brouwer et al. point out that weightings are used under extra-welfarism, they do not indicate the fundamental importance and necessity of weightings in decision-making in relation to distributional issues.155 As discussed, efficiency and equity are not separate under the extra-welfarist approach; thus equity concerns – concerns related to fairness in the allocation process and final distribution205 – need to be incorporated prior to distribution.155 2.4.4. Interpersonal comparability of relevant outcomes The final difference between welfarism and extra-welfarism as identified by Brouwer et al. is in how interpersonal comparisons are treated.134 It is not possible to interpersonally compare utilities under welfarism in a meaningful way.134, 205 As individual utilities are subjective, it is difficult to compare their utilities of different 88 individuals to measure welfare.206 However, under the extra-welfarist approach, interpersonal comparisons are of great importance and required in economic evaluation.134, 205 Under the extra-welfarist approach, welfare is measured in terms of health - e.g. QALYs - or other 'characteristics of people', where, as Brouwer et al. put it, "although one may not know for sure whether an unhealthy person has less utility than a healthy one, we do know that the unhealthy person has less health" (p.334).134 2.4.5. Cost-Consequence analysis Cost-Consequence analysis falls under the extra-welfarist framework 164 as outcomes other than utility or health can be considered in economic evaluation. CCA, rather than combining all outcomes into a single metric, presents costs and outcomes in a disaggregated manner. 164 A comprehensive list of costs and consequences is presented to the decision maker who will need to use their own system for weighting the outcomes to make decisions on the cost-effectiveness of interventions. 207 CCA has been used in economic evaluation of EOLC interventions 154 and may be a relevant framework for evaluating interventions for CYP at the EOL (including interventions for those close to them) as it allows for a more inclusive consideration of outcomes. However, outcomes are not aggregated into a single metric to provide a clear framework for decision makers, who are left to make difficult decisions on relative weighting of costs and outcomes across complex interventions for a particularly vulnerable population 153, including those close to CYP at the EOL. 89 2.5. Capability approach- beyond health and maximisation Sen's 'capability approach' has been broadly applied in the human development area 208, 209 and has formed the theoretical framework underlying the United Nations Human Development Index since 1990.210 Sen’s work has also been influential as an alternative to welfarist approaches in economic theory134, 211 and was applied by Culyer in developing the extra-welfarist approach.134 There have been a number of more recent attempts to apply the capability approach in economic evaluation in health care.212-221 Sen argued that within Welfarist approaches to economic theory, the utility focus is too narrow and the metric used focuses only on their “mental reaction” (p.218).222 Under welfarism, mental states such as happiness and desire fulfilment are used as a psychological metric to evaluation.223, 224 Sen recognises the importance of mental states in valuation but argues that "it cannot really be taken to be all there is to leading a life" (p.54).223 Sen points out that if valuation is done in terms of pleasure, then other human states would not be valued in themselves but only to the extent that they contribute to producing pleasure.223 Similar to Culyer's approach in extra-welfarism, Sen suggested that a variety of human states – ‘functionings’ and 'capabilities' – are important in themselves in evaluating welfare and have an intrinsic value to the person rather than just being a means to produce utility.223, 224 90 2.5.1. Functionings and capabilities During the same period that QALY measures developed in the health economics discipline, Sen's work on capabilities222, 225 was "originally developed for the purpose of evaluating inequality, as an answer to the question ‘equality of what?’ " (p.818).211 Sen (1979) argued that as individuals are not identical and they do not hold the same utility functions, then equity and distribution issues will arise if one attempts to distribute benefits or outcomes in an equitable way based on utility functions.222 Sen (1980) suggested that what is missing in the utility framework in relation to equal distribution is the notion of “ ‘basic capabilities’: a person being able to do certain basic things” (p.218).222 In his approach, Sen conceptualises wellbeing in terms of ‘functionings’ – what a person is and does – and ‘capabilities’– what a person is able to do and be.223-225 ‘Functionings’, consisting of beings and doings such as being healthy and having a job, relate to many different dimensions of life and are influenced by different factors, on an individual level but also on a social, institutional and environmental level.223, 226 ‘Capabilities’ are the alternative functioning combinations that a person is able and has the freedom to achieve, which contribute to the person’s wellbeing.224 What is unique to the concept of ‘capability’ is the ability and freedom a person holds to function in a particular way whether or not they choose to do so.169 The extra-welfarist approach, was influenced by Sen's notion of 'functionings'.169, 201, 202 Culyer (1990) argued that the difference between his notion of 'characteristics of people' and Sen's notion of 'basic capabilities' is that his notion is a more open and 91 inclusive way of thinking about quality of life whilst Sen’s criterion of 'basic' could involve the exclusion of some characteristics of people.201 In contrast, other academics have argued the opposite; that the capability approach is broader than Culyer's approach in extra-welfarism which in practice only focuses on health states and achieved 'functionings' rather than 'capabilities' - the ability to do certain 'functionings'.155 2.5.2. Freedom and agency In his approach, Sen argues that the freedom to achieve functionings is of intrinsic value to a person's well-being.223, 224 Sen notes that "if the achieved functionings constitute a person's well-being, then the capability to achieve functionings ... will constitute the person's freedom - the real opportunities - to have well-being." (p.40).223 Sen called these 'real opportunities' to have well-being 'well-being freedom', which it is not just concerned with a person's achievements but rather with having the freedom to choose.223 To explain this, Sen used as an example the difference between starving and fasting; whilst a person who fasts may choose to do so, a person who is starving does not choose this.224 In this example, the functioning of not eating - that is 'well-being achievement' - is the same for both the person starving and the person fasting.224 Yet, only the person fasting has the freedom to choose - which constitutes their 'well-being freedom' or capability - from two alternative options.224 Thus, Sen argues that having 'wellbeing freedom' is of direct importance to having well-being and it is an alternative way to measure well-being.223 92 In addition to 'wellbeing freedom' and 'wellbeing achievement' Sen suggested (1993) another two sources of evaluation which relate to 'agency' - that is 'agency achievement' and 'agency freedom'.224 Sen defines 'agency' as a person's ability to pursue goals that they value and they have reason to value.224 Sen notes, that 'agency' does not only refer to personal goals driven by self-interest to - at least directly - improve one's own wellbeing but could also refer to goals that relate to the wellbeing of others.224 Thus the idea of 'agency' also includes a person's role and participation within society, it relates to community values and it is a way to take societal barriers into account when assessing a person's freedom and capabilities.209, 224 'Agency achievement' differs from 'well-being achievement' in the sense that the former assesses "the person's success in the pursuit of all the objectives that he has reason to promote" (p.37)224 whilst the latter only looks at a person's achieved functionings.224 Furthermore, ‘wellbeing achievement’ is not concerned with all the goals - including altruistic goals - a person values but only with goals related to a person's own wellbeing.224 Thus, as Sen argues the assessment of 'agency achievement' consists of a broader evaluation space than 'wellbeing achievement.224 Finally, 'agency freedom' refers to the freedom (e.g., to be educated) people need in society to make decisions, to act as 'agents' and achieve the functionings they have reason to value.209 It is important to note that Sen's definition of 'agency' is very different to the common use of the term in the health economics discipline and healthcare - where an 'agent' is usually a person who acts on behalf of someone else (p.17)161. For example, a HCP who 93 acts on behalf of, and in the best interest of, the patient. Sen's concept of agency is similar to how the term is often used in the social science disciplines where agency is usually a synonym to human action and the capability for individuals to make their own choices.38 In addition to the focus on individual empowerment - as a person acts to bring about change - Sen also acknowledges external influence (e.g. economic or social institutions) to individual action and change.227 Other capability scholars, such as Nussbaum (2011) and Robeyns (2003) give more emphasis to the role of institutions in bringing about change and see governments as a key actor for change.228-232 Within the capability approach literature there are several types of factors that can influence the processes by, and extent to which, individuals are able to convert resources into functionings.223, 231, 232 These different types of conversion factors can be personal, social and environmental.227, 233 Personal factors are those related to the personal characteristics of a person that may hinder or expand their capabilities, such as their age or health. Social conversion factors include social norms and government policies, while environmental factors include infrastructure, pollution and provision of a public good.233 Health can be considered as a personal conversion factor. For instance, a person’s poor health condition may limit their ability to achieve a valuable capability, such as being able to enjoy life or have an income.233, 234 However, having good health can be also valued in itself and in this case conceptualised as a capability rather than a conversion factor.233, 234 The capability approach suggests that such differences need to be made 94 clear when conceptualising capabilities and considering whether an aspect of life is valued as an end in itself, or as a means to a valuable end.231, 233 2.5.3. Capabilities and EOLC for CYP The narrow focus of extra-welfarism implies that the goal of care is health maximisation; this however may not be appropriate in the context of EOLC 235 for CYP. As discussed in Chapter 1, the approach to EOLC is multidisciplinary and holistic and aims to improve the quality of life of both CYP at the EOL and their families. 7 Therefore, it can be argued that EOLC for CYP should be evaluated in a way that is more relevant to, and aligns with, the goals of EOLC for CYP. Using the capability approach can provide a broader framework in which the goals of EOLC are considered in economic evaluation but that allows for the inclusion of individual choice which is particularly important in EOLC. 235 Coast 235 gives the example of two individuals, with one having access to a hospice where they live but choosing not to use it and the other individual living in an area where there is no access to a hospice but who wishes to access hospice care. While neither individual is using hospice services, and so their functioning levels are considered the same, the individual with no access to the hospice but who wishes to use it is worse off in terms of their capabilities. 235 Including individual choice in economic evaluation is also important in terms of considering CYP not just as recipients of care but as active agents with their own priorities and wishes. 236 The capability approach has been used in the 95 EOLC context to evaluate capability outcomes, alongside health outcomes, for PC day services in the UK. 237 2.5.4. Critique of the Capability Approach The capability approach has been subject to criticism in terms of both theory and its applicability to economic evaluation.228 Here the focus will be on theoretical arguments whilst the following chapter will include discussion of concerns in relation to the application and measurement of capabilities in economic evaluation. 2.5.3.1 Capability approach and Individualism A critique of Sen's approach relates to individualism.209 It has been argued that the capability approach presumes people act as individuals rather than as members of a community.209, 228, 238, 239 Advocates of communitarian theory - who regard individuals as part of the broader social environment and their decisions and actions as being connected to and influenced by others - have argued that the capability approach promotes the idea that individuals are autonomous and their choices are independent of the society they are part of.228 However, this critique has been challenged based on the argument that there are different forms of individualism - ethical and ontological - and the capability approach does not endorse both.228 In his work on capabilities, Sen takes the position of ‘ethical individualism’ which supports the causal relationship between social structures and individual wellbeing 96 and at the same time looks at what happens to every individual within a society.209, 228 Thus, in his approach Sen takes a liberal stance in the philosophical sense that values individual autonomy and freedom.228 Furthermore, Sen argues that freedom has multiple aspects.224 The freedom an individual holds to live the way they would like may be assisted and enhanced by the choices others make, by public action and policy, and thus, individual active choice cannot on its own count as an individual's achievements.224 This position is different to how individualism is understood in the ontological sense - where explanation and study of social phenomena is reducible to the characteristics of individuals.38 It also differs to individualism in the neoliberal economic sense in which individuals are thought to act in a rational (thus efficient) way.228 2.5.3.2 Which human capabilities are important in defining quality of life? Another critique of Sen's approach is around questions of which capabilities are considered important in measuring wellbeing and whether there is a need for a fixed list of basic human capabilities. This is not only a theoretical, but also a methodological challenge for measuring capabilities in economic evaluation. Sen recognises that not all capabilities are of equal footing223 but does not support having a fixed list of functionings for a good life as this runs the risk of being over specified and perhaps less objective.224 In particular, Sen argues that individual choices vary by culture and thus any application of the capabilities approach needs to be culture-dependent, especially 97 when weighting different capabilities.222 Furthermore, in his critique on welfarism, Sen argues that the focus on utility results in overlooking other factors that contribute in defining, and assessing quality of life.227, 240 Other factors include: - 'personal heterogeneities' as people have different personal needs240 - 'institutional variations' depending on access people have to institutions240 - differences in 'relational perspectives' as individual disadvantage is relative to the community the individual is part of240 Sen concludes that the 'incompleteness'(p.47)224 of the capability approach is deliberate.224 Yet, there have been attempts from other academics208, 229, 232, 241 to generate lists of basic human capabilities.242 One of the most influential attempts was by Nussbaum. Nussbaum proposed a list of ten central human capabilities to help define a set of essential functionings and to establish a minimum threshold of what constitutes a ‘good life’.209, 228, 243 According to Nussbaum, these basic capabilities are necessary for individuals to be able to develop other more advanced capabilities.228 Nussbaum's list of capabilities includes being able to have good health, to be able to move freely, to be able to affiliate and interact with others and to be able to have political and material control over your environment.209 Nussbaum defends the necessity of going beyond the incompleteness of Sen's approach to avoid issues of omission – as groups may unintentionally overlook an 98 important capability and issues of power – as what people consider to be important can be the product of inequality structure.209 In response, Sen comments that Nussbaum's list – although it can be useful in identifying important capabilities – can undermine social participation and individual agency when others – in this case an academic – make judgements about what is best for people.209 Instead, Sen supports that the social choice of important capabilities should be collectively decided – through democratic processes – by the individuals affected by the relevant intervention in question.232 2.6. Shift of focus in health economics The extra-welfarist approach and the capabilities approach, developed and gained momentum within the health economics discipline during and following a time that inequalities - in health and beyond - were highlighted in various reports244-246 in the UK and abroad (1970s to late 1980s) following a period of stagflationn in the 1970s. During this time NHS financial problems worsened147 and inequalities in health – in particular different mortality rates between social classes – became more evident.247 The Black Report, published in 1980 provided evidence of persisting health inequalities in the UK and a widening gap between social classes.244 n Unemployment and Inflation 99 There was a shift of focus evident in the way both the extra-welfarist and capability approaches in theory and in practice attempted to measure welfare. This was reflected by a move away from a theoretical focus on individual utilities towards understanding welfare and development as multi-dimensional and an attempt to incorporate these multiple attributes within economic evaluation (e.g., QALYS and Capabilities).209, 211, 248 Under extra-welfarism there was a shift on valuation focus from individual preferences to community preferences (Mooney)195 and 'decision maker' (Culyer)201, 249 or community values (Sen).209 Finally, under extra-welfarism there was greater emphasis on weighting values in economic evaluation134, 155 and interpersonal comparisons were introduced.134 The above shifts have been acknowledged as attempts to deal with equity and distribution issues in economic evaluation.134, 155 More recent reports on health equity in England such as the Marmot reviews in 2010250 and 2020251 continue to highlight health inequalities including a growing health gap between wealthy and deprived areas. The application of the capability approach is growing over the recent years in relation to the health field suggesting a shift in the evaluative space from functionings to capabilities,252, 253 a focus on health and health care justice254-258 and a growing interest to include the experiences of those receiving health care interventions in health care decision making.253, 259, 260 100 2.7. Economic evaluation of child health 2.7.1. Shift of focus - from child's human capital to child indicators approach Another shift of focus has been evident in relation to measuring child wellbeing. Past literature on economic evaluation of health care interventions for children focused on the children's human capital related to their future earnings potential and the household's maximisation of utility.261 However, in the 1990s, new initiatives and reports, such as the United Nations International Children's Emergency Fund (UNICEF) report and the United Nation's convention on the rights of the child (UNCRC) contributed to the 'social indicators movement' (from 1960-70s) which argued for a need to assess the conditions of different groups in society over time in a more multi-dimensional way.262 In the late 1990s there was a shift of focus in economic evaluation concerning children; from the negative concept of 'child survival'262 and a focus on deprivation and basic needs indicators to the positive concepts of 'child development',262 'wellbeing' 262 and 'quality of life'.261-263 Ben-Arieh suggests three important normative changes that have contributed to the development of the child indicators approach: the concept of children's rights, the development of new social theories on childhood as a stage of life in and of itself, and child development theories.262 The UNCRC principles (1989) promoted children's rights 101 and particularly children's role as citizens in their own right, acknowledging that the child's own views should be taken into account in matters that affect them.262 This focus on children's rights highlighted a need to include indicators of child wellbeing that the children themselves consider important.262 Including the child's view in studies that affect them, rather than just the adult's (caregiver or HCP) perspective, does not only promote respect for the child as a person but also shows a recognition that the child's perspective might differ from that of the adult.262, 263 Furthermore, the UNCRC (1989) drew attention to children's current wellbeing and the childhood life stage in its own right rather than just child wellbeing in terms of the child's future and employability.262 2.7.2. Child's own perspective and capability approach There is a growing interest in including the child's own perspective in research and economic evaluation involving children.261, 262, 264 A number of academics have suggested that the capability approach can be particularly useful for understanding, measuring and improving child wellbeing by acknowledging children as active rather than passive agents in the process of developing their wellbeing and capabilities.265, 266 Biggeri et al. emphasize the need to incorporate children's voices on matters that affect them and not just simply consider them as recipients of freedom but as active agents who hold their own views and values.265, 266 Furthermore, Ben-Arieh et al. argue that by applying the capability approach and focusing on the child as the unit of analysis allows for the measure to reflect the life that the child is actually living - taking 102 into account personal choices, circumstances, abilities to achieve what they value and constraints - rather than the resources or means the child may have available.262 Yet, academics supporting children's participation in economic evaluation and research also recognise that this entails a number of methodological and ethical challenges including possible differences about views and priorities in different age groups, gaining informed consent and children's developmental changes and cognitive abilities.262, 265-268 These concerns are outlined in Chapter 3. 2.7.3. Concepts of wellbeing and well-becoming The concept of child wellbeing in economics, as defined by Heckman et al., envisions "a core set of capabilities as capacities to function including cognition, personality and biology." (p.3).261 However, as with the concept of adult wellbeing, there is no consensus in the literature on its definition, what the defining dimensions and 'indicators' of child wellbeing are nor how to weigh them in terms of importance.261, 264 UNICEF suggests five different child wellbeing dimensions: material wellbeing, health and safety, education, behaviour and risks, housing and environment.269 Each dimension includes several different indicators such as deprivation, housing problems and education participation.269 Whilst the concept of wellbeing is concerned with current wellbeing, the concept of well-becoming is future oriented and concerned with future opportunities the child may have as an adult.253, 262 Ben-Arieh et al. argue that the concept of well-becoming is 103 concerned with preoccupations of adults which, although a very useful approach in measuring outcomes for children, does not include measurement of the current situation of the child but rather assumes that the child's 'good life' is postponed until adulthood.262, 263 In health economics, the focus has been around current wellbeing rather than well-becoming 253 but in more recent years there has been a growing interest in using both concepts to maintain a dual perspective as each one is of value; the former is useful in understanding and measuring childhood in itself and the latter is forward looking.253, 263, 265 The capability approach agrees with this dual perspective as wellbeing can be understood in both present and future beings and doings 253, 265, 266 and capabilities experienced during childhood have a significant effect on children's capabilities as adults.265 2.7.4. Family as a unit of analysis There is also a growing consensus on taking a family approach in economic evaluations with children, rather than measuring health outcomes for the patient in isolation.261, 264, 268 The family approach considers the family as the unit of analysis in which health care decisions for members of the family would be made by the household inter-dependent collective.261, 264, 268 Children have a higher level of dependency than adults and that often significantly affects not only their health state but that of others in the family.268, 270 The effect of health care interventions on other individuals in close proximity is called a 'spillover effect'.270 Under this approach, costs such as informal care, and outcomes for individuals other than the child, such as the person with 104 parental responsibility, are included in the evaluation.268, 270 The provision of informal care can involve significant costs (e.g. time invested in care) and effects (e.g. on psychological wellbeing) to the person providing the care.270 However, there is lack of consensus on what methods are most appropriate to include such costs and effects in economic evaluation.270 Measuring spillover effects is discussed further in Chapter 3 with a focus on EOLC and child health. 2.8. Conclusion The different theoretical approaches used in the economic evaluation of health care were outlined in this chapter in order of chronological developments in order to show how focus has shifted towards new ways of assessing outcomes of health care interventions; examples of their use in evaluations of EOLC interventions for CYP were given at a number of points in the chapter. Although a large part of the health economics literature is based on welfare economics, the focus of welfare economics on individual utility determined by commodities and the consumption of goods, and particularly monetary valuation, has reduced its influence on the economic evaluation of health care interventions. Extra-welfarism, focusing on maximising health gain rather than utility is the method currently most often used within economic evaluation, but the capability approach has been suggested as offering a broader evaluative space that includes outcomes beyond utility and health. 105 Given the multi-disciplinary and family approach to care at the EOL, the application of the capability approach may be particularly useful for informing decision making for complex care interventions at the EOL. This is because the impacts of interventions at the EOL are not limited to changes in health. Interventions at the EOL include more general impacts of supportive care for the person at the EOL and impacts of support for those close to them. The empirical work of this thesis will thus draw on the capability approach to assess impacts of interventions at the EOL for young people and for those close to them. Yet, there are practical considerations for assessing impacts of interventions, and in particular around measuring outcomes at the EOL and in child health. Furthermore, when outcomes for those close to CYP at the EOL are included within economic evaluation, there is the question of how outcomes should be weighted between CYP at the EOL and those close to them. Chapter 3 moves on to consider all these practical challenges and to identify existing gaps in assessing care at the EOL for young people. 106 107 CHAPTER 3. MEASURING OUTCOMES IN HEALTH AND CARE FOR USE IN ECONOMIC EVALUATION The normative and methodological approaches used for the economic evaluation of health and care, were introduced in Chapter 2. Chapter 3 focuses on the measurement of outcomes in health and care for use in economic evaluation, starting with a general introduction to outcome measurement. The chapter then moves to the particular challenges in measuring outcomes at the EOL, for the person at the EOL but also for those close to them. Additional challenges are then outlined which are measuring outcomes in child health and challenges in using the capability approach as an alternative evaluative space to measure outcomes for use in economic evaluation. 108 Existing empirical work is then reviewed at the final part of Chapter 3 (section 3.7) to identify gaps in research in relation to measuring outcomes at the EOL for young people and those close to them. Chapter 3 then concludes with the three aims of the empirical work conducted in this thesis. 3.1. Measuring outcomes in health and social care 3.1.1. Definition of outcome measurement Outcomes research looks at ways of quantifying change between an individual’s current (baseline status) and their future outcomes over time.271 Traditionally, outcomes in health care included extreme health indicators such as mortality and life expectancy which were quite easy to measure.137 This changed with time as health provision improved and outcomes included objective clinical indicators such as blood pressure which were also relatively easy to measure.137 The focus then moved to measuring outcomes related to the impact of interventions on the patient and the patients’ own perspective on the impact of interventions (patient-oriented indicators) particularly around the patient’s ability to function (functional indicators).164 109 Finally, a need for a broader view of the impact of interventions was recognised that could be used across different health conditions: measuring change in patient outcomes in terms of impact on a patient’s quality of life (QOL).137 Measuring QOL is challenging, as indicators are multi-dimensional and aim to include a range of dimensions of life which could have an impact on a person’s wellbeing.164 QOL indicators in economic evaluation research are often restricted to those that have an impact on health, that is health-related quality of life (HRQOL) indicators.137 However, it has been argued that the focus on health indicators may result in missing other important outcomes that cannot be measured in terms of health, such as social care outcomes which may not directly contribute to health but may improve a person’s QOL.190, 195, 196, 272-275 3.1.2. Purpose of measuring outcomes in health and social care Measurement of health and social care outcomes is used for a variety of purposes. Outcome measures are widely used to provide evidence for clinical assessment of interventions in empirical research that aims to inform clinical practice.276 Outcome measures are also used by service providers and commissioners as one of the means to assess, evaluate and monitor the quality and effectiveness of interventions provided by a service.277 Finally, outcome measures are used to compare outcomes, alongside costs, across different interventions for the purpose of effective resource allocation in 110 health care.278 This chapter will focus on measuring outcomes for the latter purpose of resource allocation through economic evaluation. Although measurement of outcomes in terms of money has been used in the past in economic evaluation, measurement of outcomes in terms of improved health – rather than in monetary terms – is dominant currently in the UK and internationally.137 The conceptual and methodological reasons contributing to a shift from money to a health focus, as well as current theoretical issues in measuring improved health in economic evaluation were outlined in Chapter 2. Currently outcomes in economic evaluation tend to measure change in the patients' health status, and the impact of an intervention or set of interventions on their HRQOL.271, 276, 279-281 For the purposes of economic evaluation these outcomes are then quantified using different methods to inform decision-making for health care resource allocation.137 Decision-making bodies around the world such as NICE in England and Wales usually have formal guidelines (sometimes known as a reference case) which state what methods should be used to measure outcomes in health and social care for decision making, often through economic evaluation.281 Current NICE guidelines (2022) state that health outcomes in the UK should be measured in QALYs for the evaluation of the cost effectiveness of interventions.197 111 3.1.3. Types of outcome measures used in health and social care in the UK Outcomes for economic evaluation are standardly measured through the use of questionnaires, that patients or proxies (a family member or caregiver) complete.137 Questionnaires that aim to obtain HRQOL information from the patient are often called Patient Reported Outcome Measures (PROMS).281 There are different types of PROMs used in health care depending on what they intend to measure (generic or condition-specific) and from whom they intend to collect the information (patient or proxy). Generic measures can be used with patients with different conditions and who may be receiving care in different care settings whilst condition-specific measures can only be used for patients with a specific condition.280 3.1.4. Use of generic or specific measures in economic evaluation of health care It is important to develop or choose to use a measure that relates to its purpose and that will most adequately capture relevant changes in the population it is intended for.277 Specific measures are useful in capturing clinically important changes for a particular condition, and thus are useful for clinical purposes and individual patient care.277, 280 Generic measures are often not responsive to such small, but clinically 112 important changes nor are they able to show information about particular types of care outcomes that are specific to a particular condition.277, 280 Generic measures often capture broad dimensions of HRQOL focusing on function (for example mobility), but more often than not they fail to capture other non-health related benefits that may be valued by patients and may also be important in improving their QOL.282, 283 Generic measures can be applied consistently across different interventions and thus are useful in comparing outcomes of different interventions.277, 280 Comparing outcomes across interventions is a great advantage for the purpose of resource allocation in health care where decisions have to be made across competing alternatives for funding.253, 277, 280 A generic measure called EQ-5D (also known as EuroQOL-5Do, EQ-5D-3Lp or EQ-5D-5Lq), is commonly used in measuring cost-effectiveness in healthcare in the UK and internationally.197, 284 o 5D stands for ‘five dimensions’ p three level version q five level version 113 3.2. Methodological and practical considerations in measuring and valuing outcomes for use in economic evaluation of health and social care 3.2.1. Psychometric properties of measures An outcome measure needs to fulfil certain criteria specific to its purpose – in this case for use in economic evaluation – to be of good quality.137, 164 These criteria are mainly around the psychometric properties of the measure: feasibility, reliability, validity and responsiveness.137 3.2.1.1 Feasibility Feasibility is concerned with whether, and by how much, the measure can be practically used in the relevant context. For instance, a patient may find it too difficult to complete a measure, because they are too ill and the measure is cognitively demanding137 or the language or cognitive ability requirements are not appropriate.285 Feasibility can be assessed in relation to the ways the measure is administered, the time it takes to administer and rates of completion.164, 286 It has been suggested that using a ‘bottom up’ methodology in developing a measure can increase self-completion rates.281 For example, a child may be more likely to self-complete a child measure that uses wording determined by children.281 114 3.2.1.2 Reliability If an outcome measure gives consistent results when repeatedly measuring a situation that has not changed, then the results it provides are reliable.280 The extent to which a measure is reliable may vary between different patient populations.287 The test-retest method is often used to assess reliability: the measure is administered at two different points in time, using participants whose situations have not changed between the two time points, to look at measure score correlation.288 3.2.1.3 Validity Validity assesses the extent to which what the measure intends to measure – for instance HRQOL – is what it actually measures.280 There are different types of validity: Face and Content validity: Face validity refers to whether the measure contains dimensions that are relevant and important to the patients themselves and whether the measure seems to be assessing the desired dimensions.137, 280, 287 Content validity assesses whether the measure content and dimensions are relevant and important to the intervention.137, 280, 287 When a measure is developed, face and content validity may be ‘tested’ by patients, carers, or relevant others who make a judgement about whether, and to what extent, the measure appears to be valid for its purpose.164, 287 Criterion validity: The extent to which the measure reflects a ‘gold standard’ measure already used in the area studied.137, 280 If a ‘gold standard’ measure does not exist, 115 which is often the case for measures in health, other measures intended for the same or similar purpose that already exist are compared against the new measure developed.287 Validity through comparison can be assessed by administering existing measures alongside newly developed measures to a sample of people and checking whether correlation is strong between the two measures.287 Construct validity: How the measure’s results relate to dimensions suggested by theory.137, 280 A hypothesis can be used to assess differences between population groups expected to provide measure results at differing amounts.287 This can be achieved by applying the measure to the relevant sample population and testing whether the result is the one expected by the hypothesis.287 Still, even if the expected result is not achieved it does not necessarily mean that the measure is not valid; it is possible that the hypothesis is not valid.287 3.2.1.4 Responsiveness It is also important that the measure is responsive, or sensitive, to change and thus can measure meaningful changes over time.280 This is particularly difficult to achieve if the measure focuses on measuring mainly functioning and intends to measure change as a result of an intervention at an advanced stage of illness or a progressive illness where interventions may focus on improving areas other than functioning.280 It has been suggested that using ‘bottom up’ qualitative methods in developing a measure may improve responsiveness to change.281 This is because the measure will be more 116 patient-centred aiming to capture change relevant to the population that it is intended for.281 It is important to note that when assessing the responsiveness of a measure the purpose of the measure defines the type of change that the measure aims to capture. If the measure is for clinical purposes then measuring clinical change is what is most relevant. However, if the measure is for economic evaluation purposes, then what matters is responsiveness in changes to patient preference for the health change, rather than actual clinical change.289 3.2.2. Issues with valuing outcomes in economic evaluation Another key consideration in developing a measure for the purposes of economic evaluation is the number of items included in the measure. Each outcome should represent one item (also called dimension or attribute or question) on the measure.290 A large number of items can reduce the extent to which the measure is feasible to complete, as completion rates decrease when number of dimensions increases.286 A large number of items also makes valuation and weighting of dimensions more challenging to achieve.290 As different dimensions may not be valued equally, by patients or the general public, it is important to assess the value of each dimension.290 It has been suggested that a measure for the purposes of economic evaluation should typically have five to nine items to allow for valuation.289 117 There are also methodological and theoretical challenges to consider. It is important to choose the most appropriate method to weigh dimensions (how important each dimension is to the population that it is valued by).137 Health preferences may vary across age groups and socio-economic background.194 There is also the normative question of whose preferences in society a researcher should choose when valuing a measure, those of the general public or the patient group.194 It is more appropriate to consider the above questions by focusing on the specific purpose of a measure and the population that it is intended for. 3.3. Challenges in measuring EOLC outcomes for use in economic evaluation 3.3.1. Important outcomes of interventions may not be limited to HRQOL outcomes at EOL stage. As care needs at the EOL are complex, different types of interventions are provided to the patient to manage different needs.291 People may benefit from interventions around their social care needs, spiritual needs around dying and family needs.291 This holistic approach to a spectrum of needs makes measuring outcomes challenging. This is because outcomes are both multidimensional and subjective. Outcomes are multidimensional, consisting of a variety of QOL outcomes but not necessarily 118 exclusively concerned with improving or extending health, and subjective, measuring the outcomes that people value for a good QOL and a good death.285 As a result, there have been concerns about whether using QALY gains, that focus on outcomes in improving or extending health, in EOLC can adequately capture important outcomes specific to the EOL stage to demonstrate the cost-effectiveness of interventions at the EOL.272, 282, 292-294 There have been different responses to this challenge.295-298 One of the responses to this challenge is to not abandon the QALY measure, as it is a measure widely used and thus useful for comparing different interventions, but rather than using a generic health measure to instead develop a measure that incorporates and reflects health issues experienced specifically by those receiving PC and EOLC.282, 297 Condition or intervention specific measures have been developed and widely used in PC 280 such as the Palliative Care Outcome Scale (POS),282 a shorter version of POS called POS-E specifically for use in economic studies,282 a refined version of POS called Integrated Palliative Care Outcome Scale (IPOS) which includes additional symptoms,299 the Edmonton Symptom Assessment Acale (ESAS) for cancer patients 291 and the Support Team Assessment Schedule (STAS).280 However, these measures, with the exception of POS-E, are not intended for economic purposes because the measures are too long to be amenable to valuation.277 Another response to the challenge of QALYs not adequately capturing important outcomes of EOLC is to look beyond QALYs with the goal of measuring a person’s capabilities. QALYs focus on health-related functioning, whilst capability measures 119 allow outcomes beyond health gain to be captured to measure wellbeing. Examples of capability measures for use in economic evaluation of EOL are the ICECAP Supportive Care Measure (ICECAP-SCM)293 for adults in an EOL setting and the ICECAP Close Person Measure (ICECAP-CPM)290 for those close to the person at the EOL. There are values available for ICECAP-SCM based on best-worst scaling techniques with the general population300 and a provisional scoring system for ICECAP-CPM based on deliberative work with focus groups with members of public and policy makers.301, 302 3.3.2. QALYs and time QALY measures have also been criticised for their applicability in EOLC due to their assumption that time is additive.294-296 QALYs assume that patient preferences at any point in time are stable and each year of added life is weighed equally thus making time additive.295 Whilst in QALYs time in a health state is considered as something that has a constant value, there are situations, such as in the EOL stage, where the patient may put less or more value on time spent in a health state.183 It has been suggested that the valuation of time may increase as time is running out.295 However, it is also suggested that in PC longer illness trajectories may not necessarily be preferred to shorter trajectories.294 This assumption that time is additive could also be a potential issue for QALY measures in other health and social care settings depending on the nature of the intervention and patient needs.183 Responses to this issue in PC – within the QALY framework – 120 include the suggestion of measures such as the Palliative Care Yardstick (PaLY) which allows valuation in components of a ‘good death’183 and the suggestion of Valuation Index Palliative (VIP) which is based on the PaLY suggestion and assigns a higher value to time in proximity to death.303 A further challenge noted is the relatively low life expectancy at the EOL stage.295 When estimating QALYs for interventions the score of the length of time spent in a health state (life years gained - LY) is weighed by the quality of life score (Q) to estimate a QALY.295 For example, a year in perfect health is equivalent to one QALY whilst a year in a health state that is not perfect is worth less than one QALY.295 Yet, in EOLC, life expectancy is low and may limit the QALY gains of an intervention.295 Furthermore, one of the main goals of EOLC is to improve QOL and to provide care that will allow the patient to have a ‘good death’; this goal does not necessarily align with prolonging life which within the current QALY framework is assumed to be beneficial.295 As a result it has been argued that the interventions may not present as cost-effective if based on QALY gains, and it would not be fair to compare EOLC interventions with others and base economic decision-making for EOLC interventions on QALY gains.183, 292, 294-296 Yet, it has been suggested that there could still be QALY gains even when life expectancy levels are relatively low and the goal of the intervention is not necessarily to prolong life.297 This can be achieved by focusing on the outcomes of interventions related to improving the QOL component rather than LY component of QALY.297 121 3.3.3. Important outcomes may not be limited to patient outcomes in EOLC. Over the last 15 years, attention to the measurement and inclusion of outcomes for those close to the patient (close persons), such as informalr carers, close family and close friends, has increased considerably in economic evaluation of health care.272, 280, 290, 304-307 Interventions directed at the patient can affect the wellbeing of those close to them (spill-over effects or externalities) in two ways: the effects of caring for someone (‘caregiving effect’) and the effects of caring about someone directly on the wellbeing of those close to the patient (‘family effect’).308, 309 It is suggested that if a societal perspective is used in economic evaluation, the effects of interventions on those close to the patient should be considered and measured.308, 309 NICE guidelines (2022) suggest the inclusion of carer health related outcomess in economic evaluation where relevant197; however this is a limited interpretation on the impacts of carers.310, 311 Going beyond patient outcomes may be particularly important for EOLC to address both patient needs as well as family and carer needs.280, 302, 312 Those close to the dying person often contribute substantially to care and experience significant burden when r Also called unpaid carers. The rest of this chapter refers to ‘carers’ to mean informal/unpaid carers. s Caregiving effect 122 providing care to the patient.280 It is also relevant to measure the direct and indirect impact of interventions at the EOL on the wellbeing of those close to the dying person.290, 302, 313, 314 For instance, a close person may benefit directly from an intervention they receive themselves, such as support around bereavement.310 Furthermore, an intervention directed at the dying person that may enhance their independence, may reduce the time an informal carer needs to spend with them to provide care.313 In this case, the indirect benefits for the close person are not solely a result of providing less practical care to the patient315, 316; the close person may also benefit from knowing that their loved one has benefited from the intervention (caring externalities).310 A number of outcome measures for people receiving interventions include the needs and experiences of carers, such as POS with a question on family anxiety.277 Others are designed specifically to assess carer needs and outcomes, such as the CarerQoL,317 CES,318 ASCOT-Carer,319 and FAMCARE.320 ICECAP-CPM290 goes beyond carers to capture the experiences of those close to the person at the EOL, such as close family members and friends.301, 314 Existing HRQOL measures usually focus on QALY gains of the patient and do not include the effect of interventions on carers or close person(s).310 Not including such effects for the purposes of resource allocation may result in an under-provision of EOLC if interventions are perceived as not cost-effective.307, 310 Although existing HRQOL measures do not usually capture outcomes for carers or close person(s), this 123 does not mean they do not have the flexibility to do so.310 However, the inclusion of QALY gains for those close to the patient in economic evaluation poses methodological and normative challenges that would need to be considered.305, 310 To have the advantage of consistency and facilitate comparison for resource allocation purposes, the same measure could be used to calculate QALY gains for both patients and close person(s).310 However, if the measure focuses on a HRQOL framework then this focus might not be as relevant to capturing important benefits of interventions for close persons who have a different role to that of the patient.310, 321 A study on what people value when providing informal care suggests that important outcomes are around feeling supported, being able to maintain control and engage in other activities.318 It has been argued that measures for carer outcomes should thus cover a range of outcomes relevant to the caregiving experience and welfare of carers, especially in the PC and EOLC context.321 A suggestion is to measure ‘care-related quality of life’ (CRQOL) rather than carer HRQOL to reflect care related outcomes.305, 310, 311, 317 However, this means that a different unit is used to measure outcomes for carers (CRQOL) and outcomes for patients (HRQOL) which produces challenges in how to aggregate the two types of QOL outcomes.305, 317 Different units may imply that the carer and patient have a different ‘identity’ within interventions and economic evaluation.310 Another option is to develop a measure that adopts a wider QOL framework that is relevant to both patient and close person(s) and at the same time allows consistency 124 to be maintained in calculating QALYs.310 Such an option may be appropriate in the EOL context where interventions do not focus on improving health but rather on improving broader QOL attributes – such as maintaining dignity – for both the patient and those close to them. 3.3.3.1 Identifying close person networks One of the issues in measuring outcomes for those close to the patient is around which person(s) to consider and include in economic evaluation.310, 314, 316 Identification of close person(s) could be achieved by including ‘family networks’ where ‘family’ suggests the close network of individuals around the patient.307 Within the EOL context, a UK qualitative study has used hierarchical mapping to examine the size and structure of the patient’s informal network at the EOL stage finding that the size of the patient’s close-person network was eight individuals on average, including family and non-family members.314 The study concluded that a key starting point for the purpose of economic evaluation of EOLC would be to focus on the three individuals closest to the patient.314 However, this may vary depending on the cultural context and population, and networks may be of very different size and structure where the person at EOL is a child or young person. It has been argued that extending the measurement of outcomes beyond the patient may result in equity issues.304, 316 This is because estimates of gain in outcome may be 125 higher for patients who have a wider network of close person(s) than for those who have smaller or no informal support networks.316, 322 This may favour decision-making for conditions that usually require or involve higher levels of informal care over conditions that do not, or undervalue conditions or interventions that tend to affect more isolated individuals.323 For example, individuals suffering with mental health conditions, over individuals that are usually more connected, such as children. In addition, inclusion of close person outcomes may result in ‘double counting’ of benefits, where spillover effects may have already be included in measuring outcomes for the patient, or it may affect decision-making in benefiting the close person(s) over the patient.323 3.3.3.2 Relative weights between outcomes for the person at the end-of-life and outcomes for those close to them When outcomes of different stakeholders, such as patients and carers are incorporated within economic evaluation there is the question of how outcomes should be weighed between the different stakeholders. If using the same unit across different stakeholders, for example carer and patient HRQOL outcomes, outcomes for carers and patients can be aggregated and a decision about whether the relative weights are the same for both carers and patients should be made.305, 324, 325 A suggested framework for incorporating health outcomes of carers or family members 126 within economic evaluation where the goal is to maximise health is through the estimation of multiplier effects in relation to health outcomes gained and health outcomes being displaced by a new intervention.307 Aggregating outcomes using different units across different stakeholders, such CRQOL for carer specific outcomes and HRQOL for patient specific outcomes is more challenging.310 Empirical work on obtaining non HRQOL relative weights between different stakeholders and especially in an EOLC context is limited.305, 310, 326 A recent study obtained relative weights between carers and patients QOL from members of the public using a person trade off (PTO) task 305. Another recent study obtained relative weights between family and friends and people at the EOL from members of the public and policy makers using budget pie tasks completed using deliberative methods.326 Although methods used to elicit relative weights in the two studies were different, both showed that participants were willing to split health care resources across the two different groups of stakeholders. 3.3.4. Valuation methods for EOLC The methods used for non-monetary cardinal valuation may not be appropriate for sensitive topics and may be difficult to use with populations that struggle with cognitively demanding tasks, which may include those at EOL among others.272, 300, 327 Alternative ordinal methods of estimating values which involve less complex tasks have been suggested such as discrete choice experiments (DCEs) and best-worst 127 scaling (BWS). These methods have ordinal properties and thus are cognitively simpler to complete as they do not require the respondent to choose by how much they prefer a health state to another, but to just choose which health state they prefer (DCE) or choose the worst and best attribute within one health state (BWS).272, 328 3.3.5. Practical challenges with patient and proxy measure responses at the EOL stage The illness trajectory of patients at the EOL, involving fluctuating and deteriorating levels of functioning, can pose practical challenges that may influence measure design decisions.277, 329-331 Patients may be unable to complete an outcome measure themselves, due to cognitive impairment or advanced illness or they may have already died.285, 291, 329 A patient may also not have been identified as dying - due to the unpredictable nature of the illness trajectory - and thus may not been given the option to complete an EOLC measure.276, 329, 331 This may mean that measures are more often completed by patients who are less ill and may exclude those with less ability to complete.285 There is the question of whether it is ethically appropriate to ask a patient at the EOL to complete a measure which may add to their distress given the short time left to live.276, 285, 331. However, it has been argued that patients who are very ill should not be automatically excluded from research because of this and providing patients with the opportunity to participate in research is ethically desirable.292, 332 Evidence from 128 systematic reviews of both studies on PC and EOLC suggests that patients – including those close to death – and families were consistently willing to participate in studies and the majority of them found the experience positive.292 Other literature on EOLC research suggests that giving EOLC patients the opportunity to make their own decision about participation in research allows them to benefit from contributing something to future generations.333 Recommendations suggest that the format and length of the measures – particularly in relation to completion instructions – should be brief and straightforward to answer.277, 291, 332 Furthermore, given the sensitive topic it is important to take particular care with the way questions are framed to improve response validity. Patients or proxies may find it difficult or embarrassing to talk openly about issues – physical, emotional, financial or other – that they are experiencing in relation to their condition and care received.280 Developing or piloting the measure with involvement from the patient group that the measure is intended for may assist in improving question framing by using the respondent’s language and avoiding technical terms.280 In addition, questions that include two questions in one (double-barrelled), hypothetical or leading questions should be avoided.334 Furthermore, respondents may under or over report due to social desirability bias rather than respond based on true reflection.335 The way the measure is administered – for example self-completion – can thus improve the accuracy of responses.280, 335 129 There is general agreement that the preferred and most accurate way to measure outcomes of PC would be by asking patients to prospectively evaluate outcomes from the care they currently receive.291, 329, 331, 336, 337 However, when patients are not able to provide this information themselves, a family member or a close person (proxy respondents) may complete the measure on their behalf prospectively or retrospectively.291, 329-331, 336, 337 This helps to have a representative sample for those patients who are unable to provide that information themselves.329, 330 Yet, it is important to recognise that proxy accounts are an interpretation, rather than direct reflection, of events as experienced by proxy respondents and influenced by their own needs, interests, and feelings.276, 329 A further issue with proxy responses particularly relevant to economic evaluation, is that if some interventions are evaluated by patients and others by proxies then it may be difficult to compare the interventions for resource allocation purposes.338 The validity and reliability of proxy responses will be significantly influenced by how close the proxy respondents are to the patient and the timing of the proxy respondent completion.329, 331 An early study (1994) compared retrospectivet proxy responses for EOLC with patient and proxy responses during the last week prior to death.339 It found that retrospective proxy responses were in good agreement with prospective responses for some outcomes.339 These outcomes were practical aid, time lost that t seven months in bereavement 130 could have been avoided and communication of professionals, family and patient.339 Yet, there was poor agreement between prospective self-report and retrospective proxy responses in relation to symptom and pain control, and patient and family anxiety.339 It has been suggested that whilst proxy responses may offer good accounts in relation to service provision and more observable patient symptoms, proxy accounts related to more subjective patient experiences have lower agreement with patient-reported accounts of these experiences.291, 299, 329, 331 3.4. Challenges in measuring outcomes in child health for use in economic evaluation 3.4.1. Challenges in measuring outcomes in different developmental stages Measuring outcomes in children poses specific challenges as measures need to be specific to age and cognitive abilities.291, 340-342 A series of physical, cognitive, emotional and behavioural changes are observed across different developmental stages within childhood.268 When developing an outcome measure for children it is important to consider these changes as they affect the way a measure is developed and what it is important to measure. 131 3.4.1.1 Cognitive functioning and ability to report health status in children As discussed in Chapter 2, there is growing consensus that the child’s own perspective is important, and being content experts341-343 children should be involved both in developing and completing child health measures.342, 344 One of the challenges in measuring outcomes in children is around whether and to what extent children of different age groups are capable of adequately understanding and reporting health and wellbeing outcomes and which methods may be most appropriate for obtaining reliable and accurate outcomes from children at different developmental stages.268, 344 In order to be able to report health outcomes, children need to have a basic understanding of the concepts of health and illness and be able to recall health experiences, so they are able to comprehend questions and response options and able to pay attention to completing the measure.268 Different methods of obtaining responses from different age groups are necessary given that the level of cognitive abilities will vary across different children age groups.268 There is no standard method of obtaining children’s perspectives on their health but recommendations suggest that when designing measures for children particular attention should be paid to the vocabulary and the techniques used to maximise comprehension.268 The following table outlines change in the ability of children and adolescents to report health status according to age268, 341: 132 Table 1 - Reporting ability of health status by age Age (years) Reporting ability of health status by age 0-4 • No ability. Proxy report 4-6 • Ability to recognise face icons and images on a Likert scale with assistance. • Tendency for responses to be on the extreme ends of scales (for example, ‘never’ and ‘always’). 6-8 • Ability of verbal report with limited non-abstract constructs, such as mobility, with assistance. • Tendency for responses to be on the extreme ends of scales. • Increased ability to maintain attention but ability depends on the extend that the measure is engaging, (e.g., includes pictorial representations.) • Increased ability to recall events and accurately report timing of events from several weeks ago. 8-11 • Ability of verbal report with increasingly abstract constructs such as well-being and mood. Assistance is less likely. • Ability to distinguish between different causes of different diseases. ≥ 12 • Ability to report independently. Although it is necessary to summarise ability to report by age range so that practical recommendations can be provided for the design of measures, it is important to note that these age groups will not necessarily fit all PROMs nor the developmental stage of every child.340 It is advised that decisions about specific age ranges are made for each measure depending on what it intends to assess, its format and the characteristics of the population it is intended for.341 It is recommended to use pictorial formats with younger children (younger than 12 years) for representing health-related terms to improve understanding and maintain 133 engagement and attention.268, 340 A relatively recent (2015) systematic reviewu of generic PROMs for children reports that most questionnaires used Likert-style scales for response options and some (11 out of 35) used illustrations (e.g., facial expressions) or cartoons instead of text and tick boxes to illustrate questions.340 Recommendations for developing self-reported health outcome measures for CYP suggest that the use of face scales – such as the Wong-Baker faces scale – show better psychometric properties than analogue or Likert scales.345 Yet, cognitive interview studies show that children over the age 8 are able to use and understand 4 or 5 point Likert scales, and ability to use either Likert or face scales does not differ between healthy CYP or CYP who are unwell.345 The use of an animation format measure on a touch screen device, such as a tablet, instead of a paper text-based measure has also been suggested to enhance understanding among 4-11 year olds.346 Vocabulary used in measures also needs to be age appropriate to enhance understanding.341 In addition, the length of sentences within the measure needs careful consideration as ability to maintain attention increases with age.341 Finally, as the ability to recall past events is also age-dependent, it is not possible to have one ‘optimal’ recall period for all ages and this period will need to be specific to the age group as well as relevant to the purpose of the measure.341 u Review of PROMS since 1992 134 For ages below 8 years, child-specific health measures are usually completed by parent proxies as the child may not be cognitively capable to complete the instrument independently or even with assistance.268, 341, 345, 347 It is also likely that some children may not be capable of completing a measure for other reasons not related to their developmental stage, for example if they are too ill.341 As with proxy responses for adults, it has been suggested that parent proxy responses for children will be more reliable for observable symptoms than responses for subjective symptoms such as mood.268, 341 When a child measure is completed by a parent proxy it has been suggested that the content of the measure differs from the child version to assess dimensions that are important to parents.341, 347 This is so that parents can respond more accurately. 3.4.1.2 Changes in development, needs and context across childhood. The context and dimensions used in a child measure should not be a replication or a rewording of concepts used in an adult measure, because it is important that the measure is relevant to the population and age group it is intended for.341 Child development (physical and cognitive), health needs, and as a result important QOL dimensions, vary across childhood.268, 340 For example, the health needs of an infant will be very different from the needs of an adolescent. Thus, when measuring outcomes in child health the multiple population sub-groups within childhood should 135 be acknowledged along with the unique developmental and health needs each sub-group experiences.268, 341 One of the challenges is to identify important life stages across childhood that may differ not only in terms of age and cognitive development but also in terms of the context – for example pre-school stage – and experiences of children at a particular life stage.253 This can be achieved by conducting empirical work directly with children of different age groups to identify the dimensions that are important to a specific age group and life stage.342 For instance, an empirical study was conducted with children age 7-11 years old to develop dimensions for a preference-based measure of HRQOL for children.348 The study reported that the dimensions identified in that age group were common and concluded that a measure could be developed for that age group as a whole.348 The adolescence stage (12 years old and over) is particularly challenging in capturing what is important to this age group. Self-report measures for adolescents may not face the psychometric challenges evident in younger children, as ability to self-report in adolescents is usually adequate. However, the main challenge is for the measure to capture content relevant to this age group.341 It has been suggested that the emotional and social traits of the adolescent age group are unique and QOL dimensions that are important to adolescents – such as freedom and identity 349 - are more complex than those evident in ages below age 12 years.341, 349, 350 Still, other QOL dimensions that are important to adolescent age groups – such as peer relations and education349 – may 136 not be as important for adults and thus adult measures will not be appropriate either.341 Important dimensions within the adolescent age group may also be diverse as this age range is quite wide.341 For example, late adolescence and early adulthood (18-25 years old) may also include dimensions which are important particularly for the adolescent age group – such as independence and relationships.351 It is also important to consider cross-cultural issues when identifying important dimensions for a measure for CYP. These includes differences in educational systems and literacy levels for a specific age group across countries, differences around the type of information provided to a child about their condition and treatment and the typical activities of a child in different countries.341 The measure will need to be reassessed in terms of language, dimensions, instructions and pictorial representations to be administered to different population groups.341 3.4.2. QALYs in children QALY valuation is challenging in children for a number of reasons. Children develop rapidly and at different rates, and as a result it may be difficult to distinguish between changes due to interventions and changes due to expected development.268, 352 This issue may be managed by developing dimensions informed by age-related changes.352, 353 A methodological challenge is in using valuation methods used in adults, such as SG and TTO, with children to estimate QALYs for child measures. These methods may be especially challenging with young children, because the ability to grasp the concept of 137 time, health-related concepts and understanding of risk will be limited.268, 341, 352, 353 With adolescents there is another challenge with using SG and TTO methods, such as the tendency to prefer higher risk choices to adults.353 A suggestion is to use simpler valuation tasks, such as DCEs that young children may find easier to complete.327 As a result, valuation tasks for child measures and especially for the very young children (under 11 years old) are usually valued by proxy adult respondents, such as the parents.352 However, valuation done with proxy respondents such as the parents may result in potential bias.353 It has been suggested that a way to minimise proxy bias may be by taking a family approach, combining utilities from multiple family members and incorporating effects of child outcomes on the family.327, 353 There are also normative questions on who should value a child measure. Although child patients are experts by experience, it has been argued that since they are not yet financially contributing to the economy it would be preferable for those who do contribute, such as adult tax-payers in the general public, to value child measures instead.3, 353 The justification behind this view is that as the health care system is publicly funded, then the general public should be the ones who decide how health care costs and benefits are allocated.164 Furthermore, the extent to which a child’s ability to make informed decisions about their care and wellbeing, is sufficient, has been questioned.353, 354 138 3.4.3. Important outcomes may not be limited to child outcomes Children are more dependent on their caregiver – usually family or other adult care provider – than adults. Children will gradually become less dependent as they grow older, but they remain dependent on adult caregivers for a prolonged period of time. As a result, this will often have a significant effect on the QOL of those caring for them.268, 327, 352, 354 It has been suggested that because there is evidence of inter-dependency, it may be appropriate to take a societal perspective when measuring outcomes in child health by viewing the family as a unit and including family or household outcomes for all those significantly affected by an intervention.268, 327, 352-354 This could be achieved by developing separate measures for children and other family members or to capture inter-dependency dimensions within a single measure.268, 322 HRQOL measures such as the EQ-5D, have been used to measure health related effects of interventions directed at child patients, on parents and family carers.355 The CarerQol has also been suggested as a potential measure to assess CRQOL of family carers of CYP.270 As in the adult EOL context, there is the question of how to incorporate outcomes for family carers and those close to CYP and obtain relative weights across the different stakeholders within economic evaluation.270 While empirical work has been conducted on relative weights in the EOL context,326 there is no current work conducted on 139 obtaining relative weights in an EOL context for CYP. Therefore, if outcomes for those close to CYP are considered within economic evaluation, work will also need to be conducted to ascertain the relative weights between outcomes for CYP at the EOL and outcomes for those close to them. There is also the question of how many family members should be considered in economic evaluation and for children this may be particularly challenging, as important outcomes may be extended beyond the parents and siblings, to include extended family, close friends or others in the community such as teachers.304, 356 3.5. Measuring outcomes for use in economic evaluation of EOLC in CYP 3.5.1. Challenges specific to measuring outcomes in EOLC in CYP There are some common challenges to consider in measuring outcomes for adult and CYP patients receiving EOLC. The nature of EOLC means that both patient groups often depend on families for their care and thus a family approach to measuring outcomes has been suggested.291, 357 Furthermore, both patient groups and particularly those at the later EOL stage and young children may have limited cognitive and physical abilities 140 to report and value outcomes, thus the design and format of the measure and valuation methods used need to be carefully considered.291 Yet, measuring outcomes in EOLC for CYP involves some additional challenges specific to this patient group. Whilst it is well understood that CYP’s cognitive abilities are shaped by the developmental stage they are at; it is less clear how their experience of a LLC/LTC and EOL further influences their understanding of QOL related concepts and their preferences regarding the care they receive.358 The longer unpredictable trajectories of CYP in PPC and EOLC and broad heterogeneity in conditions – compared to those of adults – may influence what is important to CYP and their families in health care provision.357, 359 There are also ethical challenges in relation to recruiting CYP participants at the EOL stage in research, to either develop or validate measures, given that this population is particularly vulnerable and the EOLC topic is one of the most sensitive ones in health care.3, 357, 360 These ethical issues may have contributed to the fact that there is a lack of research in measuring outcomes for this population. For instance, when CHU9D was developed through qualitative work with children the researchers did not include children with very poor health or who were in hospital as they felt that it would be unethical to do so.3, 348 Still, decisions to exclude the population that the measure is intended for based on the assumption that it would not be appropriate or practically feasible for the population to be involved in research because they are vulnerable, is not thought to be an adequate justification for excluding them.361 Rather than 141 automatically exclude the population from being involved in research, it may be preferable to make reasonable adjustments in the way data are collected from research participants – taking into account the sensitive topic and the extent to which collection of data is cognitively demanding for research participants – to facilitate involvement from the most vulnerable.361 Most existing PC outcome measures have been developed for the adult population and there are increasing calls for the need to develop measures specific to CYP that use age-appropriate language and concepts, and that better capture QOL dimensions relevant to this population.3, 291, 357, 358 There is a child version of POS called the African C-POS that can be used in clinical practice, but this measure has been developed for PPC in sub-Saharan Africa 362, 363 and thus not all dimensions included in the measure may be relevant to settings of higher income countries or to CYP at the EOL phase.357 Ongoing work undertaken by the C-POS team aims to develop and validate another C-POS child version of POS, which will be relevant to the UK setting.364 As with the African C-POS, following development the UK version will aim to be used in clinical practice in PPC.364 3.5.2. Measures typically used in PPC and EOLC in CYP A HRQOL paediatric measure commonly used in PPC population is the Paediatric Quality of Life 4.0 Generic Score Scales (PedsQL).3, 360 However, a study assessing the 142 measure’s appropriateness in EOLC in terms of its psychometric properties found that the measure lacked valid psychometric properties to measure HRQOL dimensions important to this population.365 There is a range of other generic preference-based measures for CYP such as the Child Health Utility 9D (CHU9D),348 the EQ-5D-Y,366 the AQOL-6D 367 and the Adolescent Health Utility measure (AHUM).368 Yet, these have not been developed for specific use in the CYP population at the EOL stage, nor have they yet been validated in this population and thus they may not capture all the dimensions that are important in this specific context.3, 369 Furthermore, the above measures focus on HRQOL outcomes and may fail to capture other important QOL outcomes in EOL which are not health related. A relatively recent (2016) systematic review of the psychometric properties of HRQOL outcome measures for use in PPC concluded that there is currently no ‘ideal’ measure for use in PPC as not all dimensions in measures are relevant to CYP receiving PPC and their families.369 Suggested solutions were to either adapt an existing measure or to develop a more patient-centred measure that can better capture patient experience.369 Another, more recent systematic review (2019) found that information on the psychometric properties of existing measures used in PPC was limited, there is lack of consensus on which dimensions are important to measure and that CYP are usually not involved in reporting outcomes.359 The review (2019) suggested involving children in developing measures and taking a family centred approach so that support to family is 143 also considered and QoL for the whole family is addressed when developing measures.359 As suggested in section 3.3.3, a capability measure has been developed, ICECAP-CPM, which captures broader outcomes for those close to the person at the EOL for use in economic evaluation.301 Given the family centred approach to measuring outcomes in PPC suggested,359 it may be appropriate to assess use of this existing measure with those close to young people at the EOL. It is important to note that neither of the existing reviews 359, 369 focused specifically on measures for economic evaluation purposes but rather for use in clinical practice to improve the quality and effectiveness of care. Nevertheless, the focus of both reviews was on the psychometric properties of measures looking at the HRQOL359, 369 which is relevant to – and important for – measures for use in economic evaluation. 3.6. Methodological challenges in measuring capabilities in economic evaluation There are also methodological and practical challenges to consider in applying the capability approach to measuring outcomes in economic evaluation in health and social care. A consensus has not yet been reached with regards to the most appropriate way to use capability measures in economic evaluation of health and social care.221, 370, 371 A recent systematic review of capability measures (2019) found that 14 capability 144 measures have been developed to be used in economic evaluation of health care interventions but the approaches used in measuring and valuation varies between the different measures.221 CCA was used with some of the measures whilst with other measures the incremental cost-effectiveness ratio (ICER) method was used.221 The measures also differ in terms of interval properties. For example, some are anchored on death and full capability (ASCOT)372 whilst others on no capability and full capability (ICECAP-O).217 Furthermore, these measures were developed with and for different populations, such as older adults, or interventions, such as social care.221 Comparing the measures is challenging without being able to apply a common framework.221 One suggestion is to use a capability-adjusted life years (CALYs) framework, similar to the QALYs framework but with a capability equivalent of ‘complete capability’ and ‘death’.221, 371 According to Månsdotter et al., the difference between QALYs and CALYs is that whilst QALYs focuses on maximisation, CALYs focuses on a trade-off between efficiency and equity.371 This approach is only suitable however, where the capability measure includes death as an anchor. For the ICECAP measures which do not include death as an anchor, Mitchell et al., suggest the ‘sufficient capability’ approach which focuses on those in capability deprivation in a health setting and defines a minimum threshold of sufficient capability across time.221, 370 In this way the focus is on individuals reaching a minimum acceptable level of capability, after which further allocation of resources is no longer a priority, rather than focusing on maximisation.370 145 3.7. Measuring outcomes for use in economic evaluation of care at the end-of-life in young people and those close to them 3.7.1. Measuring outcomes at the end-of-life for young people A review of the literature outlined in Chapter 3 suggests that there is a need to develop measure(s) for CYP for use in economic evaluation of care at the EOL. Due to the different developmental and life stages within childhood it may be necessary to develop measure(s) that are appropriate for the different age groups of CYP and relevant to the particular life stage of CYP. As outlined in Chapter 1, there is a particular need to measure outcomes for young people going through transition to adult services (aged 14 to 25 years old) which is the age group that the empirical work will focus on. In order to decide whether it was necessary to conduct primary research to explore broader outcomes that are important to young people at the EOL, a review of existing empirical studies was conducted which is summarised below. 146 3.7.1.1 Summary of systematic review – What is important to young people at the EOL 3.7.1.1.1 Purpose of the review A systematic review was conducted of empirical studies that have used qualitative and mixed methods to explore what is important to young people (age 14-25 years old) with LTCs/LLCs at the EOL. 3.7.1.1.2 Objectives of the review The review had two objectives: (i) to explore the extent of empirical studies on what matters to young people with LTCs/LLCs at the EOL; ii) to identify any knowledge gaps that exist in current empirical studies. 3.7.1.1.3 Rationale of the review A recent (2019) systematic review was conducted on HRQOL outcomes in CYP (0-23 years old) with LLCs/LTCs and included evidence from quantitative, qualitative, and mixed method studies.373 The wide range of methods used in the studies included in the review meant that the authors could not assess the extent of concerns that mattered to the study participants.373 The study suggested that further qualitative work is necessary to capture what matters to this population and highlighted the need to increase involvement of CYP in research which was limited in the studies reviewed.373 147 The literature search for the above review was conducted in 2017,373 the topic of focus was specific to HRQOL outcomes and the review did not include CYP perspectives at the EOL.373, 374 The systematic review conducted in this thesis aimed to conduct a more up to date search of the literature (March 2023), to explore broader experiences from the perspective of CYP, rather than experiences on a pre-defined topic such as PPC provision or HRQOL outcomes. 3.7.1.1.4 Summary of the review methods A summary of the key steps of the review are outlined in this section with details of the full systematic review methods in Appendix 1. To help structure the systematic search, the PICOSv criteria recommended by the PRISMA 2020 statement checklist 375 were used but the PICOS framework was adapted to include elements relevant to this review. The inclusion criteria are outlined beloww: i) An empirical study with findings that involve both data collection and analysis using qualitative methods or mixed methods ii) Study includes the voice of young person: v Population, Intervention, Comparator, Outcomes, Study Design w Appendix 1 includes further details in relation to why the below study inclusion criteria were decided for the systematic review undertaken. 148 - through young person participants, age 14-25 years old, with any LLC or LTC; including studies focused on one specific condition (e.g., cancer) or group of LLCs/LTCs or - through family member or friends as proxy participants for those young people who are unable to participate directly because a) they do not have capacity or are too unwell or b) have died iii) Data collection is conducted in a high-income country Studies were excluded if they were: i) Quantitative studies ii) Published before 1997 (The definition of PPC was first documented in 199717. iii) Published in language other than English iv) Studies with focus on experience of specific intervention or treatment e.g., chemotherapy; and/or evaluation or experiences of specific service e.g., respite service v) Research protocols or conference abstracts Studies published in OVID Medline, Embase, PsycInfo, AMED, CINAHL and Web of Science core collection and ProQuest dissertations and theses global were considered. The screening process included two stages i) title and abstract screening ii) full text 149 screening. A proportion of studies were double screened. Citation screening was also undertaken on the studies that were included in the full text stage. The quality of the identified studies was assessed using the CASP tool by two reviewers, and data were extracted using a standard form (see Appendix 1) to capture the key characteristics of each study included in the review. A summary of the search is provided in a flow diagram in Figure 1: 150 Figure 1 - systematic review flow diagram 151 3.7.1.1.5 Characteristics of included studies and narrative synthesis of results The characteristics of the included studies are summarised in this section and a narrative synthesis of the review results is provided. The characteristics of studies were compared to ascertain the extent of empirical work conducted on what matters to young people with LLCs/LTCs at the EOL and identify any knowledge gaps in current studies. Given the diversity of studies, in terms of study aims, participants, settings and data collection method, a narrative synthesis rather than a quantitative synthesis was used.376 Results of studies and relationships between studies, were summarised in tabular form and discussed in textual form below. Twenty-six studiesx were included in the review with publications ranging from 2005 to 2022. A majority of the studies focused on specific countries: UK(n=8) 52, 377-384 and Canada(n=7).385-391 The rest of the studies were from USA (n=3)392-394; Australia (n=2)395, 396; Singapore (n=1)73, 397; Romania (n=1)398; Japan (n=1)399; New Zealand (n=1)400; Taiwan (n=1)401 and one study had an international focus (n=1).383 All 26 studies used a qualitative method, with the majority using inductive coding in their analysis(n=21),52, 378-381, 385-390, 395, 397-401 some using combined inductive and deductive coding (n=4) 377, 392-394 and one used deductive coding (n=1).384 x Reports of the included studies were 27 in total. Two reports of the same study were identified – a journal paper and a PhD thesis. These two reports are presented and discussed as one study in the systematic review. 152 The majority of studies captured experiences in relation to a specific condition (n= 17) 378, 380, 383, 385-392, 394-396, 399, 401 although some studies did cover experiences in relation to a range of LLCs/LTCs (n=9).52, 73, 379, 381, 382, 393, 397, 398, 400 Of the 17 condition specific studies the majority were cancer specific (n=14)380, 383, 385-388, 390, 392, 394-396, 399, 401 but there were also three other condition specific studies capturing experiences of other non-malignant conditions: Duchenne Muscular Dystrophy (DMD) (n=1)389; Cystic Fibrosis (CF) (n=1)378; End Stage Renal Disease (ESRD) (n=1).391 Half of the included studies were health and care needs specific, focusing on what is important for CYP in relation to health and care service provision and health outcomes (n=13) 52, 377, 379, 381-386, 388, 394, 395, 402 while others had a broader focus on what is important overall in life including aspects of life beyond the care and health context(n=13) 73, 378, 380, 387, 389-391, 393, 396, 398-401. Some of the included studies captured experiences at the EOL stage (n=14) 73, 379, 380, 385-388, 390, 391, 393-395, 399, 402 while the rest of the studies focused on experiences prior to the EOL stage (n=12).52, 377, 378, 381-384, 389, 396, 398, 400, 401 Of those studies that had an EOL stage focus the majority (n=11) captured condition specific experiences of which ten (n=10) were cancer specific 380, 385-388, 390, 394, 395, 399, 402 and one was specific to ESRD.391 The rest of the studies with an EOL stage focus captured a range of LLCs/LTCs (n=3).73, 379, 393 The age group of CYP participants (or proxies if CYP participant was unable to take part) included in the studies varied and ranged from 0-39 years old. The terms used in the studies to describe the age group of participants also varied, for instance one study described those between 15 153 to 25 years old as young adults,379 while other studies used the term ‘adolescent and young adults’ (AYAs) to describe a similar age group in their inclusion criteria ranging from 16 to 17 years old (n=5). Some studies focused on the experiences of CYP under or up to 18 years old only (n=12) 23, 73, 377, 384-387, 391, 398-401; others focused on young adults, ranging from 18 years old and up to 39 years old (n=4) 378, 388, 390, 396; while the rest of the studies included experiences of those both over and under 18 years (n=10).52, 379-381, 383, 389, 393-395, 402 Finally, some of the studies included only CYP participants (n=11),378, 383, 384, 387-391, 393, 396, 401 some both CYP and family participants as proxies (n=4),52, 382, 398, 400 some only family member participants (as proxies) (n=3), 385, 386, 395 and some included CYP, family members and HCPs as participants(n=8).73, 377, 379-381, 394, 399, 402 Overall, the availability of studies that had a broad focus on aspects of life that are important to CYP with LLCs/LTCs at the EOL stage is very limited with only two studies fitting these criteria.73, 393 These two studies however do not capture what is important within the UK context (the studies were undertaken in the USA393 and in Singapore).73 Furthermore, the latter study is a case study capturing experiences of five CYP ranging from 18 months to 17 years old through interviews with only proxy participants – that is bereaved parents and HCPs.73 The rest of the studies that have a broad focus on important aspects of life at the EOL stage are either non-malignant condition specific (ESDR)(n=1)391 or cancer specific (n=4)380, 387, 390, 399 with one of the cancer specific studies being a single case study with only one participant.387 The relationships 154 between the included studies discussed in this section are summarised in Table 2 below in tabular form. Further details of the characteristics of each study included in the review are summarised in Appendix 1. Table 2 - Relationships between studies included in the review. Author(s) and publication date Focus on care needs or health outcomes Focus on broad experiences Captures EOL Not EOL Covers range of LLCs/LTCs Condition specific Includes CYP participants Age-range of CYP (years) Ananth et al. (2021)392 17-23 Barling et al. (2014)403 13-23 Barrera et al. (2005)385 7-15 Cataudella et al. (2012)386 1–5; 8–11; 12–19 Chien et al. (2020)401 11-17 Chong (2019) and Chong et al. (2021)73, 397 1.5-17 Ciobanu et al. (2021)398 14-17 Coad et al. (2015)52 0-25 156 Coombes et al. (2022)377 0-17 Evan et al. (2012)393 9-21 Flavelle et al. (2011)387 15a Gaab et al. (2013)400 9-18 Garland et al. (2020)388 21-39 Gibson et al. (2014)389 16-27 Higham et al. (2013)378 18-29 Ito et al. (2015)399 3-16 Johnston et al. (2016)379 17-23 a Single case study 157 Kenten et al. (2019)380 16-40 Kirk et al. (2014)381 16-31 Knox et al. (2017)390 18-35 Mack et al. (2021)394 12-39 Mitchell et al. (2020)382 5-18 Nicholas et al. (2011)391 7-18 Patterson et al . (2012)396 20-25 Sodergren et al. (2018)383 14-25 Taylor et al. (2021)384 12-18 The main themes from the findings from each study were also extracted and summarised under the broad themes presented in Table 3 (see Appendix 1 for full findings of each study). ‘Physical and mental health’ was the most frequently identified theme (21 out of 26 studies) whilst the themes of ‘privacy’ (2 out of 26 studies), ‘future goals’ (3 out of 26) and ‘financial security’ (3 out of 26 studies) were the three least identified themes in the studies included. The two studies identified as capturing broad experiences at the EOL stage for CYP with a range of LLCs/LTCs 73, 393 both included the themes of ‘physical and mental health’ and ‘relationships’ while the rest of the themes identified were different with ‘Independence’ and ‘quality of health care services’ captured in Chong et al.73 and ‘activities and education’, ‘normalcy’ and ‘privacy’ in Evan et al..393 159 Table 3 - Main themes identified in included studies Author(s) Physical/mental health Psycho-social/spiritual Activities/Education Independence Relationships Identity Normalcy Privacy Future goals Quality of and access to care Financial security Ananth et al. (2021)392 Barling et al. (2014)403 Barrera et al. (2005)385 Cataudella et al. (2012)386 Chien et al. (2020)401 160 Chong (2019 - PhD thesis) and Chong et al. (2021 - journal article)73, 397 Ciobanu et al. (2021)398 Coad et al. (2015)52 Coombes et al. (2022)377 Evan et al. (2012)393 Flavelle et al. (2011)387 Gaab et al. (2013)400 Garland et al. (2020)388 161 Gibson et al. (2014)389 Higham et al. (2013)378 Ito et al. (2015)399 Johnston et al. (2016)379 Kenten et al. (2019)380 Kirk et al. (2014)381 Knox et al. (2017)390 Mack et al. (2021)394 Mitchell et al. (2020)382 Nicholas et al. (2011)391 162 Patterson et al . (2012)396 Sodergren et al. (2018)383 Taylor et al. (2021)384 163 3.7.1.1.6 Conclusion of the review Existing studies capturing broad experiences of young people at the EOL is limited and most existing studies are cancer specific. The conclusion of the review was that none of the existing studies identified in the review could be used to inform the development of a capability measure for young people at the EOL and primary research was necessary to capture views relevant in this context. 164 3.7.2. Measuring outcomes at the end-of-life for those close to young people Given that outcomes of care at the EOL go beyond the person at the EOL, there is need to measure outcomes for those close to them. As mentioned in Chapter 3 (section 3.3.1 and 3.3.3) a capability measure has been developed, ICECAP-CPM, to capture broader outcomes for those close to the dying. ICECAP-CPM may be appropriate to measure outcomes for those close to young people at the EOL. 3.7.2.1 Systematic review – Psychometric characteristics of ICECAP-CPM 3.7.2.1.1 Purpose of the review Given that ICECAPCPM may be appropriate to capture outcomes for those close to young people at the EOL a review aimed to assess current information on the psychometric characteristics of ICECAP-CPM. 3.7.2.1.2 Objective of the review A systematic review of the literature was conducted to explore whether any existing work has been undertaken to assess the psychometric properties of ICECAP-CPM. If 165 such work was identified, a secondary objective of the review would be to examine whether this work is relevant to the context of care at the EOL for young people. 3.7.2.1.3 Rationale of the review A recent systematic review was conducted (2020) to assess evidence on the development, psychometric testing, and valuation of capability measures for use in economic evaluation.404 Although the review identified a study in relation to the development of ICECAP-CPM, there was no availability of evidence on the characteristics of ICECAP-CPM for use in economic evaluation.404 Given that ICECAP-CPM was only recently developed (2016),290 a more up to date systematic search looking for studies specific to ICECAP-CPM may potentially provide new evidence relevant to psychometric testing of the measure. 3.7.2.1.4 Review methods Given that in the existing systematic review to assess work conducted on capability measures404 there were no relevant studies identified on psychometric testing for ICECAP-CPM, the search terms used for this review were deliberately broad so not to miss any potentially relevant studies. Whilst the search aimed to be as inclusive as possible it also had to be feasible and so search terms such as ‘capability’ or ‘ICECAP’ were not used as they would yield a large number of studies that may not be relevant.405 166 A search of the following electronic databases was conducted using the search terms ‘ICECAP-CPM’ and ‘ICECAP-close person measure’: Medline, Embase, CINHAL, PsycInfo, Scopus, Pubmed and ProQuest Dissertations and thesis Global. The Boolean term ‘OR’ was used to combine the two search terms. A recent systematic review on existing work conducted for another capability measure (ICECAP-Oa) suggested that many studies which reported on psychometric testing on the measure did not actually refer to the measure in the title and abstract.405 Therefore, the ICECAP-CPM search was applied to ‘all fields’ or ‘full text’, rather than being limited to ‘title, abstract, keyword’ to keep the search broad. Studies published from 2016 when ICECAP-CPM was developed, to March 2023 were included in the search. Studies published in a language other than English, research protocols and conference abstracts were excluded. There were no other exclusion criteria in the search. 3.7.2.1.5 Results The search resulted in 26 studies being identified from all electronic databases. Duplicate records of studies were removed, and the remaining 22 studies were screened. All of the remaining studies were excluded on title and abstract. a ICECAP-O and ICECAP-CPM are both from the same family of ICECAP capability measures developed for use in economic evaluation of health and care interventions. 167 3.7.2.1.6 Conclusion of the review No studies were included in the review and there is no known evidence on the psychometric properties of the ICECAP-CPM. It is therefore necessary to conduct primary research for an initial assessment of the measure’s psychometric properties and its use in the context of those close to young people at the EOL. 3.8. Conclusion Chapter 3 has highlighted challenges in measuring outcomes for use in economic evaluation of care at the EOL for CYP. Important outcomes at the EOL for CYP may not be limited to health and thus a broader evaluative space may be necessary within economic evaluation. Arguments were also made to include those close to the dying within economic evaluation of EOLC for CYP and challenges in relation to how to incorporate outcomes for those close to CYP within economic evaluation were highlighted. As there is currently no measure developed or validated for measuring outcomes for young people at the EOL, in Chapter 3 the literature was reviewed for existing work conducted with young people at the EOL that could inform the development of attributes for a new measure for young people. As evidence from existing work is limited, primary research is necessary to assess important outcomes at the EOL for young people. 168 The literature was also reviewed to assess whether any work has been conducted on the psychometric properties of ICECAP-CPM and if so, whether it has been used in the context of EOL for those close to young people. As evidence of existing work is unknown, an initial assessment of use of ICECAP-CPM in this context can be conducted. Given that the empirical work in this thesis considers the incorporation of outcomes for those close to young people within economic evaluation then empirical work will also need to be conducted to obtain relative weights between outcomes for young people and outcomes for those close to them that can be used in economic evaluation of care at the EOL. The remainder of this thesis will focus on the empirical work conducted to assess outcomes of care at the EOL for young people and those close to them. The empirical work first focuses on outcomes for young people and then moves on to focus on outcomes for those close to them. The final part of the work is focused on bringing the outcomes of both stakeholders together to assess how outcomes beyond the young person at the EOL can be incorporated within economic evaluation. The three primary aims of this thesis are: 1. To develop attributes for a capability measure for young people, 14-25 years old, at the EOL. 169 2. To assess feasibility of use and content validity of ICECAP-CPM with those close to young people, 14-25 years old, at the EOL. 3. To estimate relative weights between outcomes for young people, 14-25 years old, at the EOL and outcomes for those close to them. 170 171 CHAPTER 4. QUALITATIVE METHODOLOGY AND METHODS 4.1. Introduction Chapter 4 describes the methodology and use of methods for addressing the first two aims of the research, both of which use qualitative methods. As suggested in Chapter 3 there is need for a measure to be developed to assess outcomes for young people at the EOL that can be used in economic evaluation. Given the family approach to EOLC outlined in Chapter 1, to support both the young person and those close to them, it is important that assessment of outcomes includes those close to the person at the EOL. In Chapter 2, the capability approach was explored as an alternative evaluative space in economic evaluation of health care, that could capture broader outcomes of interventions at the EOL. This chapter starts with the use of qualitative methodology in developing measures for use in economic evaluation of health care. Qualitative methodology used to assess feasibility of use and content validity of measures is then discussed, followed by guidance on conducting qualitative research with a vulnerable population, on a sensitive topic and in the context of COVID-19. The qualitative methods used in the 172 study are then described including all stages from recruitment through to analysis and ethical considerations. 4.2. Methods for identifying attributes in measures for use in economic evaluation. 4.2.1. Identifying conceptual attributes The development of a measure for use in economic evaluation consists of several stages. The first step is to identify the conceptual attributes that will inform the questions within the measure. Different approaches have been used to develop dimensions 287 with most having been developed with the use of a ‘top down’ approach.406 A ‘top down’ approach involves adapting the relevant content of existing measures or using literature, health surveys or expert opinion to develop the measure dimensions. Generic measures, such as EQ-5D, SF-6D and HUI have been developed using a ‘top down’ approach406 and EQ-5D-Y (child measure) was adapted from EQ-5D (adult measure).366 Adapting a measure may fail to capture important dimensions for the population it is adapted for.287 Furthermore, if there is not sufficient literature available on the topic under research then using literature to develop dimensions may 173 be inadequate in capturing key attributes that are important to the relevant population.407 The use of a ‘bottom up’ approach in developing measures is growing.281 A ‘bottom up’ method involves using qualitative methods to include the views of relevant patients or proxies when developing a measure.293 Although traditionally qualitative methods have not been used in health economics, the application of these methods within a ‘bottom up’ approach to developing measures, and conceptual attributes in particular, is rising.281, 407, 408 Examples of such measures are the CES,318 ICECAP-A,218 ICECAP-O,217 ICECAP-SCM,293 ICECAP-CPM,290 CHU-9D,409 WAItE,410 and CP QOL-Teen.411 An inductive approach to attribute development assists in developing a measure that is less reliant on the views of the researcher or ‘experts’ and instead focuses on the views of the population that the measure is intended for.406 Involving the relevant population in measure development may improve the face and content validity of the measure as it will be more relevant to the population being studied.281, 408, 412 As there is currently no preference-based measure developed for the population studied in this project and given the uniqueness of the EOL experience in young people, the possibility of adapting an existing measure is limited. As discussed in Chapter 3, the possibilities for using existing literature are also limited because of the lack of such literature. The use of qualitative - rather than quantitative - methods to develop attributes can offer ‘richer’ and more informative data, especially on sensitive topics such as the EOL, as research participants are given space to express their views. 174 290, 406, 409 For these reasons, the empirical work in this study will use a ‘bottom up’ approach and qualitative methods to develop the conceptual attributes that will inform a new measure for young people at the EOL. 4.2.2. Guidance on the use of qualitative methods to develop conceptual attributes for a measure. The qualitative methods used to develop the measure attributes were informed by general guidance on the use of qualitative methods for the development of attributes for PROMs. The Professional Society for Health Economics and Outcomes Research (ISPOR) good research practice guidelines on paediatric PROMs recommend conducting qualitative research, including CYP in research to assess the content validity of a measure.341 Similarly, MORECare guidance on evaluating interventions in EOLC and Medical Research Council (MRC) guidance both encourage the use of qualitative data to improve understanding of experiences of patients receiving EOLC.293 There is limited guidance on the use of qualitative research methods on attribute development for the purpose of economic evaluation of health care.290, 408 The empirical study drew on two papers 281, 407 in developing the design of the work. Coast et al. suggest that two different phases in developing attributes should be considered.407 The first phase involves developing the attributes that will inform the measure407. The second phase involves checking the content and coverage of 175 attributes developed in the first phase and then generating wording to express the attributes in a way that is meaningful to the population that the measure is intended for.407 Coast et al. suggest it is helpful to distinguish the two phases, which can run concurrently, in terms of their role in the data collection.407 Each of these two phases involve further stages. Stevens et al. suggest that there are 5 stages in qualitative research design for the development of a measure for use in economic evaluation281: 1. Decision on who the research participants should be 2. Data collection method 3. Study design 4. Data analysis 5. Development of the descriptive system The five stages are outlined in more detail below: 4.2.2.1. Research participants The use of a ‘bottom up’ approach required consideration about who would be the most relevant population to include in the empirical study. To optimise the content validity of the measure, the choice of attributes included in the measure should be relevant to the population that the measure is intended for.412, 413 This can be achieved by collecting data to develop the measure dimensions from the population with lived 176 experience and first-hand knowledge on the topic of interest.281, 412 Carers and family can also be considered as relevant populations for measure development.281 Then the sampling frame needs to be decided. Whilst in quantitative research one of the sampling aims is representativeness, the large amounts of data generated from qualitative research often do not allow collecting a representative sample due to the amount of time and resources involved in acquiring a large sample.414, 415 Yet, sampling methods used in qualitative research to develop a measure aim to include representative experiences and views that are most important to the study aims and objectives.413, 414 The use of purposeful sampling – a selection of participants that is intentional and based upon specific inclusion criteria – can assist in including study participants with a range of characteristics and insights.413-415 In qualitative research there are a number of different purposeful sampling approaches such as maximum variation and purposeful random sampling from which a researcher could choose based on the research questions the study aims to answer.412, 414, 416 For instance, in maximum variation strategy the researcher purposefully chooses a wide range of cases to (a) include diversity and a detailed description of each case and (b) identify emerging patterns that are common and important across different cases.416 The latter is of particular value in capturing the shared views of study participants. It is suggested that for maximum variation in a small sample the researcher should initially identify diverse 177 criteria for the sample, e.g. geographical variation, and then look for common themes emerging across the heterogeneous sample.416 The choice of sampling criteria for the development of generic measures can be more challenging than for the development of specific measures. For the development of a generic measure, the sampling criteria need to cover a wide range of diagnostic conditions and achieve an in-depth exploration of experiences within different conditions.281 It is not always feasible to have a sample that covers all diagnostic groups and if the sampling criteria fail to include a number of different diagnostic groups, the measure dimensions may only be relevant to the groups included in the study.281 Stevens et al. suggest sampling from the general population and including a wide range of general health levels.281 The use of a theoretical sampling approach – where the choice of sampling criteria is guided by the emerging theory in the empirical work and revised as data collection progresses – to provide a diverse sample has also been suggested in health economics research.414 This approach is useful in capturing perspectives that are important in a particular study context.414 Owen-Smith et al. suggest that a number of different sampling approaches – such as snowballing and maximum variation sampling – can be used and combined as the data collection and analysis progresses and theoretical understanding of the topic under study evolves.414 A flexible approach to sampling may be particularly useful when recruiting participants who are from ‘seldom heard’ groups or if the study topic is 178 sensitive, because this can achieve greater participant numbers via a number of different channels.417, 418 4.2.2.2 Data collection method The second step outlined by Stevens et al. involves choosing the most appropriate data collection method to develop the measure. Interviewing has been previously used in developing measures for economic evaluation purposes.218, 290, 318, 348, 407, 419 Using in-depth interviews can assist in exploring what matters to the research participant from their viewpoint and allows them to express their views in their own terms and words.334, 416 This is useful in developing attributes, particularly if the topic under study is new to the researcher.420 Semi-structured interviews, which allow the researcher to focus on a more specific study aim, can be useful in developing the terminology used in the measure.290, 420 An alternative method often used in qualitative data collection is focus groups.407, 415 The use of a focus group allows group interaction and it may be helpful in exploring a variety of perspectives on a particular topic.334, 421 However, the focus group environment can also be challenging for participants who may not feel comfortable expressing their own views on the topic,334 especially if the topic of discussion is sensitive.407 It is also important to ascertain the sample size for the research.412 However, in qualitative research it is difficult to know in advance what would be an adequate 179 sample size.334, 415 The term ‘saturation’ is often used in qualitative studies as a criterion for sample adequacy and it is associated with the use of grounded theory.422 During data analysis when no new emerging themes occur then there is less value in collecting further data to contribute to the study findings.413 Even though there is a possibility that new themes may emerge from further data collection, when it is thought that further research will not produce new findings that are significant to the topic under study then the sample size can be thought of as adequate and that it has reached ‘saturation’.334, 412, 414 It has, however, been suggested that ‘saturation’ should not be used as a criterion in all qualitative studies as they may not all follow the same analytic procedures with their data.423 Malterud et al. propose the use of the term ‘information power’ as a guide in deciding the adequacy of the sample size in qualitative research; the larger the information power that the sample holds, the lower the number of study participants is needed.423 Malterud et al. suggest five items that affect the ‘information power’ of the sample423: 1. Study aims: the broader the aims the larger the sample size. 2. Specificity of experiences among study participants: the more specific the characteristics of participants are for the study aim, the less the number of participants needed to achieve sufficient information power. 3. Theoretical background: the higher the level of theoretical background used in the study the lower the sample size necessary. 180 4. Quality of dialogue: The analytic value of the data depends on the quality of communication between the researcher and the participant. The higher the analytic value of the data the lower the sample size necessary. 5. Analysis approach: For instance, a study using in-depth analysis of discourses will require a lower number of participants compared to a study using cross-case analysis. Hollin et al. suggest that in qualitative studies for measure development a smaller sample compared to other qualitative studies may be adequate due to the limited purpose of the study.412 Existing qualitative studies conducted to develop measures for use in economic evaluation 218, 290, 293, 348, 419, 424 have generally used samples of between 20 and 35 research participants to develop the conceptual attributes for the measure and a subsequent sample of 15 to 20 research participants to check the content and coverage of the attributes developed in the initial study phase. 4.2.2.3. Study design The design of the qualitative interview will depend on the type of interview – semi-structured or in-depth.415 The use of a topic guide in semi-structured interviews, with topics from existing literature can guide the interviewee to focus on the topic under study.334, 425 It is important to ensure that the topic guide includes all the relevant topics and that as data collection and analysis progress, the topics covered may be modified from emerging themes.334, 426 The researcher may choose to explore the topic 181 by conducting an unstructured interview without the use of existing questions. Conducting an interview without the use of a topic guide could allow the interviewees to express what is important to them; in this way a particular agenda with predetermined questions imposed on the interviewees could be avoided.281 Decisions about the level of interview structure – i.e., in-depth, unstructured, or semi-structured – are influenced by the methodology used and the research aims.334, 421 The researcher will also need to consider the setting that the interview takes place in, as well as the way the interviewer may influence the participant’s responses.334, 415 The setting should allow the participant to feel as comfortable as possible and it should be quiet and private to avoid the risk of interruptions and ensure that the data is collected in a confidential manner.334, 421 It is important that the interviewer carefully considers how to build rapport with the interviewees prior to, and during, data collection – such as the way the interviewer asks questions. It is recommended that the interviewer has undertaken training to develop their interviewing skills.334, 426 The topic of building rapport is explored in more detail later in this chapter, focusing on interviewing on sensitive topics and with young and vulnerable populations. 4.2.2.4. Interview data analysis There are different methods of qualitative data analysis.334, 412 It is suggested that the most appropriate method of analysis may not be necessarily obvious in health economics studies, as qualitative methods within this discipline have not been 182 commonly used.427 Thematic content analysis has been used to develop HRQOL dimensions and levels for a preference based child measure (CHU9D),348 but the constant comparison approach 428 has been more commonly used in health economics to develop measures 218, 290, 348, 424 and attributes for discrete choice experiments.407 The technique of constant comparison associated with the grounded theory approach422 is generally used to generate theory that is ‘grounded’ in the data collected.334 It is an iterative process in which the researcher constantly compares data collected and codes them under specific categories to develop theory emerging from the data.334 In this way data collection and analysis occur in conjunction and the researcher constantly refers back to the data and findings to develop new theory using an inductive approach.334, 422 That is, the theory is not pre-conceived. An inductive approach can be particularly useful in developing measure dimensions that focus on what is important to the participants.413 Strauss and Corbin 428 distinguish three levels in grounded theory coding: open, axial and selective coding. In open coding data are examined, compared and categorised so that the researcher can later group them and develop new categories.428 This is followed by axial coding in which the researcher makes connections between the different categories and links codes.428 The final coding procedure involves selective coding in which the researcher chooses the core categories by relating all the developed categories systematically to integrate them.428 These core categories can be used to form the measure dimensions. 183 4.2.2.5. Development of the descriptive system Following the development of the measure dimensions, a descriptive system and levels needs to be generated.281 The descriptive system is made of the questions relevant to the different dimensions of the measure. Each of these questions have a number of different levels from which respondents can choose, based on their perception of their own health or wellbeing.164 Levels are usually based on severity or frequency of each dimension, for instance ‘ a lot’ and ‘ a little’.164 Levels could have equal weighting, where intervals between the response items are of equal value and importance or items could have different weights where valuation studies are conducted to determine the scoring of each item.164 Where levels are weighted, it is suggested that dimensions should not be more than nine so that the measure is amenable to valuation.281 Qualitative data collected can inform the type level and wording used by participants to describe the dimensions and levels 281 which assists in developing a measure that is meaningful to the population the measure is intended for.407 The use of qualitative data to decide on the type of levels or wording used in the measure has been used in a number of measures for use in economic evaluation such as CHU9D,348 ICECAP-SCM,424 ICECAP- A,218 and ICECAP – CPM.290 184 4.3. Methods to assess content validity and feasibility of use of measures. As outlined in Chapter 3, following development of a measure for use in economic evaluation it is important to assess the performance and psychometric properties of the measure.137 An important criterion for the initial psychometric assessment is feasibility – the extent to which the measure can be used in practice in the relevant context. Feasibility may be assessed by looking at missing data, rates of completion, rates of difficulty reported by participants and time taken to complete the measure. All of these might suggest potential difficulties during completion.286 Another important step in the development of a new measure is to test its content validity – the extent to which the content and dimensions of the measure are relevant and important to the intervention.137, 287 It is also important to test the content validity of an existing measure if the measure is intended to be used within a different context, i.e. cultural, intervention or population group, to assess whether the measure is relevant in the particular context and whether there are topics of interest not captured in the measure.429-431 COSMIN and ISOQOL recommendations for use of PROMS, suggest that the content validity of a measure is a critical measurement property of a PROM and a lack of content validity in a measure affects the rest of the measure’s psychometric properties.430, 432 Thus, assessing the content validity of a measure prior to other measurement properties is important. 185 In some studies, statistical methods have been used to test a measure’s content validity.433 However, the use of qualitative methods, such as cognitive interviewing with patients, has been more widely used and is suggested to be a more appropriate method to assess validity, through direct communication with the people completing the measure.430, 431 In cognitive interviewing, the researcher can use different techniques, such as the think-aloud technique, to assess content validity.434 Focus groups have also been used to assess the content of a measure with relevant population groups,164 but it is suggested that when the topic is sensitive, participants may feel more comfortable in a one-to-one interview rather than a group.431 4.3.1. Use of cognitive interviewing to assess content validity and feasibility of use. 4.3.1.1 Sampling To test the feasibility of use and content of a measure, it is important to have the input of the population group that the measure is intended for as they are the primary experts with regards to the measure.430 The sample of the relevant population group should ideally include participants with a wide range of characteristics, for example, age and socio-economic background, so that the sample is purposefully selected to represent a range of perspectives within the particular population group.431 There are no specific recommendations about adequate sample sizes for cognitive interview studies 430, 431, 435 and sample size varies with participants ranging from as few as six 436 186 to as many as seventy-two.293 In studies on sensitive topics and/or with vulnerable participants, such as on the topic of PC or with older adults, the sample size has often been fewer than fifteen.436-441 4.3.1.2 Design and analysis COSMIN guidelines on the use of qualitative methods to assess the content validity of PROMs suggest that it is not possible to define the most appropriate design and analysis method that should be used.430 Design and analysis may differ if the assessment is conducted as part of the development of a new measure – where the aim may be for instance to elicit further information to finalise the measure dimensions – or as an assessment of validity of an existing measure to examine the extent that the content is relevant and captures the dimensions that are important to a particular group or context.431 In the assessment of an existing measure it is important that the interview includes a section, such as a semi-structured interview, where the researcher asks the participants whether there were topics of interest that were important to them but not captured in the measure.431 Cognitive methods used vary and include think-aloud interviewing or verbal probing techniques such as vignettes and paraphrasing.430, 431, 442-444 A cognitive interviewing method which has been widely used in health research to develop and validate measures 438, 443, 444 and in studies to assess content validity and feasibility of use of capability measures is the think-aloud technique.218, 293, 435, 444, 445 The use of the think-187 aloud technique has the advantage that the interview is driven by the respondent, rather than the researcher asking specific questions.446 During a think-aloud interview participants are asked to verbalise their thoughts as they occur whilst completing a task, such as a measure.434, 447 ‘Think-aloud’ information is then used by the researcher to understand the participant’s thought process and identify any errors or other issues the participant may have encountered during measure completion.447 The theory behind methods used in the social and cognitive psychology disciplines to test survey measures suggests that there are four processes involved in responding to a survey question: comprehension, retrieval, judgement, and response.434, 447, 448 A respondent needs to: comprehend the question as intended (comprehension), retrieve the appropriate information to be able to answer the question (retrieval); make a correct judgement on how to use information to answer the question (judgement); and give a valid response to the question using the available response options (response). These processes are not necessarily performed in sequence and it is possible that each of the four processes may interact with each other when a respondent answers a question.434 In cognitive interviewing studies, these four cognitive processes have been adopted as criteria in a coding framework to identify and rate errors.448 It is recommended that error rating is conducted by at least two reviewers who first conduct the rating independently based on agreed criteria. Ratings are then compared, and rating issues discussed to reach a consensus.430 188 4.4. Guidance on conducting qualitative research with a vulnerable population to explore sensitive topics. Sensitive topics of research include ‘taboo’ subjects, such as addiction, rape, grief and other traumatic events that could make the participant – and sometimes the researcher – feel threatened, ashamed or guilty.449 Thus, planning of research design and data collection methods needs to be carefully thought through to minimise and manage the potential risk involved.450 This is particularly important when research participants are ‘vulnerable’, a term used in research to refer to individuals or groups who may be ‘seldom heard’ and/or their situation involves limited autonomy or marginalisation.450 4.4.1. Research participants and recruitment 4.4.1.1 Palliative and end of life care One of the methodological challenges in conducting research with vulnerable populations, such as those receiving EOLC or their family and friends, is related to recruiting appropriate research participants.3, 292, 332, 451-453 Individuals receiving EOLC may be particularly frail and with complex needs 3, 451 and they are considered to be vulnerable for a number of reasons: they may have a high level of dependency on 189 others to meet their physical care needs; and patterns of illness may change and progress throughout the time the research is conducted.280, 333 Their ability or willingness to participate in research may be negatively affected by their symptoms and condition.292 Also, as their condition progresses or cognitive abilities fluctuate, the researcher may need to re-gain informed consent as the duration of the consent is not certain.280, 333 Bereaved family and friends, are also vulnerable due to the grief experienced.453 Conducting research with a vulnerable population to explore a sensitive topic can be a further barrier in recruiting and obtaining an adequate and appropriate study sample size.332, 333, 452, 453 Flexibility with – and diversification of – recruitment strategies can assist in reaching populations that are seldom heard.414 Recruitment via key informants, such as sponsors or gatekeepers (e.g. HCPs) who facilitate access to the fieldwork, can assist with potential participants who may be worried about participating.334 However, it is suggested that the role of gatekeepers in PC research often suffers from an over-protective culture and HCPs who act as gatekeepers may attempt to control – unintentionally or not – recruitment.292, 451 The use of posters with study information to advertise the study at places that potential participants may have access to, is an alternative strategy.290 Potential participants should be offered a number of alternatives in the way data is collected, for instance through the use of computer mediated or telephone communication, rather than solely face-to-face. Offering participants more options 190 may increase participant recruitment and allows participants residing in different geographical locations to take part.454, 455 Ethical considerations can be challenging. There is potential risk of distress to study participants and the question of whether the benefits of conducting research with a vulnerable population outweigh the risks.450, 452, 453 Evidence from PC studies suggest that participants, including those near the EOL and their families, were consistently willing to participate in research studies and the majority of them found the experience to be positive.292, 452, 456 This was particularly the case for research relevant to care and involving face-to-face interviews.292 Yet, some participants found parts of studies - such as reading long information sheets – distressing.292 In relation to potential distress caused to bereaved persons, there are recommendations about the appropriate timing of recruitment into a study with a minimum of six months into bereavement and avoiding special dates, such as anniversaries, recommended.452, 453 Giving participants the opportunity to make their own decision about participation in research, especially on sensitive topics, could provide them with a sense of empowerment and a chance to share their stories; taking part could thus be ethically desirable and may even have therapeutic and cathartic effects for participants.333, 450, 453, 457 191 4.4.1.2 Children and young people Disabled CYP and CYP with LLCs are often not included or consulted in research related to CYP.458, 459 This is partly due to complex issues and conflicting opinion in relation to trade-offs between CYP’s rights to participation and their rights to protection.458, 460 The different developmental stages and abilities within, and changes in needs across childhood discussed in Chapters 1 and 3 are further issues that may hinder research with CYP participants, especially if they are vulnerable because of a condition. The complex issues in relation to CYP participation have also had an impact on the type of CYP that have been more often involved in research and decision making, such as older young people with no disabilities,461, 462 because researchers are more likely to be able to use recruitment strategies and data collection methods commonly used with adult participants. Adapting these methods for younger ages can be time consuming and resource intensive.463 Economic and Social Research Council (ESRC) guidance suggests that researchers should not assume that CYP are necessarily vulnerable because of their age.464 The degree to which children are vulnerable should be considered based on the research aims and context and the characteristics of the CYP group that will take part in the study.464, 465 There are conflicting views and guidance on whether CYP are able, and have a legal right to provide informed consent to participate in research.466 Literature in the USA 192 and European Commission guidance suggest that CYP under the age of 16 do not usually have a legal right to provide informed consent and parent or guardian permission is necessary; but there are exceptions, for instance if a CYP is considered to be a ‘mature minor’ and competent enough to provide valid consent themselves (1985 Gillick ruling in the English law467).466, 468, 469 Guidance suggests obtaining informed consent and permission from the parent or guardian alongside ‘assent’ from the CYP, under 16 years old.469, 470 However, guidance on always obtaining ‘assent’ – rather than informed consent – from CYP has been contentious as it assumes that all CYP are only able to assent and are not competent enough to consent to take part in research.466 There is increasing tendency for the assent process to be personalised and tailored to the particular CYP participant, rather than using solely age as a threshold.471 Given that CYP are a very diverse group, the researcher should adapt the recruitment material – such as information sheets – appropriately and sensitively to fit the capacities and competencies of the specific CYP group that they aim to include in their study.472, 473 Recruitment material should use language that is simple and direct and documents should be shorter and more focused than adult recruitment material.458 The use of drawings, speech bubbles and/or a question and answer format to break the text into shorter sections in information sheets and recruitment adverts may also assist reading.466 Symbols based information – such as Board maker and Widgit symbol systems – has also been suggested to make study information more accessible and age-appropriate to CYP participants.458 Another way to present recruitment materials 193 to CYP is via the use of short video clips, rather than written information, to introduce information about the research.474 The use of plain language can have the risk of being simplistic and/or patronising so it needs to be adapted to be age-appropriate.466 Involving CYP in research to develop recruitment documents can help the researcher to use language that is used and preferred by CYP themselves. Patient participation, also known as patient and public involvement (PPI), is increasingly used in child health research and at different stages of the research process, including research planning and design, and dissemination of findings.475 4.4.2. Data collection method 4.4.2.1 To explore sensitive topics Interviewing is suggested to be most appropriate to explore sensitive topics in research 476 and it is frequently used in qualitative health care research.334, 450 Interviewing can encourage the participant to tell their story and participants may feel they benefit from taking part in an interview on a sensitive topic even though this may also cause them some distress.477 The one-on-one human interaction involved in interviewing can facilitate exchange of sensitive information which could otherwise be difficult to achieve through the use of other methods, such as a focus group.457 194 Developing rapport with interview participants is particularly important when exploring a sensitive topic.334 It allows the researcher to form a trusting connection with the interviewees to access their lives and to facilitate disclosure from participants.457, 478 The development of good rapport is a two-way reciprocal process in which the researcher should not only expect to receive information but also give information, for instance information on the study findings.479 Self-disclosure of personal stories by the researcher has been suggested as good practice in feminist research to create a non-hierarchical environment and enhance rapport.480, 481 Yet, the researcher will need to consider how much self-disclosure would be appropriate as this can leave them feeling vulnerable.478 Closeness between the researcher and the participant runs the risk of blurring the boundaries and confusing the roles between the researcher and participant, particularly if there are repeated contacts for data collection with the same participant.449 Contact prior to data collection, such as a phone conversation, can assist in developing initial rapport with participants.457 In the initial contact the researcher should be flexible with arrangements for data collection, aiming for a time that is most convenient for the participant and at a place that feels comfortable for them; decisions about location should also take into consideration privacy and potential distractions.450 The researcher should also consider the timing for data collection following traumatic events experienced by the participant.450 There are conflicting views in relation to 195 people’s ability to recall traumatic events in detail and the severity of the event has an impact on traumatic memory.482 During the interview, it is advised that a topic guide is used to assist introduction of discussion topics and to facilitate discussion of specific and sensitive topics in a way that has been well thought out prior to data collection.483 Yet, it is suggested that the researcher actively listens to the participant’s responses and does not use the guide in a rigid manner, for instance asking questions in a predetermined sequence, as this could feel formal to participants and may hinder discussion flow.450 Active listening involves showing empathy and care during the interview, allowing the participant time to think and respond to the questions and assessing the participant’s willingness to continue with the interview.457 It is also advised that a risk assessment and a distress protocol is developed, and the researcher is familiar with both documents prior to data collection to be able to manage potential distress of participants.293, 450 It is good practice to debrief with the participant following data collection and provide them with contact information for appropriate support services if required, to facilitate a positive closure for participants.450 There is also potential risk of distress for the researcher conducting the data collection and it is advised that the researcher also devises a plan for debrief opportunities for themselves, for instance by keeping a reflective diary to write about data collection and/or regularly meet with the supervisory team for a debrief.450 196 4.4.2.2 Data collection with children and young people participants The researcher has to consider the data collection methods that would be most appropriate to use with the particular CYP group that will take part in the study based on their competencies and capacities.484 Interviews and focus groups have been developed for, and traditionally used with, adult participants but can be used with CYP if adapted and used sensitively taking into account the CYP’s circumstances, the shorter attention span compared to adults and the need for pacing and use of simple and clear information.473, 484 Incorporating activities when interviewing CYP or conducting focus groups may assist in making data collection more enjoyable for CYP and give participants more control over the process.474 It is important that the choice of activities is carefully thought through, taking into account the resources available, the characteristics and abilities of the CYP participants, the data collection setting and whether the activity will assist in eliciting information that is useful and relevant to the research aims.474 Non-verbal and/or art based activities - such as drawing, photography, use of puppets - has been used to facilitate data elicitation with young children and disabled CYP, especially for where verbal communication was not possible or appropriate, for example when there is little or no speech.410, 458, 459, 484-487 However, if mobility of participants is limited or participants are older CYP (for example adolescents) or young 197 adults such methods may not be appropriate and/or may be interpreted as patronising and childish.458, 488, 489 If CYP are able to successfully respond to questions asked by the researcher, the use of additional play activities during data collection may be a distraction from the goal of the main data collection method used.281, 348, 490 Furthermore, the choice of a specific art-based method may be popular with some CYP, for example CYP who have a special interest relevant to the particular method used, but with other CYP it may be not; this diversity would need to be accommodated.484 CYP usually have less control over their environment and need to rely on adults to take part in research.484 The location and context in which data collection takes place is important as it can affect the CYP’s decision and willingness to take part in research; for instance if the research takes place in a school setting and school staff members have provided consent for students of a class to take part then some CYP may find it difficult to decline taking part or may interpret that this is part of ‘school work’ so that their participation is expected.474 Conducting data collection in the home environment may also affect CYP’s responses to research questions. For example, parents or guardians may want to be present which can influence the CYP’s responses.474 Although the presence of a parent or guardian may influence dynamics during data collection, some CYP may feel reassured and more comfortable if a familiar person is present during data collection.458 198 It is also important to develop good rapport with CYP participants to elicit richer data. A warm-up exercise can be used prior to starting data collection, for example by asking the CYP things that they already know such as daily activities,474 to give the CYP the opportunity to feel more comfortable and get used to the researcher asking questions.281 If the research topic is sensitive it is advised to start with less difficult questions and assess how the CYP responds, prior to asking more difficult questions.474 Non-verbal communication – such as maintaining good eye contact and head nodding – to show that you are listening is particularly important when interviewing CYP.474 4.4.3. Conducting qualitative research with a vulnerable population in the context of COVID-19 There were additional challenges in conducting qualitative research with a vulnerable population during COVID-19 where researchers needed to communicate and interact with participants without face-to-face contact so as to follow social distancing measures in place. Alternative methods can be used to collect data, such as media and virtual techniques.491 The researcher can collect data with participants through the use of computer mediated communication (synchronous and asynchronous), such as text messaging, email and video calling. These may feel more comfortable than face-to-face interviewing for some participants when disclosing personal and intimate information 454, 457 and can also allow participants with verbal communication impairment to take 199 part in research.455 The use of these methods requires less travel time and lower expense for the researcher.473 Furthermore, the location of participants does not necessarily need to be constrained by distance which allows for a more geographically diverse sample to be achieved.473 However, there are certain limitations with using computer mediated communication if used in isolation, such as failing to include those who do not have digital literacy, who are not able to communicate in written form, who do not have access to a computer or the internet or those who may not desire this form of communication.492 Social interaction between the interviewer and the interviewee may also be limited if this way of communication is used in isolation.455 Maintaining conversation flow and reading body language may be more difficult than when face-to-face, and communication and interaction relies on having a good internet connection.493 It may also be more difficult to collect data online when the researcher is not able to see the physical environment that the participant is in, which may elicit valuable information for the research. There may also be privacy issues if the researcher is not aware of other people present during data collection and these may influence the responses of participants.493 There are some ethical considerations that were specific to collecting data during COVID-19. The wellbeing of participants and the researcher should be prioritised over research; even if it is possible to move data collection from face-to-face to online interaction to follow social distancing measures, the researcher should consider 200 whether the timing is appropriate for people to take part in research.494 Normal routines for many people were disrupted during the pandemic and it left many people feeling uncertain, or ill, or with additional caring responsibilities.495 There were people who lived in an environment that did not allow them to speak in private, for example if they were at risk of harassment. Yet, it has also been suggested that, as people were more confined during the pandemic and may have felt bored or lonely, they may have shown more interest in taking part in research.495 4.4.3.1 Potential alternatives to primary research during Covid-19 Alternatives to primary research were considered given the difficult research environment during COVID-19. Possibilities considered included the use of primary data from a data archiveb and the collection of data that already exists, for example through broadcasted media 496 (e.g. television) or online sources (e.g personal blogs497, 498 or online networks499, 500). However, the use of online content that is generated by the public, for example in forums or blogs, can have additional ethical considerations in research that may not be obvious, as in online spaces it is often not clear what can be accessed as ‘public’ or ‘private’.501-503 Another possible alternative that was explored was the use of qualitative synthesis. A synthesis of qualitative studies involves a process of combining or integrating findings b UK data archive, University of Essex 201 from different studies as means of developing a ‘whole’ picture of the phenomenon under study.504 There are several different methods of synthesis used in qualitative research, such as a meta-ethnography.505, 506 A synthesis does not require conducting primary research which made it a potentially valuable option in the Covid-19 context. The method also has potential advantages for the broader research context, given that it can be particularly useful when exploring sensitive topics involving seldom heard groups. It is suggested that if there is a significant number of existing qualitative studies done on a particular topic then it would be preferable to use those studies rather than having to ‘reinvent the wheel’ by conducting further primary research on the same topic.507 A synthesis of existing studies could provide a collective – and more complete – understanding of the individual studies and the topic under research.504 Also, the use of existing studies rather than the collection of primary data does not involve potential issues with recruitment of research participants, such as having an adequate sample size and potential distress caused in data collection. The method has been used previously in eliciting attributes for economic measures.508 Although its use was explored at the start of the Covid-19 pandemic, the option was rejected because of the lack of suitable material to include in a meta-synthesis. 202 4.5. Conduct of in-depth interviews to develop the conceptual attributes for an EOL measure for young people. This section outlines the methods used to develop the conceptual attributes for a capability measure for young people at the EOL. This includes all stages in developing the conceptual attributes, from recruitment of research participants through to analysis. 4.5.1. Research Design 4.5.1.1 Sampling and research participants It was not clear how many respondents the study was likely to be able to recruit, given the vulnerable participants and the new uncertainties introduced by the Covid-19 pandemic. The recruitment strategies accounted for both possibilities of large and small number of responses from potential participants. If the number of responses from potential participants was large, then the response from a screening questionnaire (Appendix 2) would be used to purposefully sample on the basis of the respondents’ characteristics. In this case, maximum variation sampling was intended, 203 to capture different perspectives and experiences that could inform the development of the conceptual attributes for the measure and ensure that the attributes were relevant to the characteristics of the population with which it will be used. As the analysis of data continued to generate theory around what is important to the sample group, the researcher also planned to sample more participants based on the need to further explore findings emerging from the research (theoretical sampling). However, given the sensitive topic and the vulnerable participant group, the number of responses from potential participants was small and the researcher included all initial respondents, and also applied snowball sampling through existing participants. Based on previous qualitative research to develop capability measures for use in economic evaluation218, 290 it was expected that an approximate number of 20-35 in-depth interviews would produce a sufficient sample for data saturation. Data saturation for the development of the conceptual attributes was expected to be achieved when the conceptual attributes were deemed to be fully formed with no new perspectives and ideas being introduced by new participants that would change the nature of the attributes. When recruitment started in January 2021 the initial inclusion criteria for research participants was: • young people aged 14-18 years old with a LLC or LTC 204 • Those with parental responsibility for young people (14-18 years old, with a LLC or LTC) who were not able to take part in the research directly due to the condition and/or symptoms. The inclusion criterion in relation to LLCs/LTCs was decided due to the terms being used to describe the population of CYP who could benefit from receiving PPC and EOLC.16, 17 The decision to include those with parental responsibility as proxy participants aimed to capture the perceived experience of young people who were not able to take part themselves in the study. The age group criterion was decided due to the need to develop different measures for different age groups of CYP. CYP of this age, are facing the transition from childhood to adulthood, exemplified by the transition to adult care services. In the UK, YP with LLCs or LTCs prior to reaching adulthood, will prepare for transition of care to adult services, with initial discussions on transition starting from as early as the age of 14 with transition at ages 18+.107, 110 Preparing for transition may be challenging for YP and their family, involving a period of change and uncertainty in terms of future care provision and needs being met after transition to adult services.110 Potential participants who fit the above inclusion criteria but did not reside in the UK and/or were not able to understand information provided about the study and/or did not speak English were excluded given that the aim was to produce an English language measure. 205 During the recruitment and interview analysis stages – which were undertaken concurrently – the study inclusion criteria shifted as outlined in the next section below. The shift in the inclusion criteria was in part due to recruitment difficulties but in part to enable the study to better capture experiences of young people at the EOL as more was learned about the nature of conditions, the experiences of CYP and what they were – and were not – able to speak about. 4.5.1.2 Recruitment The young person and parent/guardian study posters (Appendix 3) were first circulated in January 2021 through the family involvement lead at TfSL charity who promoted study information in their family groups targeting families of young people living with LTCs/ LLCs. The first round of advertising led to only one respondent who, at the time, was not eligible to participate in the study due to not fitting the age range criteria for the initial target young person age group (14-18 years old; 25 years old). Following discussion with the family involvement lead at TfSL and the supervisory team, and in response to difficulties with recruiting participants through members of staff at the charitable organisation, an ethics amendment was requested. The first FREC ethics amendment was approved at the end of January 2021 (Appendix 4). The amendment included the addition of two new recruitment strategies. The first additional strategy involved posting of the young person and parent/guardian study posters on social media platforms – Twitter, Facebook, and Instagram – to promote 206 the study directly to young people aged 16-18 years old and parents/guardians of CYP aged 14-15 across the UK. Relevant family and young person groups and forums were identified on social media platforms to advertise the study poster on their online page. Young people aged 14-15 years old were recruited though their parent/guardian when using the above strategy. The TfSLs family involvement lead also promoted the study poster on Twitter and tagged other relevant charitable organisations (e.g., TCT) and individuals (e.g., young people with lived experience, disability advocates). The second additional strategy involved asking staff at relevant charitable organisations to assist with introducing the researcher to potential participants at relevant online events during which the researcher promoted the study by presenting the study poster information to event attendees. Events in which the study was promoted by the researcher included the TfSL Children’s Hospice Practice Education forum and monthly Bereavement sector webinars organised by the Childhood Bereavement network. In February 2021, the study was also promoted through the family involvement lead in the TfSL care newsletter and study information was sent out to TfSL transition leads. By the end of February, one other young person expressed interest to take part in the study, through the TfSL family involvement group. However, the young person was not within the initial age range group (16-18 years old). Feedback from attendees at the South-west palliative care and end-of-life care (SWPCEOLC) network meeting with research and clinical expertise in PPC and PC in 207 young adultsc, suggested that many young adults over 18 and up to the age of 25 years old continue to experience practical challenges in relation to transition to adult services and some may transition to adult care services after they turn 18 years old. Therefore, these young adults do share similar transition experiences and challenges with those under 18 years old preparing for transition. It was also suggested that the number of young people with LLC/LTCs under 18 years old who have verbal ability is limited. The young person age range criterion for the study was therefore extended to up to 25 years old both in an attempt to increase participation interest and to be able to capture fully the experiences and challenges of transition to adult services which continues past the age of 18 years. A second university ethics amendment was requested to update the inclusion criteria for interview participants to include: • young people 14-25 years old with a LLC or LTC • Those with parental responsibility for young people (14-25 years old, with a LLC or LTC) who were not able to take part in the research directly due to the condition and/or symptoms. The second ethics amendment was approved in March 2021 for the age range extension and updated study posters were re-circulated through the TfSL groups, c February 2021 network meeting 208 newsletter, and social media platforms. Circulation of the study poster through the family groups and transition leads at TfSL ended in April 2021 as the research support department of the organisation went through an overhaul and following other recruitment support requests was not able to continue recruitment support for this study. Alternative ways to recruit participants through social media platforms (Facebook, Instagram, and Twitter) were then used. Paid Facebook and Instagram adverts (see Appendix 5) were used, and relevant Facebook UK-based family and support groups were contacted and agreed to post study posters on their Facebook page. A study webpage was also created (A Quality of Life Study: Young people living with serious illness | Bristol Medical School: Population Health Sciences | University of Bristol) to refer people to if they were interested in taking part in the study or wanted further information. Twitter poster adverts were also created (Appendix 5) and relevant individuals and organisations were tagged to re-tweet with their followers. 209 Between April-May 2021, thirty-eight UK-based charitable organisations supporting families with specific LLCs/LTCsde, including rare conditions and/or offering specific types of support and care to families (e.g., music therapy; equine-assisted therapy; financial support; advocacy service) were contacted by email or phone to seek further recruitment support through these organisations. The response rate from charitable organisations was low (thirteen out of thirty-eight responded) and those who did respond were unable to support recruitment due to reasons including: - resource capacity limitations and/or COVID-19 impact on current resources available for research related support - difficulty identifying young people, or families, who they felt would be able to take part in the study at this time - the nature of the study topic is sensitive; the population is vulnerable and/or they did not want to involve young people in discussions around PC/EOLC d To identify the most appropriate condition-specific charitable organisations, the researcher used: - the Directory of LLC/LTCs (Microsoft Word - Directory of LLC v1.3.docx (icpcn.org)), - the ICD-10 diagnostic coding framework for CYP with LLCs/LTCs (2020).19 - advice from supervisory team with PPC expertise e ‘Contact A-Z medical conditions’ website (https://contact.org.uk/conditions/ ) was used to obtain contact information for relevant condition-specific charitable organisations 210 - they had recently supported recruitment for another study and on that occasion uptake was very low and/or they felt it would be too soon for the families to take part in another study - did no longer consider the specific condition as LLC/LTC due to improvements in care standards, medical technology advances and/or new drug treatments. A third FREC amendment was approved in June 2021 to allow for two additional recruitment strategies: to recruit through child and adult hospices in England (non-NHS sites) and UK-based special schools. The researcher contacted via email twenty-three children hospice services in England, five adult hospices and three local (South-West England) special schools. None of the special schools responded but eighteen hospices (sixteen children, two adult) responded and were able to support recruitment. Members of staff at hospices assisted by identifying and informing eligible participants about the study; and/or by circulating the study poster and participant information sheets to families within their hospice, their young ambassadors and parent/carer groups, and/or through their social media accounts and newsletters. Due to difficulties recruiting young people and parents/guardians of young people who are at the very EOL, in September 2021, a fourth FREC amendment was approved to include bereaved close person participants to be able to capture their perceived experiences, and what they thought was important to the young person, at the very EOL. Such experiences were limited from existing interview participants, and it was 211 decided, in discussion with the supervision team, that bereaved family and friends accounts could complement existing accounts to capture the EOL experience. Following the fourth FREC amendment the study inclusion criteria were updated to include: • young people 14-25 years old with a LLC or LTC • Those with parental responsibility for young people (14-25 years old, with a LLC or LTC) who were not able to take part in the research directly due to the condition and/or symptoms • Close person bereaved at least six months post death of a young person (14-25 years old) with LLC/LTC. Close person includes family, partner, close friends. For ethical reasons and as advised by the supervisory team, bereaved participants within six months of bereavement were excluded from the study.290, 452, 453, 509 The grief experienced by bereaved close persons of young people, and in particular bereaved parents can be particularly intense and prolonged. 186 The six-month timing criterion was chosen to avoid potential early intense grief experienced by participants and ensure that enough time had passed for them to be able to work through any issues they may have experienced. 510 Twenty-nine charitable organisations across the UK and eight England based Hospices offering bereavement support were contacted to facilitate recruitment. Six charitable 212 organisations and five hospices were able to support recruitment and circulated the study poster and/or participant information sheets to family and friends using their bereavement support service. Between January 2021, when the study information was first promoted, and until recruitment ended in September 2022, contact was maintained with the organisations and hospices assisting with recruitment to update on progress and, where appropriate and depending on capacity, to ask for the study information to be re-circulated to potential participants. In June 2022 a study newsletter (Appendix 6) was circulated on social media platforms and through the charitable organisations and hospices with information on study progress, and was used as a final call for any further expressions of interest to participate. 4.5.1.3 Data Collection Given the sensitive nature of this study and that data collection occurring during the Covid-19 pandemic, online in-depth interviews were chosen to collect data. Interviews took place at participant’s preferred time and location. Duration of interviews was subject to the participants’ preferences for up to an hour but with breaks if needed, and at a pace that the participant felt most comfortable with. Participants were given a choice to pause the interview process to allow them to take a break if they wished to or to withdraw at any time. 213 In advance of the interview, the researcher provided brief information to participants about the topics that would be covered during the interview to allow them to think in advance about the things that are important to them, or to their CYP if the participant was a family member. To build rapport with participants, the researcher also discussed in advance with the participants their preferred means of communication, such as whether they preferred a phone call or a video call interview, and the way they preferred the interview questions to be asked, such as in written form, with short breaks during the interview and having an interpreter present (e.g., Makaton interpreter). Young people were also given a choice to be interviewed with or without the presence of their parent/guardian. While most of the participants showed preference for being interviewed alone in a private space, two young people wanted to be interviewed with a parent who would facilitate the interview with the young person. In one of these interviews the parent facilitated interview discussion between the interviewer and the young person with the use of a language programme that uses signs together with speech and symbols to enable the young person to communicate. Having an interpreter who was familiar to the participant helped them feel comfortable and the parent was able to recognise speech and symbols that an independent interpreter – who was not familiar with the young person – would not have been able to interpret. In the other interview the parent was present to assist with joining online for the interview, to hold the mobile device used for the interview. This was in order that that the young person participant 214 and the researcher could see each other and to facilitate discussion. The young person participant asked if they could have the interview questions in advance as they can get quickly overwhelmed with information, having fatigue and have verbal communication that is limited due to a tracheostomy. By providing them with some of the key questions in advance the participant had the chance to think about them at their own time and their parent wrote down the young person’s thoughts at the time, so they were able to read them out loud during the interview when the young person needed a break from talking. Many of the participants also showed preference for being interviewed over the phone without the use of a video when given the option. There were twelve ‘video-call’ interviews (five young people; three parents/guardians and four bereaved close person participants) and nine ‘phone-call’ interviews (one young person; three parents and five bereaved close person participants) in total. ‘Phone-call’ interviews were conducted by using the Microsoft Teams Calls option or by turning off the camera/video option. Interviews explored what is important to young people in terms of their wellbeing to identify important capabilities. The interviews also explored the things that may be important to young people in the future, including at times that their health may change. Three separate broad topic guides were developed to conduct the in-depth interviews: one for young people, one for parents/guardians proxying for a young person and one for bereaved close persons (Appendix 7). The researcher used the 215 topic guides flexibly, with both closed and open questions to probe and allow exploration of the topics elicited in more depth. The researcher also kept reflective notes following each interview that included what went well, what not so well and what topics need to be explored further and in what way. Reflective notes were particularly useful in exploring what is important to young people near the very EOL. This was important because participants in the first three interviews did not talk about this topic in detail, and it was difficult for the researcher to find a way that they felt comfortable with to ask in more depth about this topic. The three first interviews began by asking the participant about what things were important to the young person followed by questions on how their condition affects the young person. However, following these interviews the researcher felt that it would be better to begin interviews with general questions about the participant and how their condition affects them to a) build rapport and ease the participant into the main questions about what is important b) to get the context and ask questions which were more relevant to the specific participant, their situation and condition. Questions to explore the evaluative space were then asked, including questions around transition to adult services, prognosis, experiences of advance care planning, limitations young people experience related to their condition and how the young people view their condition progressing in the future. As the research progressed, the topic guides were updated iteratively with new themes emerging from the analysis and the reflective notes, to explore further the 216 topics emerging from the interviews, such as the very EOL, transitioning to adult services and around the topic of identity (Appendix 7). 4.5.1.4 Constant Comparative Analysis With the permission of the participants, and the permission of the parent/guardian for participants under 16 years old, interviews were video and/or audio recorded and transcribed verbatim. Audio transcripts were uploaded into a qualitative data analysis software (NVivo12) where they were coded and analysed to identify emerging themes in terms of what is important to young people at the EOL. The analysis of interview responses of young people, parents/guardians and bereaved close persons was done separately, with a separate coding schedule developed for each group, to allow for the possibility of identifying different themes in each of these groups. The method of constant comparative analysis 428 was used, concurrently with data collection. Interviews took place in batches of usually two or three, dependent on the dates that participants expressed interest and were interviewed, and the analysis was conducted iteratively with a gap between each batch of interviews. A general reading through each interview transcript was first undertaken making notes next to the text on a line-by-line basis looking at what is being said.427 This was followed by a more thorough reading to undertake open coding within each batch and apply open codes to the text. The questioning technique, as described by Strauss and Corbyn (1990), was used to guide initial coding.428 The technique involved asking 217 questions such as, ‘What phenomenon is being described here?’; ‘Who is involved?’; ‘Which aspects of the phenomenon are being described and why?’. ‘In Vivo’ codes were then generated from the data, using the language of the participants, and using short descriptions, such as ‘Enjoyment’. These codes were developed in a hierarchical manner by comparing the data for similarities and differences and making connections between codes and sub-codes (axial coding). Three separate coding schedules were then developed, one for each participant group. For the first five interviews with young people and parent/guardians, open coding was conducted independently by the researcher and two members of the supervisory team with expertise in qualitative research and PPC nursing, in order to compare open coding data with the researcher and develop the initial coding schedules. Two spider diagrams were also used to map initial categories and sub-categories for the coding schedules. Following the development of the initial coding schedules, in discussion with the supervisory team, the need to capture the experiences of those at the very EOL was reiterated and it was decided that the study would benefit from recruitment of bereaved close persons to better capture these experiences. As analysis progressed, interview topic guides were updated, and emerging themes were constantly reviewed by comparing existing themes and participant responses to improve understanding and to identify and include different participant perspectives on what is important to young people. Eight descriptive accounts were developed for groups of participants – young people; parent/guardian and bereaved close persons – 218 to compare responses and develop themes related to what appears to be important to young people at the EOL. The accounts included quotes from the interview transcripts. Members of the supervisory team with research experience in the constant comparative method also provided feedback on the emerging themes in the descriptive accounts. Finally, an analytic account was generated bringing together the eight descriptive accounts to repeat comparison of all main topics that emerged from the three different participant group accounts by looking for connections between categories and data from each of the participant groups. The repeated searching of the data also assisted in checking that no new themes arose 511 on what is important to young people at the EOL from the different perspectives of the three participant groups. At the point of data saturation512 – that is when the conceptual attributes were fully developed - the recruitment, data collection and analysis ended. 4.6. Conduct of cognitive interviewing The inclusion of bereaved close person participants in in-depth interviews to develop the conceptual attributes for the young person measure also allowed the research to achieve a second research aim: to explore the appropriateness of ICECAP-CPM in this population and context. This section outlines the methods used for assessment of the feasibility of use and content validity of ICECAP-CPM in bereaved close persons of young people with LLC and LTC. 219 4.6.1. Sampling and recruitment The same sample of bereaved close person participants (n=9) was used for both the conduct of the in-depth interviews, outlined in the previous section, and the cognitive interviews outlined in this section. Data collected for the development of conceptual attributes and data collected to explore feasibility of use and content validity of ICECAP-CPM were both collected in a one-off session with bereaved close person participants. See section above for further information regarding recruitment of bereaved close person participants. 4.6.2. Research Design Given the sensitive topic of the study, the method of cognitive interviewing was used in the empirical work to assess feasibility of use and content validity of ICECAP-CPM. As with previous studies conducted to test capability measures within the ICECAP family,218, 293, 435, 445 the think-aloud technique followed by semi-structured interviewing concerning the measure was used in this work. 4.6.2.1 Data collection Interviews with close person participants were divided into two parts. Part one involved the in-depth interview (to develop conceptual attributes) outlined in the previous section, and part two involved cognitive interviewing using a think aloud technique followed by semi-structured interviewing (see Appendix 7). Part two of the 220 interview focused on the experiences of the close person, rather than the experiences of the CYP. Details of the development and content of ICECAP-CPM used in the think-aloud part of the interview are outlined below. The think-aloud task was conducted with ICECAP-CPM (Appendix 8), a capability well-being measure recently developed for use in economic evaluation in an EOL setting that captures benefits of EOLC to those close to the dying person.290 The descriptive system of the measure was derived from qualitative data collected from in-depth interviews with close persons (family and close friends) either recently bereaved or who had loved ones receiving EOLC. The close person participants included in the ICECAP-CPM development were from an older adult patient population between 40 to 80+ years old. There are no existing studies conducted to assess the properties of the measure with close persons for younger people at the EOL. ICECAP-CPM has six attributes for measuring EOL experience, each with five response levels. Close persons completing the measure are asked to think about their experience and respond based on which level best describes their situation for each of the attributes. Each attribute is presented as a statement; the statements are: ‘Communication with those providing care services’; ‘Privacy and space to be with the loved one’; ‘Practical Support’; ‘Emotional support’; ‘Preparing and Coping’ and ‘Emotional distress to you, related to the condition of the person’. Each of the attributes includes descriptors under each statement, for instance under ‘Practical support’ one of the descriptors is: ‘having been able to get practical support and help 221 with the care of the person, such as nursing help, help from social services or help from family’. The five response levels to each attribute range from either ‘all of the time’ to ‘none of the time’ or from ‘fully able’ to ‘completely unable’ depending on the question askedf. Use of the measure was registered with the ICECAP team. In the cognitive interviews, participants were first asked to complete ICECAP-CPM and think aloud to verbalise their thoughts whilst completing the measure.447 Prior to the cognitive interview, the researcher explained to participants that the aim of the task was to find out if there were any issues with the measure and so they should not worry about expressing their opinion on the measure. During the think aloud part of the interview the researcher remained silent unless the participant was silent for longer than ten seconds at which point the researcher asked the participant to keep thinking aloud. During the think aloud part, the researcher also kept notes on specific aspects of the measure that the participant seemed to find problematic – such as in relation to the measure instructions or ticking between two of the available response options – so that they could explore these issues further in the semi-structured part of the interview. Following completion of ICECAP-CPM, participants were asked to express their views about the measure, the extent to which they felt each of the measure items were f The full ICECAP-CPM can be obtained on https://www.bristol.ac.uk/population-health-sciences/projects/icecap/icecap-cpm/ 222 comprehensible and relevant to their experience as a close person, and to clarify any issues they had with completion of the measure. A copy of the ICECAP-CPM was sent via email to participants during the interview to facilitate their completion of the think-aloud task. 4.6.2.2 Analysis Interviews were video or audio recorded, depending on participant preference, and transcribed verbatim. Audio transcripts were uploaded into NVivo12 qualitative data analysis software to be coded and analysed. The think-aloud part from each interview transcript was divided into six segments which represent the six items in ICECAP-CPM for use in rating. Transcripts were then independently rated to identify interview sections that participants encountered a problem (error or struggle) when completing each question. The model proposed by Tourangeau et al.448 for survey responses was used as a rating framework to look for errors. Responses were rated based on the number of errors made during completion using the following four criteria: Question comprehension in the way the measure intended; Retrieval of appropriate information to answer question; Correct judgement on how to use recalled information to answer question; Use of information in a way that it gives a valid response to question. The addition of a ‘struggle’ category followed earlier studies of capability measures 293, 435, 445, 513 to identify areas in which there was no error and the respondent provided a response appropriately, but had difficulty arriving at a response. 223 Transcript selections along with information about how participants scored themselves against each attribute (the ‘scoring’) was presented to four independent raters in the supervisory team (I.F, J.C., P.M and S.N.). Independently, raters identified, in relation to each segment, whether responses were free of error, or any one of the four types of errors was present or there was a struggle in answering the question (see Appendix 8). Following independent rating, at two rating meetings with all raters present, responses were discussed to reach a consensus on the final number and type of errors or struggles present. The following rules were used to reach consensus in the rating meetings: - If none or only one of the four raters identified an error or struggle this was not discussed, and the item was recorded as being free of error. - If three or more raters identified the same type of error, or struggle, the item was recorded as having an error or struggle. - If fewer than three raters identified the same error or a different error type or struggle, then consensus for the type of error was reached with discussion among the raters. - If more than two raters identified an error or struggles but fewer than two identified a specific type of error, then consensus for the type of error was reached with discussion. Qualitative data from the semi-structured interviews following the ‘think aloud’ task were coded and analysed using constant comparison 422 to identify and compare 224 recurrent themes across participants to explore the extent to which each item was relevant to their experience. 4.7. Ethical issues and conduct of qualitative interviews Ethics approval was obtained from the University of Bristol by the Faculty of Health Science Research Ethics Committee (FREC) on 14/12/20 (Ethics reference number: 111822) to allow for recruitment of participants in the study for the in-depth and cognitive interviews (Appendix 4). Given the sensitive topic and vulnerable groups of participants, there were a number of ethical considerations. These are discussed below: 4.7.1. Process to obtain consent from participants. Prior to recruitment of interested potential participants, and during data collection, the researcher informed all participants aged 16 years and over, verbally and in written form using the information sheet (Appendix 9), what participation would involve. This included participants’ rights to withdraw at any point if they wish to. Prior to data collection, written informed consent (Appendix 10) was obtained from all participants aged 16 and over. Participants under 16 years old were asked to give their assent by 225 completing an assent form (Appendix 10); their parent/guardian completed and signed the consent form for the CYP’s participation. Potential participants were not given a deadline for giving consent to take part so they could decide in taking part in their own time. 4.7.2. Data Storage and Sharing A University of Bristol approved audio-visual interface (Microsoft teams) was used for all interviews and the researcher followed up to date University of Bristol guidance on data collection (https://www.bristol.ac.uk/red/research-governance/covid-19_guidance/) to use an audio-visual interface supported by the University of Bristol that is more secure than other online platforms. All interviews were audio recorded with the use of the recorder available on the audio-visual interface used for the interviews. The researcher stored and backed up all research data collected for this study in an electronic form, in the University of Bristol Research Data Storage Facility (RDSF). Access to all confidential research data were only available to the researcher and members of their supervisory team; anonymised transcripts were shared with the broader research team. Anonymised Interview transcripts were stored separately to all other data collection information, such as consent forms and each participant was allocated a personal code that corresponded to the transcript of the interview. Details of the code allocated to each participant were stored in a password protected 226 document and the password was only known to the researcher and the supervisory team. The University of Bristol approved transcription service had access only to the audio-recorded interview data and the researcher anonymised transcripts and replaced personal information so that participant information, including names, addresses and locations, cannot be identified. Care was also taken to anonymise the medical diagnosis of the conditions of participants as this information may make the participant identifiable especially when the condition is rare and the numbers of young people with any specific diagnosis is very small. This anonymisation included removing details of symptoms, treatment and/or support provided from specialised services offered only in specific location and treatment centres in the UK. Furthermore, the medical diagnosis for each condition was anonymised and replaced with the relevant diagnostic group category, for example: congenital, neurological. The ICD-10 diagnostic coding framework used in the latest report (2020) on prevalence of LLC/LTC in CYP in the UK, was used to identify and categorise conditions.19 4.7.3. Assessment and management of potential risks of the research The researcher, who has previous experience working with vulnerable adults, took appropriate safety measures to prevent risk of harm (physical or psychological) to participants and to minimise distress to participants, including anxiety, sadness or fear. Measures included managing potential risk associated with invasion of privacy, 227 confidentiality breaches and coercion into study participation by a parent/guardian or the researcher. Assessment and management of potential risks included: 4.7.3.1 Language and Wording used To manage potential risk of distress to participants and coercion into participation, the researcher considered carefully terms and wording used in all recruitment and data collection documents in terms of the emotional weight they may carry,514 for example, the use of the term EOL. Written documents were developed and reviewed in consultation with members of the researcher’s supervision team, including a supervisor with clinical nursing background in PPC to ensure that the topic was introduced in a sensitive and appropriate way. Feedback and suggestions on the wording and layout of the recruitment material were sought from staff members of TfSL charity and a young people advisory group (YPAG) in Bristol that includes young people from different socio-economic and ethnic backgrounds. Feedback on the terminology used in the recruitment documents was also sought from the ‘young experts’ involvement group in TfSL charity. The members of the group comprised young people (ages 14-18 years old) living with LLCs or LTCs. Their feedback was considered when making final decisions about the terminology used in the recruitment documents, such as the use of the term LLC which was replaced with the term ‘serious illness’ in the recruitment documents for young person and parent/guardian participants. 228 To manage potential risk of distress to participants or perceptions of being coerced into participation, the researcher considered the language she used with participants at all research study phases from recruitment and in any contact she had with participants following recruitment. This included language that may have implied an expectation to participate in the study515. 4.7.3.2 Recruitment To avoid potential risk of coercion to the participants by the researcher into participating in the study the researcher attempted to ensure that the potential participant understood what the study involved and how the study findings would be used, prior to them making a decision to take part. The reseacher went through the information sheet and consent form with participants prior to data collection and reminded them that they were able to withdraw at any time without giving a reason for this. 4.7.3.3 During Data Collection A distress protocol was developed to manage potential distress during and following data collection (Appendix 11). The distress protocol involved a contingency plan presented in a flow chart detailing the process to be followed if a participant became distressed. 229 To minimise potential risk of distress and in an attempt to make the participant feel comfortable during the interview the researcher maintained good eye contact, used facial expessions and body language to express their interest in what the participant was saying and showed empathy. The researcher also allowed the participant to respond to interview questions at their own pace and asked the participant if they wish to take a break, re-arrange data collection for another time and/or remind them that they can withdraw, if they seemed uncomfortable or struggled to proceed with the data collection process. A number of participants asked to take a break during the interview but were happy to continue with the interview following the break. One young person needed a break as they were feeling tired and three bereaved participants took breaks as they became tearful during the interview. 4.7.3.4 Following Data Collection Following data collection the researcher went through a debriefing statetement (Appendix 12) with participants, asked them how they found the data collection process and gave them the chance to ask any questions. The form included information about free national support services the participant may have wished to access following the interview. 4.7.3.5 Safeguarding and Complaints There were no safeguarding concerns with any of the participants. Prior to data collection, the researcher became familiarised with guidance from the ‘Working 230 together to safeguard children’ guidance HM Government document (July 2018) (https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/779401/Working_Together_to_Safeguard-Children.pdf) and the University of Bristol Safeguarding policy (https://www.bristol.ac.uk/media-library/sites/secretary/documents/policies-and-forms/safeguarding-policy.pdf) to be aware of when and how potential safeguarding concerns should be reported. 4.7.3.6 Risks to researcher To avoid potential risk of distress to the researcher following data collection, regular debriefs took place following interviews at the start of the process. This continued as and when the researcher felt they were required. The regular debriefs were with members of the supervisory team which included people who have experience of sensitive interviewing including at the EOL. 4.8. Summary Qualitative methodology and the qualitative methods used in this thesis have been outlined in this Chapter 4. The next two chapters report on the findings from the qualitative work conducted. Chapter 5 reports the findings from the in-depth interviews on important capabilities for young people. Chapter 6 reports the findings from the cognitive interviews to assess use of ICECAP-CPM on close persons. 231 CHAPTER 5. YOUNG PEOPLE AT THE END OF LIFE - CAPABILITIES AND CONVERSION FACTORS The qualitative methods used in this thesis were outlined in Chapter 4. Chapter 5 presents the findings of the in-depth interviews on aspects of life that are important to young people at the EOL. Informed by the Capability Approach, key themes elicited in the interviews are presented in terms of what is important in young people’s life as an end in itself (capability themes) and the underlying factors (conversion factor themes) that influence the extent to which young people can achieve these valuable ends. The capability themes derived from the interviews are: Experience and enjoy; Independence; Freedom from physical suffering; Freedom from emotional suffering; Formal care and support; Control; Identity. The conversion factor themes are: Access to care and support, Continuity and consistency in care and support, Communication between the family and services, Coordination between services and COVID-19. Discussion of the conversion factor themes is interwoven with discussion of the capability themes. First, recruitment details and the characteristics of participants are outlined in the chapter. Then emerging themes from the interviews are presented. The chapter ends with a diagram that summarises the main findings from the interviews. 232 5.1 Recruitment and participant characteristics The first two interviews were carried out in May 2021 and the total number of interviews was twenty-one by the end of September 2022 when recruitment ended. Circulation of study information through charitable organisations, hospices and social media platforms resulted in forty-five initials expressions of interest to take part in the study. Five participants were recruited through charitable organisations, six participants were recruited through Children’s Hospices, seven participants were recruited via social media platforms (Twitter post, Facebook group and paid Facebook advert) and three participants were recruited through existing study participants into the study (snowball sample). Common reasons for individuals not being included in the study following an expression of interest were that they did not fit the young people age group eligibility criteria (six); they did not have a LLC/LTC (seven) and, for bereaved close person participants, that they had been bereaved for fewer than 6 months (two). Nine individuals initially expressed interest in taking part in the study but eight did not respond to the follow up email invitation to interview and one young person chose to withdraw from the study prior to the interview due to feeling too unwell to take part. The characteristics of the young people interviewed or whom the interview was conducted about are summarised in Table 4 for each of the three participant groups. Interviews were conducted with 1) six young people, 2) six parents of young people who were not able to take part themselves, and 3) nine bereaved family member 233 participants (parents and siblings) participants. More interviews were conducted with or about female (n=12), rather than male, young people. Interviews were with or about young people from different UK regions in England and Wales. Many of the interviews (n=11) were with or about young people living in the most deprived neighbourhoodsg in the UK; those in the three lowest deciles. Interviews included experiences of young people with a wide range of conditions, including rare conditions, and a wide range of illness trajectories and types of care and support received. The diagnostic group category, such as neurological, respiratory, metabolic, rather than the full medical diagnosis of the young people’s LLC/LTC, is reported in Tables 4 and 5.19 This is to protect the anonymity of the participants given that some of the participants had rare conditions. Table 5 shows further details about the individual characteristics of interview participants. g Deprivation levels used in Table 4 and 5: 1-3 most deprived; 8-10 least deprived. From English Indices of deprivation, 2019 (https://www.gov.uk/government/statistics/english-indices-of-deprivation-2019) and Welsh Index of Multiple Deprivation, 2019 (https://statswales.gov.wales/Catalogue/Community-Safety-and-Social-Inclusion/Welsh-Index-of-Multiple-Deprivation#:~:text=The%20Welsh%20Index%20of%20Multiple,to%201%2C909%20(least%20deprived)) 234 Table 4 – Young people characteristics YOUNG PEOPLE CHARACTERISTICS YOUNG PEOPLE PARTICIPANTS (n=6) PARENT/GUARDIAN PARTICIPANTS (n=6) BEREAVED FAMILY MEMBERS PARTICIPANTS (n=9) TOTAL (n=21) AGE GROUP (YEARS) 14-17 3 1 2 6 18-25 3 5 7 15 GENDER Male 3 3 3 9 Female 3 3 6 12 ETHNICITY white British 4 4 6 14 black - - 3 3 Asian 1 2 - 3 other 1 - - 1 DEPRIVATION LEVEL 1-3 (most deprived) 3 2 6 11 4-7 1 4 1 6 8-10 (least deprived) 2 - 2 4 HEALTH CONDITION Congenital 2 3 1 6 Metabolic 1 1 - 2 Cancer 1 - 5 6 Circulatory - - 1 1 Neurological 1 1 2 4 Respiratory 1 1 - 2 RECRUIMENT METHOD Charities 1 1 3 5 Hospices 2 4 - 6 Social Media 2 1 4 7 Snowball 1 - 2 3 235 Table 5 - Descriptive characteristics of participants (n=21) Participant ID Participant age group (years) Participant Gender Participant Ethnicity Participant Deprivation level Months since bereavement Relationship to young person Young person’s health condition Young person’s age group (years) Recruitment method Young people participants (n=6) YP01 18-25 Male Asian 1-3 n/a n/a Respiratory n/a Via charity YP02 14-17 Female white British 1-3 n/a n/a Congenital n/a Via hospice YP03 14-17 Male white British 1-3 n/a n/a Congenital n/a Snowball YP08 18-25 Female white British 8-10 n/a n/a Cancer n/a Via social media YP11 18-25 Female white British 4-7 n/a n/a Metabolic n/a Via hospice YP12 14-17 Male other 8-10 n/a n/a Neurological n/a Via social media Parents/Guardians (n=6) PG01 50-59 Male Asian 1-3 n/a Father Respiratory 18-25 Via charity PG03 40-49 Female white British 1-3 n/a Mother Congenital 14-17 Via hospice PG04 50-59 Female white British 4-7 n/a Mother Congenital 18-25 Via hospice PG05 50-59 Female white British 4-7 n/a Mother Neurological 18-25 Via hospice PG10 40-49 Female Asian 4-7 n/a Aunt (guardian) Congenital 18-25 Via social media 236 PG11 50-59 Female white British 4-7 n/a Mother Metabolic 18-25 Via hospice Bereaved family members (n=9) BCP06 50-59 Female white British 1-3 12-18 Mother Cancer 14-17 Via Charity BCP07 50-59 Female white British 8-10 42-48 Mother Circulatory 18-25 Via Charity BCP09 40-49 Female white British 1-3 12-18 Mother Congenital 18-25 Via social media BCP13 50-59 Female white British 1-3 6-12 Mother Neurological 18-25 Via Charity BCP14 20-29 Male black 1-3 12-18 Brother Cancer 18-25 Via social media BCP15 20-29 Female black 1-3 12-18 Sister Cancer 18-25 Via social media BCP16 20-29 Male black 1-3 12-18 Brother Cancer 18-25 Snowball BCP17 50-59 Female white British 8-10 42-48 Mother Cancer 14-17 Via social media BCP18 20-29 Female white British 4-7 6-12 Sister Neurological 18-25 Snowball 237 5.2 Presenting interview findings Findings from the in-depth interviews conducted are presented in the rest of this chapter. Quotes from the interviews are used to illustrate the findings. Where quotes are presented, ellipses (…) are used to show where text is missing. Where participants used words which do not add to the meaning of the text, such as ‘you know’, this has been excluded from the quotes without the use of ellipses. Square brackets […] are used to clarify the context that the participant is referring to or where personal information is omitted to protect participant’s anonymity. This includes all places, names and names of conditions. Extra care was taken to exclude information about condition-specific symptoms and types of treatment due to some of these being associated with rare conditions or treatment being provided in specific treatment centres across England. Participant codes indicate the recruitment group of the person speaking: for young people participants ‘YP’ initials are used; for parents/guardians participants ‘PG’ initials are used; for bereaved close person participants ‘BCP’ initials are used. 5.3 Capabilities and conversion factor themes 238 Key themes and sub-themes that emerged in the interviews were analysed and explored further in following interviews. Through the analytic process, the final themes were later separated into two distinct types to distinguish between important capability themes (seven), which was the initial topic of interest; and conversion factor themes (five), which emerged naturally in the interviews and were further explored in addition to the capability themes. The distinction between capability themes and conversion factor themes became apparent during the last stage of the analysis in which the descriptive accounts from the three different groups of participants were brought together to develop an analytic account that enabled comparison of themes between the three participant groups. Prior to this stage, the themes identified were considered important aspects of life as an end in themselves, that is, aspects of life which were of intrinsic value to young people, thus they were identified as capability themes. However, comparison between the groups of participants in the final analytic account assisted in identifying an important distinction within the key themes captured. Some themes were about aspects of life that were of intrinsic value, and thus were identified as the final capability themes, the ‘beings and doings’.223 Other themes were about aspects of life that were important in the sense that they helped or hindered young people in achieving those ‘beings and doings’. These latter aspects of life were thus identified as conversion factors, factors that influence the young people’s ability to convert resources into valuable capabilities.231 For instance, being able to access social care 239 was important for young people – not in itself – but because if social care can be accessed when necessary and appropriate then it can enable young people to maintain their independence as an example. The extent to which a young person can have and maintain independence is affected by whether they can access necessary social care provision. Thus, access to social care provision is a conversion factor for the independence capability theme. The question ‘Is this important in itself or is this important because it enables (or prevents) the young person to be and do the things that are of intrinsic value to them?’ was used during the last stage of the analysis (analytic account) to assist with distinguishing between capability and conversion factor themes. Still, some of the themes identified appeared to be both of intrinsic value to young people (capabilities) but also of value as they enabled (or prevented) them from being and doing other aspects (conversion factor) that were of intrinsic value to them. For example, being free from physical suffering was important in itself, thus a capability, but it also influenced the extent to which they could be free from emotional suffering. In the latter sense, being free from physical suffering acted as a conversion factor. In such cases, it was decided that it would be more important to capture the capability aspect of a theme given that the aim of the analysis was to capture important capabilities to inform attributes for a measure. In the rest of this chapter, capability themes and sub-themes, and conversion factor themes relevant to each capability theme, are discussed. 240 5.3.1 Experience and enjoy The Experience and enjoy capability theme is presented in this section along with two sub-themes: ‘having normality’ and ‘activities and socialising’, which are important aspects of this capability theme. Then two conversion factor themes, that of access to care and support and the impact of COVID-19, are presented in terms of their influence on the ability of the young person to have experiences and enjoyment in life. A recurrent theme to emerge from all three groups of participants was being able to live life to the fullest and make the most of everything. Living life to the fullest for participants meant them having as many experiences and as much enjoyment in life as possible: YP03: … I’d rather just forget about the negative stuff and embrace the positive stuff. I’d rather get on with my life, and live my life to the fullest rather trying to figure out how, figure out what’s wrong with me… BCP09: …She was just brought up to enjoy life around the medical (…) as much enjoyment as we could possibly give her. Opportunities, and outings, and special moments were pushed (…) She wanted to get out of the hospital and travel the world… Yes, she wanted (…) to see new places and do things. BCP07: … she was quite interested in going out and seeing and experiencing new places and having new experiences (…) she tried to carry on with everything that she could… Participants explained that it was important to have as many enjoyable experiences as possible because their life and their future was so unpredictable. In particular, the 241 experience of frequent changes in their health and condition and as a result changes and new limitations in the activities and things they are able to do pushed the young people to do as many things as possible in the present time: YP02: … when you are disabled and we all got a serious illness, life is unpredictable. The future is unpredictable, you don’t know what’s gonna happen, you could be fine one minute and maybe ill the next, and I feel like life is very short not to do it. And I feel like if I’ve got a chance then I should do it, take it. Not everyone has that chance … YP12: ...The reason I’m doing a lot outside of school is because my condition is so unpredictable. That’s even since I’ve been discharged, I’ve been going into hospital, in and out (…) There are so many times something can just happen, and I could be in hospital. When I’m well I just make sure I’m making the most of work in school because otherwise I’ll never catch up… 5.3.1.1 Sub-theme: Normality Part of having as many experiences and as much enjoyment in life as possible is about making experiences and life feel as ‘normal’ as possible for the young people. For most participants a ‘normal’ life means a life beyond the medical, that the young person is able to enjoy as a young person, without the limitations and struggles that they experience: YP08: …We talk about normal stuff. Like, outside of medical stuff… 242 YP12: …I just really want normality, at least try and go outside, try and do different things, getting the option (…) I’m spending most of my life now in the house… PG05: … she’s got her own house, it shouldn’t be clinical, it should be a home and not a clinical house. She has lots of sensory lights and music and TV, all the normal things. It has to be like that… BCP09: So (…) not making it all about medical, and trying to, like, “We're in this crap situation, but let's do something fun”. Like if we've got a day where we are in the hospital but you're able to go out, “let’s go out and do something”… BCCP13a: … just feeling- I don’t want to use the word ‘normal’, but I can’t think of another word for it. Just the normality of the situation, singing and dancing… Having some normality was particularly important if the young person spent long periods of time in hospital: BCP07 : …When we were at [hospital name] there were a couple of nurses that took [daughter] (…) to a [coffee shop], so she could have an iced drink. [Daughter] actually cried, because it was a taste of normality (…) it felt like there was life outside that hospital bed. Because you do become quite institutionalised, particularly if you are in hospital for a long period of time. That is your world then… BCP09: …It was all about making memories, and making things as normal as possible for her, and just making the best out of the situations that we were in, like we only had about three holidays of our own in her lifetime. Most of our holidays we actually made out of [name of hospital] stays ... However, there was an exception to what normality meant in one of the accounts. A parent participant suggested that ‘normality’ for their child is what they currently 243 experience. For example, having a carer with them at all times; and explained that this is because this is how they grew up and what they had experienced their whole life: PG04: …this life is [son]’s normality, isn’t it? He doesn’t know any different. He doesn’t know that not everybody has a carer that comes with him to school. He doesn’t know that not everybody goes to bed and then someone comes and sits in his room all night awake…Having carers with him all the time is a positive thing for [son] because that’s part of his life. That’s what he’s always had. That’s what he expects… 5.3.1.2 Sub-theme: Activities and socialising A second sub-theme that emerged was around activities and socialising. Activities that are enjoyable and fun to the young people vary. Accounts included social activities such as sports, arts and crafts, learning and going to school, meeting and making friends, going on holiday, but also activities young people enjoy doing on their own such as reading, writing and listening to music: YP01: ... speaking, seeing friends (…) explore with friends with different people around the world…’ YP02: …I like reading, like I’m obsessed [giggles], I like my films too (…) I love hearing new songs like coming across the new songs and listening to that…’ PG01: …education is a big part for him because it means he can meet people and speak to people … PG05: …She loves being on the bus and loves the journey… BCP06: …. She just liked chatting and singing and just being silly (…) She looked forward to all the volunteers coming into the hospital (…) And she was able to leave her room and (…) find a friend to talk to. Or during the night she would 244 continuously call for the nurses, so she could have a chat and a midnight snack with them… BCP09: …Anybody that she met (…) if they had an accent, she was very interested to know where they came from. She was very interested in languages and where people originated from, she was very eager to learn… An important aspect of activities young people particularly enjoy is social interaction and spending time with friends. This social aspect of activities was more prominent in parent accounts, with parents suggesting that unfortunately young people often felt lonely and isolated from others and especially their friends. This was due to not feeling well enough to join in with friends but also due to friendships fading out along the way: PG05: … she doesn’t like being alone. So as long as she’s with people. If you have a party (…) she’s happy because there are noises and people and things happening… BCP07: … she felt very isolated. I think partly because she had been feeling ill for so long before that she hadn’t been able to join in with friends and parties and socialising… BCP13: ‘…[when she got unwell] she found it very isolating. Because soon, her friends disappeared (…) it made her world very small… 5.3.1.3 Conversion factor relevant to ‘Experience and enjoy’: Access to support 245 The more opportunities for activities available to the young people the more experiences and enjoyment they can have in life. Participants described situations in which young people often missed out on opportunities that were important to them. This was sometimes partly due to not feeling well enough to take part in an activity but often it was due to barriers around accessing activities. This included, for example, not being able to access an activity because of lack of funding for practical and additional support needed or lack of reasonable adjustments provided so that activities are suited to the needs and abilities of the young people: YP12: …I can’t even go for a basic walk (…) I could but I need to get wheeled because the pavements are just so hard to get round the pavements around my house… PG01: …The local council, saying they haven’t got enough funding, or they’ve got no this and no that, so for example he wants to learn music. He can’t just go into a music college and enrol there because he has to have help at the college so that’s money, that’s extra person, extra budget and it’s kind of, people not able to give the money and services that he needs, which is quite disgraceful … PG03: …the buildings [in current school] are adaptable for him to get around and I think it is more of making sure they are included, in sports [son,17] had to stand there and wave a flag in his mainstream secondary school whilst everyone else was doing sports, why should he have to do that? why could he not do the sports?... BCP13: …We found the school wasn’t helpful …Because she was already studying for her GCSEs. They literally handed me a book on…Because she was too ill to regularly attend school, and within a month, she couldn’t attend school, at all. They literally handed me this book on English, maths and science, and said, “Teach her yourself.”… BCP18: … having access to services that meant she could maybe through college meet other people… 246 Difficulties with accessing opportunities was a strong theme and participants, particularly parents, expressed frustration with how this limited opportunities for young people to enjoy life. Many parents talked about feeling like it is a constant battle they have to ‘fight’ themselves so that their child could have the practical and/or financial support they need to access opportunities that are important to them. The ‘fight’ theme was also prominent in parent accounts around access to necessary healthcare for their child so that their child can avoid further physical suffering. Quotes relevant to access to healthcare are discussed later in this chapter: PG01: …there’s a lot of children with complex needs that need the additional funding and is never there, we are just left there to do whatever they can or they are left for parents like us to just fight, fight, fight... PG11A: …it's always been absolutely abysmal. So, you fill in a form for a blue badge, it lasts five years, you have to fill in another form even though this child is never going to recover from what's wrong with her (…) It's the same for when you apply for any club, you apply for school (…) all these forms. PIP forms, as well, it's designed to put you off. It feels like it's a gatekeeping exercise… BCP13: … we kept pushing [the school], and they said, “Well, we’ve got this unit.” But we had to do it ourselves. We didn’t actually find- She wasn’t assigned anybody from anything. So, we had to look, ourselves. We had to find the courses and different people that would help her, ourselves… 247 5.3.1.4 Conversion factor relevant to ‘Experience and enjoy’: COVID-19 COVID-19 had a significant impact on the young person’s social life and limited the opportunities to have enjoyable experiences. Many parent participants suggested that, following lockdowns periods when social distancing restrictions eased and schools or hospitality was open again, because the young people were in hospital and/or ‘clinically extremely vulnerable’ they continued with periods of shielding which further limited social opportunities for an extended period of time: PG01: …he used to go out quite a lot and eat food in different restaurants and stuff and we can’t do that anymore. So he used to go out, meet people and eat with them (…) socialising was a big thing for him… BCP06: … during COVID it was really hard for her, because no-one- Because when she was in hospital she was fine as long as there were people around her. As long as there were visitors, (…) I was there all the time, so she got bored of me. She wanted other people to be there. (…) it was not possible even for her dad to come and see her… Some of the families however made the decision to stop shielding so that the young people could return to the activities that were important to them: PG05: …I wouldn’t restrict her quality of life. I’ve sent her to college (…) and other people might not do that, they might say that’s risky. But they are there with PPE in small groups and quite careful. So I’ve sent her to college, because I think that’s really important for her… 248 PG11A: …She's fully vaccinated, we all are, so we've just taken the decision that she needs to go to college. It's much better for her mental and physical health to have some social interaction… 5.3.2 Independence The independence capability theme is presented in this section. There were two important aspects identified in relation to the nature of this capability theme: ‘not relying on others’ and ‘having privacy’. The conversion factor theme on access to support and care is then presented in relation to how access to social care services and support facilitates the independence of young people. 5.3.2.1 Sub-theme: Not relying on others An important aspect of being independent was around being able to do things by themselves without having to rely on other people, including family and HCPs. This included being able to do activities of daily living (ADLs), looking after themselves, attend important medical appointments or treatment and go to places: YP01: Can’t do what I want by myself… PG03: … he still wants to do things himself but he doesn’t want to have a carer that makes him feel that he is not capable of doing anything… BCP16: …she wished to walk by herself, and do some of the work by herself. Like, eating by herself, making her bed, and going to the toilet by herself. Also 249 washing her clothes. She wished to do those things, but because of her condition it didn’t allow her… Being reliant on other people, and especially their family for ADLs, had a wide impact on different aspects of life for the young people: YP02: … I can’t get my own drink or food so that is how it can affect me. Sometimes I feel like I’m being needy, when if I could get it myself I would… YP08: …I've been completely reliant on my parents because I can't do it [travel to medical appointments] myself... YP12: …I hate getting wheeled, I just hate it. It’s like someone carrying your whole body, someone carrying you … Parent and bereaved family member participants also suggested that being able to be independent in daily life was particularly important to the young people because they should be able to have and enjoy that independence that other young people at their age are able to enjoy: BCP13: …Well, she couldn’t do anything, she couldn’t go out with friends. Initially, until she started pacing herself out, it was difficult for her to make herself food. You know, live an independent life, like a normal 15-year-old would. It was very debilitating… PG03: … they [social care services] don’t want us to have independence, because they say you can only have so many [social care package] hours, but what choice has he got when he wants to go out or he wants to go and do something it just seems like (…) they are not allowed to have a life like an eighteen year-old lot person has, can. All the eighteen-year-olds are always out they are not always at home… 250 BCP18: …she was almost sort of left in a place where she wasn’t able to always look after herself as much as she wanted to be able to. (…) she really wanted to be able to have that independence. Especially when she became very unwell in the months before her death, she wasn’t able to do much at all … 5.3.2.2 Sub-theme: Having privacy A secondary theme to emerge was around ‘having privacy’. While it was important for young people to have access to practical support and carers if necessary – being able to spend less time around carers and HCPs if their care was not needed and to be left alone if they wanted to – was also important. This was particularly important when in hospital and/or when young people needed regular care interventions or observations by HCPs: YP12: It’s like everywhere. In school I’ve got someone with me all the time, coming around, having to carry my bag, getting me around the doors, if I need to go to the toilet and stuff, I need different things because of my condition (…) Every single day I’m with someone… BCP06: …they wanted her to gain independence [following tracheostomy decannulation]. So, there wasn't so much urgency care needed (…) it was time to build on her independence and actually let her go cold turkey without a carer, which she absolutely loved… BCP09: … she used to get quite angry with [the nurses] because she was like, “Why did you come here and turn that light on?” and, “Why do you keep doing blood pressures on me all night?” She didn't understand that they have to do them to a certain… Like hourly (…) She was like, “Why can't you just leave me alone?” … 251 5.3.2.3 Conversion factor relevant to ‘Independence’: Access to social care provision While young person participants focused more on wanting to be able to do daily activities by themselves and to be less reliant on others, parents and bereaved family participants also talked about the factors that enable or limit the young person’s independence. They suggested that it was important for their child to have access to social care provision, such as house adaptations, that will facilitate mobility and improve ADLs without having to be reliant on others. Unfortunately, it was often very difficult to access necessary social care provision from the local council: PG04: ‘…He likes being independent in his chair. He’s got a power chair that he controls by himself. The [school] grounds are massive and there’s a lot of to-ing and fro-ing between areas. (…) He likes driving there himself. He just likes being independent, but in an environment where it’s safe to be independent… BCP18: … [have] access to care so that she could become more independent…. PG03: …House adaptions, it is absolutely awful out there (…) They [social care services] are trying to do things which are not suitable, and (…) [son] had to move out of our house and live with [his grandparents] in their living room with his bed, because he couldn’t live in our house. Couldn’t go to the toilet because the house wasn’t adapted to go to the toilet…. 5.3.3 Free from physical suffering The capability theme of being free from physical suffering is presented in this section. ‘Physical symptoms’ and the ‘impact of the condition and symptoms on daily life’ are 252 presented with regards to the nature of this capability theme followed by the conversion factor theme around access to necessary health care so that young people are able to be free from physical suffering. 5.3.3.1 Sub-theme: Physical symptoms Managing pain and distress was initially a theme that included both physical and emotional suffering. This was because both themes were talked about together by most participants and sometimes it was difficult to distinguish between the two especially when participants talked about the impact of physical and emotional wellbeing on other important aspects of the young person’s life, such as enjoyment and being independent. However, it was important to distinguish between physical and emotional suffering as the ways that physical and emotional suffering are managed differ and young people may not feel that both physical and emotional wellbeing is managed, or not managed, to the same degree, and they may not be of equal importance to them. There is a wide range of LLC/LTCs affecting CYPa. Although it was not possible to capture experiences of young people affected by all LLCs/LTCs in the qualitative work, the interviews included a broad set of LLCS/LTCS, including congenital, fluctuating, a The Directory of LLCs includes a list of around four hundred ICD-10 codes associated with LLCs in CYP. 253 progressive and non-progressive conditions with different illness and death trajectories. All the young people had complex health care needs and/or conditions that caused severe or multiple disabilities. Some conditions were rare and/or there was no formal diagnosis. Physical symptoms of conditions varied significantly but included: pain and soreness, exhaustion and sleeping problems, gastrointestinal problems, breathing difficulties, visual, hearing and/or mobility impairment. BCP06: …her skin was really puffy where she was holding on to fluid, and it was really sore.… BCP07: …She had been complaining about feeling very tired. She suffered from migraines. And (…) she would get quite breathless… BCP09: ‘Quality of sleep… extremely difficult because she had overnight oxygen during the night and she had [gastrointestinal symptoms] … 5.3.3.2 Sub-theme: Impact of condition and symptoms on daily life While in bereaved parent accounts details of physical symptoms were more prominent, young people and parent participants talked more about how the young person’s condition and symptoms impacted their daily life and wellbeing. Symptoms often limited what the young people could do independently, how much they could enjoy activities and as a result had a significant impact on their mental wellbeing. Young people often struggled with feeling sad, worried or angry with how their condition limited the things they could do: 254 YP02: …I worry about getting ill and getting a chest infection because I am prone to them and it does make me nervous, a lot, (…) also (…) I’ve been having trouble with my swallowing (…) and so I have not been able to eat some of types of food, that made me really sad …definitely the medical side can get me sad. It’s some things like that I can’t do (…) like go play basketball or go to a theme park, that can make me sad too… YP03: … half the problem is that anger feeling, hospital appointments aren’t the greatest thing, if there’s a change in your condition it’s not nice, because that’s not what you wanna hear… During periods that the young person’s physical health was well managed, and they could do the things they enjoy, then their mental wellbeing also improved: PG05: …But I think she can only do all that [activities and socialising] when she’s well, if that makes sense. If she’s poorly, then she’s just quiet and unresponsive… 5.3.3.3 Conversion factor relevant to ‘Free from physical suffering’: Access to health care provision Having ways to manage pain and other distressing physical symptoms was very important to young people. Pain control, medical equipment that young people could use at home, such as non-invasive ventilation, having a care plan that is being followed, access to necessary health and social care support and having regular appointments with HCPs were some of the ways participants suggested young people 255 were able to manage physical symptoms. Managing physical symptoms was important for the young people’s mental wellbeing, so that they were able to enjoy life and to do daily activities as independently as possible. Managing physical symptoms was also important so that the condition did not progress and/or to minimise the need for hospice or intensive care or an emergency hospital admission. Thus, managing physical symptoms was important so that the young person could avoid further suffering: YP01: …go see doctor regularly… it’s keeping me safe and alive… PG05: …I think pain control is really important and keeping her medically well is really important. If you do that, she can kind of respond…’ BCP06: …she had a particular care plan to do with moving her in the bed and looking after her skin... BCP14: …the family doctor, who was always there, whenever he was feeling pain…” PG11a: … we don't get to crisis point. We manage her [chemical that body produces], we manage her [glucose levels] and she doesn't need acute, critical care because of how she's looked after… PG11b: …the lack of respite provision and the lack of money to give families respite, is a very short-sighted policy because the children and the families go into crisis and they end up in permanent care, which is far more expensive … While young people struggled or were frustrated more with how their condition and symptoms impacted their daily life, parents were more frustrated with how lack of access to necessary support and care impacted the young person’s daily life and physical health. As with limited access to opportunities discussed in section 5.3.1.3, many parents talked about having to ‘fight’ to access health care so that their child can 256 have the care they need to avoid further suffering. This need to ‘fight’ for what they need was more prominent during transition to adult services, a period during which young people often had care and support gaps: PG03: … I will do my hardest for [son,17] to fight that, but why do we have to keep fighting? (…) They [social care] just said that I don’t need it. So, I had to fight and when he got to fifteen, I had to fight to get a care package (…) on me (…) but they wouldn’t do a care assessment on [son, 17] and what needs he needed. … PG04: …Trying to access equipment and vital things for your child. You’re always met with an assessment to basically see if your child has the monetary worth to get it, or not. It’s always a fight. It’s a fight for everything (…) [but] my concerns shouldn’t be his concerns. Any shortfall in care is picked up by me, so I’m hoping that [son] doesn’t feel vulnerable. He’s never been without care because I’m always there… PG11A: … that was an absolutely appalling thing to go through [transition to adult services]. (…) we were there [transition multi-disciplinary meeting], and people from the NHS and people from social services. They basically had a fight not to pay for the person. (…) So, they sat around the table and they bullied us into a corner saying, "There's nothing wrong with this, she doesn't go to an emergency, to hospital, so she's fine”… BCP09: … if you've got a problem, the only way of getting it fixed is to keep on repeating yourself like a broken record. Like, if one doctor doesn't fix it, you have to keep on at every doctor that comes to you. (…) until it's done… BCP06: …always just a constant battle … 5.3.3.4 Conversion factor relevant to ‘Free from physical suffering’: COVID-19 257 COVID-19 had an impact on accessing necessary health care to manage symptoms and health. Parent participants talked about health care treatment and medical appointments that used to be provided to their child before the pandemic but during the pandemic these were stopped or became less regular. This was due to a reduced availability of services during the pandemic but also due to an attempt to minimise risks to young people who were clinically vulnerable from getting infected with COVID-19: PG01: …he has got about ten-fifteen consultants, so he has regular appointments, but they’ve been cut down because of covid… PG03: … it just seems like Covid had a knock-on effect because all of [his] consultants stopped seeing him so his condition had deteriorated and there was nothing we could do about it because he couldn’t see the consultants so he couldn’t have medication (…) and he deteriorated more… PG05: …She used to get, once a week, hydrotherapy for a term, but obviously, you can’t risk it with the pandemic… 5.3.4 Free from emotional suffering The capability theme on being Free from emotional suffering is presented in this section. The sub-themes with regards to different aspects of the capability theme that are important to young people are: ‘expressing emotions’, ‘distraction’ and ‘helping others’. The conversion factor theme around access to support is then discussed. 258 All young people and most parents and family member participants talked about emotional wellbeing and suggested that young people often struggled with emotional distress, feeling worried, sad, frustrated or angry, overwhelmed and at times helpless and hopeless. Having a serious illness, often with an uncertain trajectory, spending long periods of time in hospital and the physical symptoms of the condition impacted their daily life and their emotional wellbeing: BCP06: …. She was feeling very sad a lot of the time … BCP09: …She was really down and just sick of being stuck in there [hospital]… BCP17 describes how being severely sight impaired made their child anxious and scared when walking: BCP17: … it made him very anxious, he couldn’t walk alone. He was always holding on to somebody, usually me. He was frightened. People would run at him or something, and he wouldn’t see them until they were right in his face. So, it must have been a very, very scary world for him….” 5.3.4.1 Sub-theme: Expressing emotions Participants talked about the different ways that young people try to cope with distress and emotional suffering. Some young people found it helpful to talk to others about how they feel – especially if they are struggling – but also to talk about other 259 things going on in their lives as well. Talking about how they feel to others and sharing their worries with others helps them offload: YP01: …speaking about things I’m struggling more … YP02: … I talk about what is going on in my life, I go to [Hospice name] (…) so they like to catch up on what’s been going on in my life, they talk about everything…boys, clothes, films, (…) Sometimes if I am feeling down they will ask me ‘What’s wrong?’ and see if they can think of stuff to cheer me up or help me out. They are very good at that…. Similarly, a parent participant wondered whether her son’s inability to express his thoughts and feelings (being non-verbal) meant that he may not be able to have an outlet to release his thoughts and feelings. PG04 :… It’s sad to think that he has no outlet really to let anybody know his innermost thoughts and feelings because he does comprehend so much. There must be a lot going on in his head that he just can’t get out… 5.3.4.2 Sub-theme: Distraction Some young people also suggested that having distractions helped them cope with difficult emotions and the difficulties they experience: YP03: … [listening to music] helps me relax and make playlists and things like that, contrasting it to thoughts on a daily basis, because it makes me forget about certain things (…) because things can get overwhelming sometimes, so I just like to let my mind loose and do what I like to do… BCP09: …[In hospital] they brought these virtual reality goggles to her, and she was able to watch, an underwater world, so basically making them feel like they're not in hospital and they're having some sort of outing. I think it could be 260 so beneficial to so many people because it gives you the feeling like you're in a different place… 5.3.4.3 Sub-theme: Helping others Some young people suggested that sharing their worries and feelings with others was also about helping others who are struggling too. YP08 has joined an online chat group where young people with LLCs and LTCs around the world can share their experiences and talk to each other. YP08: …Because it's just someone that we can offload to and vice versa, like, they'll message me because they've got a scan coming and I can talk to them about the scan and stuff like that… A young person participant found it helpful to write down about how he is feeling and his experiences living with serious illness. He wrote lyrics for songs he then shared with others online who listen to his music. He explained that this was helpful mainly for himself but also to help others who may be struggling with similar experiences: YP12: …lots of the time I just write how I’m feeling. I’d write the song completely how I’m feeling at the point (…) maybe hopefully it works with others and people maybe feel the same but it’s mainly for me, a calm feeling. It doesn’t always have to be about my time in hospital. It can be about different things…’ 261 For some young people helping and giving to others or trying to make others happy or laugh when they were around was their way of feeling useful rather than feeling like a burden to others: BCP06: …So if there was anything that she could do herself then she would. She wanted to help. (…) Like she always asked the nurses if she could help the babies and go round and check the babies with them. Or help make the beds or do this or do that. She wanted to just help where she could (…) It was important for her for everyone to be happy…. PG04: …He likes to make people smile. He likes to make people laugh…. BCP13: … some were in wheelchairs, some had other health problems, that weren’t always visible. But it did give her that look into helping people. And she actually helped people along the different lessons. Because they’d all fallen behind, when they were doing their GCSEs… Another reason why young people said it was important to them to help others – particularly those with similar experiences and difficulties – was because this was part of who they were as a person. This is discussed in section 5.3.7 under the Identity theme. 5.3.4.4 Conversion factor relevant to ‘Free from emotional suffering’: Access to informal support Being able to have emotional support from family and friends, and in particular from people they could trust was valued by young people. Having this support had a positive impact on their emotional wellbeing. Close family was the main source of consistent 262 informal support for many young people. Unfortunately, in many of the accounts, participants mentioned that young people were let down by friends and lost trust in them, especially at times when they were very poorly and needed them to be there the most: YP02: ‘…The most important thing to me right now is probably family. Because family are always there for you and they are a great support system that you can have… (…) [they] support me when I am feeling very down, cheering me up (…) and if I’m ill as well they are always there to look after me… YP12: …I never really managed to get anyone [friends] and I couldn’t really trust anyone throughout my whole time because no one would speak to me (…) It’s about people spending time, dedicating time to you, not just coming over and speaking to you for the sake, “He’s in a wheelchair. I feel bad for him.” You speak once every two weeks. It’s also building that trust with the other person… BCP07: …Her friends, unfortunately, really didn’t come through for her, and she felt quite let down by them. In the end all she wanted to see was family. And only a limited number of them. So she didn’t really find too much support outside of that group really…. BCP18: …When she had to leave school, a lot of the friends that she had didn’t really stand by her or support her, so she lost a lot of friendships when she became unwell… 5.3.5 Formal care and support The capability theme on formal care and support with the sub-themes regarding the nature of this capability theme are presented in this section. The sub-themes are to ‘be cared about by someone who cares about me’ and ‘building trust’. The conversion factor theme presented is around continuity and consistency in care and support. 263 5.3.5.1 Sub-theme: Be cared about by someone who cares about me Participants talked in detail about the young people’s experiences of formal care and support they receive or need from services. This is not limited to health care services and includes social care and assistance in educational settings. What was important to young people in terms of formal care and support was that people involved in their care – such as doctors, nurses and carers – cared about and were able to spend time with the young person to make sure that they had what they needed: YP12: ‘… They [hospital discharge team] just fling you in the deep end really (…) They just want to sort stuff in the house and that’s it, “Right, get a bed at home.” I didn’t even have good access to my toilet but decent access to my toilet. That’s it, just a room and a bed. That’s all they care about. The rest you do it yourself… YP08: … They're all [medical team] very personable. That makes a difference. I feel like they remember me when I come in. I'm sure they read my notes before, but it doesn't feel like… I feel familiar with them all (…) I feel very attached to all of my medical professionals. But that helps because obviously you trust them with a lot of stuff… BCP09: …People [in hospital] that were willing to give her the time, and sit down and spend a little bit of time with her in talking about things and getting to know her. Yes, just having the time, really, to actually find out about her … BCP06: … knowing that people [in hospital] cared and that people were there… 264 5.3.5.2 Sub-theme: Building trust A strong theme to emerge in formal care and support was around trust. Participants suggested that because the young people were so vulnerable, and/or they had to spend long periods of time in hospital with HCPs it was particularly important to the young people to be cared for and supported by people they could trust, possibly to the extent of feeling friendship towards them: YP02: some of them [hospice nurses] are very protective of me (…) cause you can’t just trust anyone …I think you have to with this disease…you have to learn to trust a lot. Because your life is basically in the hands of someone else and it is very important to be able to trust that person… BCP09: …It has been her life since she was born…She has always been brought up in that environment [hospital], so it was about finding new friends for her in the nurses that she knew that she could have a laugh with and trust... Parents also suggested that it was not only important to the young person to be able to trust people involved in their care, but it was also important to the parent too: PG05: …I would find it difficult to put an advert in the paper (…) and say, “Carer needed.” It would have to be someone I knew and trusted. You have to be able to trust someone and know them. She’s very vulnerable…’ PG04: … you’ve got to be confident. You’re leaving precious cargo. (Laughter) You’re putting all your faith in somebody who you don’t know and that’s really difficult… 265 5.3.5.3 Conversion factor relevant to ‘Formal care and support’: Continuity and consistency in care and in transition to adult services Parent participants talked about the importance of having continuity and consistency in care and support so that the young people can be cared for and supported by people who know them and that they can trust. This was particularly important during transition to adult services when young people had to get used to a new care team and a different hospital environment. In the accounts, many young people had had many different care staff and medical teams involved in their care throughout the years. This often had a negative impact on the young person when staff who were new to them needed time to get to know the young person and their individual preferences and needs: BCP09: …They’d [in child services] then assess her and then get the doctor – the paediatrician that knew her, 9 times out of 10 – to see her. Or that would be somebody beside him who knew her already, whereas, being transferred into adult services, it was then a case of taking her to A&E, and then having to go through everything with the A&E doctors… BCP06: …the nurses that really knew her and the nurses that had just met her. You could tell a massive difference.(…) when we were on the ward and she was like this they would still get her cleaned up, freshened up, put on her favourite music, still talk to her like they should have been. But we did not get that when it was other staff looking after her… PG04: … we then have to leave him with complete strangers even though there’s two of them. Sometimes they interpret his needs differently (…)There just needs to be some continuity with staff (…) [In transition to adult services] we lost all his child team and we were introduced an adult team. So it was really 266 difficult because some of his carers had been looking after him since he was nine months old (…) then all of a sudden they weren’t there anymore… PG01: … his hospital transition from child to adult… that’s happened and that was horrible. That was not really a good experience and sometimes it isn’t still (…) so he was happy seeing certain doctors all his young life and now he has to see different doctors he is trying to get used to them… PG05: … lots of her carers have known her for years and years (…) I know them [carers] and I know their backgrounds and they’ve all been careful… So, I haven’t had to worry that much about it, the fact I know them so well … 5.3.6 Control Another strong theme to emerge from the interviews around care and support was about young people feeling in control over their health and care. Feeling in control was not limited to being able to manage physical symptoms and being able to be independent in daily life, it was about being able to feel in control. The capability theme of Control is presented in this section. Aspects of Control were around being able to ‘have a say, be listened to and make choices’ in care and support. The two conversion factors that influence the young person’s ability to feel in control over their health and care, communication with services and coordination between services, are then presented. 5.3.6.1 Sub-theme: Have a say, be listened to, and make choices 267 For young people feeling in control of their care and health was about being able to have a say, be listened to and make their own choices for their care: YP01: …being able to do my own choices… BCP06 : …She wasn’t happy about being in hospital (…) She had had enough of it. But she knew she had to be there. So they just had to do things her way and that was it… BCP09: … she had control of her treatment. That was her who was in control to receive it, it wasn't… Like I said, “You can't come in here and rush it. This is something you have to take notes from me. You have to listen to me. You have to listen to her.” (…) It was just giving her back some control because everything was so forced. Like, she had no control over what she needed to receive, so, yes, it was just giving her some control over the situation… Unfortunately, some young people felt like they were not being listened to or they were not able to have a say as much as they wanted to. A young person participant explained that when they turn 18 years old, they will be able have a lot more say in what happens in their care as they will be adults: YP03: … because I wanna make sure that my existence and interests, as much as anyone else thinks what I need is best, then sometimes it’s actually helpful to ask the person who needs the help (…) I just wanna make sure that I am there to say what I want as well (…)I’d like it if there was more of me having a say in things. Because I’m not 18 just yet when I turn 18 I’ll be able to have a lot more say in what happens… Another young person found receiving their diagnosis helpful as before that even though she was unwell, she did not feel that services were taking her seriously: 268 BCP07: …From [daughter’s name]’s point of view, I think in one way she said, “At least I know now it is not in my head like everyone kept telling me.…’ Similarly, BCP18 suggested that her sister was not being listened to when she was unwell towards the end when the medical team did not know that her condition was serious: BCP18: …she was calling them [doctors] because she was very unwell towards the end and she was trying to get support for herself because she knew she wasn’t very well. (…) she felt that she wasn’t really being listened to, that she tried speaking to them and then she was given some care at a hospital but they reassured her there's nothing seriously wrong… 5.3.6.2 Conversion factor relevant to ‘Control’: Communication in care and support Participants talked about what hindered or facilitated their ability to feel in control of their care. Good communication between the family, including the young people, and health and social care services, are important so that young people can stay in control of their health and care. Both young people and parent participants suggested that it was important that health and social care services provided them with adequate information about their condition, care and treatment options. This included not only being given the 269 information, but also that the information was communicated in lay language that the young people and the family could understand and that there was a chance to ask questions: YP12: …I’ve got so many questions to ask in terms of the medication side (…) I think there are still so many issues with my health, I don’t have time when I speak to the doctors. I’ve got so many questions… PG01: …Some [doctors] talk big language that he doesn’t understand it so it’s (…) a barrier in the way he understands and that does kind of confuse him because sometimes it’s like “what are they talking about?” and he will say that “I don’t understand them…” they “don’t talk my language” (…) he always needs to ask twice or three times to kind of explain it slowly to him… BCP07: …we didn’t always understand the information we were given (…) There were terms that were used that we had never heard of before. (…) And we thought, “Well, why is that important? What does that mean?” (…) Sometimes you can be given information and you think, “You have just given that to me. I don’t know what to do with it now. Can we change it? Why is that important for us to know?”… A young person participant also talked about an online phone app that her medical care team uses so that people under the care of that team can access all the important information about their health and care if they wished to. She found that this way of accessing information when she wanted to helped her feel more in control, particularly because she had to attend several different types of medical appointments: YP08: …Everything is there. All my stuff is really accessible, all my data, my appointments (…) What's coming up in the future, you can see who I'm seeing. It's a much more, like, everything's organised for me (…) making you feel a bit more in control of stuff. I would hate to be in limbo (…) You have so many blood tests, so many scans, it’s difficult to stay on top of it…. 270 There were also exceptions to this. Some young people did not find being given detailed information about their health and care always helpful. Feeling in control does not always mean that the young person wants to be given all the information but rather that the information is for example communicated via others, such as the parents, or at the right time or in the right quantity. For example, BCP17 explained that her son found being given detailed information – such as what the HCP was about to do – overwhelming and worried about what was about to happen to him. The parent explained that because her son was autistic, the process of explaining to him what was going to happen made him more anxious: BCP17: …if the nurses or the doctors did what they needed to do and got all the stuff ready and then called him into the room and literally just took his blood or did whatever was needed, he would let them. (…) Whereas, if they’d try and talk to him and try to tell him what was happening and he would just get really anxious (…) It didn’t work at all well for him, at all. If he’d had control, he wouldn’t have had any bloods done, or scans or anything…. However, BCP17 suggested that her son found attending regular medical appointments where he knew what was going to happen less stressful than when having a long break in between appointments. It was thus important for her son to be in control of his care but in a different way to other young people: BCP17: … Although he would get very anxious about things, he did know what was happening and why and how, whereas if he had a long break in between it was almost as if he would have to reset and start again… 271 Another example was a parent who suggested that being given lots of information around PC was overwhelming for the whole family, who did not feel prepared for this. It was important to the family that HCPs checked with the family that this was the appropriate time to have that conversation and ascertain if the family wants to know, so that they have the opportunity to stop the conversation if desired: BCP07: …I can remember every time a doctor came they had this thing. They had to tell you everything. And sometimes you don’t want to hear everything. Fine, yes, you have got to make a decision, but sometimes (…) I would have appreciated being told things when I felt ready to. In fact, I think I did say, “Can you not tell me any more about that? Because I don’t know what to do with it. You are just telling me, and I can’t do anything about it. So I would rather almost not know.” And it was the same for [daughter’s name], although she did ask questions…’ In situations where the condition was rare or there was no formal diagnosis, there was a lot of uncertainty for the whole family. HCPs providing information about the condition and acknowledging that the young person is suffering was helpful. For example, for BCP13’s daughter having information about her medical condition was important so she was able to work out what was wrong and feel more in control of the situation: BCP13: …She googled and looked on medical journals and things, to try and see if she could work out what was wrong with her. (…) I think, had she have known [the diagnosis], she would have just been able to work around it and it would 272 have given her more security, and she could understand more what was going on with her (…) She’d have done everything she could… Even though the medical team involved in the care of BCP13’s daughter was unable to provide a diagnosis, her daughter valued the medical staff who were able to spend more time with her and acknowledge that even though they do not know what is wrong, they can see that she is in pain and that something is wrong: BCP13: …I think she really wanted to know what was wrong with her. Some of them were very, “Oh, I don't know what’s wrong, I’ve got no time.” And other people (…) were much more helpful to her, and they said, “Look, I’m sorry, I can’t work out exactly what it is. But it is relevant, I can see you're in pain, I can see there’s a problem. Just because we can’t put our finger on it right now doesn’t mean it doesn’t exist.”… BCP18 suggested that because her sister did not know how serious her condition was, she had less control over her situation and less chance to make important decisions about her care, for example around being proactive about her care or being able to ask the right questions: BCP18: … [Doctors] saying it can’t be anything serious, it gave [her] a false hope, feeling that she was going to be okay (…) because in one way it’s sort of like she didn't know how ill she was so it was almost like she wasn’t scared in that way because she didn't know. But then at the same time she wasn’t able to make a decision that she may have made differently. So if they’d have said that she could have something more serious (…) she may have been able to be more proactive in her care speaking to her doctors and knowing how to address her symptoms (…) So the time that she did have, she may have been able to have a 273 better quality time in that period which she would have known how to deal with it because she’d have understood a bit more what could be wrong…” 5.3.6.3 Conversion factor relevant to ‘Control’: Coordination in care and in transition to adult services Apart from good communication, good coordination between the different health and care services is also important so that young people can feel in control of their health and care. The coordination theme was more prominent in parent accounts. Parents often took a ‘care coordinator’ role to facilitate communication and coordination between the different services involved in their child’s care. Coordination involved a great degree of bureaucracy and completing multiple forms that duplicated information, in order to be able to access support and care from different services. In transition to adult services, there appears to be an increasing need for coordination by parents as the transition is not always smooth and there are fewer services available for young people once they turn eighteen years old. In addition, young people often do not feel ready, or are not able, to take a coordinating responsibility: PG05: … It was more to do with the duplication. It wasn’t what they [health and social care teams and school] discussed; it was more the fact you’d have to go through a Social Services form. And then you’d have to go through a Health form. And then you’d have to do an Education form. And they all took about two hours… PG11: …we have to fill in 42-page forms for every single thing we apply for and it's absolutely ridiculous… 274 BCP09: …[In transition to adult services] for patients that have a lot of problems and need a lot of speciality doctors, a key doctor or a key person would be really helpful to them because, especially with my little girl, she may have been 18, but she wasn't mentally 18. She wasn't ready to coordinate all the stuff by herself….I was having to coordinate everything, like her doctor's appointments, the hospital appointments, all her care, all her management, doing all her hospital admissions and all the paperwork… PG03: …because he is going to be turning eighteen, the support you get…in fact you lose all your support from the age of sixteen where it just seems that there is this gap, between sixteen to eighteen, because children’s services don’t wanna know them because, they are going to be turning into adults, and adults don’t wanna touch them because they haven’t turned eighteen… 5.3.7 Identity The capability theme of Identity is presented in this section. The sub-themes presented with regards to different aspects of this theme are: ‘maintaining identity as a young person’, ‘disability’ and ‘sense of belonging’. The conversion factor themes presented are: access to care and support and continuity and consistency in care and support. A final capability theme to emerge from the accounts was around identity. To some young people it was important to maintain their identity as a young person. That is, who they are as a young person, beyond the condition and/or disability. To others disability was an important part of their identity that formed who they were as a person now. In both cases, it appears that what is important to young people is to be 275 valued and included for who they are as a person and not to be judged or pitied by others because of the condition and/or disability. 5.3.7.1 Sub-theme: Maintaining identity as a young person There were some young people for whom their condition and/or onset of symptoms came later in life. These young people, having not had any serious health issues prior to that, talked about how much their life had changed and compared how their life was before and after their symptoms started. Part of this change included a change from who they used to be, for example ‘a runner, to who they are and are able to be now. For these young people it was important to be able to find ways to maintain their identity and personality following the onset of symptoms and condition, especially when they were very poorly or at the very EOL: YP12: … I did [enjoy playing football] but now it’s changed. I was a runner as well. I ran a lot. They were the two sports I did, football and running. I still make music now. I make music (…) I do play some wheelchair sports but I just hate every- it just doesn’t feel the same to me (…) It’s a good team. It’s just as much as I don’t want to be saying that, I just don’t like any kind of sport involving a wheelchair. I don’t find it enjoyable. I’m just doing it because I used to do sport so I just try and do some sport (…) I just don’t feel the same anymore really even though I am the same person... PG03: … he did not see anybody who is in a wheelchair because he doesn’t want to see himself…he wants to be around people who are walking and doing stuff. Because he doesn’t want to see himself in a wheelchair… BCP13: …She needed help with her hair, her hair was very long. And she wanted to keep her hair long, but it was difficult for her to look after. So, it was 276 important. So, I helped to wash her hair and things like that, so she could keep her personality… BCP06: …[Near the end of her life] they [HCPs] just saw this very sick child, just lying there, not particularly responsive, not able to move themselves, not able to talk, not able to do anything for themselves. And that wasn’t [daughter, 14] (…) She stopped becoming [daughter’s name] for all that time, and she was just a child that was very sick.(…) she lost that individuality. She lost who she was (…) So there needed to be a lot more understanding of the person that she is, not the person that they [HCPs] could see that is just lying there very, very poorly… 5.3.7.2 Sub-theme: Disability For some young people disability was an important part of their identity. It was important for them to inspire or help others, who like them, may be struggling with a condition or disability. This included using social media to tell their story and raise disability awareness to change the way societal views on disability: YP02 …I’ve been using my [social media account] to like spread awareness of disability, so trying things like being a disability advocate on [social media account] … I like that I can show people that you know we are here too, and we are not just the only ones. Because I feel like disability and disabled people do get pushed aside (…) just proving that disabled people are not all sad and gloomy…that is what people view us as in society, that we are all sad about it when we are not… BCP18: … because of her own experiences of being unwell and being limited because of that, that she maybe wanted to help other people who are going through something similar (…) She wanted people to live in an equal society and to have equal opportunities. I suppose part of it because with her education, 277 because of her accessibility needs it didn't always seem like she had equal opportunities… A bereaved parent participant also talked about her daughter’s identity changing over the years. When her daughter was younger, at the beginning of her teenage years she struggled with altered body image due to, for example, having a colostomy bag and a tracheostomy. It made her very self-conscious of her body image and she did not want to have further surgical procedures and equipment that were necessary to keep her alive: BCP09: …It got to a point when she was a teenager that she wasn't happy to have [further surgical procedures and medical equipment] (…) She got, as a teenager, quite conscious of all this (…) I was like (…) Don't feel like, ‘I want these off and I want to die,’” which is what I got from her in her teenage years. She felt very different and very, shy of them all… However, BCP09 suggested that towards her older years and near the end of her life, her daughter was more open about her health issues and wanted to tell others her story and help other people with a serious condition: BCP09: …A lot of friends [on social media] said that they actually stopped doing things or started doing things because of her and what she said. (…) [She] would message to them to say, ‘This is what I do,’ or, ‘This is how I do it.’ (…) She got very, very powerful towards her end-of-life, she was very like, “If I can do it, you can do it” (…) She did get very influential… 278 5.3.7.3 Sub-theme: Sense of belonging A strong theme to emerge from many accounts was around young people wanting to have a sense of belonging within their friendship circles and social networks. A sense of belonging involved young people feeling valued for who they are and not to be judged or pitied by others because of the condition or disability: YP12: … it’s always in the back of my mind if actually people take me seriously or they’re just pitying me (…) I’ll see when I keep releasing [music] that if people just like my music with pity or is it because it’s actually good. … YP03: … certain people think because I’m in a wheelchair I can’t do a lot of things, like people’s image of someone in a wheelchair is like they can’t walk, can’t talk, that kind of things (…) so that frustrates me when people think that because I’m in a wheelchair everything’s different – it’s not (…) It’s been all my life, like at school, whenever I go out certain people treat me different… BCP13: …she didn’t want to be invisible. Just because she was unwell didn’t mean she wasn’t relevant. And she had a value and something to say. And I think, because she had a walking stick as well, she was very worried she’d be judged for being young (…) And she was worried about being judged about that… Another important aspect in relation to belonging is around feeling included, especially by their friends and peers. Unfortunately, in the accounts many young people often felt left out from their friend groups and spend more time with their family: 279 YP12: …It’s just everyone at my age now are just going out with groups of people and stuff and I’m left out… PG04: …he’s non-verbal. He has a speaking aid on an iPad. At [mainstream] school, they were not very forthcoming with using it (…) A lot of his classmates [in the therapy-based school] also use the same app, so he’s encouraged to use it. Whereas at school, the majority were verbal and he was just on the side lines looking in really. Whereas he might feel more included, or less excluded.…’ PG03: … at the secondary school, the mainstream, they made him feel different and the other children didn’t understand anything about [son,17] or his condition, it’s like they tried to keep them separate and keep them at a little part of the school … BCP06: …he felt included as well. Because, although, you know, we weren’t all the same [in the family], he had all his issues, he could have a laugh, he could joke with us, he could just be who he was… BCP13: …She spent obviously an awful lot of time with us, if she wasn’t at college. Her friends had disappeared, they were going (…) and doing things and moving off, and she got left behind… 5.3.7.4 Conversion factors relevant to ‘Identity’: Access to age-appropriate services, Continuity and Consistency in care and support When young people turn eighteen years old, they transition from child to adult health and care services. Some young people transitioned to adult services sooner than others but in most accounts, participants describe a situation where the young people felt too old to be in child services and too young to be in adult services. It appears that this lack of age-appropriate services for this young person group has a negative impact on the young person’s identity and sense of belonging: 280 BCP07: … She was in an adult ward. And, needless to say, with it being a heart issue they were nearly all elderly. There weren’t any other young people in there. So that made her feel even more of an odd bod… BCP13: …[In child services] It was difficult for her sometimes, because she was in with tiny children, she was 16, 17 and there were little 3, 4-year-olds and babies running around, waiting for similar services and clinics. And she was sort of sitting there with little Disney characters drawn on the wall. She felt a bit out of place…. PG01: … the adult site is different to what the children site was. The adult hospitals have got no people to play with you, they’ve got no people like to take you to like play rooms or there’s not pretty stuff on the walls, it is just like a miserable hospital situation (…) Nothing carries on from the children life to the adult life… PG03: …And there are no hospices like that now for [son,17], the only places these chi…well these young adults can go to is care homes with the elderly or with dementia, and why is nothing been done? Why is that these services put a …for them? You know from eighteen to twenty-five? Because at the end of the day they are still children. You know he is going to turn eighteen, he is still a child (…) And it’s like they turn eighteen and they are thrown in the deep end … PG11B: …It's a struggle […] there are a lack of respite facilities, especially for young adults. If you're 50 and you've got dementia, you can find a place, if you're 19, 20 and you've got special needs, there's nothing…. A young person also mentioned that because other people in the UK who receive the same treatment and have the same rare condition are all older than her, it was important to her to connect with other people that have the same rare condition but are of similar age to her. She was able to connect with other young people around the world via social media. She explained that this was important because younger people are at a similar stage to her in their life: 281 YP08: …It's definitely a huge, beneficial thing to be able to have found them [young people around the world with the same condition]. I've met people here in the UK with it, but I think they've all been like [mentions people from older age groups between 50-70 years old]. And we still get on, but some of the challenges that we have are different, obviously, because they are at a different stage in their life to me. So there's still a bit of a limit (…)I've definitely found it easier to speak to people my own age, with my own condition… 5.4 Summary The findings from the in-depth interviews with young people, parents and bereaved family members were presented in this chapter. Seven capability themes and five conversion factor themes were identified as important to young people at the EOL. The capability themes were: Experience and enjoy; Independence; Freedom from physical suffering; Freedom from emotional suffering; Formal care and support; Control; Identity. The conversion factor themes were: Access to care and support, Continuity and consistency in care and support, Communication with services, Coordination between services and COVID-19. The capability and conversion factor themes identified will be discussed in Chapter 9 with regards to how these themes can be taken forward with future work to inform the development of a measure for young people at the EOL. Figure 2 below, inspired by Robeyns (2017)233 provides a graphical representation of the capability themes and the conversion factor themes identified in the accounts, and 282 how they relate. The arrows used in the diagram indicate the conversion factors that have an influence on the young people’s capabilities. Figure 2 - Capabilities and conversion factors identified Chapter 5 focused on aspects of life that are important to young people at the EOL while Chapter 6 focuses on findings on aspects of life that are important to those close to the young people at the EOL. 283 CHAPTER 6. FEASIBILITY OF USE AND CONTENT VALIDITY OF ICECAP-CPM Cognitive interview findings to assess feasibility of use and content validity of ICECAP-CPM in bereaved close persons of young people are presented in this chapter. The chapter starts with details of the characteristics of participants followed by the findings section. Types of errors and struggles identified are first discussed. The views of participants on the measure and their reactions during and following completion of the measure are then outlined. Errors and struggles identified, as well as views of respondents about the measure, are illustrated with representative quotes from the interviews. 6.1 Participant characteristics The first two interviews with bereaved family members were carried out in November 2021 and the total number of cognitive interviews was nine by the end of September 2022 when recruitment ended. Circulation of study information through bereavement 284 support charitable organisations, hospices and social media platforms resulted in seventeen initial expressions of interest to take part in the study. Reasons for individuals not being included in the study following expression of interest were due to not fitting the young person age range eligibility criteria (three) or being bereaved for less than six months (two). Three individuals who initially expressed interest to take part in the study did not respond to follow up email invitation for an interview. From the nine participants, three participants were recruited through a bereavement support charitable organisation, four participants were recruited via social media platforms – two through a Twitter post, one through a Facebook paid advert and one through a Facebook post in a local (Bristol) family support group page – and two participants were recruited through existing study participants into the study (snowball sample). Characteristics of bereaved family member participants from the interviews, along with characteristics of the young people they were close to, are summarised in Table 6. All parent participants were female/mothers. There were both parent and sibling participants and interviews included experiences of family members of young people with a range of LLCs/LTCs and illness trajectories, including rare conditions. As with the in-depth interviews for attribute development, the ICD-10 diagnostic group category under which each diagnosis fits rather than the medical diagnosis for each condition is reported in Table 6. A more detailed table with individual characteristics of interview participants was provided in Table 5 in Chapter 5. 285 Table 6 - Participant characteristics Participant characteristics Gender Male 2 Female 7 Relationship to young person Mother 5 Brother 2 Sister 2 Months since bereavement 6-11 2 12-18 5 >18 2 Recruitment method Charities 3 Social Media 4 Snowball 2 Ethnicity white British 6 black 3 Deprivation level b 1-3 (most deprived) 6 4-7 1 8-10 (least deprived) 2 Age group (years) of associated young person at death 14-17 2 18-25 7 Gender of associated young person Male 3 Female 6 Health condition of associated young person Congenital 1 Cancer 5 Circulatory 1 Neurological 2 b English Indices of deprivation, 2019 (https://www.gov.uk/government/statistics/english-indices-of-deprivation-2019) and Welsh Index of Multiple Deprivation, 2019 (https://statswales.gov.wales/Catalogue/Community-Safety-and-Social-Inclusion/Welsh-Index-of-Multiple-Deprivation#:~:text=The%20Welsh%20Index%20of%20Multiple,to%201%2C909%20(least%20deprived)) 286 6.2 Findings 6.2.1 Measure completion rating findings: errors and struggles. All nine respondents completed ICECAP-CPM. Eight errors, one struggle, and eleven possible errors or struggles were initially identified through the independent rating process. Following the consensus meetings with raters, from the eleven possible errors or struggles identified through independent rating, a further two errors and one struggle were agreed on. As each of the participants (n=9) completed all six questions on ICECAP-CPM, a total of 54 segments were created. Ten (18%) out of 54 segments were associated with an error and two (4%) with a struggle. At least one type of error or struggle was identified for six out of the nine respondents completing the measure. Most errors (five response errors) were identified in the think-aloud task completed by BCP06. Errors were from three different error types: comprehension, retrieval, and response with the majority, eight out of 10, being response errors. Only one retrieval and one comprehension error were identified. No judgement errors were identified in the measure. At least one type of error was identified in all the attributes of the measure 287 apart from the ‘Preparing and Coping’ attribute that was error free. Most errors (three) and struggles (one) were found in the first attribute on ‘Communication’ with all three errors being a response error. Information on the final number of errors and struggles against each of the six ICECAP-CPM dimensions is provided in Table 7. 288 Table 7 - Errors and struggles identified by type across ICECAP-CPM dimensions (n=9) Response errors were mainly due to respondents being unable to choose their desired response as this was missing from the response choices available. For instance, one participant responded to many of the measure questions by giving an in between response level answer: BCP06: …I would say between D and E. Honestly, most of the time I would say it was between D and E... The response error was most common for the ‘Communication’ item where the time period in relation to their close person experience was too long for them to ‘aggregate’ their experience over time and provide a single response. Communication at different time points and with different services varied. Suggestions from respondents included adjusting the question so that it specifies the time period and/or services the question Error type or struggle/ICECAP-CPM attributes Communication Privacy and Space Practical support Emotional support Preparing and coping Emotional distress Total Comprehension 0 0 1 0 0 0 1 Retrieval 0 0 1 0 0 0 1 Judgement 0 0 0 0 0 0 0 Response 3 1 1 2 0 1 8 Struggle 1 0 0 0 0 1 2 Total 4 1 3 2 0 2 12 289 is referring to and if the measure is used for experiences after death, then it may be appropriate for the measure to be adjusted so that it highlights the EOL period: BCP18: … if I was just going to call it just end-of-life, then I could answer [the communication question] based on just that. But if they want people to think about leading up to the end of life, maybe a couple of months before, it might be good for them to have an option. So like communication with providing care services perhaps as an option for different hospitals (…) and then you have the option to say how you feel the communication was for each hospital the child may have been involved in so that you can see how communication varies…. BCP17: …he had such a long journey (…) So you’d speak to one doctor and you’d have a really good conversation, and then another doctor, it would be like you were talking to someone in a different country (…) So I wouldn’t be able to answer that first question, because, some of the times it was fine. Some of the times there was no communication. Most of the time, maybe, yeah… I just wouldn’t be able to pinpoint which one of those I would really be able to say…. One of the two struggles was also related to the ‘Communication’ attribute where the respondent did give a response but found it difficult to respond due to having different experiences with communication across the different services involved in the young person’s care: BCP07: …So I would say in our situation being able to have good communication most of the time would be true. ‘Good communication’. Yes, I think it has varied about who you would be speaking to… A similar issue arose in relation to the ‘emotional support’ attribute, which had two response errors. As with communication with services, respondents had difficulty ‘aggregating’ their experience of emotional support from different people or services 290 to provide a single response to the question. For instance, one respondent suggested that they would give a different response to each of the two descriptors included under the ‘emotional support’ question, ‘being able to get emotional support through family, friends or colleagues’ and ‘being able to get emotional support through other services including charities and religion if applicable’, as each referred to different individuals or services: BCP17: … from family, friends, and colleagues, I’d say “Somewhat Able”. My family really don’t talk about [him] anymore so much, which is really heart-breaking. “Being able to get emotional support through other services, including charities”? I would say “Fully able”… In the ‘practical support’ attribute three types of errors were identified, a comprehension, a retrieval, and a response error. The response error, as with the previous response errors discussed for the ‘communication’ and ‘emotional support’ attributes, was due to difficulties of ‘aggregating’ experiences of practical support provided from different people or services into a single response to the question. The comprehension error identified was due to the respondent misunderstanding the meaning of the practical support question. Rather than responding to the question in relation to getting practical help with the care of the young person, such as having nursing help, the respondent only talked about information they were given at the hospital around practical support they could get, and not about actual support available or received: 291 BCP13: … they gave us some information (…) They sort of handed you the practical support, and then you had to go away and read it… The retrieval error identified in the ‘practical support’ attribute was because the respondent, who was a sibling, did not know the details around practical support, as it was her mum, rather than her, who was caring for her sibling most of the time: BCP15: …I wouldn’t really know because it was my mum that was basically doing all the stuff. I was usually at work… The ‘emotional distress related to the condition of the decedent’ attribute had one response error and one struggle. The struggle was due to the respondent’s experience with the sudden death of the young person: BCP18: … the emotional distress one in particular is quite a difficult one […] perhaps it’s hard to imagine… […] It may be because of the situation because it was sudden. Perhaps if I’d have been prepared I would have been, so I guess that actually now that I talk over it, it does make a little bit more sense. But on face value when I first looked at it I wasn’t really sure how I could relate to that and where I felt, like I sit within that. The response error for the ‘emotional distress’ attribute, as with most attributes, was due to the desired response being missing from the available response options: BCP06: … sorry, D and E. Yes, the last two. Yes, absolutely…It was very traumatic, incredibly traumatic for all of us… 292 6.2.2 Views of respondents on measure Views of respondents on the measure were explored in both the think-aloud and semi-structured parts of the interview. Overall most views on the measure were positive. Attributes appeared to be understood by most respondents and respondents reported that the measure was relevant to their experience as a close person: BCP14: … It made sense, it wasn't difficult. (…) I would say it was relevant, because all these, from 1 to 6, these are questions that are really important, and it’s something most common, the questions are most common… BCP16: …All of the questions are really relevant… BCP07: … I think it has covered the practical and the emotional side, so I think that is good. I can’t think of anything else to add... BCP13:… I think it was questions that needed to be asked, and some of it…it is all relevant … One respondent (sibling) felt that the way the questions are asked in the measure was quite rigid, but it was still relevant to their experience and covered the key areas that were important to them: BCP18: …It does cover the key areas, and if I was to think about just end of life, I think it would work quite well, but again it does seem quite rigid. (…) generally speaking, it does cover the main things… Some participants suggested that it was helpful to them to complete the measure as it gave them a chance to talk about their experience and feelings and to be listened: 293 BCP06: … It is good to be able to actually say what we have been feeling, what we saw, because we haven’t had the opportunity to say all that… BCP17: it [the measure] gave me some… What’s the word I’m looking for? What’s the word? I can’t think of the word, but it’s like… Erm… Validation. Yeah, it’s helped validate that it is an awful thing to have happen to you, when your child dies. And that people do care, and people do listen… Many respondents and in particular parents, found it quite difficult to complete the measure, due to the extremely sensitive topic and the need to recall traumatic events to be able to respond. Some became tearful during completion and the interview was paused so that the respondent could take a break from the task and decide whether they wished to continue with the interview. Although the interview was paused, participants were happy to continue with the interview and completed the measure in the hope that the study may improve service provision for other people in the future: BCP15: …I found it quite difficult (…) it brought back memories… BCP13: …the second [cognitive interview] part was very difficult…We weren’t ready for it [death] […] if it can help somebody else, you know, give them better support that’s a good thing… 6.3 Summary Chapter 6 outlined the qualitative findings from the cognitive interviews conducted. This work facilitated evaluation of the way respondents answered the measure questions, the type of problems and the degree to which respondents encountered 294 problems during ICECAP-CPM completion. Respondents were mostly able to complete ICECAP-CPM with the exception of one participant who found it problematic. Error issues were particularly related to the lack of a response option that was desired by respondents. Views of the respondents about the measure, and whether they felt it was relevant to their experience, were then explored. Respondents had different experiences of completion with some finding it helpful in validating their experience but some finding it upsetting to complete. The cognitive interview findings will be discussed in Chapter 9 in light of existing evidence. Chapter 7 outlines the quantitative methodology and methods used to explore relative weights in outcomes at the EOL. 295 CHAPTER 7. QUANTITATIVE METHODOLOGY AND METHODS 7.1 Introduction Qualitative methods used in the thesis were presented in Chapter 4 and qualitative findings were presented in Chapters 5 and 6. Qualitative work was undertaken with individuals who may benefit from health care interventions at the EOL; either as a young person at the EOL or as a close person to young people at the EOL. Chapter 7 now focuses on the quantitative methodology and use of methods to address the third aim of the thesis, that is to estimate relative weights between outcomes for young people (14-25 years old) at the EOL and outcomes for those close to them. Relative weights were elicited from members of the general public, rather than those who may benefit from interventions at the EOL. A secondary aim of the quantitative work was to explore factors that might have influenced these relative weights. An initial section examining possible methods for eliciting relative weights for different outcomes is provided, followed by the rationale for the study design, 296 elicitation technique choice and choice of study participants. The rest of the chapter outlines the development of the survey, sampling, data collection and analysis. 7.2 Methods for eliciting relative weights for different outcomes As outlined in Chapter 3, a holistic approach to care at the EOL includes a joint focus on interventions for both the person at the EOL and those close to them, with outcomes from interventions extending beyond the person at the EOL. If capability outcomes for both the young people and those close to them are captured and included in economic evaluation of care at the EOL, then this raises the question of how to integrate these outcomes in economic evaluation and the relative weights that should be placed on outcomes for young people and for those close to them.516 Empirical evidence obtained through studies obtaining relative weights of outcomes between young people at the EOL and those close to them for use in economic evaluation is limited. Yet, participatory deliberative methods have been used recently to explore an integrated capability framework for use in economic evaluation by Coast et al. (2021) where relative weights were obtained between outcomes for people at the EOL (of all ages) and their family and friends.326 A deliberative approach is suggested to be conceptually more aligned with Sen’s recommendations for public reasoning and consensus.225, 227, 302 297 Coast et al. conducted focus groups with members of the public and policy makers. To elicit weights using the deliberative method, participants were asked to complete a number of tasks first independently (pre-deliberation weightsc), and then share and discuss their task responses in the focus groupd. The budget pie technique (also called constant-sum measurement) was used in these tasks to elicit relative weights as it can present a constrained choice to respondents (given the scarce health care resources the task aims to imitate).302, 517 In a budget pie task, respondents are presented with a circle (pie) and are allocated a limited budget of tokens or points (budget) and asked to allocate the tokens between two options.517, 518 The weights that respondents provide can then be aggregated to obtain values. It is recommended that a ‘manageable’ number of tokens or points is allocated to respondents, such as a hundred, but a sufficient number that it allows for an adequate discrimination between options.517 Respondents may choose to split their tokens equally or unequally between groups or allocate all tokens to one group and none to other(s).517 Coast et al.’s work did not explore the impact of age of the person at the EOL on the relative weights between the person at the EOL and those close to them. Young people are more dependent than adults on their families, especially when suffering from c Although these weights were obtained prior to the focus group discussion (pre-deliberation) an introduction to decision making and about EOL was provided to participants and there were discussions in relation to EOL caring experiences of participants prior to starting the budget pie task. d Following deliberation, participants had the chance to change their task response if they wished to. 298 severe illness.268 Young people may gradually become less dependent as they grow older, but they remain dependent on their caregivers for a prolonged period of time. This can affect the wellbeing of those caring for them.268, 327, 352, 354 Others who are close to the young person – family members, friends, and peers – can also be significantly affected by the young person’s ill-health and death.52, 93, 94 Given the difference between adults and young people’s dependency on others, it is possible that there are differences in the relative weights that would be given between young people at the EOL and those close to them, relative to those that would be given to people at the EOL from across the population. It is also possible that different weights would be given at different points in time in the last year of the young person’s life, during a shift from curative care interventions to supportive care interventions affecting both the young person and those close to them.17, 58 Moreover, the study sample (n=38) of members of the public was not representative of the general population at least in terms of age, with participants willing to participate in the study being older than average.326 Although work to obtain relative weights for use in economic evaluation is very limited, there are alternative options to obtain relative weights. A recent study elicited relative weights between patients and carers quality of life (QOL) from members of the public using Person-trade-offs (PTOs).305 The PTO technique,305, 518 was considered as an alternative approach to address the third aim of the thesis but it was decided that the budget pie approach would be the best option in this context as: i) the budget pie 299 task is less cognitively demanding and instructions needed to complete a PTO task may need to be more detailed than for a budget pie task. Given that in an online survey respondents do not have the chance to ask questions to clarify task instructions, it was important that the task was as simple and comprehensible as possible 517 and ii) given the sensitive topic of EOL it was decided that it would not be appropriate to ask participants to make trade-offs between people at the EOL and those close to them. Work using DCEs to combine two evaluative spaces, health and social care related QOL (using EQ-5D and ASCOT), has also been conducted to obtain relative weights between the two measures.519 However, this work estimated relative weights between different outcomes for one group (patients) only rather than outcomes across different groups of people (e.g. patients and carers).519 Completing DCEs asking about outcomes for different groups of people may be more cognitively demanding than asking about outcomes for a single group of people and the task may involve a larger number of attributes for which weights would need to be obtained. Moreover, given that the final attributes for a young person EOL measure have not yet been developed nor validated, it would not have been possible to use DCEs for the purpose of estimating weights in this thesis. 300 7.3 Conduct of survey 7.3.1 Study design In this thesis, a survey was conducted to address the third aim of the thesis, that is to estimate relative weights of outcomes between young people at the EOL and outcomes of those close to them. A survey was chosen as an appropriate method to allow a larger and more representative population sample than that obtained by Coast et al.’s work. A larger sample would enable more exploration of relative weights given by subgroups, such as older adult participants, which is the secondary aim of this work to explore the factors that might have influenced these weights. The choice of a larger sample with a more representative sample would also enable a comparison to be made between the weights elicited in the survey with the earlier weights elicited in Coast et al.’s deliberative study. An online panel survey was used to elicit weights from the general population on the relative importance of the provision of outcomes of young people at the EOL relative to the provision of outcomes for those close to them. For consistency in methods between the survey and the deliberative work, the survey used the same elicitation technique, the budget pie task approach,517 to obtain weights. 301 7.3.1.1 Choice of participants Weights in the survey were elicited from members of the general population, rather than young people at the EOL or those close to them. As outlined in Chapter 3, this follows the standard approach within health economics of using general population values which is justified on the basis that it is citizen’s views that should be used in the allocation of scarce resources across society.137 Further, although young people at the EOL and those close to them have knowledge of the EOL experience, there are practical and ethical issues, with both those at the EOL and those close to them, in eliciting weights from them.516 These include the practical difficulties of recruiting these groups (as evidenced by the recruitment difficulties evident in the qualitative work undertaken in this thesis) and the potential for adaptation to chronic illness and disability for those approaching the EOL 164, 520 which may result in a biased response shift.521 As members of the general public may have limited knowledge about EOLC provision, respondents for the weighting tasks were provided with brief information about PC (Appendix 13). 7.3.1.2 Development of the survey To allow for direct comparison between the survey and the deliberative work, the first task of the survey (Task 1, shown in Figure 3) replicated the first task from the deliberative study by Coast et al. (Task A, shown in Figure 4). Figure 3 shows a copy of the script provided to the survey company to build the online survey. Instructions in 302 red were instructions provided to the company and they were not shown to respondents in the online survey launched. In the tasks, as shown in Figures 3 and 4, respondents were first asked to say whether policy makers should take account of the impact of interventions on family and friends, and if so, to split their one-hundred tokens between ‘family and friends’ (close persons) on the one hand and ‘people at the end of life’ (people of all ages at the EOL) on the other to show how much of their tokens they would allocate to the family and friends and how much they would allocate to the person who is at the end of their life. Although the two tasks were identical, the written instructions given to the survey respondents to complete the task were more detailed than the instructions given to respondents in the deliberative work. This was because – unlike the deliberative work where a facilitator was present – survey respondents did not have the chance to ask for clarification when completing the survey tasks. 303 Figure 3 - Task 1 (task used in the survey study) 304 Figure 4 - Task A (task used in the deliberative study by Coast et al. 2021) 305 Following Task 1 in the survey, the remaining budget pie task (ten) used the same format as in Task 1 but asked respondents to think about different scenarios, such as in Task 2 shown in Figure 5. These different scenarios aimed to explore whether relative weights given would differ based on the particular scenario. The tasks explored: - Weights given by the general population between young people at the EOL prior to transition to adult services (14-17 years old) and those close to them; and weights between young people post transition to adult services (18-25 years old) and those close to them. - Weights given by the general population between young people (14-25 years old) and those close to them at different time points in the last year of the young person’s life and for two different age groups: ‘14-17 years old’ group and ‘18-25 years old’ group. The choice of the two age group categories ‘14-17 years old’ and ‘18-25 years old’ used in the survey for the young people group is consistent with age-related changes in UK legal rights (young people 18 years old and over are treated as adults by the law) 41, 42 and the age of transition (at age 18) from child to adult services in the UK NHS health system.110 The different age group scenarios aimed to explore whether the difference between adults and young people, in terms of young people being more dependent on others, would influence weights given by the general public. Similarly, the use of scenarios presenting different time points before death aimed to explore whether the 306 extent of ‘closeness’ to death, including a shift from curative to supportive care interventions,17, 58 would influence weights given by the general public. 307 Figure 5 - Task 2 (task used in the survey study) 308 Each panel respondent was presented, in total, with eleven budget pie chart tasks to complete to obtain weights from respondents. Each pie chart consisted of two segments and respondents were asked to input their two token scores (weights), one for each segment of the pie chart. The budget pie chart tasks are presented in Table 8 (see Appendix 13 for a copy of the full survey). Table 8 – Budget pie chart tasks in survey: Task number Task Question – Please split 100 tokens between family and friends and: 1 people of all ages at the EOL 2a young people 14-17 years old at the EOL 2b young people 18-25 years old at the EOL 3a (i) young people 14-17 years old at 12 months before death 3a (ii) young people 14-17 years old at 6 months before death 3a (iii) young people 14-17 years old at 1 month before death 3a (iv) young people 14-17 years old in the last few days before death 3b (i) young people 18-25 years old at 12 months before death 3b (ii) young people 18-25 years old at 6 months before death 3b (iii) young people 18-25 years old at 1 month before death 3b (iv) young people 18-25 years old in the last few days before death 309 A socio-demographic questionnaire was also included in the survey to collect information about respondents’ socio-demographic characteristics and details relevant to caring responsibilities and their health (Appendix 13). Prior to launching the final survey, a pilot was used to inform its development. The purpose of the pilot survey and the use of free text answers within the pilot survey was to inform the content of the final panel survey. Issues in completing questions in the survey (as opposed to a deliberative) environment were captured and addressed for the final survey. The pilot survey was administered during March 2022 to staff and students at Population Health Sciences (PHS), Bristol Medical School at the University of Bristol, using the PHS mailing list which at the time included 942 staff and students. In the invitation email, PHS staff and students were provided with brief information about the purpose of the survey and a copy of the pilot survey (Appendix 14) to complete and return to the researcher via email if they wanted to take part. Pilot respondents (n=16) included staff and students from across PHS. The sample for the pilot survey was not random nor representative of the general population but contained respondents from different disciplines and backgrounds. Average response time for survey completion was estimated from four pilot respondents. Qualitative data collected from free text responses from the pilot survey were analysed with inductive coding and thematic analysis to explore key reasons that pilot survey participants gave for their responses to each of the budget pie tasks. Responses to free text questions were used to generate a set of codes that were included in the 310 final panel survey to generate multiple pre-specified responses for questions related to the reasoning used by participants in arriving at their task responses. An ‘other’ category with a free-text response box was also added in the final survey for respondents to use if they wished to provide an alternative response to those provided. Some pilot respondents commented on the survey design and suggested that the final survey would benefit from clearer task instructions. Although the wording of the written introduction and task instructions given to survey respondents to complete the survey was very similar to the introduction and instructions given verbally and in writing to the deliberative work participants (Appendix 15), small amendments were made to the wording for the final survey following feedback from pilot respondents. For details of the wording amendments made to the final survey in response to pilot feedback see Appendix 16: 7.3.2 Sampling and recruitment for panel survey Online panel survey participants were recruited through a professional online panel survey company, ‘PureProfile’e.‘PureProfile’ is an independent company with a unique panel accessed through its technology platform. Panel survey respondents on ‘PureProfile’ are compensated financially per minute of their time for their participation on a continuous scale according to estimated time for survey completion. e https://business.pureprofile.com/esomar/. 311 The estimated compensation amount and anticipated time needed for completion was stated in the survey invitation. Based on the pilot study average response time, the anticipated average completion time for this survey was twenty minutes and compensation for the time was £1. The study design was cross-sectional, and the survey was administered in English. The sample was intended to be representative of the general adult population (18+) in the UK. Given that a small sample may produce inaccurate findings and a large sample can be an unnecessary use of limited resources, it was important to determine an appropriate sample size for the survey.522 To determine the appropriate sample size for the panel survey, prior knowledge about the variability in general population responses was needed.522 Information from the previous deliberative work 326 was used in estimating the sample size for the online survey. In the deliberative study, the number of observations in the pre-deliberative work undertaken with members of the general population was 38 and the mean of the value given from all observations to ‘family and friends’ was 28 (out of 100 tokens). The standard deviation (SD) from these 38 observations was 20. Given a SD of 20 from the pre-deliberative work 326 and using the confidence interval (CI) formula to calculate the sample size,522 it would require a sample size of 384 to achieve a margin of error (MOE) of 2. Sample sizes with alternative MOEs of 3 and 4 were also calculated (Table 9). Conducting the survey through an online survey 312 company allowed for a large sample size, hence the option of a narrower MOE of 2 being chosen over a MOE of 3 or 4 for the sample size. However, a more conservative SD of 25 was used to account for the narrower range of ages in the pre-deliberative sample (where age was skewed towards older groups) compared with what would be anticipated in a sample representative of the general population,522 as intended here. In turn, this might well lead to greater variability in the planned survey study 522 compared with the pre-deliberative. To achieve a MOE of 2 with a SD of 25 a sample size of 600 was required. The MOE portion of a CI formula (x̄ ± z *σ/√n) for a population mean 522 was used to calculate sample size, where z * σ/√n is equal to the MOE, σ is the (estimated) population SD (25), n is the sample size, and z represents the appropriate z-value. A 95% confidence level with a 1.96 z-value was used for this sample size calculation: z * σ/√n=MOE z * σ/√n=2 1.96*25/√n=2 √n=1.96*25/2 √n=49/2 √n=24.5 n=24.52 n=600.25 Table 9 below shows alternative MOE values considered for a smaller sample size obtained. 313 Table 9 - Sample size by MOE value MOE Sample Size 2 600.25 3 266.78 4 150.1 It is important to note that the above CI-based sample size approach used to calculate the sample size assumes a normal distribution (as well as that the sample size is large enough to use the z-value as a close approximation to the value from the relevant t-distribution).522 There is skewness (positive) to the pre-deliberative data used but given that the skewness probability value is below 0.5 (0.05) 326 and the large sample sizes involved, any asymmetry in the distribution can be reasonably ignored.522 The sampling strategy aimed to achieve representativeness (of the general population) in the characteristics of the sample. To reflect the general adult population participants were quota sampled by: • Gender: female, male • Age group: 18-29; 30-44; 45-64; 65+ • Ethnic group: White; Mixed/Multiple ethnic groups; Asian/Asian British; Black/African/Caribbean/ Black British; Other ethnic group. • Highest degree or level of qualification completed: Below level 1; level 1; level 2; level 3; level 4 or above; NVQ or equivalent; Apprenticeship; Other 314 • UK Region/Country: England – East Midlands; East of England; London; North East; North West; South East; South West; West Midlands; Yorkshire and The Humber; Northern Ireland; Scotland and Wales Respondents who did not meet one of the quota criteria above were excluded by the panel survey company. 7.3.3 Data collection Ethics approval was obtained from the University of Bristol from the Faculty of Health Science Research Ethics Committee (FREC) on 07/03/22 (Ethics reference number: 9194) (Appendix 17). In the online survey, participants were first presented with an information sheet with details about the study and the survey tasks and how their data would be stored and used (Appendix 13). Participants were then provided with contact information for relevant support services. Participants were then asked to provide informed consent for taking part in the survey (Appendix 13) and to complete a screening questionnaire to collect information about their socio-demographic details that was used to target a representative sample of the general adult population in the UK (Appendix 13). Further socio-demographic details, relevant to the context of the survey, such as whether the participant was a carer were also collected during survey completion (Appendix 13). The further socio-demographic 315 information was used to look at how different characteristics of participants may have influenced survey responses and to explore factors associated with particular forms of decision making in the budget pie tasks. Contact support information was provided both pre- and post-completion to ensure that participants were aware of the sign-posting information even if they decided to withdraw from the study prior to completing the survey. To set the scene, panel participants were provided with a short introduction about EOLC for young people and families, brief information on the transition from child to adult Palliative and EOLC services and on the transition from curative to supportive care. Each panel participant was then presented with the survey questions that included completing eleven budget pie chart tasks as shown in Table 8. 7.3.3.1 Data Storage and Sharing ‘PureProfile’ has a secure data storage operation where the details of all survey respondents are backed up within a private security facilityf. All survey data collected from respondents are anonymous and data are only linked to a random identification number. f https://www.pureprofile.com/privacy-policy/ 316 All research data collected for this study were stored and backed up in an electronic form, in the University of Bristol Research Data Storage Facility (RDSF). 7.3.4 Analysis Quantitative data from the panel survey responses were entered into the STATA v17 statistical software package for the analysis. Data from all observations were first examined for potential inconsistencies and errors. The exclusion criterion for inconsistent responses were observations where a respondent allocated all of their tokens to one of the two population groups, either the ‘family and friends’ or ‘person at the EOL’ group, and none of their tokens to the other group for Task 1 and then the opposite allocation scoring was given by the same respondent in any of the following pie tasks. The decision for this ‘allocation’ criterion was due to such observations, in which one group is given all the tokens in the first task but none in another task or vice versa being implausible, and suggesting that respondents were randomly allocating tokens. 7.3.4.1 Estimation of weights Once inconsistent responses were omitted, the final data were examined using descriptive statistics. This included testing for normality in the pie task responses elicited. The mean/median values and variability for each of the budget pie tasks were then estimated. As there were deviations from normal distributions in the pie task responses, descriptive and inferential statistics were analysed using both parametric 317 and non-parametric approaches to compare the two versions of analysis and to further check whether the pie task weights varied between the two versions of analysis. Differences in the estimates of the mean responses given in the replicated pie task (Task A) in the deliberative work by Coast et al. 326 and the first task in the survey (Task 1) were then compared. As the survey did not involve an intervention such as deliberation, pre-deliberation weights rather than weights elicited following deliberation (focus groups discussion) were used from the study by Coast et al. to provide the best available comparison of estimated weights between the two studies. Means with appropriate standard deviations were then used to estimate weights for each of the other ten budget pie tasks (pie task scenarios presented in Table 8). 7.3.4.2 Exploration of relationships between estimated weights and other factors Independent associations between each of the eleven survey pie task scorings and the characteristics of the survey sample, including socio-demographic, health and care related sample characteristics were then explored using regression analysis. Associations between explanatory variables and pie task scorings were explored to help understand differences in weights given to each of the pie tasks by different subgroups of the sample. 318 To enable comparable regression analysis, observations from respondents who chose the ‘Prefer not to say’ option for the socio-demographic questions on gender, religion, and ethnicity were omitted. As there were eleven pie tasks, eleven separate regression analyses were conducted, one for each. The fourteen variables collected for the analyses were first placed into three conceptually similar groups shown in Table 10: Table 10 - Conceptually similar groups for explanatory variables Key socio-demographic variables (Group I) age group; gender; ethnic minority member; education; UK region; employment status; religion (have a faith) Health and Disability variables (Group II) self-assessed health; disabilityg Care-related variables (Group III) work in health or social care; look after family members, friends, or others; family member or friend received palliative or EOL care; cared for someone at the EOL; parent/guardian Placing explanatory variables into groups allowed for a staged regression analysis. A staged approach involved regression analysis conducted in stages, in order to avoid problems of multi-collinearity across a large number of variables, using an approach pursued in other research.523-526 g Disability defined as: Having a physical or mental impairment that has a ‘substantial’ and ‘long-term’ negative effect on ability to do normal daily activities. 319 At the first stage of the regression analysis, associations between each of the pie task outcomes and the explanatory variables were explored individually – that is, in univariable analyses – to determine whether there is any evidence, even if weak, of an association between the pie task outcomes and the explanatory variables. At this first stage, univariable regression models (unpaired t-tests and one way ANOVA) were conducted for each explanatory variable in turn individually. At this first stage of the analysis, in order not to miss variables which were of marginal significance, only variables with a p-value greater than 0.20 were excluded from further analysis. At the next stage of the regression analysis, variables with p-values below 0.20 were retained and used in multivariable models to investigate confounding between them. These multivariable models were first conducted with variables within the conceptually similar groups and then across groups, in order to produce a final model of independent associations between the retained explanatory variables and the budget pie chart task outcomes. A threshold p-value of 0.05 was used to retain variables in the multivariable models used for the within and across groups analysis. Non-parametric tests were also conducted at the first stage of the univariable analysis (Mann-Whitney and Kruskal Wallis tests) and p-values between parametric and non-parametric tests were compared to check whether there were any substantial differences in p-value results. As p-value results derived from parametric and non-parametric tests were similar, further regression analysis to look at associations was conducted using only parametric tests. 320 Qualitative data collected from free text responses from the panel survey were analysed with inductive coding and thematic analysis 527 to explore the reasoning that survey respondents gave for their estimates of weights in the various tasks beyond those covered in the multi-select multiple choice response options given. 7.4 Summary This chapter focused on the quantitative methodology and methods used to address the third aim of the thesis which is to estimate relative weights between outcomes for young people at the EOL and outcomes for those close to. A survey with budget pie chart tasks was undertaken to estimate relative weights with the general population. The methods used to address the secondary aim of this work, that is, to explore independent associations between estimated weights and the characteristics of the sample were then outlined. A staged approach involved regression analysis conducted in stages to explore independent associations. Chapter 8 presents the survey results. 321 CHAPTER 8. SURVEY RESULTS 8.1 Introduction Chapter 8 reports findings on the relative weights between outcomes for young people at the EOL and outcomes for those close to young people. The survey results, conducted with members of the general public, are presented in this chapter. Mean weights and standard deviations elicited are first presented along with reasoning responses given by the respondents. Then potential associations between task responses and the characteristics of the respondents are explored. 8.2 Survey results 8.2.1 Survey data and cleaning In total, 601 respondents completed the online survey. All respondents completed all survey tasks. Even though the respondents completed all tasks, more than half (55.4%) reported finding the token allocation exercise either difficult or very difficult to 322 complete and only 14.3 % reported that the exercise was easy or very easy to complete. When examined for inconsistency, 14 of the 601 observations were found to be inconsistent responses using the allocation criterion (see Chapter 7) and were omitted from further analysis. To enable regression analysis exploring associations between pie task scorings and respondent characteristics a further 21 observations were omitted because key socio-demographic information was not available, leaving the final survey sample as n=566h. In the final survey sample, respondents as a group, met the quota criteria for a representative sample of the general UK population as detailed in Chapter 7. 8.2.2 Characteristics of the respondents Across the final sample (n=566), a quarter of the respondents had cared for someone at the EOL and 18% were currently providing informal care to a close person because of a long-term condition or problems related to old age. Almost half of the h Mean weights and standard deviation values were also estimated for the sample including the 21 observations with missing key socio-demographic data (n=587) to check whether the values varied between the analyses with and without these 21 observations. The values estimated for the sample n=587 are reported in Table 25, Appendix 18. 323 respondents (43%) have had a family member or friend who has received PC or EOLC. Further details on the characteristics of the sample are reported in Table 11 below along with quotas placed by the survey company on gender, age, ethnicity, region, and highest educational attainment. 324 Table 11 - Characteristics of the survey sample Final sample characteristics (n=566) Variable Number (%) Quotai Gender Female Male 295 (52%) 271 (48%) 52% 48% Age 18-29 30-44 45-64 65+ 93 (16%) 178 (31%) 203 (36%) 92 (17%) 16% 31% 36% 17% Ethnicity White Mixed/Multiple ethnic groups Asian/Asian British Black/African/Caribbean/Black British Other ethnic group 495 (87%) 9 (2%) 44 (8%) 16 (3%) 2 (0%) 82% 3% 9% 4% 2% Region England - Northj England - Middlek England - Southl N. Ireland Scotland Wales 123 (22%) 149 (26%) 197 (35%) 18 (3%) 50 (9%) 29 (5%) 24% 16% 44 % 3% 8% 5% i Quotas reported are the point estimates provided to the survey company. Estimates for gender, age, region are in line with the latest Census 2021 England and Wales (https://census.gov.uk/census-2021-results). Census 2021 data for England and Wales are: gender (female, 51%; male, 49%), age (0-25 years, 29%; 25-39 years, 20%; 40-59 years, 26%; 60 and over, 24%), region (as shown in Quota column, Table 11). Soft quotas were placed by the survey company for ethnicity and highest educational attainment . For highest educational attainment quota Eurostat data were used (https://ec.europa.eu/eurostat/databrowser/explore/all/popul?lang=en&subtheme=educ.educ_part&display=list&sort=category&extractionId=DEMO_R_D2JAN). j North East, North West, Yorkshire and the Humber k East Midlands, West Midlands l South East, South West, London, East of England 325 Education highest obtained Early childhood education Primary education Lower secondary education 2 (0%) 4 (1%) 52 (9%) 19%m Upper secondary education Post-secondary non-tertiary education 190 (34%) 52 (9%) 40%n Higher Education: Short-cycle tertiary education Bachelor’s degree or equivalent Master’s degree or equivalent Doctoral degree or equivalent 266 (47%) 56 (10%) 152 (27%) 41 (7%) 17 (3%) 41%o Employment status Full-time employment Part-time employment Self-employed Unemployed (looking for work) Unemployed (not looking for work) Student Other 231 (41%) 85 (15%) 42 (7%) 27 (5%) 90 (16%) 16 (3%) 75 (13%) - Religion No religion Christian Buddhist Hindu Jewish Muslim Any other religion 276 (49%) 240 (42%) 6 (1%) 8 (1%) 3 (1%) 28 (5%) 5 (1%) - Self-assessed health Very good Good Fair Bad Very bad 99 (17%) 271 (48%) 151 (27%) 39 (7%) 6 (1%) - m Early childhood education, Primary education, Lower secondary education n Upper secondary education and Post-secondary non-tertiary education oShort-cycle tertiary education, Bachelor's degree or equivalent, Masters degree or equivalent, Doctoral degree or equivalent 326 Disability Yes No 128 (23%) 438 (77%) - Work in health/social care Yes No 67 (12%) 499 (88%) - Look after/ support others Yes No 103(18%) 463 (82%) - Family or friend received palliative or end of life care Yes No 246 (43%) 320 (57%) - Cared for someone at EOL Yes No 143 (25%) 423 (75%) - Parent/Guardian Yes No 342 (60%) 224 (40%) - 8.2.3 Estimation of weights 8.2.3.1 Testing for normality Histograms for the final sample (n=566) were used to display the relative weights elicited for each of the eleven pie tasks and visually inspected to see if they followed a normal distribution. All pie task histograms had visually similar distributions as shown in Figure 6. The Shapiro-Wilk test was also used to check if the task mean weights followed a normal distribution but the p-value in the test was below 0.05 (p=0.0046), showing that the data deviate from a normal distribution. From the graphs in Figure 6, it appears that deviation from a normal distribution was due primarily to greater variability in the data than would be expected, including a relatively high number of responses at the two tails (i.e. scores of 0 and 100). Most responses were found in the 327 middle (around 50) meaning an equal split across outcomes for young people and outcomes for family and friend. Figure 6 - Histograms showing weights elicited by task. 328 8.2.3.2 Budget pie chart task responses and reasoning for token allocation Most respondents (n=521, 92%) reported that it is important for policy makers to take account of benefits from interventions that go to family and friends when thinking about people of all ages who are at the EOL. Respondents then completed eleven pie chart tasks that incorporated: 1. Differences between ages 14-17 and ages 18-25; 2. Differences between time before death. Mean weights and standard deviation values, estimated for the eleven pie chart tasks are reported in Table 12 to show weights given across the sample (n=566). The median and IQR were also explored for each pie chart task for parametric and non-parametric comparison of the results (see Table 26 in Appendix 19 for non-parametric descriptive statistics of pie task scores): 329 Table 12 - Mean weights elicited by pie chart task (n=566) Pie chart tasks Mean weight elicited for person at EOL (SD) Mean weight elicited for family and friends (SD) Task 1: Person at EOL (all ages) 55.7 (26.4) 44.3 (26.4) Task 2a: Young person 14-17 yrs. at EOL 53.1 (22.8) 46.9 (22.8) Task 2b: Young person 18 -25 yrs. at EOL 55.9 (22.5) 44.1 (22.5) Task 3a: Young person 14-17yrs. at: i) 12 months before death ii) 6 months before death iii) 1 month before death iv) In the last few days before death 56.4 (22.7) 54.8 (22.5) 51.8 (25.4) 48.3 (28.1) 43.6 (22.7) 45.2 (22.5) 48.2 (25.4) 51.7 (28.1) Task 3b: Young person 18-25 yrs. at: i) 12 months before death ii) 6 months before death iii) 1 month before death iv) In the last few days before death 58.8 (22.9) 57.5 (22.3) 53.9 (24.6) 50.4 (28) 41.2 (22.9) 42.5 (22.3) 46.1 (24.6) 49.6 (28) In most pie tasks, apart from Task 3a(iv)p, respondents, on average (mean), prioritised outcomes for (gave more weight to) those at the EOL over family and friends. In Task3a(iv), however, respondents, on average, prioritised family and friends over the p SD around the mean is higher for Task 3a(iv) and Task 3b(iv) compared with the rest of the pie tasks. 330 young person at the EOL in the last few days before death. On average, weights between family and friends (45.8%) and the person at the EOL (54.2%) in all tasks, were not far from both groups being given equal weighting, with mode response across all tasks being equal across groups. Mean weights (standard deviation) elicited for Task 1 and Task 2b for the importance that should be given to family and friends relative to the importance given to the people at the EOL of all ages and 18-25 years old are very similar, with 44.3 % (26.4) and 44.1% (22.5) of the weight given to family and friends respectively. The weight given to family and friends of younger people (14-17 years old) at the EOL is slightly higher, 46.9% (22.8), than the weight given to family and friends of people of all ages (Task 1), 44.3% (26.4), and people 18-25 years old (Task 2b), 44.1% (22.5), at the EOL. Even though Task 1 replicated a task used in the study by Coast et al. ,326 outlined in Chapter 7, results for the relative weights for the same task differed markedly between the two studies. While in both studies the person at the EOL was prioritised over family and friends, the relative weight given by respondents in the Coast et al. study to family and friends was 28.5%q (20) while survey respondents allocated 44.3% (26.4) to family and friends. The difference in results between the two studies will be discussed in light of existing evidence in Chapter 9. q Pre-deliberation average scoring 331 Following completion of Task 1 in the survey, respondents were asked to explain their reasoning behind their token allocation for Task 1 from the pre-selected options and an ‘other’ category as shown in Table 13r. In their reasoning for token allocation in Task 1 a large number of respondents chose responses 1: ‘The needs of the person at the end of life have to be the priority for care as they suffer the most’. (280, 49%); and/or 2: ‘Family and friends would want the needs of the person at the end of life to be the priority’ (234, 41%) showing preference for the person at the EOL over family and friends. r Respondents were able to choose as many options as they wished for their reasoning responses. 332 Table 13 - Reasoning responses for Task 1 Response options How did you decide how to share your tokens in Task 1 Freq. (n=566) 1. The needs of the person at the EOL have to be the priority for care as they suffer the most. 280 (49%) 2. Family and friends would want the needs of the person at the EOL to be the priority. 234 (41%) 3. Care at the EOL should aim to improve both the wellbeing of the person at the EOL and the wellbeing of family and friends. 180 (32%) 4. Do not want to take too much away from the person at the EOL but there is a need to acknowledge support needs of family and friends. 155 (27%) 5. Support to family and friends is important to provide better experience and care for the person at the EOL. 148 (26%) 6. Family and friends are those left behind and the death of a loved one will have a lasting impact on family and friends. 126 (22%) 7. Any intervention that produces benefits to one group produces benefits to the other group too. 103 (18%) 8. The impact of the condition falls equally to both the person at the end of life and family and friends. 81 (14%) 9. The cost of interventions for the person at the EOL is higher than the cost for support to family and friends. 77 (14%) Main themes from free text responses ‘Other, please give brief reason’: ‘People die, family and friends survive until they die’ Want to give both groups equal weighting Would prefer more information/token allocation on a case-by-case basis 5 (1%) 1 2 2 Details of the respondents’ reasoning responses for Tasks 2a and 2b are shown in Table 14. The most frequently selected reasoning responses for Tasks 2a and 2b were options 1: ‘The age of the young person at the EOL does not affect my response to the pie task’ (176, 31%); 2: ‘The younger the person at the EOL is, the more care and 333 support needs they have’ (161, 28%), and/or 3: ‘The younger the person at the EOL is, the more support family and friends need to be able to support and care for the young person’ (176, 31%) as shown in Table 14 with around a third of respondents choosing each of these three options. 334 Table 14 - Reasoning responses for Tasks 2a and Task 2b Response options How did you decide how to share your tokens in Task 2a and 2b? Freq. (n=566) 1. The age of the young person at the EOL does not affect my response to the pie task 176 (31%) 2. The younger the person at the EOL is, the more care and support needs they have 161 (28%) 3. The younger the person at the EOL is, the more support family and friends need to be able to support and care for the young person 176 (31%) 4. The younger the person at the EOL is, the greater the impact of loss to family and friends 126 (22%) 5. Family and friends of young people at the EOL who are 17 years old or under have more support needs to family and friends of young people over 17 years old 86 (15%) 6. Young people at the EOL who are 17 years old or under have more care needs to young people who are 18 years old and over 53 (9%) Main themes from free text responses ‘Other, please give brief reason’: Both young person groups have equal support needs Younger age group has less mental capacity or is not old enough to make informed decisions 6 (1%) 3 3 For tasks focusing on different time points in the trajectory towards death (Task 3a and Task 3b), the weight given to family and friends increased as the trajectory got closer to death, from around 43.6% to 51.7% for family and friends of young people 14-17 years old and 41.2% to 49.6% for family and friends of young people 18-25 years old. As with Task 2a, in Task 3a respondents gave slightly more weight to family and friends of young people 14-17 years old than to family and friends of young people 18-25 years old (Task 3b). 335 Following Task 3a and 3b respondents were asked to give their reasoning for their token allocation in these tasks as shown in Table 15. The most frequently selected response option was option 3: ‘The closer the young person is to the EOL; the more support family and friends require to be able to support and care for the young person’ (253, 45%), suggesting weighting preference for family and friends the closer the young person is to the EOL so that they can provide care and support to the young person. 336 Table 15 - Reasoning responses for Task 3a and Task 3b Response options How did you decide how to share your tokens in Task 3a and 3b? Freq. (n=566) 1. Thinking about the different time points at the last year of life does not affect my response to the pie tasks 172 (30%) 2. The closer the young person is to the EOL; the more they may require help from end-of-life care services 187 (33%) 3. The closer the young person is to the EOL; the more support family and friends require to be able to support and care for the young person 253 (45%) 4. The closer the young person is to the EOL; the more support family and friends require to be able to deal with the suffering of imminent death and loss 171 (30%) Main themes from free text responses ‘Other, please give brief reason’: It is a difficult decision to make so all groups should be given equal tokens at all time points (50/50) The closer to EOL, the person at the EOL must take priority The focus of care must be on the person at the EOL regardless of time point and age The closer the young person is to the EOL; the less support they need 10 (2%) 6 1 2 1 337 8.2.4 Associations between pie task responses and characteristics of the sample Univariable regression models were used at the first stage of the analysis and variables with a p-value greater than 0.20 were excluded from further analysis. Table 16 shows the p-values from the parametric tests conducted for each of the pie tasks and for each of the variables for this first stage of the analysis. The explanatory variables included in the regression analysis were: • Gender: Female; Male • Age: 18-29; 30-44; 45-64; 65+ • Ethnicity: White; Ethnic minority • Region: N.I/Scotland/Wales; England Middle; England South; England North • Education: Secondary and under; Higher education • Employment: In employment; Not in employment • Religion: Have religion/faith; Have no religion/faith • Self-assessed health: Very good/Good; Fair; Bad/Very bad • Disability: Yes; No • Work in health/social care: Yes; No • Look after others: Yes; No • Family or friend received PC or EOLC: Yes No 338 • Cared for someone at the EOL: Yes; No • Parent or Guardian: Yes; No The p-values less than 0.20 are shown in bold in Table 16. Table 27 in Appendix 20 shows p-values from non-parametric tests conducted for the first stage of the analysis. Of the fourteen explanatory variables that were considered (see Chapter 7, section 7.3.4) in the univariable analyses (that is, not adjusting for other explanatory variables), having a higher education, and having a family member or friend who received palliative or EOL care had p-values below 0.20 for most tasks (seven out of eleven), giving an early indication that these groups were more likely to prioritise the person at the EOL compared to other groups in the sample. Ethnicity, looking after others, and caring for someone at the EOL were the only three factors that did not pass the 0.20 p-value threshold for any of the eleven tasks. For each of the tasks, the explanatory variables that met the criterion in univariable analyses of a p-value <0.20 (Table 16) were carried through to the next stage of analysis. 339 Table 16 - Initial analysis (Characteristics unadjusted) – Parametric tests EXPLANATORY VARIABLES/TASK 1 2a 2b 3a(i) 3a(ii) 3a(iii) 3a(iv) 3b(i) 3b(ii) 3b(iii) 3b(iv) GROUP I Gender 0.154 0.601 0.419 0.203 0.354 0.686 0.943 0.039 0.057 0.519 0.815 Age 0.202 0.551 0.781 0.164 0.657 0.164 0.250 0.824 0.350 0.775 0.468 Ethnicity 0.222 0.733 0.723 0.632 0.416 0.969 0.867 0.378 0.728 0.440 0.513 Region 0.222 0.239 0.239 0.652 0.109 0.854 0.672 0.575 0.116 0.602 0.559 Education 0.401 0.027 0.082 0.115 0.042 0.139 0.498 0.024 0.121 0.260 0.427 Employment 0.128 0.697 0.298 0.157 0.531 0.556 0.846 0.216 0.274 0.638 0.516 Religion 0.056 0.259 0.379 0.029 0.250 0.331 0.934 0.061 0.110 0.119 0.594 GROUP II Self-assessed health 0.030 0.193 0.583 0.038 0.151 0.704 0.462 0.240 0.402 0.624 0.558 Disability 0.729 0.329 0.539 0.256 0.526 0.042 0.090 0.684 0.378 0.319 0.121 GROUP III Work in health/social care 0.034 0.272 0.384 0.128 0.246 0.418 0.158 0.070 0.272 0.234 0.305 Look after others 0.624 0.219 0.808 0.621 0.976 0.619 0.743 0.623 0.553 0.313 0.722 Family/friend received PC or EOLC 0.013 0.123 0.018 0.337 0.124 0.230 0.325 0.092 0.066 0.164 0.338 Cared for someone at EOL 0.426 0.249 0.569 0.568 0.423 0.549 0.427 0.609 0.584 0.896 0.927 Parent/Guardian 0.787 0.526 0.726 0.314 0.902 0.192 0.301 0.812 0.880 0.479 0.713 340 At the next stage of staged regression analysis, that is, analysis within each of the groups, for some tasks some variables were the only ones remaining within a group. For example, in Task 1, self-assessed health was the only remaining variable from group II (health and disability variables group) after the first stage of the analysis. For such cases, the within group model was the same as the unadjusted result. For group I (key socio-demographic variables group) in Task 1, adjusting within the group led to only religion having a p-value below 0.05, while gender and employment did not pass the 5% threshold. To check that they were not confounding each other before excluding them, when adjusting just for religion neither gender nor employment reached the required threshold and so they were both excluded. In group III (care-related variables group) of Task 1, adjusting within the group led to only the ‘family or friend received PC or EOLC’ variable being retained. The same process, of adjusting each variable for the other variables within the three groups, was repeated for the within group analysis of variables for the rest of the eleven pie tasks to identify variables that were at this stage of the analysis independently associated with the outcomes for each pie task (Table 17).341 Table 17 - Analysis within groups (p-values for characteristics adjusted for others in the same group) EXPLANATORY VARIABLES/TASK 1 2a 2b 3a(i) 3a(ii) 3a(iii) 3a(iv) 3b(i) 3b(ii) 3b(iii) 3b(iv) GROUP I Gender 0.204 - - - - - - 0.028 0.067 - - Age - - - 0.199 - 0.161 - - - - - Ethnicity - - - - - - - - - - - Region - - - - 0.140 - - - 0.115 - - Education - 0.027 0.082 0.073 0.046 0.135 - 0.025 0.088 - - Employment 0.128 - - 0.080 - - - - - - - Religion 0.061 - - 0.029 - - - 0.037 0.091 0.119 - GROUP II Self-assessed health 0.030 0.193 - 0.038 0.151 - - - - - - Disability - - - - - 0.042 0.090 - - - 0.121 GROUP III Work in health/social care 0.052 - - 0.128 - - 0.157 0.070 - - - Look after others - - - - - - - - - - - Family/friend received PC or EOLC 0.013 0.123 0.021 - 0.123 - - 0.111 0.066 0.164 - Cared for someone at EOL - - - - - - - - - - - Parent/Guardian - - - - - 0.192 - - - - - 342 Following the second stage of the staged regression analysis within the three groups, a number of variables, different for most of the eleven tasks, were carried through to the analysis across the three groups. No variables were found to have p<0.05 for Task 3a(iv), Task 3b(ii), Task 3b(iii) and Task 3b(iv). The variables identified to be carried through to the next (across groups) stage of the analyses (p-values shown in bold in Table 17) were as follows: • Task 1: self-assessed health (Group II); family or friend received PC or EOLC (Group III) • Task 2a: education (Group I) • Task 2b: family or friend received PC or EOLC (Group III) • Task 3a(i): religion (Group I); self-assessed health (Group II) • Task 3a(ii): education (Group I) • Task 3a(iii): Disability (Group II) • Task 3b(i): Gender (Group I); Education (Group I); Religion (Group I) The final, across groups, analysis for each of the tasks led to the following variables remaining in the (final) regression models (p-values shown in bold in Table 18): • Task 1: self-assessed health (Group II); family or friend received PC or EOLC (Group III) • Task 2a: education (Group I) • Task 2b: family or friend received PC or EOLC (Group III) 343 • Task 3a(i): religion (Group I) • Task 3a(ii): education (Group I) • Task 3a(iii): Disability (Group II) • Task 3b(i): Gender (Group I); Education (Group I); Religion (Group I) 344 Table 18 - Analysis across thematic groups (p-values for characteristics adjusted for all other variables in the relevant final model) EXPLANATORY VARIABLES/TASK 1 2a 2b 3a(i) 3a(ii) 3a(iii) 3a(iv) 3b(i) 3b(ii) 3b(iii) 3b(iv) GROUP I Gender - - - - - - - 0.016 - - - Age - - - - - - - - - - - Ethnicity - - - - - - - - - - - Region - - - - - - - - - - - Education - 0.027 - - 0.046 - - 0.025 - - - Employment - - - - - - - - - - - Religion - - - 0.029 - - - 0.037 - - - GROUP II Self-assessed health 0.009 - - 0.051 - - - - - - - Disability - - - - - 0.042 - - - - - GROUP III Work in health/social care - - - - - - - - - - - Look after others - - - - - - - - - - - Family/friend received PC or EOLC 0.013 - 0.021 - - - - - - - - Cared for someone at EOL - - - - - - - - - - - Parent/Guardian - - - - - - - - - - - 345 Table 19 shows the variables retained in the final models along with the relevant task mean weight and sample size for each of the subgroups. Table 19 - Mean weights elicited for family and friends group by subgroups PIE TASK Explanatory variables Mean weight given to Family and Friends (95% CI) p-value Sample size (n=566) TASK 1 People of all ages Self-assessed health Very good/Good Fair Bad/Very bad 42.6 (40, 45.3) 49.1 (45, 53.1) 41.2 (31.8, 50.6) 0.009 566 370 151 45 Family or friend received PC or EOLC Yes No 41.1 (37.9, 44.3) 46.7 (43.7, 49.6) 0.013 566 246 320 TASK 2a Young people 14-17 years old Education Secondary and under Higher education 48.9 (46.3, 51.6) 44.7 (42, 47.3) 0.027 566 300 266 TASK 2b Young people 18-25 years old Family or friend received PC or EOLC Yes No 41.6 (39.0, 44.2) 46.1 (43.5, 49.0) 0.021 566 246 320 TASK 3a(i) Young people 14-17 years old, 12 months before death Religion Have religion/faith Have no religion/faith 45.6 (43.1, 48.2) 41.5 (38.8, 44.2) 0.029 566 290 276 TASK 3a(ii) Young people 14-17 years old, 6 months before death Education Secondary and under Higher education 47 (44.4, 49.6) 43 (40.63, 45.8) 0.046 566 300 266 346 TASK 3a(iii) Young people 14-17 years old, 1 month before death Disability Yes No 44.2 (39.7, 48.8) 49.4 (47.0, 51.7) 0.042 566 128 438 TASK 3b(i) Young people 18-25 years old, 12 months before death Gender Female Male 39.3 (36.8, 41.7) 43.2 (40.4, 46.1) 0.016 566 295 271 Education Secondary and under Higher education 43.2 (40.5, 45.9) 38.9 (36.2, 41.5) 0.025 566 300 266 Religion Have religion/faith Have no religion/faith 42.9 (40.3, 45.5) 39.3 (36.6, 42.05) 0.037 566 290 276 A higher mean weight was elicited for family and friends from respondents with secondary education and under in Task 2a (48.9%), Task 3a(ii) (47%) and Task 3b(i) (43.2%), compared to respondents with higher education qualifications who allocated more tokens to the person at the EOL in all three tasks (Task 2a – 44.7%; Task 3a(ii) – 43%; Task 3b(i) 38.9%). A higher mean weight was also elicited for family and friends from respondents who have a religion/faith in Task 3a(i) (45.6)and Task 3b(i) (42.9%) compared to respondents who do not have a religion/faith (Task 3a(i) – 41.5%; Task 3b(i) – 39.3%). Respondents with no experience of a family member or friend who has received PC or EOLC in Task 1 (46.7%) and Task 2b (46.1%) gave a higher mean weight for family and friends compared to respondents who did have experience of a family member or friend receiving PC or EOLC (Task 1 – 41.1%; Task 2b - 41.6%). A higher 347 mean weight for family and friends was also given by respondents with ‘fair’ health in Task 1 (49.1%) compared to respondents with either ‘Very good/Good’ health (42.6) or ‘Bad/Very bad’ health (41.2%). Respondents with no disability in Task 3a(iii) (49.4%) compared to respondents with disability (44.2%) gave a higher mean weight for family and friends. Male respondents (43.2%) also gave a higher mean weight for family and friends compared to female respondents (39.3%) in Task 3b(i). 8.3 Summary The survey results were presented in Chapter 8. The characteristics of the survey respondents were presented along with descriptive statistics for the survey tasks. Relative weights elicited through the budget pie chart tasks were then estimated and reasoning responses given by respondents were presented. Mean weights of around 44% were given to outcomes for family and friends of young people 18 to 25 years and of people of all ages at the EOL while the proportion was slightly higher at 47% given to outcomes for family and friends of young people 14 to 17 years. For tasks using different time points in the trajectory towards death, mean weights given to family and friends of young people 14 to 17 years and 18 to 25 years increased, from around 44% to 52% and from around 41% to 50% respectively, as the trajectory got closer to death. Regression analysis was then undertaken to explore associations between weights estimated and characteristics of the sample with higher mean weights identified in 348 certain subgroups of the sample for outcomes for family and friends (vs. person at the EOL). These subgroups were respondents: with secondary education; who had religion/faith; with no experience of a family member or friend who has received PC or EOLC; with ‘fair’ health; with no disability; who were male. 349 CHAPTER 9. DISCUSSION 9.1. Introduction This thesis focused on the assessment of outcomes of care interventions for young people at the EOL for use in economic evaluation. As interventions at the EOL have a holistic and family-oriented approach to care, interventions at the EOL should thus be assessed in that way.235 The empirical work in this thesis has drawn on the capability approach, which uses a broader extra-welfarist framework, to assess broader outcomes, that is capabilities, at the EOL for young people and those close to them. In this chapter, key findings are summarised and discussed in the context of relevant literature. Strengths and limitations of the empirical work are acknowledged, implications for policy are outlined and future research areas suggested. The chapter concludes with the contributions of the study. 350 9.2. Summary of findings 9.2.1. Outcomes for young people at the EOL To capture broader outcomes that are important to young people at the EOL and to inform the development of attributes for a measure, twenty-one in-depth interviews were conducted with six young people, 14-25 years old, with LLCs/LTCs; six parents of young people with LLCs/LTCs, but who were unable to participate directly because of their condition; and nine bereaved parents and siblings of young people who have died from a LLC/LTC. Data collection was conducted iteratively and analysis used constant comparison. The capability themes identified as important to young people at the EOL, and which can inform attribute development for a measure, were seven: Experience and enjoy; Independence; Freedom from physical suffering; Freedom from emotional suffering; Formal care and support; Control and Identity. The final conversion factor themes identified that influence important capabilities were five: COVID-19; Access to care and support; Communication with services; Coordination between services and Continuity and Consistency in care and support. 351 9.2.2. Outcomes for those close to young people at the EOL To assess feasibility of use and content validity of ICECAP-CPM for those close to young people at the EOL, nine semi-structured interviews using the think-aloud technique were conducted with bereaved parents and siblings of young people who died from a LLC/LTC. Measure completion was assessed in terms of the types and numbers of errors and struggles identified. The following error coding criteria were used to identify errors: comprehension; retrieval; judgment; response. Seventy-eight percent of responses were free of error or struggle, with identified errors in 18% of the responses and 4% of responses identified as struggles. Errors identified included comprehension, retrieval, and response errors but the majority, eight out of 10, were response errors. Views of the respondents about the measure were overall positive. Measure attributes appeared to be understood by the majority of participants and respondents suggested the measure was mostly relevant to their experience. Some found completion of the measure a positive experience. However, the completion error rate was high, and some respondents found it difficult to complete due to the sensitive nature of the topic captured by the measure. 352 9.2.3. Relative weights of outcomes To estimate relative weights given by the general public to the provision of outcomes between young people (14-25 years old) at the EOL and those close to them, an online survey was conducted. The sample (n=566) was representative of the adult UK population in terms of gender, age, ethnicity, region, and educational attainment. Budget pie tasks were used in the survey to obtain relative weights from respondents. Each respondent was presented with eleven budget pie tasks and asked to allocate tokens between people at the EOL and those close to them (family and friends). The task was repeated for different age groups (all ages; 14-17 years old; 18-25 years old) and for different time points in the trajectory towards death (over the twelve months preceding death). Most respondents (92%) allocated at least one of their tokens to those close to the dying person. Mean weights elicited for the importance that should be given to the provision of outcomes to close persons (vs. person at the EOL) were similar across tasks on different age groups. Weights of around 44% were given to outcomes for close persons of young people 18 to 25 years and of people of all ages at the EOL, while the proportion was slightly higher at 47% for outcomes for close persons of young people 14 to 17 years. For tasks using different time points in the trajectory towards death, mean weights given to close persons of young people 14 to 17 years and 18 to 25 years increased, from around 44% to 52% and from around 41% to 50% respectively, as the trajectory got closer to death. 353 Staged regression analysis was used to understand the association between characteristics of the sample, and their token allocations. Factors associated with responses in some of the tasks were: health, disability, experience of close person receiving PC or EOLC, educational attainment, religion, and gender. 9.3. Discussion of findings 9.3.1. Outcomes for young people at the EOL 9.3.1.1 What matters to young people at the EOL? The capability themes and sub-themes identified in the accounts were broadly the same across interview participants. This is reassuring with regards to using these data to inform the attributes for a capability measure for young people at the EOL by capturing perspectives from different participant groups. Some of the themes identified in the interviews are evident in studies included in the systematic review (Chapter 3) although the study focus, and/or participant characteristics identified in previous studies differ from the focus and participants’ characteristics in this thesis. Themes similar to existing studies are in relation to: physical and emotional wellbeing,52, 73, 377, 378, 381, 383-394, 398, 399, 401, 528 independence,73, 379-381, 385, 387-389, 391, 394, 398, 399 identity,379, 381-384, 386-389, 528, 529 normality,377, 378, 380, 384-386, 391, 393, 394, 398, 399, 529 and privacy.393, 399 Previous studies also include theme(s) in line 354 with the conversion factor themes identified in this thesis. These themes are around access to health care services.52, 73, 377, 378, 380-384, 388, 391, 392, 394, 395, 399, 528, 529 Yet, previous studies, including ongoing work to develop a child version of POS (C-POS) for measuring PPC outcomes for use in clinical practice,364, 377 do not appear to explicitly distinguish between what is of intrinsic value to young people (capability themes), such as enjoyment, and the factors that influence their ability to achieve what is of intrinsic value to them (conversion factor themes), such as accessing services. This distinction is particularly important for the purpose of developing a capability measure as it is the capability themes that will form its attributes rather than the means through which capabilities are achieved.424, 530 There were some differences across the three groups of participants – young people, parents, and bereaved family members – about the importance given to some of the themes identified. The importance of socialising was more prominent in parent accounts than those of young people and bereaved family member accounts. Details of physical symptoms were more prominent in bereaved parent accounts. It is possible that physical symptoms were more severe near the EOL17 and this may explain this finding in two ways: first, that it was more important for the family to be able to manage symptoms at the very EOL stage, and second, that in the interviews conducted where the young person was still alive, this period of more severe physical symptoms had not been reached. 355 Conversion factor themes, in particular access to and coordination between services, were more prominent in parent accounts (both pre- and post-bereavement). Pre-bereavement, parents were frustrated about having to ‘fight’ to access necessary services for their child. The ‘fighting’ reported by parents is evident in existing work with parents of CYP with LLCs/LTCs531 and work on the ongoing ‘battles’ reported by mother96 and father caregivers.532 Bereaved parents expressed similar frustration generally with regards to accessing care and support for their child, and particularly towards the very end, however they did not tend to express this frustration as being a ‘fight’ or ‘battle’. This may be due to these accounts being retrospective whilst in pre-bereavement parents’ accounts, participants talked about current and ongoing events. Parents in both situations also reported having to often take the role of a ‘care-coordinator’ to facilitate communication between services, especially during transition to adult services. This has been previously reported as a concern from parents of CYP with chronic illness transitioning to adult services.107 These differences in accounts across the groups of participants reflect the different experiences and challenges involved in transitioning from adult to child services and at different time points along the illness trajectory and up to the very EOL. Capturing these diverse experiences assists in developing attributes for a measure that captures outcomes for young people at different time points in the trajectory towards death. Capability themes were largely the same across the range of LLCs/LTCs in the study, including malignant and non-malignant conditions. Even though there is great 356 heterogeneity in LLCs/LTCs in CYP, and experience of living with different conditions may vary, what these young people have in common, is the unpredictable nature of LLCs/LTCs60 and the resulting life uncertainty, particularly when conditions are rare.13 Perhaps this uncertainty may partly explain why what is of intrinsic value to young people does not seem to vary much across the different conditions. Indeed, participants suggested that as life is uncertain it is important to live life to the fullest now, while they are able to do so. PPC symptoms and concerns have been previously reported to be shared across a range of LLCs/LTCs.373, 377 Beyond shared PPC symptoms and concerns, this thesis provides evidence that broader aspects of life that are important to young people are also shared across a range of LLC/LTCs. Uncertainty may also partly explain why there were no ‘future’ related themes identified in the study, in contrast to other studies, including ongoing work to develop C-POS for measuring PPC outcomes,364, 377 which have suggested that CYP with LLCs,377 young men with muscular dystrophy389 and adolescents and young adults (AYAs) with cancer 383 wish to make future or long term plans. This contrast in findings may be due to the EOL focus of this thesis and the inclusion of accounts from bereaved participants. It is also possible that it relates to potential differences between participants across the different studies, with curative treatment being possible for some but not for others20 or due to capturing experiences at different stages along the illness trajectory.17 None of the participants in this thesis talked about curative treatment. It is difficult to draw a strong conclusion about whether this was due to 357 participants consciously accepting and adapting to the situation or whether participants had subconsciously adapted their views and preferences to the limited options that they perceived were available to them (adaptive preferences).520 Differences in accounts between young people who had a condition present from birth (congenital) and young people who were diagnosed later in childhood17 showed in the meaning attached to some themes. For most, normality meant to live life beyond the medical, such as having experiences outside a care setting; this is similar to previous evidence of CYP wanting to live life as a child rather than a sick person.377, 384, 386 Yet, for one young person with a congenital condition, normality included having carers present 24/7, and when this was not possible during COVID-19, their normality was disrupted. Being aware of the different meanings attached to the capability themes reiterates the importance of including a second phase in the development of the final attributes for a measure to check content and develop wording that is meaningful to the relevant population.407 9.3.1.2 Comparison with child measures There are no measures available that capture what is important to young people at the EOL and thus the themes identified in this thesis cannot be directly compared with dimensions from measures relevant to this context. Yet it is possible to compare themes with dimensions of a HRQOL child measure commonly used in PPC, PedsQL 3, 533 and other generic preference-based CYP measures commonly used with CYP, such 358 as CHU-9D348 and EQ-5D-Y.366 An ICECAP capability measure has been recently generated for CYP aged 11 to 15 for use in economic evaluation (2023).534 Although assessment of its psychometric properties and valuation work is ongoing,534 given that ICECAP-CYP aims to capture important capabilities for CYP, it is useful to compare the capability themes identified in the thesis with the capability themes identified for the ICECAP CYP measure (for ages 11 to 15). The dimensions of PedsQL, CHU-9D and EQ-5D-Y are presented in Table 20 along with the ICECAP-CYP capability themes and the capability themes identified in this thesis. As seen in the table, many aspects of life reported as important to young people at the EOL are not captured in the existing CYP measures (PedsQL, CHU-9D, EQ-5D-Y) which tend to focus on health-related aspects of life and do not appear to include support related dimensions or dimensions relevant to the EOL. Important outcomes may thus be excluded within economic evaluation if using these measures in the EOL context. For instance, if an intervention at the EOL helped a young person to be more in control of their care, this outcome would not be captured by these existing measures. Another difference between the measure dimensions presented in Table 20 and the capability themes identified in the thesis, is that dimensions such as ‘activities’ or ‘school’ are included in the child measures as being of intrinsic value. Whilst activity related themes were indeed identified in the thesis work accounts, they were not important to young people in themselves but, rather, they were important because of what they could get out of attending an activity; that is to have experiences and 359 enjoyment. As this thesis has drawn on the capability approach to identify what is of intrinsic value248 to young people it is not surprising that the themes identified differ in this way to the dimensions of existing child measures. This distinction may be particularly important in terms of interventions at the EOL, where a young person may not be able to ‘do usual activities’(EQ-5D-Y) such as attend school or do sports, because they are for example in hospital, but they may have alternative opportunities to have experiences and enjoyment, such as mentioned in one of the accounts where a young person was given virtual reality goggles in hospital so she was ‘able to watch an underwater world’. Compared with the existing CYP measures presented in Table 20, ICECAP-CYP shares more similarities with the capability themes identified in the thesis. The ICECAP-CYP themes ‘Fun and enjoyment’; ‘Learning and experiencing’ and ‘Identity and choice’, are similar to the thesis themes: Experience and enjoy and Identity. This is not surprising given that the thesis work and the ICECAP CYP study have both drawn on the capability approach to identify important capabilities for young people.534 Yet, given the EOL context in the thesis, there are also some differences between the ICECAP-CYP themes and the thesis themes. The themes Physical suffering, Emotional suffering and Independence are not captured in the ICECAP-CYP study. This is likely due to differences in life experiences between young people at the EOL and CYP who are not at the EOL. Furthermore, it is likely that Independence was a prominent theme in the thesis accounts, due to the accounts capturing experiences of an older age 360 group, aged 14 to 25 years, which is likely to have stronger feelings around wanting to be independent,49 compared to the ICECAP-CYP work, capturing experiences of 11 to 15 year old. This difference reiterates the need to develop a number of CYP measures which are relevant to different CYP age groups, rather than attempting to capture important capabilities for all CYP within one measure. Furthermore, it is interesting that the theme of ‘Aspiration’ identified in the ICECAP-CYP study was not captured in the thesis accounts. The ‘Aspiration’ theme focuses on aspects of life that are important to CYP for the future, such as, having education now so that CYP can get a good job in the future.534 Previous work on CYP’s capability to aspire is also in line with this ICECAP-CYP capability.535 Perhaps the life uncertainty experienced by young people with LLCs/LTCs13 and the importance of living life to the fullest now, while they are able to do so, may partly explain why ‘Aspiration’ was not identified in the thesis accounts. Overall, the differences between the ICECAP-CYP study and the thesis work reiterate the need to develop a new capability measure for young people at the EOL. A new measure can capture important capabilities for the specific life stage and context that these young people are at. 361 Table 20 - Comparing dimensions of CYP measures with capability themes identified. PedsQL EQ-5D-Y CHU-9D ICECAP-CYP (ages 11-15) Capability themes identified in thesis • Physical • Emotional • Social • School functioning • Mobility • Looking after myself • Doing usual activities • Having pain or discomfort • Feeling worried, sad or unhappy • Worry • Sadness • Pain • Tiredness • Annoyance • School • Sleep • Daily routine • Activities • Fun and enjoyment • Learning and experiencing • Attachment • Emotional security and support • Achievement • Identity and choice • Physical safety • Aspiration • Experience and Enjoy • Independence • Physical suffering • Emotional suffering • Care and support • Control • Identity 362 9.3.1.3 Comparison with adult PC and EOL measures Table 21 presents the dimensions of POS-E282 and ICECAP-SCM424 developed for use in economic evaluation in PC and an EOL setting respectively. While POS-E dimensions focus on health-related aspects of life, such as physical symptoms, ICECAP-SCM dimensions include support-related attributes, such as ‘Being supported’ and ‘Choice’. These support-related attributes are similar to the capability theme identified in the thesis work around Care and Support. The support-related focus on outcomes captured in both ICECAP-SCM and the thesis work fits well with aspects of care at the EOL where the main focus is not on curative treatment and includes broader aspects of wellbeing such as dignity.53, 54 Furthermore, the theme of ‘Choice’ in ICECAP-SCM, which is about having a say and being able to make decisions about life and care,424 is similar to the capability theme of Control in this thesis which is around young people being able to have a say, be listened to and make choices in care they receive. Other similarities between the thesis work and ICECAP-SCM in terms of the capability themes identified are the focus on Physical suffering and Emotional suffering which do not appear in the ICECAP-CYP version discussed in section 9.3.1.2. 363 Table 21 - Dimensions of adult PC and EOL measures POS-E dimensions ICECAP-SCM dimensions • Family anxiety • Pain • Other symptoms • Depression • Anxiety • Practical matters • Feeling good • Choice • Love and affection • Physical suffering • Emotional suffering • Dignity • Being supported • Preparation Yet, there are some differences between the thesis work and the ICECAP-SCM work. While in the thesis accounts participants talked about the importance of young people being able to have as many experiences in life as possible, the theme of Experience and enjoy does not appear in ICECAP-SCM. Similarly, the ICECAP-SCM dimension on ‘Preparation’ around making the preparations they want at the EOL does not appear in the capability themes in the thesis work. In cognitive interviewing work undertaken with older peoplea in hospices using ICECAP-SCM, respondents often reflected on their lives and what they had done, with some suggesting that they had already had a good life, rather than having a sense that they still had lots that they wanted to do in life.520 This difference between older people and young people at the EOL may be partly due to the life stage people are at, with young people wanting to have new experiences in a Over 50 years 364 life and older people wanting to be able to make the preparations they want at the EOL. Furthermore, the capability theme of Identity identified in the thesis work does not appear in ICECAP-SCM. Yet the ICECAP-SCM theme ‘Dignity’, around being able to maintain dignity and self-respect, has similarities to Identity which is about young people being able to maintain their identity as a young person, have a sense of belonging and feel valued for who they are. Perhaps the focus on Identity, which also appears in the ICECAP-CYP measure, relates to the adolescent change in which young people are predominantly concerned with developing identity and a sense of direction in life.49 These differences in capability themes between ICECAP-SCM and the thesis accounts, further reiterate that important capabilities may differ across the life-course and therefore ICECAP-SCM does not adequately capture what matters to young people at the EOL. Another difference between the PC and EOL measures presented in Table 21, is that POS-E includes a dimension for family outcomes – ‘family anxiety’ – while ICECAP-SCM does not include dimensions on outcomes beyond the patient within the same measure. Yet, ICECAP-CPM, which can be administered alongside ICECAP-SCM as a companion measure in an EOL setting, has been developed to capture outcomes for close persons. Likewise, the capability themes identified in this thesis inform the development of a young person measure which does not include family outcomes 365 within the same measure. Having a companion measure, such as the ICECAP-CPM, allows capturing broader outcomes for both the young person and the close person with fewer constraints in terms of the number of attributes included in a measure that is amenable to valuation. 9.3.2. Outcomes for those close to young people at the EOL This is the first study to explore use of ICECAP-CPM and therefore evidence with which to compare these findings directly is limited. Findings can be compared with work conducted using similar methods (cognitive interviewing) to assess use of other measures, such as carer measures and PC measures. Given that ICECAP-SCM and ICECAP-CPM have both been developed using qualitative methods to identify capabilities in an EOL setting and are both on a sensitive topic it is useful to compare work conducted on these two measures. While the majority of ICECAP-CPM completion responses in this thesis work were free of error or struggle the percentage of errors was relatively high (18%) compared to existing think-aloud work. Studies conducted with ICECAP-SCM to assess feasibility of use had fewer self-completion errors among patients receiving hospice care (3.9%)293 and patients experiencing end-stage organ failure (5.7%).435 Although self-completion of ICECAP measures has not been error free, the higher error rate in ICECAP-CPM could indicate that the measure may not be appropriate for use in the EOL context for close 366 persons of young people. Yet, ICECAP-CPM participants suggested that the measure was relevant and appropriate to their experience. The numbers in the sample for the ICECAP-CPM work were smaller than in previous studies, and one respondent made errors throughout their completion of the measure. Given the smaller sample, it is possible that this one respondent has increased the number of errors. Most completion errors in ICECAP-CPM were of the response type which is evident in both ICECAP-SCM studies293, 435 although for different reasons. For instance, response type errors in one ICECAP-SCM study were due to questions not being completed as respondents had not yet considered particular dimensions, such as ‘Preparation’.435 ICECAP-CPM respondents reported that the time period in relation to their experience was too long to be able to provide a single response option over time; and thus, ticked between two response options for some of the dimensions. For example, experience of communication with different services varied, there were many services involved and/or at different time points. This response error may be partly influenced by differences evident between adult PC and PPC with regards to the timing of care offered and/or received with adults often receiving PC and EOLC later along the illness trajectory and for shorter periods of time, than CYP.13, 16 Further instructions highlighting the timing of the EOL experience the measure is looking at (for example adding, ‘Thinking about the last few days of the young person’s life…’) may assist with these type of errors. 367 During measure completion some participants became tearful and found ICECAP-CPM upsetting to complete reflecting back on their experiences around EOL and the impact of the condition on their loved ones. This has been reported in previous studies with patients435 and close persons293 completing ICECAP-SCM and PC measures for clinical use.436 There are previous recommendations on using a distress protocol536 with respondents completing PC and EOLC measures to support respondents if needed.435, 436 This study reiterates the use of a distress protocol when completing ICECAP-CPM, and particularly with bereaved close persons of young people, given that the grief experience of bereaved close persons of CYP can be particularly intense.186 Although respondents found ICECAP-CPM completion somewhat upsetting the overall experience of measure completion was positive with some suggesting that they felt their experiences and feelings were acknowledged through ICECAP-CPM. This is similar to carers’ reports in cognitive interviews for the assessment of a carer’s experience survey for social care interventions which later informed attribute development for ASCOT-carerb.319, 537, 538 Carers in the survey study were surprised that questions were regarding their own QOL rather than the person they cared for, which was their previous experience when completing questionnaires in a health care setting. The ICECAP-CPM respondents’ views on being heard and acknowledged further supports b ASCOT-carer is a carer social care-related QOL measure for use in economic evaluation of social care interventions to capture QOL aspects for family carers. 368 the importance of capturing outcomes beyond the patient in economic evaluation of health care interventions. 9.3.3. Relative weights of outcomes at the EOL One of the main findings of the survey work was that the majority of participants were willing to account for close person outcomes at the EOL. This is in line with existing work on accounting for outcomes for close persons at the EOL for adult patients 326 and carer QOL outcomes.305 There is limited literature with which to directly compare the survey results. The study by Coast et al.c, explored relative weights of outcomes at the EOL using deliberative methods.326 In Chapter 8, an explicit comparison between the relative weight results in the first task of the study by Coast et al. (Task A) and the first task in the survey work in this thesis (Task 1) was made. In both studies respondents prioritised the person at the EOL (adult) over close persons. Similar reasoning arguments were given by respondents in the two studies on their token allocation to close persons including the caring role of close persons and the impact of death of a loved one on those close to them.326 Yet, average weights for the same task (person at the EOL of all ages vs. close persons) differ markedly with survey c Study details outlined in Chapter 7 369 respondents allocating 44.3% and deliberative work respondents allocating 28.5%d to close persons.326 The different results between the two studies may be due to underlying factors regarding respondents taking part. While the survey was completed by a representative sample of the general UK population, in the deliberative work sample there was an over-representation of older adults and males with 58% of the sample being 65 years old and overe and 63% malef.326 Yet, the absence of associations between survey task responses for Task 1 and the respondents’ age and gender suggests that this difference does not help explain the difference in average weights given between the two studies. What may offer a partial explanation is that weights elicited by survey respondents with a family or friend having received PC or EOLC (43% of total sample) tended to give lower weights to close persons than respondents with no PC or EOL experience. In the study by Coast et al. there was no data collected on respondents experience of family d Pre-deliberation average scoring e The survey sample percentage for the same age group is 17%. According to the latest Census 2021 report, 19% of the population in England and Wales is 65 years old and over (https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/ageing/articles/voicesofourageingpopulation/livinglongerlives) f The survey sample percentage was 52% female. According to the latest Census 2021 report, 51% of the population in England and Wales are female (https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/populationandhouseholdestimatesenglandandwales/census2021#age-and-sex-of-the-population) 370 or friend having received PC or EOLC. Yet, many respondents spoke about their previous experiences of caring for a loved one at the EOL and there was a high number of respondents who had been recently bereaved. It is thus possible that the lower weight given to close persons in the Coast et al. study was associated with the respondents’ bereavement and EOL caring experience. The different methods used to elicit weights in the two studies could also partly explain the difference in results. Even though the pre-deliberation weights in the deliberative work study are used for comparison with this study, these weights were elicited following an introduction on the topics of EOL and decision making and respondents having already shared their own experiences of EOLC prior to completing the task.326 There was thus considerable discussion of the EOL decision making context before the task was completed. Survey respondents were briefly introduced to these topics but there was no discussion involved given the survey method used. The different timing of data collection (pre and post COVID-19) between the two studies may also partly explain the difference in weights elicited. It is suggested that visiting restrictions in EOLC inpatient settings during COVID-19 impacted the provision of emotional support to relatives.539 It is possible that, within this context and especially the inability of family and friends to be with their loved ones near the EOL due to social distancing measures, public views on EOLC may have shifted and the importance of support provision to those other than the dying person highlighted. 371 Different weights were obtained for tasks on different age groups and for tasks on different points in the trajectory towards death. It is possible that this difference in weights between the tasks was influenced by changes in the views of the respondents depending on whether the close person is close to an adult (over 18 years) at the EOL or close to a CYP (under 18 years). Given that the grief experienced by bereaved parents is known to be more intense and prolonged compared to grief experienced by bereaved spouses and bereaved adult children540, 541 this may have influenced the respondents’ change in weights given across the different tasks. Higher mean weights were identified in certain subgroups of the sample for outcomes for close persons (vs. person at the EOL). These subgroups included respondents: who were male; with secondary education and under (as opposed to respondents with higher education); with no experience of a family member or friend who has received PC or EOLC. Previous studies suggest that being female,542-544 more highly educated542 and previous experience of a family member or friend having received PC543, 544 were factors associated with a higher level of awareness around PC. Awareness of PC in these studies included an understanding of EOLC interventions: a. not being limited to pain relief for the patient and b. aiming to improve quality of life for both the patient and the family.543 This evidence from previous studies may offer a partial explanation as to why the weights identified by these particular subgroups (male; with secondary education and under; with no experience of PC or EOLC) in the survey tended to be higher for close persons (vs. person at the EOL). Identifying such factors that are most 372 strongly associated with differences in weights given by the general public may help inform decision making to prioritise benefits and allocate resources at the EOL between young people and those close to them. 9.4. Reflections on strengths and limitations of the work 9.4.1. Reflections on qualitative work There were a number of strengths and limitations with regards to the qualitative work conducted. A strength to the methodology used to assess outcomes at the EOL lies within the qualitative bottom-up approach used to identify the capability themes for attribute development. A number of measures for use in economic evaluation have been developed using a bottom-up approach more recently 218, 290, 348, 419, 424 and there have been increasing calls to adopt this approach as it can improve the content validity of measures developed.281 Another important strength of the bottom-up methodology used was that the in-depth interviews were conducted with the use of open and broad questions, for instance ‘What is important to you right now?’. The use of such questions aimed to capture 373 aspects of life that are important to young people using a broad focus rather than pre-defining the exact topic of discussion, for instance with a focus on health care needs. A pre-defined topic of research was evident in many existing qualitative studies identified in the systematic review (Chapter 3) and there was a dearth of qualitative studies looking at broader experiences of this vulnerable group with the majority of such studies focused only on broad experiences of young people with cancer. A further strength is the direct primary research conducted with young people with LLC/LTC, who are a very vulnerable group. This allowed capturing the voice of vulnerable young people in research and in matters that affect them.1, 545 However, given the sensitive nature of the topic and that the young people participants, as well as the parents and bereaved family member participants were vulnerable, conducting primary research with such vulnerable groups had significant challenges and limitations. Furthermore, all primary research was conducted during the COVID-19 pandemic, which had a significant impact on the ability to recruit participants into the study. 9.4.1.1 Recruitment Recruitment for the in-depth and semi-structured interviews was relatively slow. Expressions of interest were few and resulted from numerous re-circulations of study information through a number of recruitment avenues over an extended period of time (January 2021-September 2022). 374 Participants were recruited through charitable organisations, hospices, through social media and with the use of snowball sampling. The use of different recruitment avenues is an important strength of this study to allow the inclusion of participants that have access to a range of services and support. However, one of the study limitations was that none of the participants were recruited through the statutory (NHS) sector which may have resulted in not reaching families who have access solely to statutory care. Yet, it is possible that recruitment through social media may have assisted in reaching these young people and families via online routes. As the study was conducted during the pandemic with NHS services being understaffed and overwhelmed with other priorities, it was decided that seeking recruitment support from NHS services would be extremely difficult given the context of the pandemic as well as potential gatekeeping issues with staff.546 Yet, the statutory (NHS) sector relies heavily on the support of charitable organisations to deliver some of its PPC core services, such as hospice care.80 This means that the boundary between NHS and charitable organisations is less clearly defined, and they are likely to share the same pool of users. Thus, it is likely that the charitable organisations and hospices that supported recruitment for this study, share a similar cohort of users as NHS services. The recruitment strategies that were planned accounted for both possibilities of a large and small number of potential participants. If the number of expressions of interest to take part in the study had been large, recruitment would have been purposive to sample for maximum variation on the basis of respondents’ 375 characteristics. As the number of potential participants was small, all respondents were included in the study and snowball sampling was used. Even though this is acknowledged as a recruitment limitation, participants ,were from a range of backgrounds and experiences which is discussed in the next section. It is also worth noting the significant amount of time required to apply for FREC amendments (four in total) to get the necessary approvals for recruitment. Furthermore, due to hospices in England including a combination of NHS and non-NHS services and staff, there was some confusion between different organisations as to what the ethics requirements were to recruit through hospices. This confusion resulted in waiting and time spent in communication with the different organisations. Advice was sought from the Health Research Authority (HRA) regarding ethics application requirements which confirmed that there was no requirement for HRA approval and for an NHS REC review to recruit through hospices based in England. This was because participants were not being recruited by virtue of them being an NHS patient. FREC approval was obtained along with an internal ethics approval granted by each hospice that supported recruitment. 9.4.1.2 Participants Even though recruitment was very challenging, an important strength of the study was that the qualitative work included participants with a range of characteristics and backgrounds. This included a good representation of participants from other than 376 white groups relative to the UK population and participants from different regions in England. Many participants were from the most overall deprived neighbourhoods. Although it is common for vulnerable groups, such as those from deprived neighbourhoods, to be underrepresented in health research,547, 548 perhaps the highest prevalence of LLCs/LTCs in CYP being among the most deprived groups, including those living in areas of higher deprivation,19 may partly explain why many participants were from a deprived background. Interviews also captured experiences of young people with a range of conditions, including rare conditions, illness trajectories and/or who received different types of care and support. Given that there are around 400 LLC/LTCs affecting CYP,29 it would not have been possible to include all of them but a strength of the study was that it captured experiences of young people with conditions across six (from a total of nine) diagnostic groups of LLCs/LTCs affecting young people19 and included experiences of young people with non-malignant conditions (n=15). Although the number of interviews with female and male young people was equal, more in-depth interviews were conducted with female, rather than male, proxy participants (parents and bereaved family members) and most semi-structured interviews were with female bereaved mothers. There were no bereaved father participants. This is an important limitation in particular to the work conducted to assess use of ICECAP-CPM and thus further work to assess appropriateness of the measure is necessary. This is discussed in the future research section. The higher 377 proportion of female participants is not surprising and has been evident in many other studies in PPC where mothers constitute the majority of the sample.549 It is suggested that women may tend to be more altruistic with giving more time than men and perhaps this may mean that they are more willing to give time to participate in research.550 Most participants in the study did suggest that their main reason for taking part in the study was so that they could help others with similar experiences; however, this was equally true for both male and female participants. It is interesting to mention that none of the male participants but the majority of female participants suggested that sharing their experience, although difficult, had been ‘therapeutic’, and that it had been good to have the chance to talk to someone about their experience. This was partly due to participants sometimes feeling that the topic made their own friends and family uncomfortable, and so they were less likely to talk about it. The challenges of recruitment were most evident in recruiting young people with LLCs/LTCs, and in particular, with regards to recruiting young people at the very EOL. As many young people with LLC/LTC are non-verbal and are not able to share their experiences in a meaningful way, parents were included as proxies for those young people who were not able to take part themselves. Although it is acknowledged that the voices of young people can never be replaced with those of others, it was decided that family members and friends would be the most appropriate group of proxy participants to complement the voices of these young people. Family members and friends, rather than other groups, such as HCPs and teachers who may be involved in 378 the young people’s care, are likely to have a more holistic experience of the young person’s life and needs, and thus what is of intrinsic value to them overall in life. Interviews with professionals may risk failing to adequately capture non-health or non-education related aspects of life which are important to young people. Focusing on these professional views could perhaps bias the focus on what is important to the young people purely to specific topics, such as physical health or education. However, recruiting parents of young people at the very EOL was also challenging. To manage this limitation and to be able to capture what is important to young people at the very EOL, the theoretical sampling approach enabled an adjustment to the sampling strategy following analysis of the initial interviews with young people and parents. An additional group of participants, bereaved family members or friends, was then included in the interviews. The inclusion of bereaved participants at the time the study took place (during the first two years of COVID-19), also meant that it was possible to capture experiences of families prior to the pandemic which appears to have had a significant impact on experiences with regards to accessing services. Although it was not possible to recruit friends of young people, interviews included siblings who were of similar age to the young people when they died. Not being able to include friends in the study is an important limitation, as a friend may have a different relationship with the young person, a different role and bereavement experience to parents and siblings; this is discussed further in the future research section. 379 9.4.1.3 Data collection All interviews were conducted online due to the COVID-19 social distancing restrictions that were in place at the time of data collection. Although conducting interviews online allowed experiences of people in different regions and areas to be captured, the use of this data collection method with a vulnerable group and young people had some limitations with regards to providing practical support for participants. For instance, some of the young people taking part needed practical support to join an online meeting and/or for someone to hold the deviceg in front of them so we could see each other during the interview. This meant that a third person had to be present during the interview. However, the young people were asked about their preferences prior to the interview, and they were happy for a member of their family to facilitate the interview. There is risk of potential influence in the information collected in the young people interviews where there is a third-party present551; however, it would not have been possible to conduct these interviews without the third person and thus the views of these young people would not otherwise have been captured. Online interviews with bereaved parents were also particularly difficult where some of the participants became tearful. Perhaps if the interview had been conducted face to face it may have been easier to ‘take a break’ from the interview, for instance by g The majority of participants (n=19) used a mobile phone to join the interview meetings 380 suggesting having a tea break. Yet even though the interviews were conducted remotely, it was possible to quickly build rapport with participants and participants suggested that, although they became tearful, this was okay and that they expected the interview would be difficult for them. Given that the cognitive interviewing part was a difficult topic to cover, involving asking specific questions around the EOL, by having this element of the interview at the end, it did not influence the in-depth interview work and it was also helpful in that some rapport had already been built by this point. Yet, due to the in-depth interviews with bereaved family members being quite long in duration, the semi-structured interviews that followed, to assess use of ICECAP-CPM, were kept as short as possible. Although the ICECAP-CPM assessment work may have benefited from asking respondents to complete other carer or family measures to allow for comparison between measures, due to the sensitive nature of the topic, it was important to avoid the risk of prolonging the interview process and overwhelming participants with additional measure information. Due to the recruitment challenges, it was also not possible to collect qualitative data for the in-depth and semi-structured interviews at different time points so that the ICECAP-CPM assessment would be conducted at a separate data collection point. Furthermore, as data collection was conducted during the pandemic, this can be both a strength and a limitation in relation to the context in which the data were collected. Some aspects of the final capability and conversion factor themes identified in the first 381 part of the empirical work were specific to the context of the pandemic, such as the impact of the pandemic on accessing necessary services and its impact on having enjoyment and experiences. These context specific themes cannot be applied to a different context and thus caution needs to be taken with generalising the particular findings. Yet, the study was able to capture these unique experiences beyond the initial aim of the interviews at a time when recruitment for any qualitative study may have proved challenging. 9.4.1.4 Data saturation 9.4.1.4.1 Development of capability themes The number of participants to inform the development of a measure was within what was expected to produce a sufficient sample for data saturation.218, 290, 424 Yet, the number of young people participants was smaller (n=6) than anticipated. This was a limitation in terms of being able to reach saturation with regards to what is important to young people at the EOL from only the young people participant accounts. However, due to not being able to interview young people at the very EOL, or who were too unwell to take part, for practical and pragmatic reasons, the study used an overview of different perspectives with an overall aim of capturing what is important to include in an outcome measure for young people at the EOL. The inclusion of these different perspectives ultimately resulted in a more rounded approach that could incorporate the perspectives of young people, both verbal and non-verbal, and across the EOL trajectory, including at the very EOL. Saturation was thus reached in terms of 382 the primary goal of the research, that is, what it would be important to include in an EOL measure for young people. Saturation was achieved when no new perspectives and themes were introduced by new participants that would change the nature of the attributes. 9.4.1.4.2 Assessment of use of ICECAP-CPM As discussed in Chapter 4, there are no recommendations on an adequate sample size for cognitive interview studies, and existing studies vary in their sample size with those on sensitive topics and vulnerable participants tending to have small sample sizes with up to fifteen participants.436-438, 440, 441 Due to the sample size in this study being quite small (n=9) it was difficult to identify potential differences between mothers (n=5) and sibling participants (n=4). One sibling participant suggested that the ‘practical support’ attribute was difficult to answer as it was her mum who dealt with the practical support for her sibling. However, it was difficult to draw any strong conclusions from only one participant. Furthermore, it was not possible to explore whether the measure is appropriate for use with fathers and friends of young people. It is also possible that the high completion error rate was due to the small number of participants and a high response error rate identified from one particular respondent. 383 9.4.1.5 Data analysis With regards to the analysis of in-depth interviews, the involvement of the supervisory team, with experience in qualitative research and research expertise in PPC, in the analysis helped to broaden the interpretations from the interviews by bringing different perspectives. The involvement of the supervisory team is described in detail in Chapter 4 including their involvement in initial open coding of interview transcripts, in informing coding schedules and providing feedback on the emerging themes. The presence of three additional independent raters for the think-aloud interview rating of errors in the measure completion, also reduced the reliance on the interpretation of a single individual. The rating process is outlined in detail in Chapter 4. 9.4.1.6 Reflexivity: in-depth and semi-structured interviews self-reflection. My experience and views influenced the interview process and data analysis. 384 Although I had previous experience with conducting interviews with vulnerable groupsh, I did not have previous experience with conducting interviews with the population relevant to this study. This lack of experience may have influenced the data collected. To assist with the conduct of sensitive interviews I attended a two-day short course on ‘Managing Challenging interviews’ run by the Social Research Association. The course, being interactive and with a focus on difficult interviews, was useful in terms of preparing me on how to deal with a number of difficult situations, for instance what to do if a participant became distressed. Given the sensitive nature of the study, I also received regular debriefs following initial interviews, subsequent debriefs were then less frequent when I felt more comfortable with undertaking the interviews but were available when I felt I needed them. I also used ‘reflective memos’ immediately after each interview, to take notes on what went well and not so well during the interviews, including initial impressions. This assisted in identifying any gaps in my way of questioning during the interviews which I could then explore in following interviews. Writing the memos also assisted in reiterating the need to adjust the theoretical sampling so that experiences at the very end of EOL could be captured. Memos taken during the three initial interviews with young people and parent participants highlighted that the topic of what was important h Adults receiving support from: homelessness support services; secondary NHS mental health services; drugs and alcohol services. 385 at the very EOL appeared to be not discussed by both myself, as the interviewer, but also by the interviewees. However, with the inclusion of bereaved participants I then found it easier to ask about this topic and this group of interviewees was more comfortable in discussing this. Furthermore, given my lack of experience in conducting cognitive interviews, this may have influenced the richness of the data collected for the think-aloud interviews. Following the first three cognitive interviews conducted I sought advice from those with previous experience in cognitive interviewing so that my interviewing skills could improve. Another potential influence on how data collection and analysis was conducted was my previous work experience in a support role in a statutory secondary mental health service. To minimise potential influence of my work background with regards to the data collected, such as the interview focus being limited to service related topics, the interview topic guide for the in-depth interviews was broad and interviews were mostly led by interviewees with the use of open-ended questions to capture the topics that were important to the participants.552 9.4.2. Reflections on quantitative work 9.4.2.1 Methods The methods used to conduct the survey have some strengths and limitations. 386 An important strength of the survey was that prior to launching it online, it was first piloted to a small sample of participants from Bristol Medical School and re-designed using pilot feedback. This was done to ensure that the survey design and task instructions were as clear as possible to respondents completing the survey. This was particularly important as the survey was conducted online, and respondents would not have the chance to ask for clarifications during survey completion. Despite this, a high proportion of respondents (55.4%) reported that the survey tasks were difficult to complete which is an important limitation to consider. The respondents were not provided with an option, such as a free-text box, to clarify the reason(s) why the tasks were either difficult or easy to complete so it is hard to know why this was the case and in retrospect it would have been helpful to include this. A potential reason for this may be due to the degree of acceptability of the method used (budget pie technique) to respondents. Although the tasks may not have been cognitively demanding, perhaps the length of the survey and the repetitive nature of the tasks may have influenced the respondents’ ability to stay interested in the survey.164 It is also possible that the tasks may have caused uneasiness to respondents given the sensitive nature of the topic and respondents having to take the role of a decision maker who needs to trade off resources between two very vulnerable groups (people at the EOL Vs close persons).164 The presence of a researcher to assist respondents with completing the survey may have ameliorated such limitations. However, if the survey was administered with a 387 researcher present the sample size would have had to be a lot smaller given constraints on time, and thus may not have been representative of the general population, which are two of the important strengths of the quantitative methods used for this work. The high proportion of weights elicited in tasks with a 50/50 scoring between outcomes for the two groups is also interesting. This may be due to respondents following a ‘simple’ equality rule when responding to the survey context, making a quick or (perceived to be) neutral decision with regards to their response given.553 Another potential reason may relate to the difficulty around making a trade-off decision between groups of people, with respondents perhaps making a more deliberated decision to give equal weighting to outcomes for different groups. Another limitation to consider was in relation to variability in the task scoring data, which was greater than expected, with a high number of responses at the two tails of 0 and 100 (token scores in task) and a tendency for digit preference, where respondents allocated a number of tokens ending with 0 or 5, such as 10 or 25, rather than any other numbers ranging from 0 to 100. Work in relation to willingness to pay (WTP) tasks has been conducted which suggests that respondents may find it difficult to estimate a precise value in a task, and/or it may not be worth the time, and thus choose a ‘prominent number’ for their given value from a limited number of values available.554 Thus, this tendency for digit preference may be due to ‘lack of precision rather than existence of bias’.555 388 Although questions with multi-select response options and free text boxes were provided following the pie tasks to ask respondents to justify their reasoning behind their token allocation, respondents were able to choose as many reasoning options as they wished. Respondents tended to tick more than one option which made it difficult to understand what their main reason for their token allocation was. An alternative question design, such as a question in which respondents are asked to choose only one option that best describes their reasoning, may have been more appropriate to better understand the respondents’ thinking. Finally, an important strength of the survey design was that the replication of the first task from the deliberative study by Coast et al.326 allowed for direct comparison between the survey and the deliberative work. Without being able to do a direct comparison for the first task asking about people at the EOL of all ages, and undertaking the survey work with tasks asking only about young people at the EOL, interpretations of differences between the survey work and the deliberative work could have all been interpreted as being related to differences across adults and young people at the EOL, rather than differences between the methods used in the two studies. 389 9.5. Implications for policy The purpose of this thesis was, ultimately, to provide a way of assessing broader outcomes of care for young people at the EOL, and include outcomes for those close to them, for use in economic evaluation. It is thus important to consider the implications of the thesis work in the current EOL policy context. 9.5.1. Including capabilities as broader outcomes in economic evaluation of EOL While ICECAP-SCM captures important capability outcomes for adults at the EOL it does not appear to adequately capture the outcomes which are important to this specific young people groupi, such as being able to experience and enjoy life to the full. Therefore, the thesis findings provide empirical evidence to suggest that the use of ICECAP-SCM with this young people group is not recommended in practice and reiterate the need to develop a new capability measure for use in economic evaluation of care at the EOL for young people. The capability themes identified in the qualitative work suggest that important outcomes for young people at the EOL go beyond health-related outcomes, to include i Comparison of the capability themes identified in this thesis with the ICECAP-SCM attributes were discussed in section 9.3.1.3 390 other aspects of life which are of intrinsic value to young people. Furthermore, the conversion factor themes identified highlight a focus on access to necessary care and support so that young people and those close to them are able to achieve both health-related and non-health related outcomes, rather than a focus on cure, such as having options for curative treatment. This need to access necessary care and support to be able to address both health related and non-health related needs is in line with current TfSL recommendations for policy change in the PPC sector. TfSL recommendations emphasize the need for families to be able to access care and services, such as access to short breaks for families and access to 24/7 PPC, especially when CYP are approaching the EOL and being cared for at home.89 A need to access support provided by other HCPs, such as family social workers78 and access to affordable childcare have also been highlighted by TfSL so that CYP with LLC/LTC can see progress in their wellbeing through opportunities to socialise, learn and take part in activities in a supportive environment.556 In the UK, the need for a broader evaluative framework has been highlighted by NICE but only for the assessment of social care outcomes.197 Yet, there is evidence of a policy move through the recent restructuring (2022) of NHS decision making bodies with ICSs replacing the local CCGs.87 ICSs focusing on a system-wide integration of different health care services – i.e., primary and specialist care, physical health, mental health and social care services – aim to tackle the wider determinants of ill-health, 391 reduce inequalities and improve population health.557 This policy move, aiming to improve population health outcomes by addressing wider issues, suggests the need to capture broader outcomes to assess the impact of health care interventions. Given this shift to a broader system focus, ICSs can potentially give greater consideration to how the different services work together, communicate and fund care at the EOL, to ensure that families do not have to ‘fight’ to access necessary services, which was a prevalent theme in the parent accounts and especially in relation to transition to adult services. Unlike the use of ICECAP-SCM, the views of respondents on ICECAP-CPM suggest that the measure covered broad areas that were important to them and included both the practical and emotional aspects of their experience. Therefore, although ICECAP-CPM was not developed for specific use in the young people population at the EOL, its use with close persons of young people looks promising. ICECAP-CPM can be continued to be used in practice whilst collecting further data about its use and psychometric properties, both for close persons of adults at the EOL, and also specifically for close persons of CYP. Overall, the qualitative findings from the thesis provide empirical support for the appropriateness of a broader evaluative space in economic evaluation of care at the EOL for young people as well as those close to them. Measuring capability outcomes using a broader evaluative space, would allow for the cost-effectiveness of these 392 interventions to be assessed more fairly by capturing aspects of life that are important to young people at the EOL and those close to them. 9.5.2. Including capability outcomes for those close to young people at the EOL in economic evaluation Another important finding is that most survey respondents suggested that it is important to account for EOL outcomes for those close to young people. Thus, the study provides empirical evidence of a general agreement within the public that those close to young people at the EOL should be accounted for in decision making. Although the survey was conducted in the UK, it is possible that the public’s willingness to account for close person outcomes may be similar in other country contexts where the goals of care at the EOL include support provision to those close to young people. The inclusion of outcomes beyond the patient is not done routinely within economic evaluations558 and thus it is likely to be challenging. NICE guidance states that economic evaluations should consider all health outcomes whether that is for patients or carers.558 However, this acknowledgment by NICE is not reflected in its methods for economic evaluation and it does not extend beyond carers to include the wider close person network that has been suggested to be important for interventions at the EOL.290 393 The fact that survey respondents were willing to trade off a large part of the resources for close person outcomes may potentially raise ethical and practical implications with regards to how much weight policy makers should give to outcomes for close persons. It raises the question about whether a threshold would need to be introduced in decision making, given that young people are the main focus of EOLC interventions, to ensure that the provision of outcomes for young people also remains the main focus in resource allocation decisions. The relative weights elicited in the survey tasks can be used to develop an integrated framework in economic evaluation of EOL outcomes for young people. 9.6. Implications for service planning and clinical decision making Although this work was focused on assessing broader outcomes of care for young people at the EOL and their families for use in economic evaluation, much of what was discussed by the CYP and those close to them could also be informative for service planning and in the clinical decision-making context. The data collected here suggest that, to ensure that services are efficient and effective it is important that decision makers in service planning understand what matters to young people at the EOL, the impact that service planning decisions have on them and their families and the importance of considering support for families in service 394 planning. The voices of those who are ‘experts by experience’ 559 in terms of using, or caring for a young person using, services at the EOL have been included in this study to better understand what matters to them. The findings suggest the need for a holistic approach to care, with a greater focus on delivering care and support not only across primary, secondary and tertiary health care services but also across other settings such as in education and in the community (e.g. transport). Better planning for accessing necessary care and better planning for transition to adult services were also highlighted by participants. These voices, which have been very seldom heard, can provide invaluable and compelling empirical evidence that can help to influence how services are designed and co-ordinated so that they truly meet the needs and preferences of young people at EOL and those of their families. Including the voices of ‘experts by experience’ in this qualitative work may also help inform quality of care and clinical decision making, by raising awareness among health and care professionals to assess needs more holistically and include support for those close to the young person. The findings can enable professionals to better identify the priorities of this population and to enhance understanding of the potential impact of current practice and current concerns of those using services. Rather than focusing solely on health concerns and medical needs, professionals should consider additional concerns in the context of what matters to the young person to improve overall wellbeing and enhance a meaningful life. In addition, an approach to care that is focused on the wellbeing of young people needs to incorporate support for the family. 395 This is important, as this study suggests that families often struggle with accessing care and support for the young person, and for their own wellbeing, and parents often feel pushed into taking a role as care-coordinator to advocate for their child. 9.7. Implications for research The work done here, involving vulnerable young people and their families in qualitative research on the sensitive topic of EOL, supports the view that it is feasible and necessary for these individuals to have the opportunity to participate in research and for their voices to be heard. This is particularly important in research undertaken for measure development and validation with vulnerable populations where accessing these perspectives can be challenging and measure development often relies only on proxy involvement and reports. Fear of causing distress should not be a reason for researchers and measure developers not to include the perspectives of those for whom the measure is intended for. In this qualitative work, a distress protocol was developed, participants were offered a debrief following data collection and they were provided with contact information for appropriate support services to manage potential distress but still enable participants to give their views and facilitate a positive closure for them. Young people with LLC/LTC have unique and important views about what matters to them, and the study has demonstrated that they are able to be involved in qualitative interviews in an online setting in a meaningful way both in terms of providing useful information for measure development but also in terms of 396 young people benefiting from participation in research, feeling useful and giving back to others who may be struggling. Furthermore, the online interview setting during the COVID-19 pandemic enabled a clinically vulnerable group to safely participate in qualitative research. 9.8. Future research to develop the capability measure In this study the first steps have been undertaken for the assessment of broader outcomes of care for young people at the EOL and for those close to them from a capability perspective. Further empirical work is necessary to be able to assess these outcomes for use in economic evaluation. 9.8.1. Development of a young person EOL measure The next step towards the development of a young person EOL measure is to develop the final conceptual attributes and a descriptive system for the measure.281 The capability themes identified in Chapter 5 can be taken forward to determine the final conceptual attributes that would form the measure dimensions.287 Information about the factors identified as conversion factors can be collected alongside future capability measures to explore the influence of these factors on the capabilities of different individuals. Information collected about the conversion factors is likely to aid 397 in interpretation of the capability information and to understand how different factors may influence the level of capability for individuals. As with previous qualitative work conducted to develop capability measures for use in economic evaluation,218, 290, 424 semi-structured interviews with the same population used in the qualitative work in this thesis can be undertaken to check the content and coverage of the capability attributes and descriptors identified in Chapter 5. Interviewees can be presented with draft attributes and asked to feedback on them in their own words to improve the content and generate wording that is meaningful to and easily understood by young people completing measures. (A summary of the methods that would be followed to develop the young person EOL measure is provided in Appendix 21). Following development of the measure the next step would be to assess the psychometric properties of the measure. This could include work to assess feasibility of use, reliability and validity of the measure.292 Valuation work will also need to be conducted to estimate weights for each of the attributes and levels of the new measure so that it can be used in economic evaluation. An important consideration for this work would be with regards to who should value the measure and whether it is more appropriate to elicit values from the patient population or adopt the same approach taken in the third empirical part of the thesis where societal values were explored on relative weights between different outcomes. Given the ‘bottom up’ approach to attribute development in this thesis and evidence suggesting that 398 adolescents can provide reliable estimates for valuation tasks (BWS)560 it may be appropriate to elicit values from the same population used for attribute development. 9.8.2. Outcomes for those close to young people With regards to the work conducted focusing on outcomes for close persons, further work to assess use of the measure in this context is necessary. For a more thorough assessment of the response errors identified due to the difficulty of ‘aggregating’ experiences into one single response, it is necessary to conduct further interviews with a larger sample. A larger sample may help to understand whether the higher number of response errors was due to the measure not being appropriate for use in this context or due to errors made by a particular respondent. Further cognitive interviews are also necessary with bereaved fathers and friends of young people as there were no participants included from these two groups and it is possible that their relationship and/or caring role differs to mothers and siblings. Conducting cognitive interviews with younger siblingj participants and friends who are under 18 years old is also needed to assess whether there are any age-related differences between adults and CYP close person respondents. j Bereaved sibling participants were between 20-29 years old. 399 The work undertaken with ICECAP-CPM did not attempt to address other psychometric properties of the measure for close persons of CYP, yet future work around validity, reliability and sensitivity to changes is also necessary. If including close persons in economic evaluation, it is also worth exploring whether close person networks of young people differ from those of older adults at the EOL so as not to miss any significant outcomes. Existing work suggests that on average, close person networks of older adultsk at the EOL consist of eight people and include family and non-family members.314 It is possible that networks for younger people may differ in size or consist of different people given the different context and life course stage.561 For instance, it is likely that at this life stage young people will not have children of their own whilst older adults at the EOL may do and potentially be included within their close person network. These networks could be explored in interviews using hierarchical mapping methods with those close to young people at the EOL.314 9.8.3. Relative weights between outcomes Given that the survey was conducted prior to the development of a young person measure and the use of ICECAP-CPM has not yet been fully established in this context, further work with regards to weighting outcomes may be necessary. It is possible that the societal weights may be different if respondents are presented and familiarise k Those close to people at the EOL from 40 years old and over 400 themselves with the relevant capability measures for young people and those close to them, prior to completing the budget pie tasks. Furthermore, as the weights elicited differ to those obtained from previous work, it may be useful to conduct further research to better understand factors influencing this difference in values elicited. Potential work could involve conducting the same survey tasks but using in-depth deliberative work to allow discussion and therefore explore in more depth reasons behind task allocations given. 9.8.4. Outcomes for other CYP age groups at the EOL This thesis focused on outcomes at the EOL for young people aged 14-25 years old and those close to them. This age group, moving from late childhood to early adulthood, has unique experiences of transition to adult services.110 Given the lack of appropriate measures for CYP at the EOL for use in economic evaluation,359, 369 future research is necessary to explore outcomes for younger age groups of CYP at the EOL. It is possible that the different life stages and developmental changes in childhood44 may influence what aspects of life are important to CYP of different ages and by how much. Future qualitative work can explore what is important to CYP under 14 years old and assess whether, and how many, measure(s) are necessary and for what age groups. Initial work could involve testing use of an existing young person EOL measure, if available, with CYP under 14 years old. Validity work to assess use of ICECAP-CPM for those close to CYP under 14 years old may also be necessary, given potential 401 differences in experiences of closer persons of younger CYP, as well as further work to explore societal values for relative weights between outcomes for CYP under 14 years old and those close to them. 9.9. Next steps The findings of the study will be disseminated to all relevant and interested stakeholders at the end of the PhD study. A draft dissemination plan was presented at the Wellcome 4th Advisory Group meeting on 15th November 2023 (Bristol). Feedback from the advisory group was considered in deciding the final dissemination plan to different audiences based on their information needs and preferences (See Appendix 22 for the final dissemination plan of the study findings). 9.10. Conclusion This thesis has made three novel contributions in the area of assessing outcomes of interventions at the EOL: i) the development of preliminary conceptual attributes for a new a capability measure for young people at the EOL; ii) a first assessment of the feasibility of use and content validity of ICECAP-CPM with bereaved close persons of young people; iii) estimation of population weights between outcomes for people at the EOL and for those close to them from a representative sample of the general public. 402 As discussed in Chapter 3, there is a dearth of measures that can be used to measure outcomes of care at the EOL for young people for use in economic evaluation. The empirical work in this thesis has addressed this gap within the literature by conducting qualitative work to inform the development of a new measure. This is the first qualitative study in the UK to capture broader aspects of life that are of intrinsic value to young people at the EOL, finding that the key capabilities that are valuable to young people at the EOL are: Experience and enjoy; Independence; Freedom from physical suffering; Freedom from emotional suffering; Formal care and support; Control and Identity. These capabilities are affected by a number of conversion factors: COVID-19; Access to care and support; Communication with services; Coordination between services and Continuity and Consistency in care and support. Whilst some of these conversion factors are likely to have enduring impact, others were particularly specific to the timing of the work, around the COVID-19 pandemic. These findings provide the preliminary conceptual information to inform the development of a new capability measure for use in economic evaluation of EOL for young people, an area where there is paucity of both health and broader measures. With further development, the work reported here provides the first step in the direction of broadening the evaluative space for CYP at EOL. Ultimately, it could produce a more appropriate means of assessing outcomes within cost-effectiveness analyses of interventions and thus contribute to a fairer decision-making process within the context of limited resources. 403 Beyond capturing important capabilities for young people, this is the first study to assess use of ICECAP-CPM with those close to young people who have been bereaved. In Chapter 3 it was suggested that an existing capability measure for those close to people at the EOL, ICECAP-CPM, may be appropriate but its use with and relevance for those close to young people at the EOL would need to be assessed. The thesis work found that the measure was considered to be acceptable, capturing appropriate dimensions. Although challenging for some participants to complete, others found its completion helped to validate their experience. Furthermore, although the error rate was found to be relatively high compared to error rates for other capability measures, the research identified that the primary cause of errors was in the response options, suggesting that this may be an area to address in future development of the measure. Taken together, the findings suggest that the measure appears to be promising for use in this context, albeit with further research being needed. Finally, this research provides statistically robust estimates of relative weights for a general EOL population from a representative adult UK population sample: weights of 56% were given to the provision of outcomes for the person at the EOL and weights of 44% were given to the provision of outcomes for close persons. This research also enables the assessment of differences in the estimates of relative weights for different young people EOL populations. These quantitative findings can inform an integrated framework in economic evaluation which considers capability outcomes for close persons within funding decision-making about EOL interventions. Furthermore, the 404 work provides empirical evidence that the general public is willing to trade off resources for close person outcomes at the EOL. Thus, this work can inform a decision-making process which better reflects what the general public feels it is important to consider when allocating resources between interventions at the EOL. Overall, the three key contributions of this research together, have helped to move forward the area of economic evaluation for health and care interventions for young people at the EOL. The thesis work has moved this area forward in a significant way both in better understanding the capabilities that are important to young people at the EOL and how these capabilities, which will inform development of a new EOL measure for young people, might be used alongside information about the broader impacts of interventions on close persons in economic evaluation. Finally, the survey work on how the general population weights outcomes between people at the EOL and close persons, has significantly enhanced the potential for an integrated framework for economic evaluation. 405 406 Appendices Appendix 1: Systematic review methods Appendix one provides further information on the systematic search method used to review existing studies. Search Strategy Search strategies used: Three search strategies were used to avoid missing studies, due to different indexing terms and title contents which may be potentially relevant.504, 562 The first search strategy to identify relevant studies in English was to search (title and abstract) across a number of electronic bibliographic databases. The databases used were MEDLINE (use of MeSH terms), CINAHL (use of MeSH terms), PsycINFO, Web of Science Core Collection, AMED and EMBASE. The choice of databases was based on the different specialisms each database covers, including nursing, psychology and PC, that may include studies relevant to the topic under study. Web of Science, a multi-disciplinary database, was included to cover social sciences journals and books which are not indexed in the rest of the databases chosen. Alongside electronic databases, following up of reference lists of identified papers was also used in case the online search was not sensitive enough or the relevant articles 407 did not include adequate keywords. Following up reference lists also assisted in identifying studies published in books which may not have been indexed in electronic databases. Finally, ProQuest dissertations and Theses Global (PQDT) was searched for relevant theses. Duplications found across databases and supplementary searching strategies were removed at this stage. Database search terms: The PRISMA 2020 checklist 375 and CRD376 guidelines were used as guidance to conduct the search in this review. The PICOS search, recommended by both CRD and PRISMA, was used in this review but given the lack of an ‘intervention’ and a ‘comparator’ included in the PICOS search, additional inclusion criteria were considered that were more relevant for this specific study. The search terms used to conduct the search on databases are outlined in the Table22. Key words and medical subject headings (MeSH) terms along with Boolean operators depending on the database were used (Table 22) below shows the terms used in Medline database search as an example). The InterTASC Information Specialists’ Sub-group (ISSG) search filter resource was used to choose keywords relevant to Qualitative and mixed methods research and age groups563 and advice from the 408 subject librarianl was also sought on the choice of key words and MeSH terms used. The search strategy was designed to be first applied on MEDLINE and then was refined, if necessary, and adapted for use with the remaining databases. l Subject librarian for Biochemistry, Population Health Sciences (Bristol Medical School) and Psychological Science, University of Bristol 409 Table 22 - Search terms used in Medline database. Term domain MeSH terms (includes use of explode option) Key word terms (in title and abstract) Participants (P) Child OR pediatrics OR adolescent OR young adult Child* OR adolescent* OR P?ediatric OR youth OR teen*; young AND Type of care or condition that benefits from interventionm n Hospice and Palliative care nursing OR palliative care OR palliative medicine OR terminal care OR hospices End of lifeo OR supportive care OR life limitingp OR life threateningq OR palliative care OR terminal care OR hospice AND Study Design (S): methodology used to collect and analyse data; Qualitative research or Mixed Methods qualitative research OR interview OR observation OR personal narrative OR grounded theory OR focus group qualitat* OR interview* OR observation OR discourse OR grounded theory OR phenomenol* OR ethno* OR focus group OR narrative OR mixed method* OR multimethod* OR multiple method* OR mixed design* m MeSH terms added in Psycinfo database search: ‘death and dying’; ‘child death’. n Key word term added in Psycinfo database search (title and abstract): ‘dying’. o ‘End-of-life’ added as keyword in the rest of the databases searched. p ‘Life-limiting’ added as keyword in the rest of the databases searched. q ‘Life-threatening’ added as keyword in the rest of the databases searched. 410 Study Inclusion: Studies found from literature searching were screened by examination of their title and abstract against the inclusion criteria outlined in Chapter 3. At the title and abstract screening stage, if information was not enough to ascertain whether the publication fitted the inclusion criteria, then the full text version was screened. The search included studies from 1997 to 2023. A search in all electronic databases was first conducted in June 2022 and towards the end of the study an update of the literature search was conducted in March 2023 to ensure that not recent papers were missed. Although relevant studies done outside this time period may have been missed it is necessary to set limits on a potentially large number of relevant studies available. In 1997, a first edition of the “Guide to the development of children’s palliative care services”564 was published by ACTr. The guide categorised key characteristics of different conditions to define the concept of LLC in children.29, 564 As the definition of children’s palliative care was documented in this guide for the first time 17 the literature search will start from this date when the field of CYP’s palliative care was newly emerging. Studies using only quantitative methods were excluded. Whilst studies using quantitative methods could offer useful information, it is suggested that qualitative r Association for CYP with LTCs or LLCs 411 research is focal in gaining understanding of PPC experiences,374 and allows the inclusion of patients’ perspectives.565 This thesis takes an inductive approach to developing measures based on the perspective of the population the measure is intended for, while quantitative studies would include themes pre-specified by the researcher. Furthermore, the review aimed to include studies capturing the experience and perspectives of young people, either through research conducted with young people participants or family members as proxies. Studies including only the perspective of HCP’s were excluded. Finally, only studies conducted with research participants in a high-income countrys were included as the thesis is focused on the assessment of the cost-effectiveness of interventions and services in high-income countries. It is thus more appropriate to only include findings of studies conducted in settings with similar health care resources as low and middle-income countries face different challenges, in relation to access to health care resources and PPC needs, to high-income countries. 567 At the title and abstract screening stage, a second independent reviewer from the supervisory team (SN), with PPC expertise, screened a proportion of the publications (5% of all records screened in Medline database) to increase the rigour of the study. s The World Bank Group Data list of high-income countries was used [566] 412 Disagreements about study inclusion were resolved through discussion between the two reviewers. At the full text screening stage, a second researcher (LP) with previous experience in conducting systematic reviews for health economics studies, reviewed independently 5% of full text publications against the inclusion criteria. At this stage of the screening process, the full text for one of the studies identified was not available online and was sourced as an inter-library loan.386 At the screening stage, it became apparent that the number of adolescent and young adult (AYA) specific studies or studies that included a substantial proportion of AYA participants was limited thus all studies with at least one AYA participant within the age range of 14-25 were included to be able to capture as much data as possible. Data extraction Data from each of the 26 studies included were extracted by the researcher and summarised using a data extraction form to ensure consistency and clarity in the data extraction process. The data extraction form was designed for this systematic review and the selection of study characteristics included in the form was based on the Cochrane Data Collection form for intervention reviews.568 The following information was extracted for each publication and summarised in Table 23 below: study details; country; study aims; study design; participant characteristics; study findings. At the data extraction stage, a second researcher (LP) reviewed independently the data 413 extraction forms completed by the researcher for each publication to check for accuracy and completeness. 414 Systematic review – study characteristics Table 23 - Characteristics of included publications in systematic review. Author(s) and publication date Aims of study Study design Sampling and setting Participant Inclusion criteria Participants CYP participants age CYP participants i) gender ii) ethnicity CYP Participants i) primary diagnosis ii) prognosis i) Data collection method ii) presence of parent/carer during data collection with CYP Findings - themes 415 Ananth et al. (2021) To explore EOLC priorities for children with cancer and their families, and to examine relevance of existing adult EOLC quality measures Qualitative/ Thematic analysis Conducted at the Yale Cancer Center/ New Haven Children’s Hospital in Connecticut, and Boston Children’s Cancer Center in Massachusetts. AYAs with advanced cancer, ages 16-25 years, parents of children with advanced cancer (PCCs) of any age, bereaved parents whose children 10 AYAs, 25 parents of children with advanced cancer including 12 bereaved parents, and 19 healthcare Age, median (range) 20 (17−23 years) AYA participants n (%) i) female 5 (50.0) ii) Non-Hispanic White 6 (60.0), Non-Hispanic Black 2 (20.0), Hispanic and/or Latin 2 (20.0) i) cancer ii) advanced (metastatic, relapsed, progressive, or refractory cancer, or cancer where curative options are limited); does not provide further i) audio-recorded in person or by phone semi-structured interviews and in person focus groups ii) one on one Four overarching EOLC themes: direct communication with a child and family, interdisciplinary care, symptom management, and honoring family preferences for location of death 416 for children with cancer. Screened inpatient and outpatient rosters, solicited referrals from HCPs, and reviewed an internal database of deceased patients to identify participants died of cancer 6-60 months prior, and interdisciplinary HCPs who care to children with cancer professionals details on prognosis interviews Barling et al. (2014) To investigate the experience that accompanies the stages of diagnosis, treatment, dying and Qualitative/Narrative Inquiry Participants were a self-selected purposeful sample of family members who have experienced the death Family members: Parents, siblings and grandparents aged 12–65, who had lost an AYA to cancer when the 26 family members 13-23 years old i) not stated ii) not stated i) cancer: leukaemia (6), rhabdomyosarcoma (3), osteosarcoma (1), melanoma (2), glioblastoma (2), alveolar i) audio recorded open ended interviews, 16 family members had individual interviewplace of treatment; hospital experience; not fitting in; confronting illness and death 417 death of an AYA from the perspective of family members. of an AYA due to cancer. Participants were recruited through the media: Local newspapers, local television stations and a magazine. AYA was aged 16–25 years; Their AYA had died over 12 months previously to taking part in the research soft part sarcoma (1), rare sarcoma (1) ii) n/a s, 4 were interviewed as a couple, and the other 6 as families ii) n/a Barrera et al. (2005) To investigate HRQOL in children eligible for Phase 1 trials and the reasons why families consider participatinQualitative/ Content analysis Participants recruited from a children’s hospital, eligible for Phase 1 trials. Families of children with recurrent disease treated at a children’s hospital, eligible for Phase 1 trials. Families of 9 families 7-15 years old i) 4 female, 5 male (3 CYP males were interviewed, the rest were proxy interviews with parents. ii) not stated i) brain tumor (5), Neuroblastoma (3), and Acute Lymphoblastic Leukemia (1) ii) progressive, terminal condition, time interval i) semi-structured interviews, audio recorded ii) not stated physical, psychological, social, and spiritual 418 g in these trials children identified as being too ill or too overwhelmed were not approached, and children under age 7 were excluded across children from date of interview to date of death: from 1 to 19 months (M =7.11, SD=5.67) Cataudella et al. (2012) To better understand the psychological experiences of children dying from a brain tumor in their last month of life Qualitative/ Retrospective study/Thematic analysis Use of patient tumor registry, children who died of a brain tumor in Ontario Parents of children diagnosed at less than 18 years of age with a brain tumor, and who died between 2 and 12 years prior. Five families were excluded due to 24 parents (17 families/17 CYP) 1–5 years (3) 8–11 years (3) 12–19 years (11) i) 12 males, 5 females ii) not stated i) brain tumour: Brainstem glioma (4) Supratentorial primitive neuroectodermal tumor (4) High-grade astrocytoma (3) Low-grade astrocytoma (2) Medulloblasti) three semi-structured focus group interviews; audio and video recorded. Ii) n/a Emotional and cognitive changes; Awareness of impending death; Interpersonal interactions; Being treated as normal—not sick— by others; Remaining connected with others; Posttraumatic growth 419 either living more than 3 hours away or because the child’s oncologist determined they were not suitable to approach. oma (2) Ependymoma (1) Choroid plexus carcinoma (1) ii) n/a Chien et al. (2020) To explore the lived experiences of adolescents with cancer Qualitative/ Husserl’s descriptive phenomenologi cal method Participants were recruited at a medical centre using purposive sampling Aged 11 to 18 years; diagnosed with cancer in the pathological report; able to identify the disease they suffered from; ability to communicate in Taiwanese or 16 adolescents 11-17 years i) 11 males , 5 females ii) not stated i) cancer: leukemia (50%), osteo sarcoma (18.8%), brain tumor (12.5%), lymphoma (12.5%), and histiocytosis (6.2%). ii) not stated i) face-to-face, non structured interviews ii) all participants were interviewed alone suffering (caused by symptoms, therapy, cancer recurrence, the feeling of being trapped in hospital, burden of food therapy, delayed education, absence of peer relationships, and limited activities), emotional reactions (fear, 420 Mandarin, and parental consent and willingness to participate in the trial. Adolescents with cognitive disorders were excluded. worry, sadness, anger, and loneliness), and rebalancing (strong family relationships, optimistic thinking, and healthy lifestyles) Chong (2019 - PhD thesis) and Chong et al. (2021 - journal article) Explore how a good death for children can occur in the real-world context and identify factors influencing it. Qualitative/ Framework analysis Children’s hospital in Singapore. Invitation packs were sent to informal caregivers. Family caregivers were asked to suggest other caregivers Child’s age between 1 and 18 years at time of death; with a life shortening condition, (3) Died between 6 and 24 months before 5 cases: 8 bereaved parents and 14 health and social care providers from hospital, and a community case 1: 17 years; case 2: 16 years; case 3: 6 years; case 4: 11 years; case 5: 1.5 years i) 3 females, 2 males ii) not stated i) life shortening illness: cancer or non cancer ii) N/A i) audio recorded, in depth, semi structured interviews; artefacts (photographs, videos, physical items left Antecedents: Letting go, Acknowledging the child, Closure Determinants: Suffering, Control, Systems and processes Attributes: Comfort, Dying not prolonged 421 as respondents. palliative care service. behind as keepsakes); clinical notes (care plans and DNR forms) ii) N/A Ciobanu et al. (2021) To advance the understanding of the experiences of children with a LTC or LLC Qualitative/Thematic analysis Purposive selection: Children and the same children’s parents were recruited from a PC unit. Participants identified through the Children: Aged between 14 and 17 years; diagnosed with LLC/LTC; Who were willing to be interviewed; Who did not have a severe level of cognitive impairment; 20 individuals: 10 children with their mothers 14-17 years old; median age 16.5 years i) 6 males, 4 females ii) not stated i) Muscular dystrophy; Epidermolysis bullosa; Autoimmune thrombocytopenia; Mixed connective tissue disease; Dystrophic epidermolysis bullosa; Duchenne muscular dystrophy; i) audio recorded dyadic interviews, in depth, semi structured ii) mother present in all interviews with CYP awareness of illness, death and dying, spiritual response to illness, emotional response to illness, striving for normality and independence and coping strategies 422 database of a PPC unit. Who received consent from their parents; Who gave their assent to participate in the study; Who agreed to be voice-recorded during the interview;Who were not actively dying. Exclusion criteria, Children who were: unable to speak, with a severe level of intellectual Stg IV lymphoma; Abdominal lymphangioma; Acute lymphoblastic leukaemia (relapsed); Non-Hodgkin lymphoma (relapsed) ii) not stated 423 delay; in a terminal phase; distressed at the time of the interview; not Romanian speakers. 424 Coad et al. (2015) To explore perceived met and unmet needs of PPC services and care. Qualitative/ Appreciative Inquiry and framework analysis, Potential participants were invited via letter/ recruitment posters distributed through the charity, website and medical practitioners/care teams. The team also approached all health services such as hospitals and hospices and local schools with the Inclusion criteria: CYP up to 25 years with a diagnosis of a LLC and their families living in the defined study area Exclusion criteria • Children and families outside the study defined area and not using the services • Young adults not in receipt of paediatric/ children’s health 18 CYP, 59 individuals from 51 families (parents, legal guardians, family carers and siblings) 0-25 years old CYP i) female 45% ii) white British: 67% , British Asian: 11%, other: 9% other Asian background: 7%. white Irish: 2%, (Not Stated: 4%) i) Life Threatening/Limiting Condition: Congenital and Chromosomal: 21% Static Encephalopathy: 21% Neuromuscular: 17% Cancer: 13% Prematurity: 4% Other: 4% Pulmonary: 2% (Not Stated: 19%) ii) prognosis not stated i) digitally recorded semi-structured, in-depth interviews and focus groups ii) interviews were one to one medical/nursing needs were well met, provision needed for broader financial, social and emotional support alongside more responsive specialist therapies 425 information. services • College and universities serving needs of young people/adults 426 Coombes et al. (2022) To identify the symptoms, concerns, and care priorities of children with LLCs and their families Qualitative/Framework analysis Participants were recruited from six hospitals and three children’s hospices in the UK. CYP (5–17 years) with any LLC; parents/carers with a child < 18 years old with a LLC; siblings (5–17 years) of children with a life-limiting condition; HCPs with > 6 months experience of caring for children with LLC; commissioners of UK PPC services. exclusion criteria - CYP: unable 26 children 40 parents 13 siblings 15 health and social care professionals, 12 commissioners CYP 5-17 years (26), Parents of CYP 0-17 years (40) i) 17 females, 9 males ii) not stated i) LLCs - CYP participants: Cancer (6); Congenital (3); Gastrointestinal (10); Metabolic (1); Neurological (5); Respiratory (1); Parent participants of CYP with LLC: Cancer (6); Congenital (7); Gastrointestinal (4); Genitourinary (1); Infectious i) Semi-structured interviews ii) not stated Physical symptoms and concerns; spiritual and existential; emotional and psychological; social concerns; practical concerns; normality 427 to communicate via interview, using ‘draw and talk’ or play methods or via their parents; speak a language not supported by NHS translation services; currently enrolled in another study; unable to give consent/assent. disease (2); Metabolic (9) Neurological (10); Perinatal (1); Sibling participants of CYP with LLC: Congenital (3) Gastrointestinal (2); Metabolic (1); Neurological (7); ii) not stated 428 Evan et al. (2012) To determine what is quality of life from the perspective of pediatric patients with advanced disease Qualitative/Grounded theory approach (secondary analysis of larger dataset) Participants recruited from two children’s hospitals and a hospice. Children, ages 9-21 years, with advanced disease. Exclusion criteria: children unable to speak English fluently enough to participate and parents who were not mentally competent. 29 paediatric patients child mean age 14.7 years, range 9-21 years i) 17 (59%) female and 12 (41%) male ii) 17 (59% ) Caucasian, 9 (31%) Hispanic, and 1 (3%) Asian, 1 (3.4%) Other, 1 (3.4%) not stated i) Cardiology 14 (48.3%) Oncology 13 (44.8%) GI 1 (3.4%) Missing Values 1 (3.4%) ii) advanced disease with less than 20% chance of survival beyond 3 years - Post-BMT 2 (6.9%), Multiple Relapse 11 (37.9%), End Stage 4 (13.8%), Pre-transplant 3 (10.3%), Post-transplant 3 (10.3%), i) semi-structured interviews, audio and video taped ii) children interviewed without parent present three domains of QOL (physical, psychological, social) and lifestyle values 429 Congenital Heart Disease 4 (13.8%), Missing Values 2 (6.9%) Flavelle et al. (2011) To provide a unified description of an adolescent’s experience of living with and dying of cancer. Qualitative/ Phenomenological methods Health Centre in Canada, Permission to use the journal was given by patient before his death. n/a 1 child patient 15 years old i) 1 male ii) not stated i) advanced cancer: osteosarcoma in 2003 ii) n/a i) Analysis of his 90-page journal that spanned 3 months just before his death, last entry was written less than a week adolescent development, escape from illness, changing relationships, symptoms, and spirituality 430 before his death ii) n/a Gaab et al. (2013) To describe self identified factors that affect 9-to-18-year-old PPC patients and their siblings during the process of receiving PPC. This is to understand the concerns, cares, and Qualitative/ Ethnograhic study - Thematic Analysis Participants were recruited through the PPC team at a Children’s Hospital, and three condition specific Foundation centres in New Zealand. Most of the families were English-speaking, within the designated age range, and with the cognitive ability to verbalise their experiences 16 PPC patients and their siblings: 7 patients, 3 brothers, and 6 sisters from 8 families 9-18 years old i) does not state gender of PPC patient participants, gender is only stated for siblings ii) New Zealand European, Maori, other European, or Pacific Islander i) cancer, heart conditions, dystrophies ii) not stated i) audio and written diary accounts for one to four weeks ii) n/a special treatment patients received, spending time with families, feelings of being judged or discriminated against, sense of being understood themselves and of understanding others, and mortality 431 interests of PPC patients and their siblings. recruited through the PPC team, whose members identified potential participants. Garland et al. (2020) To explore empowerment in the context of young adults’ healthcare experiences who are now in advanced stages of cancer Qualitative/ Thematic analysis Participants recruited from cancer centre in Quebec. Convenience sample used with local community organizations with metastatic, chronic, or recurrent cancer of stage III or IV; with ability to complete interviews in French or English; and being 18-39 years of age. Exclusion criteria were: 12 young adults 21-39 years old; median age 33 i) 10 females, 2 males ii) not stated i) Cancer: Brain (2), Breast (4), Colorectal (1), Hodgkin's Lymphoma (2), Pancreatic (1), Thyroid (1) ii) stage III (2) , stage IV (8), did not know or recurrence (2) - no information provided in i) In-depth, individual interviews, audio recorded ii) interviews were individual one-on-one processes of waiting, managing, acting, and revisiting, notions of body ownership, facing obstacles to care, optimizing health, and (re)considering their legacy 432 having physical or cognitive limitations or difficulties with communica tion; being at imminent EOL. relation to prognosis. Gibson et al. (2014) To explore intersectionality of gender, disability, and emerging adulthood Qualitative/Critical Qualitative study - Bourdieusian lens and Arthur Frank’s notion of the narrative habitus Participants were recruited from a children’s rehabilitation center Young males with DMD 11 young men 16-27 years old i) 11 males ii) parents region of birth was provided: Canada; Europe; Latin America; (Indo) south America; south east Asia; south Asia; China; north Africa i) Duchenne muscular dystrophy (DMD) ii) not stated i) interviews, photo-elicitation, and solicited narrative diaries. Interviews were audio-recorded. ii) not stated Dominant life trajectories: school - college - work; Maintaining a positive attitude; Independent living; Doing identities: disability, adulthood, masculinities 433 Higham et al. (2013) To explore the hopes and fears of young adults with cystic fibrosis (CF). Qualitative/Grounded theory Participants were recruited at a University Hospital, at outpatient appointments Diagnosed with CF in childhood; had the Delta F508 mutation; and had a Forced Expiratory Volume in 1 s (FEV1) of >40 % 15 young adults 18-29 years old;median 24 years i) 8 females, 7 male ii) not stated i) cystic fybrosis ii) not stated i) semi-structured interviews, audio recorded ii) not stated living with unpredictable health and fear of death and dying; hopes for normality; hopes for a normal relationship and/or marriage; hopes for having children; and hopes for a normal work life 434 Ito et al. (2015) to explore the characteristics of a good death for children with cancer. Qualitative/ Content analysis Participants identified from electronic medical records. Parents who had lost a 16-year-old or younger child to cancer, a half year after their child has died and able to communicate in Japanese.Exclusion criteria: severe mental or cognitive disorder or disqualification by their child’s oncologist 10 pediatric cancer survivors, 10 bereaved family members and 20 medical professionals Median age at death (years) – 9 (3–15.9) i) 9 males ii) not stated i) Cancer: Leukemia (10), Brain tumor (4), Musculoskeletal tumor (4), Lymphoma (1) Hepatoblastoma (1) ii)n/a i) audio-recorded, in-depth, semi structured interviews ii) not stated sufficient opportunities to play freely, peer supporters, continued access to the patient’s usual activities and relationships, assurance of privacy, respect for the patient’s decisions and preferences, a sense that others acknowledge and respect the patient’s childhood, comfort care to minimize distressing symptoms, hope, not aware of the patient’s own 435 impending death, constant dignity, strong family relationships, no sense of being a burden to family members and good relationships with medical staff 436 Johnston et al. (2016) (1)To identify the nature of relationship transitions experienced by young adults with life-limiting illnesses, (2) to understand how these transitions inform the end of life clinical needs of young adults, and (3) to establish how palliative care providers may respond to Qualitative/ Thematic analysis Participants recruited through children’s hospice services in Scotland which cater for young people and their families. Inclusion criteria: Young adults who were diagnosed as having PC needs and are considered to be within the last 5 years of life; aed between 15 and 25 years ; able to provide consent; have the means to communicate with the researchers; be able to participate. Exclusion criteria:Too 12 young adults and 22 significant others including staff 17-23 years old i) not stated ii) not stated i) LLCs: Duchenne Muscular Dystrophy (n=6); Cerebral Palsy (n=2); Congenital heart disease (n=1); Rare conditions (n=3) ii) young adults considered to be within the last 5 years of life as determined by their medical practitioner i) digitally recorded (or written notes taken if permission to record was not granted) semi-structured interviews and case note review ii) alone with exception of participants with communication difficulties Dependence dichotomy, In it together, Biographical uncertainty, Conserving integrity 437 these needs. unwell to participate 438 Kenten et al. (2019) To understand the experiences of young adults with cancer for whom cure is not likely. Qualitative/ Realist analysis Purposive: Three cancer centres and associated PC services across England. Recruitment for Cohort 1 began via a national cohort study. Cohort 2 was recruited from three specialist cancer services and three hospices in England. People with cancer 1) aged 16–24 years for whom there may be specialist cancer centres and 2) 16–40 years cared for through general adult services. 33 people with cancer, 14 family carers , 22 HCPs Cohort 1: 16-24 years (3) and Cohort 2: 25-40 years (30); median (range) years: 32 (16-39) i) 14 men, 19 female ii) 22 white British, 4 other white background, 7 Asian/Asian British/black/African/Caribbean/black British i) Cancer: (19) carcinomas; (2) blood cancers; (8) sarcoma; (4) central nervous system tumours ii) a clinician-estimated prognosis of <12 months i) 69 audio-recorded semi-structured interviews with people with cancer and 6 audio-recorded workshops with bereaved family and HCPs ii) not stated seven key themes: loss of control; maintenance of normal life; continuity of care; support for professionals; support for families; importance of language chosen by professionals; and financial concerns 439 Kirk et al. (2014) (1) To examine how young people with LLCs and their parents experience transition. (2) To identify families’ and hospice staff’s perceptions of family support needs during transition. (3) To identify the implications for children’s hospices. Qualitative/Grounded theory Participants recruited from a children’s hospice. Young people aged over 16 years, not at an EOL stage, from a children’s hospice in the UK i) 14 male and 11 female ii) north American (17 participants), Asian (4), Caribbean (2), and Middle Eastern (1) 16–17 years old (4), 18–22 years old (8), 23–27 years old (2), 28–31 years old (2) (Mean age 20.5 years) i) female 7, male 9 ii) not stated i) Duchenne muscular dystrophy( 4) Other nervous system condition (4) Spinal muscular atrophy (3) Down’s syndrome (2) Congenital condition (2) Metabolic condition (1) ii) not stated i) (35) in-depth, semi-structured interviews, audio recorded ii) young people participants had the choice to be interviewed alone or accompanied Transition planning; equivalent adult health/social Services; building/support model recognises their adult status; emotional support and accessing information/services to realise their aspirations. 440 Knox et al. (2017) To understand the experience of younger adults with advanced cancer. Qualitative/ Thematic analysis (Methodical Hermeneutics approach) Participants recruited from a cancer center in Ontario, Canada. Eligible participants were identified through the screening process for a psychotherapy trial, from clinic lists and from referral by clinicians The inclusion criteria were age 18 to 39 years, fluency in English,and advanced cancer in any tumor site, as documented by the primary oncologist. Cognitive impairment was an exclusion criterion. 10 AYAs 18-35 years old; mean age 26 years AYAs i) female 7/10 ii) ethnicity not stated i) cancer: Tumor type • Sarcoma 3/10 • Breast 1/10 • Pancreas 1/10 • Brain 1/10 • Nasopharynx 1/10 • Ovarian 1/10 • Lung 1/10 • Mediastinal paraganglioma 1/10 ii) advanced; further details on prognosis not stated i) open ended discovery oriented in person interviews, audio-recorded ii) Not stated isolation in multiple dimensions, cancer is inconceivable, developmentally arrested, meaning-making and the problem of time, and staring at mortality 441 Mack et al. (2021) To identify priority domains for EOLC from the perspectives of AYAs, family caregivers, and clinicians, and to propose candidate quality indicators reflecting priorities. Qualitative/ Content analysis Purposive: Eligible participants were identified from 3 clinic sites, administrative databases, and/or referrals by clinicians and online via an AYA cancer support and advocacy organisation. Eligible participants included English- or Spanish-speaking AYA patients aged 12 to 39 years with advanced (stage IV or recurrent) cancer; family caregivers of living AYAs with advanced cancer, or of AYAs with cancer who had died within the past 5 years; and clinicians 23 AYAs, 28 family caregivers, and 29 clinicians 12-39 years; mean [SD] age, 29.3 [7.3] years AYA participants i) 12 men [52%] ii) 18 white participants [78%] i) cancer ii) Advanced (Stage IV or recurrent); does not provide further details on prognosis i) in person or by phone, audio-recorded in-depth interviews ii) not stated Identified seven EOL domains: physical symptoms, QOL, psychosocial/spiritual care, communication/decision-making, relationships with clinicians, care and treatment, independence. QOL included global quality of life; attainment of life goals, legacy, and meaning; support of personal relationships; and normalcy 442 who care for AYAs with advanced cancer. 443 Mitchell et al. (2020) To examine the delivery of healthcare, including PC, for children with LLCs/LTCs and their families and to provide understanding into how PC is delivered most effectively, and when, leading to policy relevant recommendations. Qualitative/ Thematic analysis, followed by narrative and cross-case analysis Participants approached through: 1. direct invitation via their clinical team 2. leaflets and posters displayed in public areas in the hospital 1. Children aged 5–18 years (school age) with a LLC/LTC under the care of the community children’s nursing team and their families. Exclusion criteria: Children aged <5 years and >18 years.; Families of children <5 years and >18 years old.; Children and families with whom 31 participants from 14 families; 10 children with life-limiting or life-threatening conditions and 21 family members. 5-18 years old; median age 9 years i) 9 males, 5 females (including family proxy reports) ii) 10 white British, 3 Asian, 1 African (including family proxy reports) i) cancer; congenital conditions ii) not provided i) 41 audio recorded, longitudinal interviews over a 12-month period ii) interviewed individually or together with family member depending on CYP preference To feel respected, heard and supported by healthcare services; delivery of PC depends on the development of trusted relationships between the child, their family, and their healthcare professionals 444 the research team has clinical contact.; Children and/or families who do not wish to participate.; Children who are too unwell, but their family members may participate; Children unable to participate in a conversational interview, but their family members 445 may participate; Children and families who are unable to provide informed consent in English will not be approached for interview. 446 Nicholas et al. (2011) To examine experiences and perceptions of children with end-stage renal disease (ESRD) Qualitative/ Ethnography - content analysis Participants were i informed of the study by a known health care provider. Children and adolescents ≤18 years of age undergoing ESRD treatment 25 children and adolescents 7-18 years old, mean age of 13.4 years i) 14 male and 11 female ii) north American (17 participants), Asian (4), Caribbean (2), and Middle Eastern (1) i)ESRD ii) not stated i) ethnographic semi structured interviews, audiotaped ii) not stated not feeling “normal”; Knowledge about ESRD and Involvement in Treatment; frequent absence from school; gaining responsibility for ESRD care; relying on family, friends, and health care providers; adjusting despite adversity 447 Patterson et al. (2012) To inform understanding of the needs of emerging adults with a diagnosis of cancer from a developmental perspective that appreciates the key transitional tasks of emerging adulthood. Qualitative/ Content analysis Eligible participants were drawn from a list of patient members of a national organisation Adults (aged 18 - 24 years) who have had a cancer diagnosis regardless of their current treatment status and if their cancer diagnosis had occurred after they were 11 years of age. No distinction made between participants who were on treatment or 14 young people mean age 22 years (range from 20 to 25 years i) 8 female, 6 male ii) not stated i) cancer: brain, leukemia, lymphoma, Hodgkin’s lymphoma, lung, and non-Hodgkin lymphoma ii) not stated i) focus groups (9 participants) and telephone interviews (5 participants), sessions were recorded in writing and were not audio recorded ii) not stated information, healthcare provision, daily living, interpersonal support, identity renegotiation, and emotional distress 448 posttreatment Sodergren et al. (2018) To capture the HRQoL issues described by AYAs with cancer Qualitative/Thematic analysis Participants recruited from research centres across France, Israel, Norway, Poland, the Netherlands and UK. AYAs aged between 14 and 25 years receiving treatment or PC for cancer. Those who had completed treatment were excluded from the study. 45 AYAs 14-25 years old; mean age 20.3 years i) males (53%) and females (47%) ii) not stated i) 12 different cancer types with leukemia (28%) and lymphoma (18%) the most predominant diagnoses ii) majority of participants currently on curative treatment (78%) i) audio-recorded, semi structured interviews ii) all participants were interviewed alone Symptoms (pain, nausea, vomiting); activity limitations (education, leisure time activities); disrupted life plans; social (loss of friends, family life); emotional (depression, anxiety); body image; self- appraisals (greater 449 maturity, braver); outlook on life (altered priorities, increased motivation to achieve); lifestyle (restricted diet, avoidance of infections), treatment-related (absence of age-appropriate information, treatment burden); fertility; and financial concerns 450 Taylor et al. (2021) To inform clinical guidelines for PPC Qualitative/ Thematic analysis Young people were recruited by a national charity. Initial convenience sample and purposeful sampling toards the end of data collection. Young people (age 12-18) diagnosed with a LLC, who were aware their condition was potentially life-shortening, were able to communicate verbally and had capacity to consent 14 young people 12-18 years old i) 7 females/7 males ii) 3 young people from ethnic minority background i) LLCs or LTCs: congenital and chromosomal, neuromuscular, cancer, and pulmonary and respiratory. Ii) condition potentially life-shortening but not further information on prognosis stated i) 3 focus groups (with 2-7 participants) and 2 individual interviews. All focus groups and 1 interview was audio recorded; extensive notes were taken following the second interview that was not audio-recorded ii) Parent/cabeing treated as individuals with individual needs and preferences; quality of care more important than place; emotional well-being; living as a young person 451 rers were not present during focus groups or interviews. 452 453 Data extraction form for studies included in systematic review Date form completed: Name of person extracting data: Paper details: Author: Title: Year: Country of origin: Publication type (e.g. journal article, Thesis) : Study characteristics: Aim of study: Is focus of study broad, on what matters in life, or specific to what matters in care?: Research design/Methodology: Sampling/Setting: Participant inclusion and exclusion criteria: Participants (with number of): Age of CYP: CYP Gender and CYP ethnicity: Primary CYP diagnosis and prognosis: Data collection method and presence of parent/carer at data collection with CYP: Inductive or Deductive coding: Findings (themes): 454 455 Assessment of quality: Prior to providing a narrative synthesis of the included studies it was important to assess the quality of the studies identified to establish a quality threshold and exclude any poor-quality studies below that. The quality appraisal of the included studies was undertaken for each included publication by the researcher and a second reviewer (LP). Given that all included studies used qualitative data collection and analysis methods, the Critical Appraisal Skills Programme (CASP) Qualitative research checklist was used to assess the quality of the studies.569 The CASP tool was chosen for the appraisal as it includes ethics as one of the appraisal items to consider and it is recommended for use with novice researchers as it is an easy tool to understand.570 Given the sensitive research topic and vulnerable population, it was important to include ethical issues as a quality criterion in the assessment of studies. Furthermore, the systematic review was undertaken by a novice researcher who did not have experience of conducting a systematic review prior to this study. The reviewers completed independently a CASP Qualitative checklist form (2018) for each publication.569 The checklist consists of 10 questions to assist with the assessment of qualitative studies in a systematic way. As suggested by the CASP checklist guidelines, the first two questions were used as screening questions to decide whether it is worth proceeding with a full assessment or exclude a study. Both reviewers agreed that all publications identified in the review were eligible for a full assessment 456 following the completion of the two screening questions and no publications were excluded. The completed checklist forms were compared, and consensus was reached with discussion between the two reviewers to resolve disagreements made due to some oversights. A summary from the CASP assessment is provided for each of the publications included in Table 24 below. As CASP has been developed for use as an educational tool, the use of a scoring system is not recommended with this tool.569 457 Table 24 - Quality assessment summary against CASP checklist criteria Author(s) and publication date Quality assessment summary Ananth et al. (2021) Minimal information on sampling, data collection and analysis. The researcher explains how participants were selected and the eligibility criteria but not why the participants selected were the most appropriate. There is no other discussion around recruitment e.g. why some people chose not to take part. The setting for data collection is provided but no justification given. It is clear how data were collected, but no justification is provided for the methods chosen. Justifications are lacking, but given the level of detail provided, it is possible to make a judgement that the data appeared to be collected in a way that addresses the research issue. There is no mention of the researcher critically examining their own role, potential bias, and influence during formulation of the research questions or data collection. The researchers did consider how their own perspectives might influence the data analysis. The researcher states that ethical approval was obtained ‘by respective institutional review boards at each institution’ but no further specifics of these institutions are mentioned. Verbal consent was obtained by participants but there is no mention of written informed consent and the ethical process is not clear. Barling et al. (2014) Good level of information provided on recruitment, data collection and analysis. The analysis process is discussed in detail and the author used a figure to illustrate the different steps of the analysis process. The author also discussed and justified amendment in ethics process following request by participants. However, the author does not discuss or justify why some of participants did not fit the age group inclusion criteria. Discussion on potential bias during data collection and analysis is limited and there is no mention on who/or how many researchers were involved in the analysis process. There is discussion of the evidence for the author's arguments, but no discussion of evidence against. However, the researcher has discussed the credibility of their findings (triangulation). The researcher discusses the contribution the study makes to existing knowledge and broadly mentions areas where research is necessary. The researchers do not discuss whether or how the findings can be transferred to other populations. 458 Barrera et al. (2005) Good level of information on recruitment and data collection, with details on why some people did not take part in the study, what the interviews included and differences between interviews conducted with families and young people. Data saturation issues were discussed in the study limitations. Ethics approval obtained and process of obtaining consent/assent was discussed. Findings discussed in detail with use of quoting from the interviews. Good level of information on the data analysis process and contradictory data are considered. However, there is no mention of the researcher critically examining their own role, and influence during analysis and selection of data. Only one researcher was involved in analysing the data. Study contribution and transferability of data was discussed but discussion on recommendations for research is limited. Cataudella et al. (2012) Good level of information provided on recruitment, data collection and analysis. Author discussed how the participants were recruited and why some families were excluded from the study but not why the participants selected where the most appropriate. Background information provided for the focus group facilitators and relationship with participants mentioned (data collection conducted in neutral environment). Ethics approval obtained and ethics process of obtaining consent discussed but it is not clear whether ethical standards were maintained (e.g. potential participants were provided with information verbally only; consent is mentioned but not informed consent). Analysis process discussed in detail including background information on the researchers conducting the analysis. Findings discussed with use of exemplar quotes for each theme. Limitations and contributions of research findings mentioned. New areas where research is necessary mentioned but there is no discussion on whether or how the findings can be transferred to other populations or considered other ways the research may be used. Chien et al. (2020) Good level of information provided on recruitment and data collection and detailed information on ethical considerations. In-depth description of the analysis process. Sufficient data are presented to support the findings and contradictory data are considered. The researcher discusses the contribution the study makes to existing knowledge and identify new areas where research is necessary. However, the researchers do not discuss whether or how the findings can be transferred to other populations or considered other ways the research may be used. 459 Chong (2019 - PhD thesis) and Chong et al. (2021 - journal article) In-depth information was provided for recruitment, data collection, ethical considerations and analysis. Findings were supported with the use of quoting, diagrams and tables summarising and comparing themes for each of the five cases. The interview guide, informed by study objectives, underwent iterations as data were analysed. The author, as an 'insider researcher' at the setting that data collection took place, considered potential power influence and bias and reported ways he attempted to minimise potential issues raising from this. The thesis also contains a section on 'Perspective of the researcher (p75) and a section on reflexivity (p150). A research journal documented emerging ideas. The author also considered transferability of findings and how the study contributes to existing knowledge and understanding. Ciobanu et al. (2021) There is sufficient information on the recruitment process but no justification for the methods used is provided and information on data collection and analysis is limited with no sufficient information on the topics covered during the interviews. There is use of quoting to support the findings and the findings are discussed in relation to the research question. The relationship between the researcher conducting data collection, as an insider researcher, and participants is not adequately considered. Credibility of findings is not mentioned. Transferability of findings to other populations is mentioned briefly. Coad et al. (2015) Good level of information is given on recruitment and ethics process but there is limited information on data collection and how the interviews were conducted. There is no information on data collection from focus groups. There is good level of information on the analysis which was undertaken by two researchers. Findings and their credibility are clearly stated and the study contribution and recommendations for service improvement are discussed in detail. However, there is no discussion of the evidence both for and against the researcher's arguments and the researchers do not appear to identify new areas where research is necessary. 460 Coombes et al. (2022) Good level of information provided on how the data was collected, how the topic guide for the interviews was developed and reviewed, on the analysis process and on the ethics process. However, the author does not give a justification for the method and setting used to recruit participants and there is no information as to why the participants selected were the most appropriate to provide access to the type of knowledge sought by the study. The researcher does not appear to critically examine their own role, potential bias, and influence during (a) formulation of the research questions (b) data collection, including sample recruitment and choice of location. Findings are explicit and there is discussion of the evidence both for and against the author’s arguments. The author discussed credibility of their findings and there is adequate discussion of the evidence both for and against the researcher’s arguments. Contribution of the study to existing knowledge discussed areas where research is necessary identified. There is no discussion on whether or how the findings can be transferred to other populations or context. Evan et al. (2012) Good level of information provided on recruitment, sampling, data collection and analysis. The author provides information on the ethics process and there is detailed discussion on how the researchers attempt to minimise bias during data collection and analysis with reliability checks of transcripts and multiple coders for the analysis stage. Coders used a preliminary codebook to ensure consistency between them and to eliminate individual coder bias. The author provided detailed information on the response rate for the interviews and the reasons why some people who expressed interest did not take part in the study. Findings are stated clearly, and themes summarised in a table. The author discusses in depth the contribution of the study and new areas where future research is necessary. Although this is a qualitative paper, certain aspects of it may be more appropriate for a quantitative study: e.g., frequency of themes mentioned. Flavelle et al. (2011) There is no information on recruitment, sampling, and data collection as this was a single case study were the author analysed data from a patient's journal. However, the author justifies their decision to use a single case study and why this would be appropriate for the aims of the research. As the study involves journal data, consideration of relationship between participant and researcher during data collection, sampling and recruitment, are not relevant for this study. Good level of information is provided in the data analysis process, including how themes were developed. However, information on how the author critically examined their role and potential bias during analysis is very brief. Findings are 461 explicit with use of quoting of text and drawings from the patient's journal. The author also illustrated a schematic representation of the findings. The researcher discusses the contribution the study makes to existing knowledge but does not appear to identify new areas where research is necessary. The researcher does not discuss the value of single case studies. Gaab et al. (2013) Good level of information provided on recruitment and data collection, but participant characteristics were not stated individually for each participant, e.g. the gender, ethnicity and condition for each individual PPC patient participant was not mentioned. Detailed information was provided on the context of recruitment material (copy of materials in Appendix) how they were designed. Ethical approval was obtained and details on how participants were approached for the study was provided. In depth information on the data analysis process was provided with more than one analyst being involved in coding and a Maori researcher involved in analysis to ensure 'cultural sensitivity. However, the researcher does not discuss saturation. Findings are explicit with use of quoting from the data to support the findings. Transferability of findings was discussed briefly and information on the study contribution to existing knowledge was provided along with suggestions for future research. Garland et al. (2020) Good level of information provided on recruitment, sampling, data collection and analysis. However, there is no justification provided for choice of methods. The findings are clearly stated and supported with quoting and a diagram presenting the themes. The researcher considered their own role and bias during formulation of the research question and interview topic guide, during data collection and analysis with more than one researcher being involved in data analysis. Gibson et al. (2014) Very little detail is provided about how participants were recruited and no eligibility criteria are provided. The authors do not state why the participants they selected were the most appropriate for the aims of the study. There is no other discussion around recruitment e.g., why some people chose not to take part. The setting for the data collection is provided but no justification is given. It is clear how data were collected, and justification is provided for the methods chosen. The researcher has made the methods explicit (e.g., detailed information about the instructions provided to participants and data was collected). Saturation is not discussed. Ethical approval was granted for the research but there are no details provided about how the research was explained to participants. There is an in-depth description of the analysis process and findings are explicit. Sufficient data are presented to support the findings. Contradictory data are considered. There is no mention of the researcher critically examining their own role, potential bias and influence during analysis and selection of data for presentation. The researcher discusses 462 the contribution the study makes to existing knowledge and new areas where research is necessary are broadly mentioned. Higham et al. (2013) Good level of information on how participants were recruited and reasons for exclusion criteria given. Good level of information on data collection with details on how/why data collection method was updated during the study. Good level of information on analysis with more than one person taking part in the analysis independently. Topic guide was provided. Ethical approval obtained but no details included on how the study was explained to participants and ethical considerations not discussed. Credibility of findings discussed, and one more than one analyst was involved. Contribution of study to existing understanding and transferability of findings discussed and new areas where research is necessary were identified. Ito et al. (2015) Sufficient level of information on recruitment, sampling and analysis including detailed table summarising all the findings. Data collection details are brief and the author does not provide details on the topic guide used in the interviews and whether the methods were modified during the study. The research has considered potential institutional bias as one of the limitations of the study. Contribution of the study to existing knowledge is discussed extensively but the author did not identify explicitly new areas where research is necessary. Johnston et al. (2016) Good level of information given on recruitment, data collection and analysis. Potential ethical risks during data collection were considered. There is involvement of a research steering group throughout the study and all study documents have been reviewed by members of the public. Use of mind mapping to manage potential bias in analysing data. However, background information on who was conducting data collection and analysis was not made explicit. The findings are explicit but there is no discussion of the evidence both for and against the researcher's arguments. The researcher discusses the contribution the study makes to existing knowledge and identify new areas where research is necessary. The researchers do not discuss whether or how the findings can be transferred to other populations or considered other ways the research may be used 463 Kenten et al. (2019) Good level of information given on sampling, data collection and in-depth information on analysis. However, there is no information given on the research team conducting data collection and analysis and there is no mention on consideration of their own role and potential bias in data collection and analysis. The number of participants aged 16-24 (3) is very low in contrast to number participants over 24 yo. (30). However, the low number of younger participants was made explicit in the limitations section of the study. Ethical approval was granted for the research. Details of how the research was explained to participants suggest ethical standards were maintained (e.g. potential participants were provided with an information sheet and consent form). Written consent was obtained from participants. However, the researcher did not discuss any other issues raised by the study (e.g., effects of the study on the participants during and after the study). Findings are stated clearly, and credibility discussed. The researcher discusses the contribution the study makes to existing knowledge and identify new areas where research is necessary. The researchers do not discuss whether or how the findings can be transferred to other populations. Kirk et al. (2014) Details provided on how participants were recruited and why some people chose not to take part. However, information on data collection is brief, with limited information on what the interview included. Copy or examples of the topic guide questions were not provided. Information about the researcher conducting the interviews is brief but author states that the researcher was unknown to the participants. Ethical approval was obtained, anonymity/confidentiality was considered, and the author states that distress and safeguarding protocols were developed. Very little information on data analysis process, authors state that both authors were involved in data analysis to develop categories iteratively from the data but there is no other information provided on the analysis process. Contribution to knowledge, transferability of findings and recommendations for research are discussed. Knox et al. (2017) Good level of information is given on recruitment, data collection and analysis and relationship between researchers and participants was considered. The authors state that the study was conducted in accordance with the ethical standards of the research committee, but no further information is given on how ethical standards were considered in the study. Sufficient data are presented to support the findings and findings are explicit. Contradictory data are taken considered. The researcher critically examines their own role, potential bias and influence during analysis and selection of data for presentation. The researcher discusses 464 the contribution the study makes to existing knowledge but does not appear to identify new areas where research is necessary. The researcher discusses whether or how the findings can be transferred to other cultural settings. Mack et al. (2021) Good level of information given on sampling, data collection and analysis. But justification for research design was not explicit. No information on ethical approval process given and only mentions that written consent was obtained from all participants. Although information on research team is given it does not state explicitly whether team critically examined their own role or potential bias and influence during analysis/presentation. Findings are explicit. Study contribution to existing knowledge is discussed and new areas of research identified. There is no discussion on how the findings can be transferred to other populations or considered other ways the research may be used Mitchell et al. (2020) The author provides detailed information for sampling, recruitment, data collection and analysis in this thesis. Justification for research design and methods are made explicit. In-depth description of the analysis process. Quoting from interviews was provided to support the findings. Ethical considerations were discussed in detail. The author critically examines their own role and potential bias during data collection and analysis. Study contribution discussed in detail and new areas where research is necessary identified. Nicholas et al. (2011) Information on analysis process discussed and researchers involved in the analysis process sought to ensure trustworthiness., authenticity and reliability by using several methods - such as critical checking and ongoing refinement of findings, reviewing of coding by two independent reviewers (10% of the data). However, there is very little information given on the research team, setting, recruitment and data collection process. The author provides examples of some of the questions asked during the interview, but no copy of the full topic guide is provided. Ethics and informed consent discussed briefly. Contribution of the study discussed, and new areas of research identified. 465 Patterson et al. (2012) Detailed information given on recruitment and data collection, including why some people chose not to take part and details on how data was collected. Information on data analysis process was brief but the author mentioned that to enhance credibility, two researchers analysed the data, and a third independent research peer then reviewed the data analysis process. Findings were discussed in depth with use of table to summarise findings and use of quoting to support findings. Discussion on contribution of the study and new areas where research is necessary was included. Transferability of findings was discussed. Ethics approval obtained and ethical considerations stated Sodergren et al. (2018) Adequate level of information on data collection and setting, details on how data were collected, and a copy of the interview template provided. Background information provided for the interviewers and relationship with participants considered. Ethics approval obtained and process of obtaining consent/assent was discussed. Good level of information provided on the analysis process. Five researchers were involved in the analysis process and an audit trail of all research team discussions was recorded to enhance rigour. Data saturation was discussed. Findings were explicit and presented with use of quoting examples. Contribution and transferability of study to existing knowledge was discussed, recommendations for research were stated but new areas where research is necessary do not appear to be identified. Taylor et al. (2021) Extensive information given on recruitment, data collection and analysis. Data saturation was not mentioned in the study. The analytical process was discussed and a table to summarise different steps taken in the analytical process was provided. The analysis was carried out by different members of the research team each with different background. Quality of the consultation/study was tested against several criteria: dependability; credibility; trustworthiness and examples given on how they tested against these criteria. Information on ethics, such as consent, anonymity and confidentiality process was provided. Findings were discussed in relation to the research questions and with use of quoting to support themes that emerged but there is no discussion of the evidence both for and against the author’s argument. Contribution of the study discussed but new areas of research where not identified and there was no discussion on transferability of findings to other populations or consideration of other ways the findings may be used. 466 467 Appendix 2: Screening questionnaires Quality of life for young people living with illness Background information I have read the information sheet and am interested in taking part in the above study. Name: ………………………………………… Postal address: ………………………………… Telephone number: Email address: ………………………….. Please circle preferred way of contact: Post Telephone Email Have you been bereaved at least six months ago? Yes No Relationship to the young person ……………….. Age at which the young person died. ……………… Are you: Male Female Prefer not to say Prefer to self-describe ………………… Young person: Male Female Prefer not to say Prefer to self-describe ………………… 468 How would you describe your ethnicity? ………………………………. How would you describe the young person’s ethnicity? ………………………………. What was the young person’s diagnosis? ……………………………………………………… Did the young person receive palliative and/or end-of-life care for their condition? Yes No If so, please provide a few further details: ……………………………………………………………………………………………………………………………………………………………………………………………………………… Through which organisation did you find out about this study? …………………………… Thank you very much for completing this form. Please return the form to Isabella.floredin@bristol.ac.uk . 469 Quality of life for young people living with illness Background information I have read the information sheet and am interested in taking part in the above study. Name: …………………………………………. .............................................. Postal address: ………………………………… Telephone number: Email address: ………………………………….. Please circle preferred way of contact: Post Telephone Email What is your relationship to the young person? ……………….. How old is the young person? ……………… Date of birth: ……………. Are you: Male Female Prefer not to say Prefer to self-describe ………………… Is the young person: Male Female Prefer not to say Prefer to self-describe ………………… How would you describe your ethnicity? ………………………………. How would you describe the young person’s ethnicity? ………………………………. 470 What is the young person’s diagnosis? ……………………………………………………… Does the young person receive care for their condition? Yes No If so, please provide further details: ……………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………… Who does the young person usually live with e.g., mother? …………………………………………… Are you currently in paid employment? Yes, Full time Yes, Part time Not working Through which organisation did you find out about this study? …………………………… Thank you very much for completing this form. Please return the form to Isabella.floredin@bristol.ac.uk . 471 Quality of life for young people living with illness Background information I have read the information sheet and am interested in taking part in the above study. Name: …………………………………………. .............................................. Postal address: ………………………………… Telephone number: Email address: ………………………………….. Please circle preferred way of contact: Post Telephone Email Date of birth: ……………… How old are you? ……………… Are you: Male Female Prefer not to say Prefer to self-describe ……………………. How would you describe your ethnicity? ………………………………. What is your diagnosis? ……………………………………………………… Do you receive care for your condition? Yes No If so, please provide further details : ……………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………… 472 Who do you usually live with e.g., mother? ……………………………………………………………. Through which organisation did you find out about this study? ……………………………………... Thank you very much for completing this form. Please return the form by email at isabella.floredin@bristol.ac.uk 473 Appendix 3: Study posters Study posters for recruitment circulated in Charitable organisations, hospices, Facebook and Instagram: 474 475 476 477 Appendix 4: FREC approval letters 478 479 13.10.2021 Dear Ms Isabella Floredin Ref: 7773 Title: Quality of life for young people with life-threatening and life-limiting conditions Thank you for submitting your ethics application for the above-named study. The Faculty of Health Science Research Ethics Committee (FREC) has reviewed your research ethics amendment request and we can confirm that your ethics application has received a favourable ethical opinion. Please note that the Faculty of Health Science Research Ethics Committee (FREC) expects to be notified of any changes or deviations in the study. Dr Allison Fulford Faculty of Health Science Research Ethics Committee (FREC) 480 07.03.2022 Dear Ms Isabella Floredin Ref: 9194 Title: Quality of life for young people with life-threatening and life-limiting conditions - Phase 3 Thank you for submitting your ethics application for the above-named study. The Faculty of Health Science Research Ethics Committee (FREC) has reviewed your ethics application and we can confirm that your ethics application has received a favourable ethical opinion. Please note that the Faculty of Health Science Research Ethics Committee (FREC) expects to be notified of any changes or deviations in the study. If you have any queries regarding your research ethics review please contact research-ethics@bristol.ac.uk Yours sincerely, Ingrid Hoxha Research Ethics Assistant pp Dr Vikki Wylde Faculty of Health Science Research Ethics Committee (FREC) 481 Appendix 5: Study webpage and twitter posters Study webpage: 482 Twitter posters 483 484 Appendix 6: Newsletter 485 486 Appendix 7: Interview topic guides PhD research study Topic Guide for close person Prior to interview: • Go through information sheet with participant • Check that participant is happy with the information provided in the information sheet and that they have signed the consent form • Give the participant a chance to ask questions • Inform the participant that with their permission the interview will be audio-recorded PART 1 – IN-DEPTH INTERVIEW: Introduction • Could you tell me a bit about yourself? • Could you tell me a bit about the young person you were close to? o Relationship to young person; age of young person; time spent with them, and type of support/care provided to young person Questions on young person’s condition: • Could you tell me a bit about the condition the young person had? o When did you/they find out about it and how? o How did the condition affect their day-to-day life? 487 o What support and care did they receive? Questions about things that are important in the young person’s life: • Could you tell me about what you feel was important to the young person near the end of their life? • You talked about… o How important was that to the young person / Why? o What else was important in their life? Why? • Do you think that what was important to the young person changed over time? /How? PART 2 – ‘THINK ALOUD’ TECHNIQUE AND SEMI-STRUCTURED INTERVIEW: • Provide the participant with the ICECAP-CPM and explain that the questionnaire asks about the quality of life for Bereaved close persons, and it has been developed for use in older adult population, so we are interested to know if it can be used for those who have been close to a younger person who has died. • Ask the participant to verbalise their thoughts by thinking out loud whilst completing the measure. Also ask the participant to say which choice they would pick for each question on the questionnaire whilst thinking out loud. Explain that we are interested in what they are thinking while they are responding to the questions • Remain silent unless the participant is silent for longer than 10 seconds at which point ask the participant to keep thinking aloud • Following completion of ICECAP-CPM ask the participant: 488 o How did you find completing the questionnaire? Researcher will explore thoughts the participant expressed whilst completing the questionnaire o Have you found any aspects of the questionnaire challenging? If so, please explain. Researcher will explore and clarify any issues that came up during ‘think aloud’ task. o How relevant was each question to your experience as a close person? Is there a question that you would add to the questionnaire? What would that be? / Is there any question you think is not relevant to you? Why? End of interview: • Thank participant for taking part in the interview • Ask participant how they found taking part and go through debriefing form with them • Ask if they would be happy to potentially take part in a second interview (for phase 1) • Ask whether they would like to hear about the results of the study 489 PhD research study Topic Guide for parent/guardian Prior to interview: • Go through information sheet with participant • Check that participant is happy with the information provided in the information sheet and that they have signed the consent form • Give the participant a chance to ask questions • Inform the participant that with their permission the interview will be audio-recorded Introduction • Could you tell me a bit about yourself/your child? (Hobbies/Interests) Questions on young person’s condition: • Could you tell me a bit about the condition your child has? o When did you/they find out about it and how? o How does the condition affect their day-to-day life? o What support have they been receiving? Questions about things that are important in the young person’s life: • Could you tell me about what you feel is important to your child right now… • You talked about… o How important is that to your child? / Why? 490 o What else is important in their life right now? Why? • Do you think that what is important to your child has changed since….? /How?/ Do you think that what is important to your child may change going forward? • Do you think that the things that are important to your child have changed since the pandemic started? How? • What would they find helpful right now? How would it help? End of interview: • Thank participant for taking part in the interview • Ask participant how they found taking part and go through debriefing form with them • Ask if they would be happy to potentially take part in a second interview (for phase 1) • Ask whether they would like to hear about the results of the study 491 PhD research study Topic Guide for young person Prior to interview: • Go through information sheet with participant • Check that participant is happy with the information provided in the information sheet and that they, or their parent/guardian (if under 16 years old) have signed the consent form • Give the participant and parent/guardian (or other person present at interview) a chance to ask questions • Inform the participant that with their permission the interview will be audio-recorded Introduction: • Could you tell me a bit about yourself? (Hobbies/interests/Likes and Dislikes) Questions on young person’s condition: • Could you tell me a bit about the condition you have? o When did you find out about it/how? o How does the condition affect your day-to-day life? o What support have you been receiving? Questions about things that are important in the young person’s life: • Could you tell me about what is important to you in life right now….. • You talked about … o How important is that to you? / Why? o What else is important to you in your life right now? Why? 492 • Do you think that what is important to you has changed since….? /How?/ Do you think that what is important to you may change going forward? • Do you think that the things that are important to you have changed since the pandemic started? How? • What would you find helpful right now? How would it help? End of interview: • Thank participant for taking part in the interview. • Ask participant how they found taking part and go through debriefing form with them. • Ask if they would be happy to potentially take part in a second interview (for phase 1) • Ask whether they would like to hear about the results of the study. 493 Appendix 8: Cognitive interviewing Error rating sheet template for ICECAP-CPM: 494 495 496 ICECAP-CPM: 497 498 Appendix 9: Information sheets Close person Information Sheet for taking part in study Study title: Quality of life for young people living with illness Researcher name: Isabella Floredin This is an invitation to take part in an interview for a research study. Please take time to read the following information and feel free to ask for more information if something is not clear. What is the study about? This study aims to explore the things that are important to young people (14-25 years old) with life-limiting and life-threatening conditions. Some young people are taking part themselves, but we would also like to ask those close to a young person what they think has been important to the young person. The study findings may help inform care and support for young people based on what young people and those close to them view as important. Am I eligible to take part? We would like you to participate in the study if you have been close to a young person, aged 14-25 years old, with a life-limiting or life-threatening condition, who died at least six months ago as a result of this condition. What does taking part involve? • We will ask you to complete and return the consent form enclosed. • Then, the study researcher will be in touch with you to arrange a virtual interview, for example via Skype or a similar online platform, at a time that is most convenient to you. If you do not have access to an electronic device to use for the interview, please let the researcher know. • The interview will last for up to an hour but with breaks if needed and with your permission it will be audio and video recorded. 499 • You are free to ask for the interview or recording to stop at any time. You can withdraw from the study at any time if you change your mind and you do not have to answer all the questions if you do not want to. • The researcher will give you a £10 voucher as a small thank you • If you would like to withdraw from the study after your interview, please let us know within a week of the interview and your data will then be destroyed. If you notify us after this time period, it is possible that the data may already have been analysed and used to inform future interviews and thus it will not be possible to withdraw your data from the study • We may not be able to interview all those interested in taking part in the study. In that case, the researcher will contact you following receipt of your consent form to let you know and your consent form will be securely destroyed by the researcher. What will the interview be about? The interview will be divided into two parts. In the first part of the interview, we would like to ask you about the things that you feel have been important to the young person you were close to. In the second part of the interview, we would like you to complete a short questionnaire asking about your own wellbeing and think aloud whilst completing the questionnaire. We will then ask you what you thought of the questionnaire. During the interview, it is possible that sensitive topics may arise that are upsetting or difficult to answer. If you feel uncomfortable in any way during or after the interview, please let the interviewer know. Will my participation and personal information be kept confidential? • All personal information collected from you will be kept confidential and your name will be removed from any information connected to you. Any names and places you mention will be anonymised so that you cannot be recognised from this. • The interview will be written up and anonymised and the interview audio and video recording will be kept on a secure computer network at the University of Bristol. Only those directly involved in the study will be able to access this information, that is Isabella Floredin and her academic supervisors. • Any paper copies including your personal information will be uploaded electronically and then paper copies will be destroyed. • If you tell us that you are at risk to yourself or others then it is a statutory requirement to report this as part of safeguarding measures If I decide I would like to take part: 500 • Please complete and return the consent form enclosed. This information sheet is for you to keep • Then, Isabella will get in touch with you to arrange a time for the interview. What if I have a concern or complaint? If you have a concern or complaint as a participant in this study, you may contact the Faculty of Health Science Research Ethics Committee, via the Research Governance Team, research-governance@bristol.ac.uk Where can I get more information? If you would like more information about this study or a copy of your interview transcript, please contact Isabella Floredin at - Address: Population health sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies Road, Bristol, BS8 1NU, Tel: +441174282096, Email: isabella.floredin@bristol.ac.uk This study is part of a University of Bristol project funded by the Wellcome Trust and has been approved by the University of Bristol Research Ethics Committee. 501 Parent/Guardian Information Sheet for taking part in study Study title: Quality of life for young people living with illness Researcher name: Isabella Floredin This is an invitation to take part in an interview for a research study. Please take time to read the following information and feel free to ask for more information if something is not clear. What is the study about? This study aims to explore the things that are important to young people (14-25 years old) who are seriously ill. This will include asking parents/guardians of young people who are currently not able to take part in the study directly about the things that matter to their child. The study findings may help inform the way care for young people is funded based on their needs and what young people view as important. Am I eligible to take part? We would like you to participate in the study if you are a parent/guardian of: a young person, 14 – 25 years old, who is seriously ill and who is currently not able to take part in the study directly. What does taking part involve? • We will ask you to complete and return the questionnaire and consent form enclosed. • Then, the study researcher will be in touch with you to arrange a virtual interview, for example via Skype or a similar online platform, at a time that is most convenient to you. If you do not have access to an electronic device to use for the interview, please let the researcher know. • The interview will last for up to an hour but with breaks if needed and with your permission it will be audio and video recorded. 502 • You are free to ask for the interview or recording to stop at any time. You can withdraw from the study at any time if you change your mind and you do not have to answer all the questions if you do not want to. • The researcher will give you a £10 voucher as a thank you • The interviewer may invite you to take part in a second interview at a later date but you do not have to take part. • If you would like to withdraw from the study after your interview, please let us know within a week of the interview and your data will then be destroyed. If you notify us after this time period, it is possible that the data may already have been analysed and used to inform future interviews and thus it will not be possible to withdraw your data from the study • We may not be able to interview all those interested in taking part in the study. In that case, the researcher will contact you following receipt of your questionnaire and consent form to let you know and your questionnaire and consent form will be securely destroyed by the researcher. What will the interview be about? In the interview you will be asked about the things that you feel are important to your child. During the interview, it is possible that sensitive topics may arise that are upsetting or difficult to answer. If you feel uncomfortable in any way during or after the interview, please let the interviewer know. Will my participation and personal information be kept confidential? • All personal information collected from you will be kept confidential and your name will be removed from any information connected to you. Any names and places you mention will be anonymised so that you cannot be recognised from this. • The interview will be written up and anonymised and the interview audio and video recording will be kept on a secure computer network at the University of Bristol. Only those directly involved in the study will be able to access this information, that is Isabella Floredin and her academic supervisors. • Any paper copies including your personal information will be uploaded electronically and then paper copies will be destroyed. • If you tell us that you are at risk to yourself or others then it is a statutory requirement to report this as part of safeguarding measures If I decide I would like to take part: • Please complete and return the questionnaire and consent form enclosed. This information sheet is for you to keep 503 • Then, Isabella will get in touch with you to arrange a time for the interview. What if I have a concern or complaint? If you have a concern or complaint as a participant in this study, you may contact the Faculty of Health Science Research Ethics Committee, via the Research Governance Team, research-governance@bristol.ac.uk Where can I get more information? If you would like more information about this study or a copy of your interview transcript, please contact Isabella Floredin at - Address: Population health sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies Road, Bristol, BS8 1NU, Tel: +441174282096, Email: isabella.floredin@bristol.ac.uk This study is part of a University of Bristol project funded by the Wellcome Trust and has been approved by the University of Bristol Research Ethics Committee. 504 Young person Information Sheet for taking part in study Study title: Quality of life for young people living with illness Researcher name: Isabella Floredin This is an invitation to take part in an interview for a research study. Please take time to read the following information and feel free to ask for more information if something is not clear. What is the study about? This study aims to explore the things that are important to young people (14-25 years old) who are seriously ill and/or expected to live a shorter life. This would involve asking young people about the things that matter to them. The study findings may help inform the way care for young people is funded based on their needs and what young people view as important. Am I eligible to take part? We would like you to participate in the study if you are a young person, 14 – 25 years old, who is seriously ill and/or expected to live a shorter life What does taking part involve? • We will ask you to complete and return the questionnaire and your parent/guardian to complete and return the consent form. • Then, the study researcher will be in touch with you to arrange a virtual interview, for example via Skype or another online platform, at a time that is most convenient to you. If you do not have access to an electronic device to use for the interview let the researcher know. • The interview will last for up to an hour but with breaks if needed and with your permission it will be audio and video recorded. 505 • You are free to ask for the interview or recording to stop at any time. You can withdraw from the study at any time if you change your mind and you do not have to answer all the questions if you do not want to. • The researcher will give you a £10 voucher as a thank you • The interviewer may invite you to take part in a second interview at a later date, but you do not have to take part. • If you would like to withdraw from the study after your interview, please let us know within a week of the interview and your data will then be destroyed. If you notify us after this time period, it is possible that the data may already have been analysed and used to inform future interviews and thus it will not be possible to withdraw your data from the study. • We may not be able to interview all those interested in taking part in the study. In that case, the researcher will contact you following receipt of your questionnaire and consent form to let you know and your questionnaire and consent form will be securely destroyed by the researcher. What will the interview be about? In the interview you will be asked about the things that are important to you in life. During the interview, it is possible that sensitive topics may arise that are upsetting or difficult to answer. If you feel uncomfortable in any way during or after the interview, please let the interviewer know. Will my participation and personal information be kept confidential? • All personal information collected from you will be kept confidential and your name will be removed from any information connected to you. Any names and places you mention will be anonymised so that you cannot be recognised from this. • The interview will be written up and anonymised and the interview audio and video recording will be kept on a secure computer network at the University of Bristol. Only those directly involved in the study will be able to access this information, that is Isabella Floredin and her academic supervisors. • Any paper copies including your personal information will be uploaded electronically and then paper copies will be destroyed. • If you tell us that you are at risk to yourself or others then it is a statutory requirement to report this as part of safeguarding measures If I decide I would like to take part: • Please complete and return the questionnaire enclosed. Please also ask your parent/guardian to also complete and return the consent form enclosed. This information sheet is for you to keep • Then, Isabella will get in touch with you to arrange a time for the interview. 506 What if I have a concern or complaint? If you have a concern or complaint as a participant in this study, you may contact the Faculty of Health Science Research Ethics Committee, via the Research Governance Team, research-governance@bristol.ac.uk Where can I get more information? If you would like more information about this study or a copy of your interview transcript, please contact Isabella Floredin at - Address: Population health sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies Road, Bristol, BS8 1NU, Tel: +441174282096, Email: isabella.floredin@bristol.ac.uk This study is part of a University of Bristol project funded by the Wellcome Trust and has been approved by the University of Bristol Research Ethics Committee. 507 Appendix 10: Consent and assent forms Population Health Sciences University of Bristol 1-5 Whiteladies Road Bristol BS8 1NU Study Title: Quality of life for young people living with illness Researcher name: Isabella Floredin INTERVIEW - CONSENT FORM FOR CLOSE PERSON PARTICIPANT Please initial all boxes next to the statements below if you agree with them and consent to take part in the interview 1. I have read and understood the study information sheet and I have had the opportunity to ask questions and have had these answered. 2. I understand that my participation in the interview is voluntary and that I am free to withdraw at any time prior to and during the interview without giving a reason. 508 3. I understand that if I decide to withdraw my data from the study after the interview, I need to notify the researcher within a week of the interview. I understand that if I notify the researcher after this time period, it is possible that the data may already have been used to inform future interviews and thus it will not be possible to withdraw my data from the study. 4. I understand that the interview will be recorded with an audio and video recording device 5. I understand that the researcher will write a report about the research findings 6. I understand that my personal information (such as my name) will only be held by the researcher within the terms of GDPR and subsequent legislation on UK data protection post Brexit on the 1st January 2021.This information will be anonymised and no personal information will be attached to anything I say. 7. I agree to my research data becoming “controlled access”, meaning that it will be stored in an online database and be accessed and used by other researchers via requests to a Data Access Committee for a period of 20 years following completion of the study. 8. I understand that direct quotations of what I say may be used in publications, but I will not be identifiable from these. 9. I agree to take part in the above study. Participant’s signature: Date: Participant’s name in BLOCK CAPITALS Researcher’s signature: Date: 509 Population Health Sciences University of Bristol 1-5 Whiteladies Road Bristol BS8 1NU Study Title: Quality of life for young people living with illness Researcher name: Isabella Floredin INTERVIEW – PARENT/GUARDIAN CONSENT FORM FOR YOUNG PERSON PARTICIPANT UNDER 16 YEARS OLD Please initial all boxes next to the statements below if you agree with them and consent to take part in the interview 10. I have read and understood the study information sheet and I have had the opportunity to ask questions and have had these answered. I understand the purpose of the interview. 11. I understand that my child’s participation in the interview is completely voluntary and that my child does not have to answer any questions they do not wish to. I understand that I am free to withdraw my child from the study at any time without giving a reason. 12. I understand that if I, or my child, decide to withdraw my data from the study after the interview, I need to notify the researcher within a week of the interview. I understand that if I notify the researcher after this time period, it is possible that the data may already have been analysed and used to inform future interviews and thus it will not be possible to withdraw my data from the study. 510 13. I understand that my child’s interview will be recorded with an audio and video recording device and then written up. 14. I understand that the researcher will write a report about the research findings. 15. I understand that my family’s personal information (such as names) will only be held by the researcher within the terms of GDPR and subsequent legislation on UK data protection (e.g. data protection act and future amendments) post Brexit on the 1st January 2021. This information will be anonymised and no personal information will be attached to anything my child says. 16. I agree to my child’s research data becoming “controlled access”, meaning that it will be stored in an online database and be accessed and used by other researchers via requests to a Data Access Committee for a period of 20 years following completion of the study 17. I agree to be contacted by the researcher about taking part in future research. 18. I agree to allow the information that my child gives to be used for educational purposes in the future. 19. I understand that direct quotations of what my child says may be used in publications, but they will not be identifiable from these. 20. I consent to my child taking part in the above study. Parent/Guardian signature: Date: Parent/Guardian name in BLOCK CAPITALS Young Person participant name in BLOCK CAPITALS Researcher’s signature: Date: 511 Population Health Sciences University of Bristol 1-5 Whiteladies Road Bristol BS8 1NU Study Title: Quality of life for young people living with illness Researcher name: Isabella Floredin INTERVIEW - CONSENT FORM FOR PARENT/GUARDIAN PARTICIPANT Please initial all boxes next to the statements below if you agree with them and consent to take part in the interview 21. I have read and understood the study information sheet and I have had the opportunity to ask questions and have had these answered. I understand the purpose of the interview 22. I understand that my participation in the interview is completely voluntary and that I do not have to answer any questions I do not wish to. I understand that I am free to withdraw at any time without giving a reason. 23. I understand that if I decide to withdraw my data from the study after the interview, I need to notify the researcher within a week of the interview. I understand that if I notify the researcher after this time period, it is possible that the data may already have been analysed and used to inform future interviews and thus it will not be possible to withdraw my data from the study . 512 24. I understand that the interview will be recorded with an audio and video recording device and then written up. 25. I understand that the researcher will write a report about the research findings. 26. I understand that my personal information (such as my name) will only be held by the researcher within the terms of GDPR and subsequent legislation on UK data protection (e.g. data protection act and future amendments) post Brexit on the 1st January 2021.This information will be anonymised and no personal information will be attached to anything I say. 27. I agree to my research data becoming “controlled access”, meaning that it will be stored in an online database and be accessed and used by other researchers via requests to a Data Access Committee for a period of 20 years following completion of the study. 28. I agree to be contacted by the researcher about taking part in future research. 29. I agree to take part in the above study. 30. I agree to allow the information that I give to be used for educational purposes in the future. 31. I understand that direct quotations of what I say may be used in publications, but they will not be identifiable from these. Participant’s signature: Date: Participant’s name in BLOCK CAPITALS Researcher’s signature: Date: 513 Population Health Sciences University of Bristol 1-5 Whiteladies Road Bristol BS8 1NU Study Title: Quality of life for young people living with illness Researcher name: Isabella Floredin INTERVIEW - CONSENT FORM FOR YOUNG PERSON PARTICIPANT 16 TO 25 YEARS OLD Please initial all boxes next to the statements below if you agree with them and consent to take part in the interview 32. I have read and understood the study information sheet and I have had the opportunity to ask questions and have had these answered. I understand the purpose of the interview 33. I understand that my participation in the interview is completely voluntary and that I do not have to answer any questions I do not wish to. I understand that I am free to withdraw at any time without giving a reason. 34. I understand that if I decide to withdraw my data from the study after the interview, I need to notify the researcher within a week of the interview. I understand that if I notify the researcher after this time period, it is possible that the data may already have been analysed and used to inform future interviews and thus it will not be possible to withdraw my data from the study. 514 35. I understand that the interview will be recorded with an audio and video recording device and then written up. 36. I understand that the researcher will write a report about the research findings. 37. I understand that my personal information (such as my name) will only be held by the researcher within the terms of GDPR and subsequent legislation on UK data protection (e.g. data protection act and future amendments) post Brexit on the 1st January 2021. This information will be anonymised and no personal information will be attached to anything I say. 38. I agree to my research data becoming “controlled access”, meaning that it will be stored in an online database and be accessed and used by other researchers via requests to a Data Access Committee for a period of 20 years following completion of the study. 39. I agree to be contacted by the researcher about taking part in future research. 40. I agree to take part in the above study. 41. I agree to allow the information that I give to be used for educational purposes in the future. 42. I understand that direct quotations of what I say may be used in publications, but they will not be identifiable from these. Participant’s signature: Date: Participant’s name in BLOCK CAPITALS Researcher’s signature: Date: 515 Assent form Population Health Sciences University of Bristol 1-5 Whiteladies Road Bristol BS8 1NU Study title: Quality of life for young people living with illness Researcher name: Isabella Floredin INTERVIEW - ASSENT FORM FOR YOUNG PERSON Young person (or if unable, parent/guardian on their behalf) to tick YES if they agree with: YES NO Do you understand what the project is about? Have you asked all the questions you want? 516 Have you had your questions answered? Do you understand you can stop the study at any time? Are you happy to take part? If you have answered ‘no’ in any question or you don’t want to take part, don’t sign your name If you do want to take part, you can write your name below: Your name: Date: Researcher’s signature: Date: 517 Appendix 11: Distress protocol 518 519 520 521 Appendix 12: Debriefing statements Debriefing statement Study title: Quality of life for young people living with illness Researcher name: Isabella Floredin If you found that this study brought up any feelings or emotional difficulties, there are some helpful sources of support. The services below are free to access for support: • CRUSE Bereavement Care: Offers emotional support to anyone affected by bereavement. National Helpline number: 0808 808 1677 • Child Bereavement UK: Offers confidential bereavement support to anyone affected by the death of a child of any age. National Helpline number: 0800 02 888 40. Email: support@childbereavementuk.org • Samaritans helpline: Provides confidential emotional support 24 hours a day. Helpline number (24/7): 116 123. Email help service: jo@samaritans.org. Website: http://www.samaritans.org • The Compassionate Friends: Offers support and understanding to bereaved parents and their families. Helpline: 0345 123 2304 (open 522 10 am – 4 pm, 7 – 10 pm every day); helpline@tcf.org.uk; website: www.tcf.org.uk If you would like a follow up call at a week from today the researcher can be in touch at your preferred time. Phone number to contact: Date/Time of call: If you have any further questions regarding the study or a copy of your interview transcript, please contact Isabella Floredin at - Address: Population health sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies Road, Bristol, BS8 1NU, Tel: : +441174282096, Email: isabella.floredin@bristol.ac.uk If you have any questions or concerns as a participant in this study, you may contact Professor Joanna Coast at – Address: Population Health Sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies road, Bristol, BS8 1NU, Email: jo.coast@bristol.ac.uk 523 Debriefing statement Study title: Quality of life for young people living with illness Researcher name: Isabella Floredin If you found that this study brought up any feelings or emotional difficulties, there are some helpful sources of support. The services below are free to access for support: • Together for short lives family helpline: Provides a confidential service for families caring for a seriously ill child. Helpline is open from 10am to 4pm, Monday to Friday. Helpline number: 0808 8088 100. • Samaritans helpline: Provides confidential emotional support 24 hours a day. Helpline number (24/7): 116 123. Email help service: jo@samaritans.org. Website: http://www.samaritans.org If you would like a follow up call at a week from today the researcher can be in touch at your preferred time. Phone number to contact: Date/Time of call: If you have any further questions regarding the study or a copy of your interview transcript, please contact Isabella Floredin at - Address: Population health sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies Road, Bristol, BS8 1NU, Tel: : +441174282096, Email: isabella.floredin@bristol.ac.uk 524 If you have any questions or concerns as a participant in this study, you may contact Professor Joanna Coast at – Address: Population Health Sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies road, Bristol, BS8 1NU, Email: jo.coast@bristol.ac.uk Thank you for taking part in this study 525 Appendix 13: Copy of final survey (with PureProfile script) 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 Appendix 14: Copy of pilot survey 552 553 554 555 556 557 558 559 Appendix 15: Deliberative work instructions Figure 7 - Deliberative work introductory instructions for task completion by Coast et al. 2021. 560 Appendix 16: Amendments to final survey Wording amendments made to the final survey in response to feedback from the pilot survey. - In the introduction section and task instructions: replacing the word ‘impact’ to ‘benefits’ and use of shorter sentences to improve readability; more clear task instructions to facilitate respondent’s understanding of how to complete each task. - In the introductory section: text has been amended as some of the pilot survey respondents felt that the way the paragraph was phrased was leading and this would have an impact on their responses. - Throughout the survey: The word ‘patient’ has been replaced with ‘person’ throughout the panel survey document where pilot participants showed preference in referring to people at the end of life as ‘persons’ and not ‘patients’. - Budget pie task added: An additional pie task exercise was added in the final survey as some of the pilot respondents suggested that they would have responded differently to the task on different time points before death if the task was on young people who were under 18 and young people over 18 years 561 old. They suggested to split the task in two separate tasks, one task for those young people under 18 and a second task for young people over 18, rather than have only one task that presents young people 14-25 years old as one age group. 562 Appendix 17: FREC approval letter 25.05.2022 Dear Ms Isabella Floredin Ref: 9194 Title: Quality of life for young people with life-threatening and life-limiting conditions - Phase 3 Thank you for submitting your ethics application for the above-named study. The Faculty of Health Science Research Ethics Committee (FREC) has reviewed your research ethics amendment request and we can confirm that your ethics application has received a favourable ethical opinion. Please note that the Faculty of Health Science Research Ethics Committee (FREC) expects to be notified of any changes or deviations in the study. Yours sincerely, Dr Vikki Wylde Faculty of Health Science Research Ethics Committee (FREC) 563 Appendix 18: Mean weights elicited for sample n=587 Table 25 - Mean weights elicited by pie chart task (n=587) Pie chart tasks Mean weight elicited for person at EOL (SD) Mean weight elicited for family and friends (SD) Task 1: Person at EOL (all ages) 55.7 (26.3) 44.4 (26.3) Task 2a: Young person 14-17 yrs. at EOL 53.1 (22.6) 46.9 (22.6) Task 2b: Young person 18 -25 yrs. at EOL 56.1 (22.4) 43.9 (22.4) Task 3a: Young person 14-17yrs. at: v) 12 months before death vi) 6 months before death vii) 1 month before death viii) In the last few days before death 56.5 (22.6) 54.8 (22.5) 51.7 (25.3) 48.3 (28.1) 43.5 (22.6) 45.2 (22.5) 48.3 (25.3) 51.7 (28.1) Task 3b: Young person 18-25 yrs. at: v) 12 months before death vi) 6 months before death vii) 1 month before death viii) In the last few days before death 58.9 (22.8) 57.7 (22.3) 53.9 (24.6) 50.4 (28) 41.1 (22.8) 42.3 (22.3) 46.1 (24.6) 49.6 (28) 564 Appendix 19: Non-parametric descriptive statistics Table 26 - Pie task Median (IQR) weights elicited – non-parametric descriptive statistics Pie chart tasks Median weight (IQR) elicited for person at end of life Task 1: Person at end of life (all ages) 50 (40, 75) Task 2a: Young person 14-17 yrs. at end of life 50 (40, 70) Task 2b: Young person 18 -25 yrs. at end of life 50 (50, 70) Task 3a: Young person 14-17yrs. at: i) 12 months before death ii) 6 months before death iii) 1 month before death iv) In the last few days before death 50 (50, 75) 50 (50, 70) 50 (30,70) 50 (25, 70) Task 3b: Young person 18-25 yrs. at: i) 12 months before death ii) 6 months before death iii) 1 month before death iv) In the last few days before death 58.5 (50,75) 55 (50, 75) 50 (40, 70) 50 (30, 70) 565 Appendix 20: Non-parametric tests – initial analysis Table 27 - Initial analysis – p-values for characteristics unadjusted (non-parametric tests) EXPLANATORY VARIABLES/TASK 1 2a 2b 3a(i) 3a(ii) 3a(iii) 3a(iv) 3b(i) 3b(ii) 3b(iii) 3b(iv) Gender 0.124 0.539 0.518 0.211 0.400 0.819 0.915 0.053 0.062 0.580 0.809 Age 0.173 0.400 0.653 0.199 0.607 0.154 0.282 0.903 0.391 0.788 0.546 Ethnicity 0.134 0.718 0.611 0.638 0.371 0.947 0.809 0.314 0.744 0.638 0.512 Region 0.304 0.228 0.186 0.576 0.058 0.848 0.797 0.452 0.108 0.736 0.604 Education 0.350 0.014 0.037 0.108 0.028 0.095 0.395 0.021 0.106 0.278 0.366 Employment 0.188 0.598 0.232 0.133 0.451 0.620 0.819 0.190 0.255 0.633 0.466 Religion 0.042 0.101 0.186 0.014 0.136 0.292 0.887 0.019 0.028 0.080 0.529 Self-assessed health 0.024 0.366 0.364 0.072 0.421 0.819 0.507 0.302 0.709 0.868 0.607 Disability 0.883 0.390 0.340 0.172 0.559 0.058 0.084 0.606 0.285 0.404 0.116 Work in health/social care 0.028 0.511 0.468 0.225 0.452 0.557 0.174 0.203 0.502 0.324 0.377 Look after others 0.617 0.262 0.941 0.747 0.634 0.497 0.795 0.424 0.348 0.201 0.673 Family/friend received PC or EOLC 0.010 0.150 0.016 0.665 0.165 0.188 0.198 0.124 0.138 0.178 0.222 Cared for someone at EOL 0.522 0.504 0.799 0.893 0.774 0.823 0.378 0.895 0.912 0.646 0.948 Parent/Guardian 0.763 0.341 0.671 0.306 0.914 0.171 0.401 0.774 0.842 0.531 0.803 Appendix 21: Summary of methods to develop a young person EOL measure Study design: Following the development of attributes (Phase 1) a second phase (Phase 2) is necessary to check the content and coverage of attributes developed in Phase 1 and to generate meaningful wording for the population the measure is intended for. 407 Semi-structured interviews will be used in Phase 2 to allow the researcher to focus on the more specific study aims of Phase 2. 290, 420 Sampling and recruitment: The sampling inclusion criteria for Phase 2 will be as follows: • young people 14-25 years old with a LLC or LTC. • those with parental responsibility for young people (14-25 years old, with a LLC or LTC) who are not able to take part in the research directly due to the condition and/or symptoms. • Close person bereaved at least six months post death of a young person (14-25 years old) with LLC/LTC. Close person includes family member, partner, close friend. The inclusion of these three participant groups will aim to capture the perspectives of young people, both verbal and non-verbal, and across the EOL trajectory, including at the very EOL. The sampling approach will be purposeful 571 aiming for maximum variation to sampling 572 including participants from different socio-economic backgrounds, gender, ethnicity, health conditions and with experience of different care and support 567 services (information on participant characteristics will be collected with the use of a screening questionnaire). If the number of responses from potential participants is small the study will include all initial respondents and snowball sampling 573 will also be used through existing participants. Phase 2 participants will be recruited using multiple recruitment strategies in a similar manner to that used in Phase 1 to maximise the potential number of expressions of interest from potential participant. Study adverts and participant information sheets will be circulated across the UK via: - Adult and children’s hospices (online and/or in-person through hospice staff) - Charitable organisations offering advice and/or support to young people and/or close persons (online and/or in-person through staff) - Social media platforms (Twitter, Instagram and Facebook using the Facebook ‘targeted ads feature’) The hospices and charitable organisations which supported Phase 1 recruitment will be recontacted and asked if they are able and willing to support recruitment in the Phase 2. Drawing on previous qualitative research done to develop capability measures for use in economic evaluation 218, 290 it is expected that an approximate number of 15 semi-structured interviews will produce a sufficient sample for data saturation. However, the sufficient number to achieve data saturation will depend on the views of participants and any potential differences in the views found. Data saturation will be achieved when no new terminology is introduced by new participants to express the conceptual attributes and no new feedback is suggested by new participants in relation to refining the content of the measure. Data collection: 568 Semi-structured interviews will take place online or in-person depending on the participant’s preference and depending on whether the interview can take place in-person at a setting that is within reasonable distance for the researcher to travel to (e.g. South-West of England). Alternatively, if travelling to the setting is not possible an online interview will be offered. Participants will be presented with the conceptual attributes developed in Phase 1 (i.e. 1. Experience and enjoy; 2. Independence; 3. Freedom from physical suffering; 4. Freedom from emotional suffering; 5. Care and support; 6. Control; 7. Identity) and will be asked: • whether they think that all aspects of life which are important to young people aged 14-25 with a LLC/LTC have been captured within these attributes and whether there is something missing that they would like to add. This is to check the coverage of the attributes developed in Phase 1. • What each attribute word means to them to check whether their understanding and interpretation matches the intended meaning of the attribute (Participants with parental responsibility and close persons will be asked what the word means to them and how they think that the young person will interpret the wording used) • To feedback on the wording used for the levels that will accompany each attribute (feedback on levels will be sought once the wording for attributes has been established from feedback given in initial interviews). Data collection and analysis will run concurrently with interviews undertaken in rounds, and the interview topic guide will be updated and adapted as analysis progresses. Initial rounds of interviews will be based on brainstorming and the final rounds of interviews will focus on establishing the final wording for attributes and attribute levels. 569 Data analysis: Semi-structured interviews will be audio-recorded and transcribed, and transcripts will be coded in NVivo12 qualitative data analysis software. Interviews will be undertaken and analysed iteratively by using constant comparison methods. 428 The iterative interview process will allow for the wording presented to new participants to be constantly refined following earlier participant feedback. Data collection will end following repeated interviews which do not raise any issues in terms of the conceptual attributes being meaningful and there are no issues with the wording and format used for the measure Ethics: The topic of this study is sensitive and the participants are particularly vulnerable.291 The researcher will take reasonable and appropriate safety measures to prevent risk of harm (physical or psychological) to participants. Measures to assess and manage potential risk in Phase 1 (as described in detail in section 4.7.3) will be followed in Phase 2 and will include managing potential risk associated with invasion of privacy, confidentiality breaches and coercion into study participation by a parent/guardian or the researcher. As in Phase 1, decision in relation to study participation, including ability and willingness to take part, will be made in conjunction with the young person and parent/guardian. Participants under 16 years old will be asked to assent to taking part in the study and their parent/guardian will provide informed written consent for young person’s study participation. 570 Appendix 22: Study findings dissemination plan 1. Study participants and patient and public involvement (PPI) group Audience: - Interview participants: Young people (14-25 years old) with a LLC/LTC and family members, including bereaved parents and siblings. - ‘Young experts’ involvement group in TfSL charity: PPI group which provided feedback on the terminology used in the study recruitment materials for the qualitative work undertaken. Members of the group comprised young people between the ages of 14 to 18 years old with a LLC/LTC. Purpose of dissemination: - To keep study participants and PPI group well informed about the progress and findings of the study. All study participants and the PPI group facilitator were happy to be contacted at the end of the study for an update on the qualitative study findings. - To provide opportunities to receive feedback on the findings from study participants and PPI group. - To thank the participants and PPI group for their participation and involvement in the study. Message to be disseminated: 571 The message to be disseminated to this audience will focus on the qualitative findings of the thesis. Interview participants will be informed about the qualitative findings relevant to the part of the study that they took part in and findings for which they were happy to and expected to be contacted about at the end of the study. A link to the study webpage (A Quality of Life Study: Young people living with serious illness | Bristol Medical School: Population Health Sciences | University of Bristol) will also be provided for individuals who may wish to look at further information on study findings and any future work undertaken. The message to be disseminated will be split into three parts: 1. What was the study about? 2. What did we find? 3. What does this mean? Key messages to be disseminated will be: • Brief details on the purpose of the study and key information on data collection (e.g., when data collection took place, number of participants per participant group). 572 o This information will be disseminated to all study participants and PPI group. • Qualitative findings: o A summary of the conceptual attributes and conversion factors reported in Chapter 5. ▪ This information will be disseminated to all study participants and PPI group. o A summary of parent and sibling participant views on use of ICECAP-CPM reported in Chapter 6. The focus here will be on the overall views of participants on how relevant the measure is to their experience. ▪ This information will be disseminated to all study participants that took part in this part of the empirical work, i.e., bereaved parents and siblings. • Brief information on the study implications and contributions o This information will be disseminated to all study participants and PPI group. Method and timing: The message will be disseminated to the target audience through a report using non-technical language within a month following successful completion of the PhD. 573 The report will be as brief as possible, and the message will be clear and easy to understand. Given the sensitive nature of the topic studied the language used and format of the report will be carefully considered prior to disseminating so that it is appropriate and tailored to the target audience. The terminology used will follow the terminology used in the participant recruitment material. Advice from the ‘Young Experts’ PPI group was sought to decide the terminology used in the recruitment material and the material has been approved by the Faculty Research Ethics Committee (FREC) (ref no: 111822) to undertake the study. Prior to disseminating findings, the message and format of the report will be reviewed and approved by a number of experts including the PhD supervisory team with research expertise in PPC, Qualitative methods and health economics. 2. Hospices and charitable organisations Audience: - Hospices and charitable organisations facilitating recruitment of study participants Purpose of dissemination: - To keep hospices and charitable organisations well informed about the findings of the study 574 - To thank them for supporting recruitment and provide opportunities to receive feedback - To share findings with a wider audience (general public) through the relevant hospices and organisations - To reach out to people, such as health and social care professionals, who may be interested in study findings and may wish to promote further the dissemination of findings to other interested stakeholders. Message to be disseminated: The message to be disseminated to this audience will focus on the qualitative findings of the thesis. Hospices and organisations will be informed of the qualitative findings relevant to the part of the study that they supported recruitment and findings for which they expect to be contacted about at the end of the study. A link to the study webpage (A Quality of Life Study: Young people living with serious illness | Bristol Medical School: Population Health Sciences | University of Bristol) will also be provided for members of staff and other individuals who may wish to look at further information on study findings and any future work undertaken. Key messages to be disseminated will be: 575 • Details on the purpose of the study and key information on data collection (e.g., when data collection took place, number of participants per participant group). o This information will be disseminated to all hospices and organisations supporting recruitment. • Qualitative findings: o A summary of the conceptual attributes and conversion factors reported in Chapter 5. ▪ This information will be disseminated to all hospices and organisations supporting recruitment. o A summary of parent and sibling participant views on use of ICECAP-CPM reported in Chapter 6. The focus here will be on the overall views of participants on how relevant the measure is to their experience. ▪ This information will be disseminated to all hospices and organisations supporting recruitment of bereaved close persons. • Information on the study implications and contributions o This information will be disseminated to all hospices and organisations supporting recruitment. 576 Method and timing: Hospices and organisations may wish to share the message with a wider audience, for example to service users and their families and other relevant organisations via their social media. Therefore, it is important that the language and format of the dissemination report circulated to this audience is consistent with the language and format used in the report circulated to study participants and PPI group. The report will be brief and will use non-technical language that is easy to understand. Prior to dissemination to this target audience, the message and format of the report will be reviewed and approved by a number of experts including the PhD supervisory team. Dissemination for this audience will take place within a month following completion of the PhD and concurrently with the dissemination for study participants and PPI group. 3. Research community Audience: Research community across the different disciplines relevant to the study findings, i.e. health economists, researchers with expertise in paediatric palliative care or bereavement research, researchers using the capability approach in outcome measurement, Qualitative researchers. Purpose of dissemination: 577 - To ensure visibility and uptake of findings in the research community - To share findings that can be used in future research - To provide opportunities to receive feedback from the research community and relevant topic experts. - To share and discuss joint issues and challenges in undertaking qualitative research on the sensitive EOL topic with vulnerable participants. Message to be disseminated, method and timing: The qualitative and quantitative study findings (inc. preliminary findings), reported in Chapters 5 and 8 have been disseminated to the research community in the form of oral presentations at a number of academic conferences and webinars with a focus on health economics, the capability approach or palliative care: Events with a health economics focus: Floredin I., Husbands S., Neilson S., Mitchell P., Coast J. Exploring Capability Wellbeing in Young People Expected to Live a Shorter Life, Using a Constant Comparative Approach. International Health Economics Association Congress, online, July 2021 Floredin I, Mitchell P, Husbands S, Neilson S, Coast J. Does the age of People at the End-of-Life Influence how Much Weight People Provide to Their Outcomes Compared to Their Family and Friends? A Budget pie Study in the UK General 578 Population. International Health Economics Association Congress, Cape Town, July 2023. Floredin I, Husbands S, Neilson S, Mitchell P, Coast J. What Capabilities Matter to Young People with Life-Limiting and Life-Threatening Conditions? A Qualitative Study From the UK. International Health Economics Association Congress, Cape Town, July 2023. Events with a capability approach focus: Floredin I., Husbands S., Mitchell P., Neilson S., Coast J. Quality of life for young people with life-threatening and life-limiting conditions. Health and Disability Thematic Group, Human Capability and Development Association, International online webinar, November 2020. Floredin I., Husbands S., Mitchell P., Neilson S., Coast J. Exploring capabilities in young people with life-limiting and life-threatening conditions. Human Capability and Development Association Conference, Antwerp, Belgium, September 2022. Floredin I, Husbands S., Mitchell P., Neilson S., Coast J. What capabilities matter to young people at the end of life and what influences their ability to achieve them? A qualitative study from the UK. Human Development and Capability Association conference, Sofia, Bulgaria, September 2023. 579 Events with a palliative care focus: Floredin I., Husbands S., Mitchell P., Neilson S., Coast J. Quality of life for young people living with illness, Southwest Palliative and End of Life Care Research Network Meeting. Bristol (online), February 2021. Floredin I., Husbands S., Mitchell P., Neilson S., Coast J. Care at the end of life: Measuring outcomes for young people for use in economic evaluation, University of Birmingham PhD Palliative Care Presentations, Birmingham (online), July 2023. The following academic journal publications will be pursued: • Conceptual attributes for a new end of life measure for young people and conversion factors influencing capability achievement. o Target journal: Social Science and Medicine o Backup journal: Journal of Human Development and Capabilities o Initial submission date: April 2024 • Survey on relative weights elicited from the general public between outcomes for young people and outcomes for close persons. o Target journal: Health Economics o Backup journal: The European Journal of Health Economics o Initial submission date: June 2024 580 • ICECAP-CPM feasibility of use and content validity with bereaved close persons of young people. o Target journal: Death Studies o Backup journal: BMC Palliative Care o Initial submission date: August 2024 • Systematic review: Qualitative and mixed methods studies on what is important to young people at the end of life. o Target journal: BMJ Supportive and Palliative Care o Backup journal: Palliative Medicine o Initial submission date: November 2024 BIBLIOGRAPHY 1. Nations U. Convention on the Rights of the Child. 1989. 2. NHS. The NHS long term plan NHS England; 2019. 3. Round J. Care at the End of Life: An Economic perspective. Switzerland: Springer International Publishing 2016. 4. Curie M. What are palliative care and end of life care? 2018 [Available from: https://www.mariecurie.org.uk/help/support/diagnosed/recent-diagnosis/palliative-care-end-of-life-care. 5. Krau SD. The Difference Between Palliative Care and End of Life Care: More than Semantics. Nursing Clinics of North America. 2016;51(3):ix-x. 6. Reid CM, Gibbins J, McCoubrie R, Forbes K. Palliative care is not same as end of life care. BMJ. 2011;342:d2735. 7. (WHO) WHO. WHO Definition of Palliative Care [Available from: https://www.who.int/cancer/palliative/definition/en/ 8. APPG. End of life care: strengthening choice. Together for short lives; 2018. 9. NHS. What end of life care involves: End of life care 2018 [Available from: https://www.nhs.uk/conditions/end-of-life-care/what-it-involves-and-when-it-starts/. 10. Chu C, White N, Stone P. Prognostication in palliative care. Clin Med (Lond). 2019;19(4):306-10. 11. Chelazzi C, Villa G, Lanini I, Romagnoli S, Latronico N. The adult and pediatric palliative care: differences and shared issues. Journal of Anesthesia, Analgesia and Critical Care. 2023;3(1):1. 12. network Icspc. What is Children’s Palliative Care? : ICPCN; 2015 [Available from: http://www.icpcn.org/about-icpcn/what-is-childrens-palliative-care/ 13. Hynson JL, Gillis J, Collins JJ, Irving H, Trethewie SJ. The dying child: how is care different? Medical Journal of Australia. 2003;179(S6):S20-S2. 14. Katherine Stevens JR. Measuring and valuing health-related quality of life in children with terminal and life-limiting illness. In: Round J, editor. Care at the End of Life: an economic perspective Switzerland: Springer International Publishing 2016. 582 15. Fraser LK, Miller M, Hain R, Norman P, Aldridge J, McKinney PA, et al. Rising national prevalence of life-limiting conditions in children in England. Pediatrics. 2012;129(4):e923-e9. 16. NICE. End of life care for infants, children and young people with life-limiting conditions: planning and management. NICE guideline [NG61]. NICE; 2019. 17. Lives TfS. A Guide to Children’s Palliative Care Supporting babies, children and young people with life-limiting and life-threatening conditions and their families. 2018 18. Liben S, Papadatou D, Wolfe J. Paediatric palliative care: challenges and emerging ideas. The Lancet. 2008;371(9615):852-64. 19. Lorna K Fraser DG-S, Stuart Jarvis, Paul Norman, Roger Parslow. ‘Make Every Child Count’ Estimating current and future prevalence of children and young people with life-limiting conditions in the United Kingdom. Martin House Research Centre, University of York; 2020. 20. Knapp C, Madden V, Fowler-Kerry S. Pediatric Palliative Care: Global Perspectives2012. 21. Norman P, Fraser L. Prevalence of life-limiting conditions in children and young people in England: Time trends by area type. Health & Place. 2014;26:171-9. 22. Namisango E, Bristowe K, Murtagh FE, Downing J, Powell RA, Abas M, et al. Towards person-centred quality care for children with life-limiting and life-threatening illness: Self-reported symptoms, concerns and priority outcomes from a multi-country qualitative study. Palliative Medicine. 2020;34(3):319-35. 23. Mitchell S, Morris A, Bennett K, Sajid L, Dale J. Specialist paediatric palliative care services: what are the benefits? Archives of Disease in Childhood. 2017;102(10):923. 24. Downing J, Boucher S, Daniels A, Nkosi B. Paediatric Palliative Care in Resource-Poor Countries. Children. 2018;5(2). 25. Knapp C, Woodworth L, Wright M, Downing J, Drake R, Fowler-Kerry S, et al. Pediatric palliative care provision around the world: A systematic review. Pediatric Blood & Cancer. 2011;57(3):361-8. 26. Fraser LK, Parslow R. Children with life-limiting conditions in paediatric intensive care units: a national cohort, data linkage study. Archives of disease in childhood. 2018;103(6):540-7. 27. Lorna K Fraser1 MM, Jan Aldridge2, Patricia A McKinney3 & Roger C Parslow3. Prevalence of life-limiting and life-threatening conditions in young adults in England 2000-2010. Together for Short Lives; 2013. 28. Devins Ha. Directory of Life-Limiting conditions. 2011. 583 29. Hain R, Devins M, Hastings R, Noyes J. Paediatric palliative care: development and pilot study of a 'Directory' of life-limiting conditions. BMC Palliat Care. 2013;12(1):43-. 30. Fraser L, Lidstone V, Miller M, Aldridge J, Norman P, McKinney P, et al. PATTERNS OF DIAGNOSES AMONGST CHILDREN AND YOUNG ADULTS WITH LIFE-LIMITING CONDITIONS; IMPLICATIONS FOR SERVICE PROVISION. BMJ Supportive & Palliative Care. 2014;4(Suppl 1):A21.2-A. 31. Murray SA, Kendall M, Boyd K, Sheikh A. Illness trajectories and palliative care. BMJ (Clinical research ed). 2005;330(7498):1007-11. 32. Hughes B, O’Brien M, Flynn A, Knighting K. Views and experiences of young people, their parents/carers and healthcare professionals of the advance care planning process: A summary of the findings from a qualitative study. Palliative Medicine. 2022;36(5):841-54. 33. The dying child: how is care different? Medical Journal of Australia. 2003;179(S6):20. 34. Constantinou G, Garcia R, Cook E, Randhawa G. Children’s unmet palliative care needs: a scoping review of parents’ perspectives. BMJ Supportive & Palliative Care. 2019;9(4):439-50. 35. Shaw KL, Brook L, Mpundu-Kaambwa C, Harris N, Lapwood S, Randall D. The Spectrum of Children's Palliative Care Needs: a classification framework for children with life-limiting or life-threatening conditions. BMJ Supportive & Palliative Care. 2015;5(3):249. 36. Birmingham Uo. The Spectrum of Children’s Palliative Care Needs: A prognosis based framework for children and young people who have palliative care needs Togethe for Short Lives; 2012. 37. Birmingham Uo. The Spectrum of Children’s Palliative Care Needs Introductory Leaflet. Together for Short Lives, England.; 2012. 38. Scott J. A dictionary of sociology. Oxford: Oxford University Press; 2014. Available from: http://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=1791127 https://doi.org/10.1093/acref/9780199683581.001.0001 http://www.oxfordreference.com/view/10.1093/acref/9780199683581.001.0001/acref-9780199683581 http://VH7QX3XE2P.search.serialssolutions.com/?V=1.0&L=VH7QX3XE2P&S=JCs&C=TC0001545430&T=marc&tab=BOOKS http://0-dx.doi.org.pugwash.lib.warwick.ac.uk/10.1093/acref/9780199683581.001.0001. 39. Law J. A dictionary of law. Oxford, United Kingdom: Oxford University Press; 2018. Available from: http://dx.doi.org/10.1093/acref/9780198802525.001.0001 584 http://www.oxfordreference.com/view/10.1093/acref/9780198802525.001.0001/acref-9780198802525 https://purl.missouristate.edu/library/e-books/9780191840807 https://login.proxy.bib.uottawa.ca/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780198802525.001.0001/acref-9780198802525 http://proxy.library.carleton.ca/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780198802525.001.0001/acref-9780198802525 http://ezproxy.eui.eu/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780198802525.001.0001/acref-9780198802525. 40. acts UPG. Children Act 1989. In: families Posfcat, editor. 1989. 41. Acts UPG. Children Act 2004. 2004. 42. acts UPG. Family Law Act 1996. 1996. 43. Goldman A, Hain R, Liben S. Oxford textbook of palliative care for children. Oxford ;: Oxford University Press; 2007. 44. Teater B. Social Work Theory. In: Wright JD, editor. International Encyclopedia of the Social & Behavioral Sciences (Second Edition). Oxford: Elsevier; 2015. p. 813-20. 45. Pillow BH. Learning About Cognitive Activities. In: Pillow BH, editor. Children’s Discovery of the Active Mind: Phenomenological Awareness, Social Experience, and Knowledge About Cognition. New York, NY: Springer New York; 2012. p. 1-11. 46. Piaget J. The origin of intelligence in the child. London: Routledge & Kegan Paul Ltd; 1953. 47. Piaget J. The Moral Judgment Of The Child: The Free Press; 1932. 48. Piaget J. Judgment And Reasoning In The Child: Harcourt, Brace And Company; 1928. 49. Erikson EH. Childhood and society. London: Imago; 1950. 50. Bluebond-Langner M. The private worlds of dying children. Princeton, N.J.: Princeton University Press; 1978. 51. Pravin RR, Enrica TEK, Moy TA. The Portrait of a Dying Child. Indian J Palliat Care. 2019;25(1):156-60. 585 52. Coad J, Kaur J, Ashley N, Owens C, Hunt A, Chambers L, et al. Exploring the Perceived Met and Unmet Need of Life-Limited Children, Young People and Families. Journal of Pediatric Nursing. 2015;30(1):45-53. 53. Field MJ, Behrman RE, Board on Health Sciences Policy S, Palliative, End-of-Life Care for C, Their Families C, et al. When Children Die : Improving Palliative and End-of-Life Care for Children and Their Families. Washington, D.C., UNITED STATES: National Academies Press; 2003. 54. Stone MJ. Goals of care at the end of life. Proc (Bayl Univ Med Cent). 2001;14(2):134-7. 55. Field MJ, Behrman RE, Institute of Medicine . Committee on P, End-of-Life Care for C, Their F. When children die : improving palliative and end-of-life care for children and their families. Washington, D.C.: National Academy Press; 2003. Available from: http://site.ebrary.com/id/10032349 http://site.ebrary.com/id/10049189 http://www.nap.edu/books/0309084377/html/index.html http://www.nap.edu/books/0309084377/html/ http://www.nap.edu/catalog.php?record_id=10390 http://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=3375163 https://www.ncbi.nlm.nih.gov/books/NBK216211 https://www.ncbi.nlm.nih.gov/books/NBK220818 http://www.ncbi.nlm.nih.gov/books/NBK220818 http://site.ebrary.com/lib/rcc/Doc?id=10032349 http://gold.worcester.edu/login?url=http://site.ebrary.com/lib/worcesteruniv/Doc?id=10032349 https://ebookcentral.proquest.com/lib/ualberta/detail.action?docID=3375163 http://ebookcentral.proquest.com/lib/northernlakes-ebooks/detail.action?docID=3375163 http://ebookcentral.proquest.com/lib/norquest-ebooks/detail.action?docID=3375163 http://ebookcentral.proquest.com/lib/oldscollege/detail.action?docID=3375163 http://ebookcentral.proquest.com/lib/concordiaab-ebooks/detail.action?docID=3375163 586 http://ebookcentral.proquest.com/lib/gprc-ebooks/detail.action?docID=3375163 http://ebookcentral.proquest.com/lib/macewan-ebooks/detail.action?docID=3375163 http://ebookcentral.proquest.com/lib/rdc/detail.action?docID=3375163 http://site.ebrary.com/lib/aunigeria/Doc?id=10032349 http://public.eblib.com/choice/publicfullrecord.aspx?p=3375163 http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=87226 http://ebookcentral.proquest.com/lib/columbia/detail.action?docID=3375163 http://belhaven.idm.oclc.org/login?url=http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=87226 http://ebookcentral.proquest.com/lib/warw/detail.action?docID=3375163 https://login.lacollegelibrary.idm.oclc.org/login?url=http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&AN=87226 https://ebookcentral.proquest.com/lib/ucm/detail.action?docID=3375163 https://ebookcentral.proquest.com/lib/columbia/detail.action?docID=3375163 http://www.netLibrary.com/urlapi.asp?action=summary&v=1&bookid=87226 https://ebookcentral.proquest.com/lib/stanford-ebooks/detail.action?docID=3375163 http://0-search.ebscohost.com.librarycatalog.vts.edu/login.aspx?direct=true&scope=site&db=nlebk&AN=87226 http://0-ebookcentral.proquest.com.librarycatalog.vts.edu/lib/vtu/detail.action?docID=3375163 http://proxy.library.carleton.ca/login?url=http://www.netLibrary.com/urlapi.asp?action=summary&v=1&bookid=87226 http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=e000xna&AN=87226 http://trmproxy.mnpals.net/login?url=http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=87226 https://ebookcentral.proquest.com/lib/ben/detail.action?docID=3375163 587 http://uclibs.org/PID/102800 https://login.gbcprx01.georgebrown.ca/login?url=https://ebookcentral.proquest.com/lib/georgebrown-ebooks/detail.action?docID=3375163 https://www.nap.edu/10390 http://www.vlebooks.com/vleweb/product/openreader?id=none&isbn=9780309502153. 56. Rodriguez KL, Young AJ. Patients’ and healthcare providers’ understandings of life-sustaining treatment: Are perceptions of goals shared or divergent? Social Science & Medicine. 2006;62(1):125-33. 57. Gardiner C, Ingleton C, Gott M, Ryan T. Exploring the transition from curative care to palliative care: a systematic review of the literature. BMJ Supportive & Palliative Care. 2015;5(4):335-42. 58. Michelson KN, Steinhorn DM. Pediatric End-of-Life Issues and Palliative Care. Clin Pediatr Emerg Med. 2007;8(3):212-9. 59. Mack JW, Wolfe J. Early integration of pediatric palliative care: for some children, palliative care starts at diagnosis. Current Opinion in Pediatrics. 2006;18(1). 60. Nicholl HaT, C. Children's palliative care: developing a service. World of irish nursing & midwifery. 2015;23(3):55-6. 61. Kaur J, Mohanti BK. Transition from curative to palliative care in cancer. Indian J Palliat Care. 2011;17(1):1-5. 62. Shaw KL, Watanabe A, Rankin E, McDonagh JE. Walking the talk. Implementation of transitional care guidance in a UK paediatric and a neighbouring adult facility. Child: Care, Health and Development. 2014;40(5):663-70. 63. Bertaud S, Lloyd DFA, Laddie J, Razavi R. The importance of early involvement of paediatric palliative care for patients with severe congenital heart disease. Archives of Disease in Childhood. 2016;101(10):984-7. 64. Zadeh S, Pao M, Wiener L, Sender L, Adlard K, Begino C, et al. Adolescents and young adults (AYA) at the end of life: Using voicing my choices to enhance communication and guide planning. Psycho-Oncology. 2016;2):5-6. 65. Reynolds J, Croft S. How to implement the Gold Standards Framework to ensure continuity of care. Nurs Times. 2010;106(32):10-3. 66. collaborative Caypsacp. Guide to using the child and young person's advance care plan: Child and young person's advanced care plan collaborative 588 [Available from: http://cypacp.uk/wp-content/uploads/2017/01/CYPACP-guide-Final-v1.-6.pdf. 67. Seymour J, Almack K, Kennedy S. Implementing advance care planning: a qualitative study of community nurses' views and experiences. BMC Palliat Care. 2010;9(1):4. 68. Advance care planning for paediatric patients. Paediatr Child Health. 2008;13(9):791-805. 69. Weston C. An answer to the question of what is a good death? 2015 [Available from: https://www.mariecurie.org.uk/blog/what-is-a-good-death/48655. 70. Meier EA, Gallegos JV, Thomas LPM, Depp CA, Irwin SA, Jeste DV. Defining a Good Death (Successful Dying): Literature Review and a Call for Research and Public Dialogue. Am J Geriatr Psychiatry. 2016;24(4):261-71. 71. Chong PH, Walshe C, Hughes S. Perceptions of a Good Death in Children with Life-Shortening Conditions: An Integrative Review. J Palliat Med. 2018;22(6):714-23. 72. Kehl KA. Moving Toward Peace: An Analysis of the Concept of a Good Death. American Journal of Hospice and Palliative Medicine®. 2006;23(4):277-86. 73. Chong PH, Walshe C, Hughes S. A good death in the child with life shortening illness: A qualitative multiple-case study. Palliative Medicine. 2021;35(10):1878-88. 74. Woods S. A Good Death? Law and Ethics in Practice. London Routledge; 2013. 75. Hynson JL, Sawyer SM. Paediatric palliative care: distinctive needs and emerging issues. J Paediatr Child Health. 2001;37(4):323-5. 76. Hendrickson K, McCorkle R. A dimensional analysis of the concept: good death of a child with cancer. J Pediatr Oncol Nurs. 2008;25(3):127-38. 77. Taib F. Is There 'A Good Death' Among Children With Life Limiting Illnesses? Malaysian Journal of Paediatrics and Child Health. 2011. 78. Together for Short Lives E. A workforce in crisis: children's palliative care in 2019. 2019. 79. Together for Short Lives E. Statutory funding for children's hospice and palliative care charities in England 2018/19. Together for Short Lives; June 2019. 80. Together for Short Lives E. Children's hospice funding in 2021/2022. Together for Short Lives; 2022. 589 81. Hunt A, Brown E, Coad J, Staniszewska S, Hacking S, Chesworth B, et al. 'Why does it happen like this?' Consulting with users and providers prior to an evaluation of services for children with life limiting conditions and their families. J Child Health Care. 2015;19(3):320-33. 82. Davies J, Carter B. Looking for an ordinary life. Journal of Child Health Care. 2013;17(3):217-8. 83. Corkin DAP, Price J, Gillespie E. Respite care for children, young people and families - are their needs addressed? International Journal of Palliative Nursing. 2006;12(9):422-7. 84. Together for Short Lives E. Commissioning children's palliative care in England: 2017 edition. Together for short lives; 2017. 85. Lives TfS. Corporate partnerships 2020 [Available from: https://www.togetherforshortlives.org.uk/get-involved/company-giving/corporate-partnerships/. 86. Together for Short Lives E. Family Factsheets: Children's Hospice Services. TfSL; 2021. 87. Charles A. Integrated care systems explained: making sense of systems, places and neighbourhoods: The King's fund; 2022 [ 88. Bureau NCs. Children’s palliative care and children’s hospice provision in the UK Hospice UK and NCB; 2018 2020. 89. Lives TfS. 24/7 access to children's palliative care in England. England: Together for Short Lives, England. ; 2022. 90. Boddy J, Chenoweth L, McLennan V, Daly M. It's just too hard! Australian health care practitioner perspectives on barriers to advance care planning. Aust J Prim Health. 2013;19(1):38-45. 91. McCulloch DR. All Party Parliamentary Group for Children who need Palliative Care ROYAL COLLEGE OF PAEDIATRICS AND CHILD HEALTH; 2018. 92. Lohman D, Cleary J, Connor S, De Lima L, Downing J, Marston J, et al. Advancing Global Palliative Care Over Two Decades: Health System Integration, Access to Essential Medicines, and Pediatrics. Journal of Pain and Symptom Management. 2022;64(1):58-69. 93. Knapp CA, Contro N. Family support services in pediatric palliative care. The American journal of hospice & palliative care. 2009;26(6):476-82. 590 94. Mastroyannopoulou K, Stallard P, Lewis M, Lenton S. The Impact of Childhood Non-malignant Life-threatening Illness on Parents: Gender Differences and Predictors of Parental Adjustment. Journal of Child Psychology and Psychiatry. 1997;38(7):823-9. 95. Hudson BF, Oostendorp LJM, Candy B, Vickerstaff V, Jones L, Lakhanpaul M, et al. The under reporting of recruitment strategies in research with children with life-threatening illnesses: A systematic review. Palliative Medicine. 2017;31(5):419-36. 96. Fisher V, Atkin K, Fraser LK. The health of mothers of children with a life-limiting condition: A qualitative interview study. Palliative Medicine. 2022;36(9):1418-25. 97. Fraser LK, Murtagh FE, Aldridge J, Sheldon T, Gilbody S, Hewitt C. Health of mothers of children with a life-limiting condition: a comparative cohort study. Arch Dis Child. 2021;106(10):987-93. 98. Sloper P, Turner S. Risk and Resistance Factors in the Adaptation of Parents of Children with Severe Physical Disability. Journal of Child Psychology and Psychiatry. 1993;34(2):167-88. 99. Hobfoll SE. Gender differences in stress reactions: Women filling the gaps. Psychology & Health. 1991;5(2):95-109. 100. Robertson J, Hatton C, Wells E, Collins M, Langer S, Welch V, et al. The Impacts of short break provision on families with a disabled child: an international literature review. Health & Social Care in the Community. 2011;19(4):337-71. 101. Meyer EC, Ritholz MD, Burns JP, Truog RD. Improving the quality of end-of-life care in the pediatric intensive care unit: parents' priorities and recommendations. Pediatrics. 2006;117(3):649-57. 102. Mack JW, Hilden JM, Watterson J, Moore C, Turner B, Grier HE, et al. Parent and physician perspectives on quality of care at the end of life in children with cancer. J Clin Oncol. 2005;23(36):9155-61. 103. Weidner NJ, Cameron M, Lee RC, McBride J, Mathias EJ, Byczkowski TL. End-of-life care for the dying child: what matters most to parents. Journal of Palliative Care. 2011;27(4):279-86. 104. Patel R, Coad J, Murray S. Communication barriers experienced by paediatric palliative care staff when interacting with children, young people and families: a literature review. BMJ Supportive & Palliative Care. 2012;2(Suppl 1):A21-A. 105. Beringer AJ, Heckford EJ. Was there a plan? End-of-life care for children with life-limiting conditions: a review of multi-service healthcare records. Child: Care, Health & Development. 2014;40(2):176-83. 591 106. Hillis R, Brenner M, Larkin PJ, Cawley D, Connolly M. The Role of Care Coordinator for Children with Complex Care Needs: A Systematic Review. International journal of integrated care. 2016;16(2):12-. 107. Heath G, Farre A, Shaw K. Parenting a child with chronic illness as they transition into adulthood: A systematic review and thematic synthesis of parents’ experiences. Patient Education and Counseling. 2017;100(1):76-92. 108. Kelly K. Standards framework for children’s palliative care. Bristol, UK; 2015. 109. Jarvis SW, Roberts D, Flemming K, Richardson G, Fraser LK. Transition of children with life-limiting conditions to adult care and healthcare use: a systematic review. Pediatric Research. 2021. 110. Chambers L. Stepping Up: A guide to enabling a good transition to adulthood for young people with life-limiting and life-threatening conditions. Together for Short Lives; 2015. 111. Fegran L, Hall EO, Uhrenfeldt L, Aagaard H, Ludvigsen MS. Adolescents' and young adults' transition experiences when transferring from paediatric to adult care: a qualitative metasynthesis. Int J Nurs Stud. 2014;51(1):123-35. 112. Heery E, Sheehan AM, While AE, Coyne I. Experiences and Outcomes of Transition from Pediatric to Adult Health Care Services for Young People with Congenital Heart Disease: A Systematic Review. Congenit Heart Dis. 2015;10(5):413-27. 113. Kerr H, Price J, O’Halloran P. A cross-sectional survey of services for young adults with life-limiting conditions making the transition from children’s to adult services in Ireland. Irish Journal of Medical Science (1971 -). 2020;189(1):33-42. 114. NICE. End of life care for adults: Quality standard [QS13]. Published 2011, Updated 2017. 115. NICE. End of life care for infants, children and young people: Quality standard [QS160] 2017. 116. Association BM. Children and young people tool kit. In: Association BM, editor. 2019. 117. Acts L-UPG. Mental Capacity Act 2005. 2005. 118. excellence NIfhac. NATIONAL INSTITUTE FOR HEALTH AND CARE EXCELLENCE - NICE GUIDELINE NG108 - Decision-making and mental capacity NICE; 2018. 119. Price A, Barry C. Mental Capacity and End of Life Decision-Making. In: Holland A, Gunn M, Jacob R, editors. Mental Capacity Legislation: Principles and Practice. 2 ed. Cambridge: Cambridge University Press; 2019. p. 91-101. 592 120. Boyle G. The Mental Capacity Act 2005: promoting the citizenship of people with dementia? Health & Social Care in the Community. 2008;16(5):529-37. 121. Waldman E, Frader J. Charlie Gard: How Did Things Go Wrong? Current Pediatrics Reports. 2018;6(2):173-7. 122. Caplan A, Folkers KM. Charlie Gard and the Limits of Parental Authority. Hastings Center Report. 2017;47(5):15-6. 123. Dyer C. Children’s hospital must be allowed to withdraw life support from Alfie Evans, court rules. BMJ. 2018;361:k1773. 124. Dyer C. Hospital trust asks High Court to rule whether it can withdraw girl’s life support. BMJ. 2019;366:l5097. 125. Bluebond-Langner M, Henderson EM, Langner R, Hargrave D. 'I have to live with the decisions I make': Laying a foundation for decision making for children with life-limiting conditions and life-threatening illnesses. Archives of Disease in Childhood. 2016. 126. Bluebond-Langner M. A child's view of death. Current Paediatrics. 1994;4(4):253-7. 127. CHILD COTROT. Convention on the Rights of the Child United Nations; 2009. 128. Heckford E, Townend N. Spiritual care: supporting the voice, and the courage, of a 13-year-old girl. Archives of Disease in Childhood. 2022;107(2):202-3. 129. Kirsch RE, Balit CR, Carnevale FA, Latour JM, Larcher V. Ethical, Cultural, Social, and Individual Considerations Prior to Transition to Limitation or Withdrawal of Life-Sustaining Therapies. Pediatric Critical Care Medicine. 2018;19(8S):S10-S8. 130. Conti A, Capasso E, Casella C, Fedeli P, Salzano FA, Policino F, et al. Blood Transfusion in Children: The Refusal of Jehovah's Witness Parents'. Open Med (Wars). 2018;13:101-4. 131. Carnevale FA, Macdonald ME, Razack S, Steinert Y. Promoting Cultural Awareness: A Faculty Development Workshop on Cultural Competency. The Canadian journal of nursing research = Revue canadienne de recherche en sciences infirmieres. 2015;47(2):18-40. 132. Kenny KE. From “Health for All” to “Health as Investment:” The Role of Economic Rationalities in the Transition from International to Global Health 1978–2013. In: Bek-Thomsen J, Christiansen CO, Gaarsmand Jacobsen S, Thorup M, editors. History of Economic Rationalities: Economic Reasoning as Knowledge and Practice Authority. Cham: Springer International Publishing; 2017. p. 121-31. 133. Blaug M. Economic Theory in Retrospect. 5th ed. Cambridge Cambridge University Press 1996. 593 134. Brouwer WBF, Culyer AJ, van Exel NJA, Rutten FFH. Welfarism vs. extra-welfarism. Journal of Health Economics. 2008;27(2):325-38. 135. Forget EL. Contested Histories of an Applied Field: The Case of Health Economics. History of Political Economy. 2004;36(4):617-37. 136. Culyer T. The normative economics of health care finance and provision. Oxford Review of Economic Policy. 1989;5:34-58. 137. Morris S, Devlin NJ, Parkin D. Economic analysis in health care. Chichester: John Wiley; 2007. 138. Black J, Hashimzade N, Myles G. A Dictionary of Economics: Oxford University Press; 2009 2009. 139. Mushkin SJ. Toward a definition of health economics. Public Health Rep. 1958;73(9):785-93. 140. Shiell A, Donaldson C, Mitton C, Currie G. Health economic evaluation. J Epidemiol Community Health. 2002;56(2):85-8. 141. Wagstaff A, Culyer AJ. Four decades of health economics through a bibliometric lens. J Health Econ. 2012;31(2):406-39. 142. Kogan BA. Enterprise in Public Health Economics by Labor Organizations: University of Michigan.; 1945. 143. Economics IoDH. Committee Reports of the Institute on Dental Health Economics, June 26 Through July 1, 19441944. 144. Economics UoMBoPH, Health UoMSoP. Public Health Economics: The University; 1948. 145. Arbor A. The president's report for 1946-7: University of Michigan official publication; 1947. 146. Blumenschein K, Johannesson M. Economic Evaluation in Healthcare. PharmacoEconomics. 1996;10(2):114-22. 147. Rivett G. Health & social care explained: NHS reform timeline Nuffield Trust Website: Nuffield Trust; 2019 [Available from: https://www.nuffieldtrust.org.uk/health-and-social-care-explained/nhs-reform-timeline. 148. Gorsky M. The British National Health Service 1948–2008: A Review of the Historiography. Soc Hist Med. 2008;21(3):437-60. 594 149. Hain R, Heckford E, McCulloch R. Paediatric palliative medicine in the UK: past, present, future. Archives of Disease in Childhood. 2012;97(4):381. 150. Carter R. Cicely Saunders—Founder of the Hospice Movement: Selected Letters 1959-1999. J R Soc Med. 2003;96(3):149-51. 151. Rudmik L, Drummond M. Health economic evaluation: important principles and methodology. Laryngoscope. 2013;123(6):1341-7. 152. Johnston BM. The Application of Economic Evaluation Techniques to Studies of Palliative and End-of-Life Care. In: Round J, editor. Care at the End of Life: An Economic Perspective. Cham: Springer International Publishing; 2016. p. 19-30. 153. Hinde S, Weatherly H, Walker G, Fraser LK. What Does Economic Evaluation Mean in the Context of Children at the End of Their Life? International Journal of Environmental Research and Public Health. 2021;18(21):11562. 154. Mathew C, Hsu AT, Prentice M, Lawlor P, Kyeremanteng K, Tanuseputro P, et al. Economic evaluations of palliative care models: A systematic review. Palliative Medicine. 2020;34(1):69-82. 155. Coast J. Maximisation in extra-welfarism: A critique of the current position in health economics. Social Science & Medicine. 2009;69(5):786-92. 156. Praag BMSv. The Relativity of the Welfare Concept in The Quality of Life. Sen MCNaA, editor. Oxford: Oxford University Press; 1993. 157. Sen A. Personal Utilities and Public Judgments: Or What's Wrong with Welfare Economics. Economic Journal. 1979;89(September). 158. Johansson PO. In: Johansson P-O, editor. An Introduction to Modern Welfare Economics. Cambridge: Cambridge University Press; 1991. 159. Nicholas K. Welfare Propositions of Economics and Interpersonal Comparisons of Utility. The Economic Journal. 1939;49(195):549-52. 160. Hicks JR. The Foundations of Welfare Economics. The Economic Journal. 1939;49(196):696-712. 161. Culyer AJ. The Dictionary of Health Economics, Second Edition Northampton Edward Elgar Publishing Inc.; 2010. 162. Hurley J. Chapter 2: An overview of the normative economics of the health sector. Handbook of Health Economics. 12000. p. 55-118. 163. Broome J. Trying to value a life. Journal of Public Economics. 1978;9(1):91-100. 595 164. Brazier J, Ratcliffe J, Saloman J, Tsuchiya A. Measuring and valuing health benefits for economic evaluation. Second edition. ed. Oxford: Oxford University Press; 2017. 165. Applied Health Economics for Public Health Practice and Research. Tudor Edwards R, McIntosh E, editors: Oxford University Press; 2019 01 Mar 2019. 166. Nick Hex JH. CHILDREN’S HOSPICES ACROSS SCOTLAND: Economic Evaluation of Hospice Services. York, UK: YHEC; 2021. 167. Williams A. Primeval health economics in Britain: A personal retrospect of the pre-HESG period. Health Economics. 1998;7(S1):S3-S8. 168. Lees DS. Health through choice. Hobart Paper 14. October 1961. 169. Coast J, Smith RD, Lorgelly P. Welfarism, extra-welfarism and capability: The spread of ideas in health economics. Social Science & Medicine. 2008;67(7):1190-8. 170. Culyer T. Medical care and the economics of giving. Economica. 1971;38:295-303. 171. Culyer AJ. The Nature of the Commodity 'Health Care' and its Efficient Allocation. Oxford Economic Papers. 1971;23(2):189-211. 172. Culyer AJ. Need and the National Health Service. London: M. Robertson; 1976. 173. Gray AM, Wilkinson T. Economic evaluation of healthcare interventions: old and new directions. Oxford Review of Economic Policy. 2016;32(1):102-21. 174. Emery DD, Schneiderman LJ. Cost-Effectiveness Analysis in Health Care. The Hastings Center Report. 1989;19(4):8-13. 175. Neumann PJ, Goldie SJ, Weinstein MC. Preference-Based Measures in Economic Evaluation in Health Care. Annual Review of Public Health. 2000;21(1):587-611. 176. Sinclair JC, Torrance GW, Boyle MH, Horwood SP, Saigal S, Sackett DL. Evaluation of neonatal-intensive-care program. N Engl J Med. 1981;305(9):489-94. 177. Svensson M, Hultkrantz L. A Comparison of Cost-Benefit and Cost-Effectiveness Analysis in Practice: Divergent Policy Practices in Sweden. Nordic Journal of Health Economics. 2017;5:41. 178. Stone PW, Schackman BR, Neukermans CP, Olchanski N, Greenberg D, Rosen AB, et al. A synthesis of cost-utility analysis literature in infectious disease. The Lancet Infectious Diseases. 2005;5(6):383-91. 179. MacKillop E, Sheard S. Quantifying life: Understanding the history of Quality-Adjusted Life-Years (QALYs). Social Science & Medicine. 2018;211:359-66. 596 180. NICE. The guidelines manual: process and methods (PMG6). NICE; 2012. 181. ZorginstituutNederland. Guidelines for economic evaluations in healthcare. 2016. 182. (NICE) NIfCaHE. Glossary 2019 [Available from: https://www.nice.org.uk/Glossary?letter=Q. 183. Normand C. Measuring Outcomes in Palliative Care: Limitations of QALYs and the Road to PalYs. Journal of Pain and Symptom Management. 2009;38(1):27-31. 184. Kinghorn P, Coast J. Appropriate frameworks for economic evaluation of end of life care: A qualitative investigation with stakeholders. Palliat Med. 2019;33(7):823-31. 185. de Wolf-Linder S, Dawkins M, Wicks F, Pask S, Eagar K, Evans CJ, et al. Which outcome domains are important in palliative care and when? An international expert consensus workshop, using the nominal group technique. Palliat Med. 2019;33(8):1058-68. 186. Hynson J, Aroni R, Bauld C, Sawyer S. Research with bereaved parents: a question of how not why. Palliative Medicine. 2006;20(8):805-11. 187. Cohen J. Putting a Different Spin on QALYs: Beyond a Sociological Critique. New Economics and Its History 1998:143–72. . 188. Bush JW, Fanshel S, Chen MM. Analysis of a tuberculin testing program using a health status index. Socio-Economic Planning Sciences. 1972;6(1):49-68. 189. Spencer A, Rivero-Arias O, Wong R, Tsuchiya A, Bleichrodt H, Edwards RT, et al. The QALY at 50: One story many voices. Social Science & Medicine. 2022;296:114653. 190. Ryan M, Gerard K. Inclusiveness in the health economic evaluation space. Social Science & Medicine. 2014;108:248-51. 191. Mackillop E. Valuing life: Exploring the history of Quality-Adjusted Life-Years (QALY) 2017 [Available from: https://remedianetwork.net/2017/05/03/valuing-life-exploring-the-history-of-quality-adjusted-life-years-qaly/. 192. Williams A. Discovering the QALY, or how Rachel Rosser changed my life. Personal Histories in Health Research The Nuffield Trust, London. 2005:191-206. 193. Enthoven AC. Reflections on the management of the national health service An international perspective on the NHS. . 1985. 194. Whitehead SJ, Ali S. Health outcomes in economic evaluation: the QALY and utilities. British Medical Bulletin. 2010;96(1):5-21. 597 195. Mooney G. What else do we want from our health services? Soc Sci Med. 1994;39(2):151-4. 196. Mooney G. Beyond health outcomes: The benefits of health care. Health Care Analysis. 1998;6(2):99-105. 197. NICE. NICE health technology evaluations: the manual. NICE; 2022. 198. Dolan P, Shaw R, Tsuchiya A, Williams A. QALY maximisation and people's preferences: a methodological review of the literature. Health economics. 2005;14(2):197-208. 199. Lancsar E, Gu Y, Gyrd-Hansen D, Butler J, Ratcliffe J, Bulfone L, et al. The relative value of different QALY types. Journal of Health Economics. 2020;70:102303. 200. Sakowsky RA. Disentangling the welfarism/extra-welfarism distinction: Towards a more fine-grained categorization. Health Economics. 2021;30(9):2307-11. 201. Culyer T. Commodities, characteristics of commodities, characteristics of people, utilities and the quality of life. In: Baldwin S, Godfrey C, Propper C, editors. The Quality of Life: Perspectives and Policies. London: Routledge; 1990. p. 9-27. 202. Stephen B, Donaldson C. Valuing the benefits and costs of health care programmes: Where's the 'extra' in extra-welfarism? Social science & medicine (1982). 2003;56:1121-33. 203. Brouwer and Hurley J. Welfarism, extra-welfarism and evaluative economic analysis in the health care sector. In: Barer ML, Getzen, T.E., Stoddard,, G.L., editors. Health, Health Care and Health Economics: Perspectives on Distribution. Chichester: John Wiley & Sons Ltd; 1998. 204. Seixas BV. Welfarism and extra-welfarism: a critical overview. Cadernos de Saúde Pública. 2017;33. 205. Bleichrodt H. Health utility indices and equity considerations. Journal of Health Economics. 1997;16(1):65-91. 206. Don R, Harold K, Ken B. Interpersonal Comparison of Utility. Oxford University Press; 2009. 207. Mauskopf JA, Paul JE, Grant DM, Stergachis A. The Role of Cost—Consequence Analysis in Healthcare Decision—Making. PharmacoEconomics. 1998;13(3):277-88. 208. Alkire S. Valuing Freedoms: Sen's Capability Approach and Poverty Reduction: Oxford University Press; 2002. 209. Shahani SDwL. An Introduction to the Human Development and Capability Approach: Freedom and Agency. Shahani SDwL, editor. London earthscan; 2010. 598 210. Stanton EA. The Human Development Index: A History. University of Massachusetts-Amherst, Amherst 2007. 211. Cookson R. QALYs and the capability approach. Health Economics. 2005;14(8):817-29. 212. Lorgelly PK, Lorimer K, Fenwick EAL, Briggs AH, Anand P. Operationalising the capability approach as an outcome measure in public health: The development of the OCAP-18. Social Science & Medicine. 2015;142:68-81. 213. Simon J, Anand P, Gray A, Rugkåsa J, Yeeles K, Burns T. Operationalising the capability approach for outcome measurement in mental health research. Social Science & Medicine. 2013;98:187-96. 214. Karimi M, Brazier J, Basarir H. The Capability Approach: A Critical Review of Its Application in Health Economics. Value in Health. 2016;19(6):795-9. 215. Anand P, Hunter G, Carter I, Dowding K, Guala F, Van Hees M. The Development of Capability Indicators. Journal of Human Development and Capabilities. 2009;10(1):125-52. 216. Anand P, Hunter G, Smith R. Capabilities and Well-Being: Evidence Based on the Sen–Nussbaum Approach to Welfare. Social Indicators Research. 2005;74(1):9-55. 217. Coast J, Flynn TN, Natarajan L, Sproston K, Lewis J, Louviere JJ, et al. Valuing the ICECAP capability index for older people. Social Science & Medicine. 2008;67(5):874-82. 218. Al-Janabi H, N Flynn T, Coast J. Development of a self-report measure of capability wellbeing for adults: the ICECAP-A. Quality of Life Research. 2012;21(1):167-76. 219. Bailey C, Orlando R, Kinghorn P, Armour K, Perry R, Coast J. MEASURING THE QUALITY OF END OF LIFE USING ICECAP SCM: FEASIBILITY AND ACCEPTABILITY. BMJ Supportive & Palliative Care. 2014;4(1):112-. 220. Kinghorn P, Robinson A, Smith RD. Developing a Capability-Based Questionnaire for Assessing Well-Being in Patients with Chronic Pain. Social Indicators Research. 2015;120(3):897-916. 221. Helter TM, Coast J, Łaszewska A, Stamm T, Simon J. Capability instruments in economic evaluations of health-related interventions: a comparative review of the literature. Quality of Life Research. 2019. 222. Sen A. Equality of What? In: McMurrin S, editor. Tanner Lectures on Human Values, Volume 1. Cambridge: Cambridge University Press; 1980. 223. Sen A. Functionings and Capability. Inequality Reexamined. New York Russel Sage Foundation 1992. 599 224. Sen MCNaA, editor. Capability and Wellbeing in The Quality of Life. New York: Oxford University Press; 1993. 225. Sen A. Commodities and Capabilities. Amsterdam: North-Holland; 1985. 226. Sen A. Rationality and Freedom MA: Harvard University Press, Belknap Press; 2002. 227. Sen A. Development as freedom. New York: Oxford University Press; 1999. 228. Robeyns I, Open Book P. Wellbeing, freedom and social justice : the capability approach re-examined. Cambridge: Open Book Publishers; 2017. Available from: http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=1682818 http://www.jstor.org/stable/10.2307/j.ctt1zkjxgc http://www.oapen.org/download?type=document&docid=641879 http://dx.doi.org/10.11647/OBP.0130 http://www.openbookpublishers.com/shopimages/products/cover/682 https://www.openbookpublishers.com/product/682 https://www.openbookpublishers.com/reader/682 http://0-dx.doi.org.pugwash.lib.warwick.ac.uk/10.11647/OBP.0130 http://VH7QX3XE2P.search.serialssolutions.com/?V=1.0&L=VH7QX3XE2P&S=JCs&C=TC0001957299&T=marc&tab=BOOKS http://VH7QX3XE2P.search.serialssolutions.com/?V=1.0&L=VH7QX3XE2P&S=AC_T_B&C=Wellbeing%2C%20Freedom%20and%20Social%20Justice%20:%20The%20Capability%20Approach%20Re-Examined&T=marc&tab=BOOKS http://liverpool.idm.oclc.org/login?url=https://www.openbookpublishers.com/product/682 https://login.gbcprx01.georgebrown.ca/login?url=http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=e000xna&AN=1682818 https://nl.idm.oclc.org/login?url=http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&AN=1682818 http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&AN=1682818 http://ezproxy.eui.eu/login?url=http://dx.doi.org/10.11647/OBP.0130. 600 229. Nussbaum MC. Creating capabilities : the human development approach. Cambridge, Massachusetts: The Belknap Press of Harvard University Press; 2011. 230. Robeyns I. The Capability Approach in Practice*. Journal of Political Philosophy. 2006;14(3):351-76. 231. Robeyns IaB, Morten Fibieger. The Capability Approach. In: Zalta EN, editor. The {Stanford} Encyclopedia of Philosophy: Metaphysics Research Lab, Stanford University; 2011. 232. Robeyns I. SEN'S CAPABILITY APPROACH AND GENDER INEQUALITY: SELECTING RELEVANT CAPABILITIES. Feminist Economics. 2003;9:61-92. 233. Robeyns I. Wellbeing, Freedom and Social Justice: The Capability Approach Re-Examined: Open Book Publishers; 2017 234. Greco G, Skordis-Worrall J, Mkandawire B, Mills A. What is a good life? Selecting capabilities to assess women's quality of life in rural Malawi. Soc Sci Med. 2015;130:69-78. 235. Coast J. Strategies for the economic evaluation of end-of-life care: making a case for the capability approach. Expert Review of Pharmacoeconomics & Outcomes Research. 2014;14(4):473-82. 236. Oduro GY, Otsin MNA. "Abortion-It is my own body": Women's narratives about influences on their abortion decisions in Ghana. Health Care for Women International. 2014;35(7-9):918-36. 237. Mitchell PM, Coast J, Myring G, Ricciardi F, Vickerstaff V, Jones L, et al. Exploring the costs, consequences and efficiency of three types of palliative care day services in the UK: a pragmatic before-and-after descriptive cohort study. BMC Palliat Care. 2020;19(1):119. 238. Evans P. Collective capabilities, culture, and Amartya Sen’sDevelopment as Freedom. Studies in Comparative International Development. 2002;37(2):54-60. 239. GORE C. Irreducibly social goods and the informational basis of Amartya Sen's capability approach. Journal of International Development. 1997;9(2):235-50. 240. Pogge T. A critique of the capability approach. 2010:17-60. 241. Alkire S, Black R. A practical reasoning theory of development ethhics: furthering the capabilities approach. Journal of International Development. 1997;9(2):263-79. 242. Clark DA. The Capability Approach: Its Development, Critiques and Recent Advances Global poverty research group - ESRC. 243. Nussbaum M. NATURE, FUNCTION, AND CAPABILITY: 601 ARISTOTLE ON POLITICAL DISTRIBUTION Oxford Studies in Ancient Philosophy. 1988;14584. 244. Gray AM. Inequalities in health. The Black Report: a summary and comment. Int J Health Serv. 1982;12(3):349-80. 245. Mack SLaJ. A more unequal country? : Poverty and Social Exclusion 2013 [Available from: https://www.poverty.ac.uk/editorial/more-unequal-country 246. trust TE. How Has Inequality Changed? Development of UK Income Inequality [Available from: https://www.equalitytrust.org.uk/how-has-inequality-changed. 247. by) RLaWGe. Interpreting Official Statistics. Andover, Hampshire SP10 5BE Routledge; September 1996. 248. Alkire S. Why the Capability Approach? Journal of Human Development and Capabilities. 2005;6:115-35. 249. Brouwer WBF, Van Exel NJA, Baltussen RMPM, Rutten FFH. A Dollar Is a Dollar Is a Dollar—or Is It? Value in Health. 2006;9(5):341-7. 250. Michael Marmot PG, Jessica Allen , et al. Fair Society, Healthy Lives (The Marmot Review). Institute of health equity 2010. 251. Marmot M AJ, Boyce T, Goldblatt P, Morrison J. Health Equity in England: The Marmot Review 10 Years On.: Institute of Health Equity; 2020. 252. Mitchell PM, Roberts TE, Barton PM, Coast J. Applications of the Capability Approach in the Health Field: A Literature Review. Soc Indic Res. 2017;133(1):345-71. 253. Coast J. Assessing capability in economic evaluation: a life course approach? The European Journal of Health Economics. 2019;20(6):779-84. 254. Venkatapuram S. Health justice: An argument from the capabilities approach: John Wiley & Sons; 2013. 255. Ruger JP. Millenium Development goals for health: building human capabilities: Round Table Discussion. Bulletin of the World Health Organization: the International Journal of Public Health 2004; 82 (12): 951-952. 2004. 256. Ruger JP. Ethics and governance of global health inequalities. Journal of Epidemiology & Community Health. 2006;60(11):998-1002. 257. Ruger JP. Health, capability, and justice: toward a new paradigm of health ethics, policy and law. Cornell JL & Pub Pol'y. 2005;15:403. 602 258. Lavie-Ajayi M, Moran GS, Levav I, Porat R, Reches T, Goldfracht M, et al. Using the capabilities approach to understand inequality in primary health-care services for people with severe mental illness. Israel Journal of Health Policy Research. 2018;7(1):49. 259. Entwistle V, Firnigl D, Ryan M, Francis J, Kinghorn P. Which experiences of health care delivery matter to service users and why? A critical interpretive synthesis and conceptual map. Journal of health services research & policy. 2012;17(2):70-8. 260. Ryan M, Kinghorn P, Entwistle VA, Francis JJ. Valuing patients' experiences of healthcare processes: towards broader applications of existing methods. Social science & medicine. 2014;106:194-203. 261. Gabriell Conti JJH. The Economics of Child Well-Being IZA. 2012;Discussion Paper No. 6930 262. Ben-Arieh A. The Child Indicators Movement: Past, Present, and Future. Child Indicators Research. 2008;1(1):3-16. 263. Ben-Arieh A. Beyond Welfare: Measuring and Monitoring the State of Children – New Trends and Domains. Social Indicators Research. 2000;52(3):235-57. 264. Chase JSaE. Childhood Wellbeing: A brief overview. Childhood Wellbeing Research Centre 2010. 265. Biggeri M, Ballet, J., Comim, F. (Eds.) Children and the Capability Approach: Palgrave Macmillan 2011. 266. Domínguez-Serrano M, del Moral-Espín L, Gálvez Muñoz L. A well-being of their own: Children’s perspectives of well-being from the capabilities approach. Childhood. 2018;26(1):22-38. 267. Ben-Arieh A. Where are the Children? Children’s Role in Measuring and Monitoring Their Well-Being. Social Indicators Research. 2005;74(3):573-96. 268. Ungar W, & Gerber, A. . The uniqueness of child health and challenges to measuring costs and consequences. In Economic Evaluation in Child Health. Oxford University Press. (2009-10-08). 269. Adamson P. Child well-being in rich countries A comparative overview. Florence, Italy: United Nations Children’s Fund (UNICEF); 2013. 270. Brouwer W, van Exel, N., & Tilford, J. . Incorporating caregiver and family effects in economic evaluations of child health. In Economic Evaluation in Child Health. 2009. 271. Marcinko DE, Hetico H. Dictionary of Health Economics and Finance. New York, UNITED STATES: Springer Publishing Company; 2006. 603 272. Coast J, Flynn T, Sutton E, Al-Janabi H, Vosper J, Lavender S, et al. Investigating Choice Experiments for Preferences of Older People (ICEPOP): Evaluative spaces in health economics. Journal of Health Services Research & Policy. 2008;13(3_suppl):31-7. 273. Mooney G. QALYs: Are They Enough? A Health Economist's Perspective. Journal of Medical Ethics. 1989;15(3):148-52. 274. Ryan M. Using conjoint analysis to take account of patient preferences and go beyond health outcomes: an application to in vitro fertilisation. Social Science & Medicine. 1999;48(4):535-46. 275. Payne K, Eden M. Measuring the economic value of genetic counselling. European Journal of Medical Genetics. 2019;62(5):385-9. 276. Tilden VP, Tolle S, Drach L, Hickman S. Measurement of Quality of Care and Quality of Life at the End of Life. The Gerontologist. 2002;42(suppl_3):71-80. 277. Bausewein C DB, Benalia H,Simon ST,Higginson IJ. Outcome Measurement in Palliative Care The Essentials Prisma; 2009. 278. Drummond MF. Methods for the Economic Evaluation of Health Care Programmes1997. 279. Porta MS, Last JM. A dictionary of public health. [Oxford] ;: Oxford University Press; 2018. Available from: http://dx.doi.org/10.1093/acref/9780191844386.001.0001 http://www.oxfordreference.com/view/10.1093/acref/9780191844386.001.0001/acref-9780191844386 https://ezproxy.spl.org/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780191844386.001.0001/acref-9780191844386 http://ezproxy.eui.eu/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780191844386.001.0001/acref-9780191844386 http://proxy.library.carleton.ca/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780191844386.001.0001/acref-9780191844386 https://purl.missouristate.edu/library/e-books/9780191844386 https://login.proxy.bib.uottawa.ca/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780191844386.001.0001/acref-9780191844386 http://liverpool.idm.oclc.org/login?url=http://www.oxfordreference.com/view/10.1093/acref/9780191844386.001.0001/acref-9780191844386. 604 280. Addington-Hall JM. Research methods in palliative care. Oxford ;: Oxford University Press; 2007. 281. Stevens K, Palfreyman SSTHNHSFTSUK. The Use of Qualitative Methods in Developing the Descriptive Systems of Preference-Based Measures of Health-Related Quality of Life for Use in Economic Evaluation. Value in Health. 2012;15(8):991-8. 282. Dzingina M, Higginson IJ, McCrone P, Murtagh FEM. Development of a Patient-Reported Palliative Care-Specific Health Classification System: The POS-E. Patient. 2017;10(3):353-65. 283. Pietersma S, van den Akker-van ME, de Vries M. Generic quality of life utility measures in health-care research: Conceptual issues highlighted for the most commonly used utility measures. International Journal of Wellbeing. 2013;3(2). 284. Devlin NJ, Brooks R. EQ-5D and the EuroQol Group: Past, Present and Future. Applied Health Economics and Health Policy. 2017;15(2):127-37. 285. Johnston B, Flemming K, Narayanasamy MJ, Coole C, Hardy B. Patient reported outcome measures for measuring dignity in palliative and end of life care: a scoping review. BMC Health Services Research. 2017;17(1):574. 286. Badia X, Monserrat S, Roset M, Herdman M. Feasibility, Validity and Test-Retest Reliability of Scaling Methods for Health States: The Visual Analogue Scale and the Time Trade-Off. Quality of Life Research. 1999;8(4):303-10. 287. Streiner DL, Norman GR, Cairney J. Health Measurement Scales : A Practical Guide to Their Development and Use. Oxford, UNITED KINGDOM: Oxford University Press, Incorporated; 2015. 288. Carmines EG, Zeller RA. Reliability and validity assessment. Beverly Hills London: Sage Publications; 1979. 289. Brazier J, Deverill M. A checklist for judging preference-based measures of health related quality of life: Learning from psychometrics. Health Economics. 1999;8(1):41-51. 290. Canaway A, Al-Janabi H, Kinghorn P, Bailey C, Coast J. Development of a measure (ICECAP-Close Person Measure) through qualitative methods to capture the benefits of end-of-life care to those close to the dying for use in economic evaluation. Palliative Medicine. 2016;31(1):53-62. 291. Bausewein C, Daveson BA, Currow DC, Downing J, Deliens L, Radbruch L, et al. EAPC White Paper on outcome measurement in palliative care: Improving practice, attaining outcomes and delivering quality services – Recommendations from the European Association for Palliative Care (EAPC) Task Force on Outcome Measurement. Palliative Medicine. 2015;30(1):6-22. 605 292. Higginson IJ, Evans CJ, Grande G, Preston N, Morgan M, McCrone P, et al. Evaluating complex interventions in End of Life Care: the MORECare Statement on good practice generated by a synthesis of transparent expert consultations and systematic reviews. BMC Medicine. 2013;11(1):111. 293. Bailey C, Kinghorn P, Orlando R, Armour K, Perry R, Jones L, et al. 'The ICECAP-SCM tells you more about what I'm going through': A think-aloud study measuring quality of life among patients receiving supportive and palliative care. Palliative medicine. 2016;30(7):642-52. 294. Normand C. Setting priorities in and for end-of-life care: challenges in the application of economic evaluation. Health economics, policy, and law. 2012;7(4):431-9. 295. Wichmann AB, Adang EMM, Stalmeier PFM, Kristanti S, Van den Block L, Vernooij-Dassen MJFJ, et al. The use of Quality-Adjusted Life Years in cost-effectiveness analyses in palliative care: Mapping the debate through an integrative review. Palliative Medicine. 2017;31(4):306-22. 296. Coast J, Bailey C, Kinghorn P. Patient centered outcome measurement in health economics: beyond EQ-5D and the Quality-Adjusted Life-Year—where are we now? Annals of Palliative Medicine. 2018:S249-S52. 297. Round J. Is a QALY still a QALY at the end of life? Journal of Health Economics. 2012;31(3):521-7. 298. Sampson CJ. Identifying Objects of Value at the End of Life. In: Round J, editor. Care at the End of Life: An Economic Perspective: Adis; 2016. p. 103-22. 299. Murtagh FEM, Ramsenthaler C, Firth A, Groeneveld EI, Lovell N, Simon ST, et al. A brief, patient- and proxy-reported outcome measure in advanced illness: Validity, reliability and responsiveness of the Integrated Palliative care Outcome Scale (IPOS). Palliative Medicine. 2019;33(8):1045-57. 300. Huynh E, Coast J, Rose J, Kinghorn P, Flynn T. Values for the ICECAP-Supportive Care Measure (ICECAP-SCM) for use in economic evaluation at end of life. Soc Sci Med. 2017;189:114-28. 301. Canaway A. Capturing the impacts of end of life care on those close to the dying for use in economic evaluation [Doctoral]. University of Birmingham: University of Birmingham; 2015. 302. Kinghorn P, Canaway A, Bailey C, Al-Janabi H, Coast J. A Deliberative Approach to Valuing Capabilities: Assessing and Valuing Changes in the Well-Being of those Close to Patients Receiving Supportive End of Life Care. Journal of Human Development and Capabilities. 2022;23(3):455-76. 303. Chochinov HM. Death, Time and the Theory of Relativity. Journal of Pain and Symptom Management. 2011;42(3):460-3. 606 304. Prosser LA, Wittenberg E. Advances in Methods and Novel Applications for Measuring Family Spillover Effects of Illness. PharmacoEconomics. 2019;37(4):447-50. 305. Al-Janabi H, Wittenberg E, Donaldson C, Brouwer W. The relative value of carer and patient quality of life: A person trade-off (PTO) study. Social Science & Medicine. 2022;292:114556. 306. Al-Janabi H, Van Exel J, Brouwer W, Trotter C, Glennie L, Hannigan L, et al. Measuring Health Spillovers for Economic Evaluation: A Case Study in Meningitis. Health Econ. 2016;25(12):1529-44. 307. Al-Janabi H, van Exel J, Brouwer W, Coast J. A Framework for Including Family Health Spillovers in Economic Evaluation. Med Decis Making. 2016;36(2):176-86. 308. Bobinac A, van Exel NJ, Rutten FF, Brouwer WB. Caring for and caring about: disentangling the caregiver effect and the family effect. J Health Econ. 2010;29(4):549-56. 309. Bobinac A, van Exel NJ, Rutten FF, Brouwer WB. Health effects in significant others: separating family and care-giving effects. Med Decis Making. 2011;31(2):292-8. 310. Al-Janabi H, Flynn TN, Coast J. QALYs and carers. PharmacoEconomics. 2011;29(12):1015-23. 311. Hoefman RJ, van Exel J, Brouwer W. How to include informal care in economic evaluations. Pharmacoeconomics. 2013;31(12):1105-19. 312. Haycox A. Optimizing decision making and resource allocation in palliative care. J Pain Symptom Manage. 2009;38(1):45-53. 313. Goodrich K, Kaambwa B, Al-Janabi H. The Inclusion of Informal Care in Applied Economic Evaluation: A Review. Value in Health. 2012;15(6):975-81. 314. Canaway A, Al-Janabi H, Kinghorn P, Bailey C, Coast J. Close-Person Spill-Overs in End-of-Life Care: Using Hierarchical Mapping to Identify Whose Outcomes to Include in Economic Evaluations. PharmacoEconomics. 2019;37(4):573-83. 315. Tilford JM, Payakachat N. Progress in measuring family spillover effects for economic evaluations. Expert Review of Pharmacoeconomics & Outcomes Research. 2015;15(2):195-8. 316. Dixon P, Round J. Caring for Carers: Positive and Normative Challenges for Future Research on Carer Spillover Effects in Economic Evaluation. Value in Health. 2019;22(5):549-54. 317. Brouwer W, Van Exel N, Van Gorp B, Redekop W. The CarerQol instrument: a new instrument to measure care-related quality of life of informal caregivers for use in economic evaluations. Quality of Life Research. 2006;15:1005-21. 607 318. Al-Janabi H, Coast J, Flynn TN. What do people value when they provide unpaid care for an older person? A meta-ethnography with interview follow-up. Social science & medicine. 2008;67(1):111-21. 319. Rand S, Malley J, Netten A. Identifying the impact of adult social care (IIASC): Interim technical report. Canterbury: Personal Social Services Research Unit, University of Kent. 2012. 320. Neko C. FAMCARE and FAMCARE-2 Guidelines. 321. Hoefman R, Al-Janabi H, McCaffrey N, Currow D, Ratcliffe J. Measuring caregiver outcomes in palliative care: a construct validation study of two instruments for use in economic evaluations. Qual Life Res. 2015;24(5):1255-73. 322. Basu A, Meltzer D. Implications of spillover effects within the family for medical cost-effectiveness analysis. J Health Econ. 2005;24(4):751-73. 323. Wittenberg E, James LP, Prosser LA. Spillover Effects on Caregivers’ and Family Members’ Utility: A Systematic Review of the Literature. PharmacoEconomics. 2019;37(4):475-99. 324. Hoefman RJ, van Exel J, Brouwer W. How to Include Informal Care in Economic Evaluations. PharmacoEconomics. 2013;31(12):1105-19. 325. Hoefman RJ, van Exel J, Brouwer WBF. Measuring Care-Related Quality of Life of Caregivers for Use in Economic Evaluations: CarerQol Tariffs for Australia, Germany, Sweden, UK, and US. PharmacoEconomics. 2017;35(4):469-78. 326. Coast J, Bailey C, Canaway A, Kinghorn P. “It is not a scientific number it is just a feeling”: Populating a multi-dimensional end-of-life decision framework using deliberative methods. Health Economics. 2021;30(5):1033-49. 327. Ungar WJD. Challenges in Health State Valuation in Paediatric Economic Evaluation : Are QALYs Contraindicated? PharmacoEconomics. 2011;29(8):641-52. 328. Flynn TN, Louviere JJ, Peters TJ, Coast J. Best–worst scaling: What it can do for health care research and how to do it. Journal of Health Economics. 2007;26(1):171-89. 329. Thompson GN, Chochinov HM. Methodological Challenges in Measuring Quality Care at the End of Life in the Long-Term Care Environment. Journal of Pain and Symptom Management. 2006;32(4):378-91. 330. Pereira SM, Hernández-Marrero P. Ethical challenges of outcome measurement in palliative care clinical practice: a systematic review of systematic reviews. Annals of Palliative Medicine. 2018:S207-S18. 608 331. Mayland C, Williams E, Ellershaw J. How well do current instruments using bereaved relatives' views evaluate care for dying patients? Palliative medicine. 2008;22(2):133-44. 332. Evans CJ, Harding R, Higginson IJ. ‘Best practice’ in developing and evaluating palliative and end-of-life care services: A meta-synthesis of research methods for the MORECare project. Palliative Medicine. 2013;27(10):885-98. 333. Field MJ, Cassel CK, Institute of Medicine . Committee on Care at the End of L. Approaching death : improving care at the end of life. Washington, D.C.: National Academy Press; 1997. Available from: http://site.ebrary.com/id/10040959 http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=916 http://0-search.ebscohost.com.librarycatalog.vts.edu/login.aspx?direct=true&scope=site&db=nlebk&AN=916 http://books.google.com/books?id=-jVrAAAAMAAJ http://catalog.hathitrust.org/api/volumes/oclc/37310651.html http://www.myilibrary.com?id=208217 http://www.nap.edu/catalog.php?record_id=5801 https://www.nap.edu/catalog.php?record_id=5801 http://libproxy.uwyo.edu/login/?url=http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=916 http://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&AN=916 http://whel-primo.hosted.exlibrisgroup.com/openurl/44WHELF_ABW/44WHELF_ABW_VU1 ?u.ignore_date_coverage=true&rft.mms_id=9912085872902418 https://openurl.ac.uk/ukfed:bat?u.ignore_date_coverage=true&rft.mms_id=991003535923102761 http://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=3375542 https://elib.southwest.tn.edu:3443/login?url=http://www.netlibrary.com/urlapi.asp?action=summary&v=1&bookid=916 http://www.netlibrary.com/urlapi.asp?action=summary&v=1&bookid=916 http://www.vlebooks.com/vleweb/product/openreader?id=none&isbn=9780309518253. 609 334. Bryman A. Social research methods. Fifth edition. ed. Oxford ;: Oxford University Press; 2016. 335. Fowler FJ, Jr. Improving Survey Questions: Design and Evaluation1995. 336. Gott M, Robinson J, Moeke-Maxwell T, Black S, Williams L, Wharemate R, et al. ‘It was peaceful, it was beautiful’: A qualitative study of family understandings of good end-of-life care in hospital for people dying in advanced age. Palliative Medicine. 2019;33(7):793-801. 337. Nekolaichuk CL, Bruera E, Spachynski K, MacEachern T, Hanson J, Maguire TO. A comparison of patient and proxy symptom assessments in advanced cancer patients. Palliative Medicine. 1999;13(4):311-23. 338. Orgeta V, Edwards RT, Hounsome B, Orrell M, Woods B. The use of the EQ-5D as a measure of health-related quality of life in people with dementia and their carers. Quality of Life Research. 2015;24(2):315-24. 339. Higginson I, Priest P, McCarthy M. Are bereaved family members a valid proxy for a patient's assessment of dying? Social Science & Medicine. 1994;38(4):553-7. 340. Janssens A, Thompson Coon J, Rogers M, Allen K, Green C, Jenkinson C, et al. A Systematic Review of Generic Multidimensional Patient-Reported Outcome Measures for Children, Part I: Descriptive Characteristics. Value in Health. 2015;18(2):315-33. 341. Matza LS, Patrick DL, Riley AW, Alexander JJ, Rajmil L, Pleil AM, et al. Pediatric Patient-Reported Outcome Instruments for Research to Support Medical Product Labeling: Report of the ISPOR PRO Good Research Practices for the Assessment of Children and Adolescents Task Force. Value in Health. 2013;16(4):461-79. 342. Willis J, Zeratkaar D, ten Hove J, Rosenbaum P, Ronen GM. Engaging the Voices of Children: A Scoping Review of How Children and Adolescents Are Involved in the Development of Quality-of-Life–Related Measures. Value in Health. 2021;24(4):556-67. 343. Stewart JL, Lynn MR, Mishel MH. Evaluating content validity for children's self-report instruments using children as content experts. Nurs Res. 2005;54(6):414-8. 344. Ravens-Sieberer U, Erhart M, Wille N, Wetzel R, Nickel J, Bullinger M. Generic health-related quality-of-life assessment in children and adolescents: methodological considerations. Pharmacoeconomics. 2006;24(12):1199-220. 345. Coombes L, Bristowe K, Ellis-Smith C, Aworinde J, Fraser LK, Downing J, et al. Enhancing validity, reliability and participation in self-reported health outcome measurement for children and young people: a systematic review of recall period, response scale format, and administration modality. Quality of Life Research. 2021. 610 346. Guerriero C, Jaume NA, Diaz-Ordaz K, Brown KL, Wray J, Ashworth J, et al. Using Animation to Self-Report Health: A Randomized Experiment with Children. The Patient - Patient-Centered Outcomes Research. 2020;13(2):175-88. 347. Janssens A, Rogers M, Thompson Coon J, Allen K, Green C, Jenkinson C, et al. A Systematic Review of Generic Multidimensional Patient-Reported Outcome Measures for Children, Part II: Evaluation of Psychometric Performance of English-Language Versions in a General Population. Value in Health. 2015;18(2):334-45. 348. Stevens KJ. Working with children to develop dimensions for a preference-based, generic, pediatric, health-related quality-of-life measure. Qual Health Res. 2010;20(3):340-51. 349. Edwards TC, Huebner CE, Connell FA, Patrick DL. Adolescent quality of life, part I: conceptual and measurement model. J Adolesc. 2002;25(3):275-86. 350. Swerts C, De Maeyer J, Lombardi M, Waterschoot I, Vanderplasschen W, Claes C. “You Shouldn’t Look at Us Strangely”: An Exploratory Study on Personal Perspectives on Quality of Life of Adolescents with Emotional and Behavioral Disorders in Residential Youth Care. Applied Research in Quality of Life. 2019;14(4):867-89. 351. Arnett JJ. Emerging adulthood. A theory of development from the late teens through the twenties. Am Psychol. 2000;55(5):469-80. 352. Griebsch I, Coast J, Brown J. Quality-Adjusted Life-Years Lack Quality in Pediatric Care: A Critical Review of Published Cost-Utility Studies in Child Health. Pediatrics. 2005;115(5):e600. 353. Prosser L, Hammitt J, Keren R. Measuring Health Preferences for Use in Cost-Utility and Cost-Benefit Analyses of Interventions in Children: Theoretical and Methodological Considerations. PharmacoEconomics. 2008;25(9):713-26. 354. Keren R, Pati S, Feudtner C. The Generation Gap. PharmacoEconomics. 2004;22(2):71-81. 355. Poley MJ, Brouwer WBF, van Exel NJA, Tibboel D. Assessing health-related quality-of-life changes in informal caregivers: an evaluation in parents of children with major congenital anomalies. Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation. 2012;21(5):849-61. 356. Lavelle TA, D’Cruz BN, Mohit B, Ungar WJ, Prosser LA, Tsiplova K, et al. Family Spillover Effects in Pediatric Cost-Utility Analyses. Applied Health Economics and Health Policy. 2019;17(2):163-74. 357. Harding R, Chambers L, Bluebond-Langner M. Advancing the science of outcome measurement in paediatric palliative care. Int J Palliat Nurs. 2019;25(2):72-9. 611 358. Ullrich C, Morrison RS. Pediatric palliative care research comes of age: what we stand to learn from children with life-threatening illness. J Palliat Med. 2013;16(4):334-6. 359. Friedel M, Aujoulat I, Dubois A-C, Degryse J-M. Instruments to Measure Outcomes in Pediatric Palliative Care: A Systematic Review. Pediatrics. 2019;143(1):e20182379. 360. Knapp C, Madden V. Conducting outcomes research in pediatric palliative care. The American journal of hospice & palliative care. 2010;27(4):277-81. 361. Turner N, Almack K. Recruiting young people to sensitive research: turning the "wheels within wheels'. International Journal of Social Research Methodology. 2017;20(5):485-97. 362. Downing J, Ojing M, Powell R, Ali Z, Marston J, Meiring M, et al. A palliative care outcome measure for children in sub-Saharan Africa: Early development findings. European Journal of Palliative Care. 2012;19:292-5. 363. Downing J, Namisango E, Harding R. Outcome measurement in paediatric palliative care: lessons from the past and future developments. Ann Palliat Med. 2018;7(Suppl 3):S151-s63. 364. Coombes L, Braybrook D, Roach A, Scott H, Haroardottir D, Bristowe K, et al., editors. Priority outcomes for children and young people with life-limiting conditions: an in-depth qualitative interview study2020 2020/9/21. 365. Huang IC, Shenkman EA, Madden VL, Vadaparampil S, Quinn G, Knapp CA. Measuring quality of life in pediatric palliative care: challenges and potential solutions. Palliat Med. 2010;24(2):175-82. 366. Wille N, Badia X, Bonsel G, Burstrom K, Cavrini G, Devlin N, et al. Development of the EQ-5D-Y: a child-friendly version of the EQ-5D. Qual Life Res. 2010;19(6):875-86. 367. Moodie M, Richardson J, Rankin B, Iezzi A, Sinha K. Predicting time trade-off health state valuations of adolescents in four Pacific countries using the Assessment of Quality-of-Life (AQoL-6D) instrument. Value Health. 2010;13(8):1014-27. 368. Beusterien KM, Yeung JE, Pang F, Brazier J. Development of the multi-attribute Adolescent Health Utility Measure (AHUM). Health Qual Life Outcomes. 2012;10:102. 369. Coombes LH, Wiseman T, Lucas G, Sangha A, Murtagh FE. Health-related quality-of-life outcome measures in paediatric palliative care: A systematic review of psychometric properties and feasibility of use. Palliat Med. 2016;30(10):935-49. 370. Mitchell PM, Roberts TE, Barton PM, Coast J. Assessing sufficient capability: A new approach to economic evaluation. Social Science & Medicine. 2015;139:71-9. 612 371. Månsdotter A, Ekman B, Feldman I, Hagberg L, Hurtig AK, Lindholm L. We Propose a Novel Measure for Social Welfare and Public Health: Capability-Adjusted Life-Years, CALYs. Applied health economics and health policy. 2017;15(4):437-40. 372. Netten A, Burge P, Malley J, Potoglou D, Towers AM, Brazier J, et al. Outcomes of social care for adults: developing a preference-weighted measure. 2012;16:16. 373. Namisango E, Bristowe K, Allsop MJ, Murtagh FEM, Abas M, Higginson IJ, et al. Symptoms and Concerns Among Children and Young People with Life-Limiting and Life-Threatening Conditions: A Systematic Review Highlighting Meaningful Health Outcomes. Patient. 2019;12(1):15-55. 374. Ghirotto L, Busani E, Salvati M, Di Marco V, Caldarelli V, Artioli G. Researching children's perspectives in pediatric palliative care: A systematic review and meta-summary of qualitative research. Palliative & Supportive Care. 2019;17(1):107-18. 375. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. 376. Dissemination CfRa. Systematic Reviews. University of York: CRD; 2009. 377. Coombes L, Braybrook D, Roach A, Scott H, Harðardóttir D, Bristowe K, et al. Achieving child-centred care for children and young people with life-limiting and life-threatening conditions-a qualitative interview study. Eur J Pediatr. 2022;181(10):3739-52. 378. Higham L, Ahmed S, Ahmed M. Hoping to live a "normal" life whilst living with unpredictable health and fear of death: Impact of cystic fibrosis on young adults. [References]: Journal of Genetic Counseling. Vol.22(3), 2013, pp. 374-383.; 2013. 379. Johnston B, Jindal-Snape D, Pringle J, Gold L, Grant J, Dempsey R, et al. Understanding the relationship transitions and associated end of life clinical needs of young adults with life-limiting illnesses: A triangulated longitudinal qualitative study. SAGE Open Medicine. 2016;4:2050312116666429. 380. Kenten C, Ngwenya N, Gibson F, Flatley M, Jones L, Pearce S, et al. Understanding care when cure is not likely for young adults who face cancer: a realist analysis of data from patients, families and healthcare professionals. BMJ Open. 2019;9(1):e024397. 381. Kirk S, Fraser C. Hospice support and the transition to adult services and adulthood for young people with life-limiting conditions and their families: A qualitative study. [References]: Palliative Medicine. Vol.28(4), 2014, pp. 342-352.; 2014. 382. Mitchell S, Slowther AM, Coad J, Dale J. Experiences of healthcare, including palliative care, of children with life-limiting and life-threatening conditions and their families: a longitudinal qualitative investigation. Archives of Disease in Childhood. 2021;106(6):570-6. 613 383. Sodergren SC, Husson O, Rohde GE, Tomasewska IM, Vivat B, Yarom N, et al. A life put on pause: An exploration of the health-related quality of life issues relevant to adolescents and young adults with cancer. [References]: Journal of Adolescent and Young Adult Oncology. Vol.7(4), 2018, pp. 453-464.; 2018. 384. Taylor J, Murphy S, Chambers L, Aldridge J. Consulting with young people: informing guidelines for children's palliative care. Archives of Disease in Childhood. 2021;106(7):693-7. 385. Barrera M, D'Agostino N, Gammon J, Spencer L, Baruchel S. Health-related quality of life and enrollment in phase 1 trials in children with incurable cancer. Palliative & Supportive Care. 2005;3(3):191-6. 386. Cataudella DA, Zelcer S. Psychological experiences of children with brain tumors at end of life: parental perspectives. J Palliat Med. 2012;15(11):1191-7. 387. Flavelle SC. Experience of an adolescent living with and dying of cancer. Archives of Pediatrics & Adolescent Medicine. 2011;165(1):28-32. 388. Garland R, Ahmed S, Loiselle CG. Taking control over our health: Empowerment as perceived by young adults living with advanced cancer. Canadian Oncology Nursing Journal. 2020;30(2):93-112. 389. Gibson BE, Mistry B, Smith B, Yoshida KK, Abbott D, Lindsay S, et al. Becoming men: Gender, disability, and transitioning to adulthood. Health. 2014;18(1):95-114. 390. Knox MK, Hales S, Nissim R, Jung J, Lo C, Zimmermann C, et al. Lost and stranded: the experience of younger adults with advanced cancer. Supportive Care in Cancer. 2017;25(2):399-407. 391. Nicholas DB, Picone G, Selkirk EK. The Lived Experiences of Children and Adolescents With End-Stage Renal Disease. Qualitative Health Research. 2011;21(2):162-73. 392. Ananth P, Mun S, Reffat N, Li R, Sedghi T, Avery M, et al. A Stakeholder-Driven Qualitative Study to Define High Quality End-of-Life Care for Children With Cancer. Journal of Pain & Symptom Management. 2021;62(3):492-502. 393. Elana E Evan* ECB, Jolene Tan BS and Lonnie K Zeltzer. A Qualitative Approach to Understanding Quality of Life in Pediatric Palliative Care. Pediatric Palliative Care J Palliative Care Med S1:005. 2012. 394. Mack JW, Fisher L, Kushi L, Chao CR, Vega B, Rodrigues G, et al. Patient, Family, and Clinician Perspectives on End-of-Life Care Quality Domains and Candidate Indicators for Adolescents and Young Adults With Cancer. JAMA Network Open. 2021;4(8):e2121888. 614 395. Barling JA, Stevens JA, Davies KM. The reality of hospitalisation: Stories from family members of their hospital experience for adolescents and young adults living with and dying from cancer. [References]: Contemporary Nurse. Vol.46(2), 2014, pp. 150-160.; 2014. 396. Patterson P, Millar B, Desille N, McDonald F. The unmet needs of emerging adults with a cancer diagnosis: A qualitative study. [References]: Cancer Nursing. Vol.35(3), 2012, pp. e32-e40.; 2012. 397. Chong PH, Walshe C, Hughes S. Perceptions of a Good Death in Children with Life-Shortening Conditions: An Integrative Review. J Palliat Med. 2019;22(6):714-23. 398. Ciobanu E, Preston N. Hearing the voices of children diagnosed with a life-threatening or life-limiting illness and their parents' accounts in a palliative care setting: A qualitative study. Palliative Medicine. 2021;35(5):886-92. 399. Ito Y, Okuyama T, Ito Y, Kamei M, Nakaguchi T, Sugano K, et al. Good death for children with cancer: a qualitative study. Japanese Journal of Clinical Oncology. 2015;45(4):349-55. 400. Gaab EM, Owens RG, MacLeod RD. The voices of young New Zealanders involved in pediatric palliative care. [References]: Journal of Palliative Care. Vol.29(3), 2013, pp. 186-192.; 2013. 401. Chien CH, Chang YY, Huang XY. The Lived Experiences of Adolescents With Cancer: A Phenomenological Study. Clin Nurs Res. 2020;29(4):217-25. 402. Ananth P, Mun S, Reffat N, Li R, Sedghi T, Bilodeau M, et al. What Defines High Quality End-of-Life Care for Children with Cancer? A Stakeholder-Driven Qualitative Study (GP706). Journal of Pain & Symptom Management. 2020;60(1):249-. 403. Barling JA, Stevens JA, Davies KM. The reality of hospitalisation: Stories from family members of their hospital experience for adolescents and young adults living with and dying from cancer. Contemporary Nurse: A Journal for the Australian Nursing Profession. 2014;46(2):150-60. 404. Helter TM, Coast J, Łaszewska A, Stamm T, Simon J. Capability instruments in economic evaluations of health-related interventions: a comparative review of the literature. Qual Life Res. 2020;29(6):1433-64. 405. Proud L, McLoughlin C, Kinghorn P. ICECAP-O, the current state of play: a systematic review of studies reporting the psychometric properties and use of the instrument over the decade since its publication. Qual Life Res. 2019;28(6):1429-39. 406. Pickles K, Lancsar E, Seymour J, Parkin D, Donaldson C, Carter SM. Accounts from developers of generic health state utility instruments explain why they produce different QALYs: A qualitative study. Social Science & Medicine. 2019;240:112560. 615 407. Coast J, Al-Janabi H, Sutton EJ, Horrocks SA, Vosper AJ, Swancutt DR, et al. Using qualitative methods for attribute development for discrete choice experiments: issues and recommendations. Health Economics. 2012;21(6):730-41. 408. Ratcliffe J, Cameron I, Lancsar E, Walker R, Milte R, Hutchinson CL, et al. Developing a new quality of life instrument with older people for economic evaluation in aged care: study protocol. BMJ Open. 2019;9(5):e028647. 409. Stevens K. Assessing the performance of a new generic measure of health-related quality of life for children and refining it for use in health state valuation. Appl Health Econ Health Policy. 2011;9(3):157-69. 410. Oluboyede Y, Hulme C, Hill A. Development and refinement of the WAItE: a new obesity-specific quality of life measure for adolescents. Qual Life Res. 2017;26(8):2025-39. 411. Davis E, Shelly A, Waters E, Mackinnon A, Reddihough D, Boyd R, et al. Quality of life of adolescents with cerebral palsy: perspectives of adolescents and parents. Dev Med Child Neurol. 2009;51(3):193-9. 412. Hollin IL, Craig BM, Coast J, Beusterien K, Vass C, DiSantostefano R, et al. Reporting Formative Qualitative Research to Support the Development of Quantitative Preference Study Protocols and Corresponding Survey Instruments: Guidelines for Authors and Reviewers. Patient. 2020;13(1):121-36. 413. Lasch KE, Marquis P, Vigneux M, Abetz L, Arnould B, Bayliss M, et al. PRO development: rigorous qualitative research as the crucial foundation. Qual Life Res. 2010;19(8):1087-96. 414. Coast AO-SaJ. Understanding sampling and recruitment. In: Coast J, editor. Qualitative methods for heatlh economics: Rowman & Littlefield Publishers; 2017. 415. Flick U. An introduction to qualitative research. Edition 5. ed. Los Angeles: Sage; 2014. 416. Patton MQ. Qualitative research & evaluation methods : integrating theory and practice. Fourth edition. ed. Thousand Oaks, California: SAGE Publications, Inc.; 2015. 417. Bonevski B, Randell M, Paul C, Chapman K, Twyman L, Bryant J, et al. Reaching the hard-to-reach: a systematic review of strategies for improving health and medical research with socially disadvantaged groups. BMC medical research methodology. 2014;14:42-. 418. The SAGE Handbook of Qualitative Data Collection. 2018 2020/11/13. London: SAGE Publications Ltd. Available from: https://methods.sagepub.com/book/the-sage-handbook-of-qualitative-data-collection. 419. Grewal I, Lewis J, Flynn T, Brown J, Bond J, Coast J. Developing attributes for a generic quality of life measure for older people: preferences or capabilities? Soc Sci Med. 2006;62(8):1891-901. 616 420. Coast J. Qualitative Methods for Health Economics. London, UNITED KINGDOM: Rowman & Littlefield International; 2017. 421. Ritchie J, Lewis J, McNaughton Nicholls C, Ormston R, McNaughton Nicholls C, Ormston R. Qualitative research practice : a guide for social science students and researchers. Second edition. ed. Los Angeles ;: SAGE; 2014. 422. Glaser BG, Strauss AL. The discovery of grounded theory : strategies for qualitative research. London: Weidenfeld and Nicolson; 1968. 423. Malterud K, Siersma VD, Guassora AD. Sample Size in Qualitative Interview Studies:Guided by Information Power. Qualitative Health Research. 2016;26(13):1753-60. 424. Sutton EJ, Coast J. Development of a supportive care measure for economic evaluation of end-of-life care using qualitative methods. Palliative Medicine. 2014;28(2):151-7. 425. Silverman D. Doing qualitative research. Fifth edition. ed. Los Angeles: SAGE Publications; 2017. 426. Coast AO-SaJ. Understanding data collection: Interviews, focus groups and observation. Qualitative methods for health economics: Rowman & Littlefield Publishers; 2017. 427. Jackson JCaL. Understanding primary data analysis. In: Coast J, editor. Qualitative Methods for Health Economics: Rowman & Littlefield Publishers; 2017. 428. Strauss A, Corbin JM. Basics of qualitative research: Grounded theory procedures and techniques. Thousand Oaks, CA, US: Sage Publications, Inc; 1990. 270- p. 429. Gustafsson S, Hörder H, Hammar IO, Skoog I. Face and content validity and acceptability of the Swedish ICECAP-O capability measure: Cognitive interviews with 70-year-old persons. Health Psychol Res. 2018;6(1):6496-. 430. Terwee CB, Prinsen CAC, Chiarotto A, Westerman MJ, Patrick DL, Alonso J, et al. COSMIN methodology for evaluating the content validity of patient-reported outcome measures: a Delphi study. Qual Life Res. 2018;27(5):1159-70. 431. Brod M, Tesler LE, Christensen TL. Qualitative research and content validity: developing best practices based on science and experience. Qual Life Res. 2009;18(9):1263-78. 432. Reeve BB, Wyrwich KW, Wu AW, Velikova G, Terwee CB, Snyder CF, et al. ISOQOL recommends minimum standards for patient-reported outcome measures used in patient-centered outcomes and comparative effectiveness research. Qual Life Res. 2013;22(8):1889-905. 617 433. Magasi S, Ryan G, Revicki D, Lenderking W, Hays RD, Brod M, et al. Content validity of patient-reported outcome measures: perspectives from a PROMIS meeting. Qual Life Res. 2012;21(5):739-46. 434. Collins D. Pretesting survey instruments: an overview of cognitive methods. Qual Life Res. 2003;12(3):229-38. 435. Nwankwo H, Coast J, Hewison A, Kinghorn P, Madathil S, Bailey C. A think-aloud study of the feasibility of patients with end-stage organ failure completing the ICECAP-SCM. Palliat Med. 2022;36(10):1559-69. 436. Ahmed N, Bestall JC, Payne SA, Noble B, Ahmedzai SH. The use of cognitive interviewing methodology in the design and testing of a screening tool for supportive and palliative care needs. Supportive Care in Cancer. 2009;17(6):665-73. 437. Bailey C, Kinghorn, P., Orlando, R., & Coast, J. Using 'think-aloud' and interview data to explore patient and proxy completion of health and capability measures at the end of life. . In: Coast J, editor. Qualitative methods for health economics Rowman & Littlefield; 2017. p. 231-44. 438. Schildmann EK, Groeneveld EI, Denzel J, Brown A, Bernhardt F, Bailey K, et al. Discovering the hidden benefits of cognitive interviewing in two languages: The first phase of a validation study of the Integrated Palliative care Outcome Scale. Palliat Med. 2016;30(6):599-610. 439. van Leeuwen KM, Jansen APD, Muntinga ME, Bosmans JE, Westerman MJ, van Tulder MW, et al. Exploration of the content validity and feasibility of the EQ-5D-3L, ICECAP-O and ASCOT in older adults. BMC Health Services Research. 2015;15(1):201. 440. Luckett T, Virdun C, Rao A, Daly L, Hackl N, Willems A, et al. Improving the methods for patient-reported experience measures in palliative care: findings from a cognitive interview study. Annals of Palliative Medicine. 2022;11(7):2275-84. 441. Murtagh FE, Addington-Hall JM, Higginson IJ. The value of cognitive interviewing techniques in palliative care research. Palliat Med. 2007;21(2):87-93. 442. Bowden A, Fox-Rushby J, Nyandieka L, Wanjau J. Methods for pre-testing and piloting survey questions: illustrations from the KENQOL survey of health-related quality of life. Health Policy and Planning. 2002;17(3):322-30. 443. Bowden A, Fox-Rushby JA, Nyandieka L, Wanjau J. Methods for pre-testing and piloting survey questions: illustrations from the KENQOL survey of health-related quality of life. Health Policy Plan. 2002;17(3):322-30. 444. Greco G, Skordis-Worrall J, Mills A. Development, Validity, and Reliability of the Women's Capabilities Index. J Human Dev Capabil. 2018;19(3):271-88. 618 445. Al-Janabi H, Keeley T, Mitchell P, Coast J. Can capabilities be self-reported? A think aloud study. Social science & medicine (1982). 2013;87:116-22. 446. Mallinson S. Listening to respondents: a qualitative assessment of the Short-Form 36 Health Status Questionnaire. Soc Sci Med. 2002;54(1):11-21. 447. Willis G, editor Cognitive Interviewing: A Tool for Improving Questionnaire Design2004. 448. Tourangeau R, Rips LJ, Rasinski K. The psychology of survey response. 2000. 449. Lee RM. Doing Research on Sensitive Topics: Sage Publications (CA); 1993. 450. Dempsey L, Dowling M, Larkin P, Murphy K. Sensitive Interviewing in Qualitative Research. Research in Nursing & Health. 2016;39(6):480-90. 451. Chen EK, Riffin C, Reid MC, Adelman R, Warmington M, Mehta SS, et al. Why is high-quality research on palliative care so hard to do? Barriers to improved research from a survey of palliative care researchers. J Palliat Med. 2014;17(7):782-7. 452. Butler AE, Hall H, Copnell B. Bereaved parents’ experiences of research participation. BMC Palliat Care. 2018;17(1):122. 453. Hynson JL, Aroni R, Bauld C, Sawyer SM. Research with bereaved parents: A question of how not why. [References]: Palliative Medicine. Vol.20(8), 2006, pp. 805-811.; 2006. 454. East L, Jackson D, O’Brien L, Peters K. The benefits of computer-mediated communication in nursing research. Contemporary Nurse. 2008;30(1):83-8. 455. Ison NL. Having their say: email interviews for research data collection with people who have verbal communication impairment. International Journal of Social Research Methodology. 2009;12(2):161-72. 456. Hynson JL, Aroni R, Bauld C, Sawyer SM. Research with bereaved parents: a question of how not why. Palliative Medicine. 2006;20(8):805-11. 457. Elmir R, Schmied V, Jackson D, Wilkes L. Interviewing people about potentially sensitive topics. Nurse Res. 2011;19(1):12-6. 458. Mitchell W. 'I know how I feel': Listening to young people with life-limiting conditions who have learning and communication impairments. [References]: Qualitative Social Work: Research and Practice. Vol.9(2), 2010, pp. 185-203.; 2010. 459. Mitchell W, Sloper P. Making choices in my life: Listening to the ideas and experiences of young people in the UK who communicate non-verbally. Children and Youth Services Review. 2011;33(4):521-7. 619 460. Moran-Ellis J. Reflections on the Sociology of Childhood in the UK. Current Sociology. 2010;58(2):186-205. 461. Franklin A, Sloper P. Supporting the Participation of Disabled Children and Young People in Decision-making. Children & Society. 2009;23(1):3-15. 462. Hill M, Davis J, Prout A, Tisdall K. Moving the participation agenda forward. Children & Society. 2004;18(2):77-96. 463. Mitchell W, Sloper P. Making choices in my life: Listening to the ideas and experiences of young people in the UK who communicate non-verbally. [References]: Children and Youth Services Review. Vol.33(4), 2011, pp. 521-527.; 2011. 464. Council EaSR. Research with children and young people 2020 [Available from: https://esrc.ukri.org/funding/guidance-for-applicants/research-ethics/frequently-raised-topics/research-with-children-and-young-people/. 465. UKRI. Research with potentially vulnerable people: UKRI; 2022 [Available from: https://www.ukri.org/councils/esrc/guidance-for-applicants/research-ethics-guidance/research-with-potentially-vulnerable-people/. 466. The Ethics of Research with Children and Young People: A Practical Handbook. London2011. Available from: https://methods.sagepub.com/book/ethics-of-research-with-children-and-young-people. 467. ER. GvWNaWAA. GILLICK RESPONDENT AND WEST NORFOLK AND WISBECH AREA HEALTH AUTHORITY FIRST APPELLANTS AND DEPARTMENT OF HEALTH AND SOCIAL SECURITY SECOND APPELLANTS. WEST NORFOLK AND WISBECH AREA1985. 468. Council E. Clinical Trials Directive: European Council; 2001 [Available from: https://ec.europa.eu/health/human-use/clinical-trials/directive_en. 469. Ross LF. Children in medical research : access versus protection. Oxford New York: Clarendon Press ; Oxford University Press; 2006. Available from: http://site.ebrary.com/id/10329698 https://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=157243 http://www.myilibrary.com?id=116453 http://rave.ohiolink.edu/ebooks/ebc/0199273286 http://www.oxfordscholarship.com/oso/public/content/philosophy/0199273286/toc.html 620 https://doi.org/10.1093/0199273286.001.0001 http://www.library.yorku.ca/e/resolver/id/2389598 http://www.oxfordscholarship.com/oso/public/content/philosophy/9780199273287/toc.html https://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=679563 http://www.library.yorku.ca/e/resolver/id/2389597 http://0-doi.org.pugwash.lib.warwick.ac.uk/10.1093/0199273286.001.0001 http://www.library.yorku.ca/eresolver/?id=991576 http://www.library.yorku.ca/e/resolver/id/1865230. 470. HRA N. Planning and improving research: Policies, Standards & Legislation: Research involving children: NHS Health Research Authority 2021 [Available from: https://www.hra.nhs.uk/planning-and-improving-research/policies-standards-legislation/research-involving-children/. 471. Waligora M, Dranseika V, Piasecki J. Child’s assent in research: Age threshold or personalisation? BMC Medical Ethics. 2014;15(1):44. 472. Mitchell C. 'Margaret's' children remember. Journal of Clinical Ethics. 2006;17(4):349-57. 473. Whale K, editor The use of Skype and telephone interviews in sensitive qualitative research with young people: experiences from the ROCCA continence study2017. 474. Fargas-Malet M, McSherry D, Larkin E, Robinson C. research with children: methodological issues and innovative techniques. Journal of Early Childhood Research. 2010;8(2):175-92. 475. Molloy EJ, Mader S, Modi N, Gale C. Parent, child and public involvement in child health research: core value not just an optional extra. Pediatric Research. 2019;85(1):2-3. 476. Elam G, Fenton KA. Researching sensitive issues and ethnicity: lessons from sexual health. Ethn Health. 2003;8(1):15-27. 477. Biddle L, Cooper J, Owen-Smith A, Klineberg E, Bennewith O, Hawton K, et al. Qualitative interviewing with vulnerable populations: Individuals’ experiences of participating in suicide and self-harm based research. Journal of Affective Disorders. 2013;145(3):356-62. 478. Dickson-Swift V, James EL, Kippen S, Liamputtong P. Doing sensitive research: what challenges do qualitative researchers face? Qualitative Research. 2007;7(3):327-53. 621 479. Booth T, Booth W. The use of depth interviewing with vulnerable subjects: Lessons from a research study of parents with learning difficulties. Social Science & Medicine. 1994;39(3):415-24. 480. Oakley A. Interviewing Women Again: Power, Time and the Gift. Sociology. 2016;50(1):195-213. 481. Reinharz S, Davidman L. Feminist methods in social research. New York, NY, US: Oxford University Press; 1992. viii, 413-viii, p. 482. Porter S, Birt AR. Is traumatic memory special ? A comparison of traumatic memory characteristics with memory for other emotional life experiences. Applied Cognitive Psychology. 2001;15(7):S101-S17. 483. Doody O, Noonan M. Preparing and conducting interviews to collect data. Nurse Res. 2013;20(5):28-32. 484. Carter B, Ford K. Researching children's health experiences: The place for participatory, child-centered, arts-based approaches. Research in nursing & health. 2013;36 1:95-107. 485. Ronen GM, Rosenbaum P, Law M, Streiner DL. Health-related quality of life in childhood epilepsy: the results of children's participation in identifying the components. Developmental Medicine & Child Neurology. 1999;41(8):554-9. 486. Morris C, Liabo K, Wright P, Fitzpatrick R. Development of the Oxford ankle foot questionnaire: finding out how children are affected by foot and ankle problems. Child Care Health Dev. 2007;33(5):559-68. 487. Markham C, van Laar D, Gibbard D, Dean T. Children with speech, language and communication needs: their perceptions of their quality of life. Int J Lang Commun Disord. 2009;44(5):748-68. 488. Cameron L, Murphy J. Enabling young people with a learning disability to make choices at a time of transition. British Journal of Learning Disabilities. 2002;30(3):105-12. 489. Arbuckle R, Abetz-Webb L. "Not just little adults": qualitative methods to support the development of pediatric patient-reported outcomes. Patient. 2013;6(3):143-59. 490. Canaway Aea. Incorporating novel qualitative methods within health economics: The use of pictorial tools. In: Coast J, editor. Qualitative Methods for Health Economics: Rowman & Littlefield Publishers; 2017. 491. Collecting Qualitative Data: A Practical Guide to Textual, Media and Virtual Techniques. Cambridge: Cambridge University Press; 2017. 622 492. Mann C, Stewart F. Internet communication and qualitative research : a handbook for researching online. London ;: Sage Publications; 2000. 493. Dodds S, Hess AC. Adapting research methodology during COVID-19: lessons for transformative service research. Journal of Service Management. 2020;ahead-of-print(ahead-of-print). 494. Cardel MI, Manasse S, Krukowski RA, Ross K, Shakour R, Miller DR, et al. COVID-19 Impacts Mental Health Outcomes and Ability/Desire to Participate in Research Among Current Research Participants. Obesity (Silver Spring, Md). 2020;28(12):2272-81. 495. Barclay. SGGaK. ADAPTING RESEARCH METHODOLOGIES IN THE COVID-19 PANDEMIC: Resources for researchers. University of Washington: Nippon Foundation Ocean Nexus at EarthLab; 2020. 496. Clarke V, Kitzinger C. Lesbian and gay parents on talk shows: resistance or collusion in heterosexism? Qualitative Research in Psychology. 2004;1(3):195-217. 497. Suzuki LK, Beale IL. Personal Web Home Pages of Adolescents With Cancer: Self-Presentation, Information Dissemination, and Interpersonal Connection. Journal of Pediatric Oncology Nursing. 2006;23(3):152-61. 498. Keim-Malpass J, Adelstein K, Kavalieratos D. Legacy Making Through Illness Blogs: Online Spaces for Young Adults Approaching the End-of-Life. Journal of Adolescent and Young Adult Oncology. 2015;4(4):209-12. 499. Thompson CM, Crook B, Love B, Macpherson CF, Johnson R. Understanding how adolescents and young adults with cancer talk about needs in online and face-to-face support groups. Journal of Health Psychology. 2016;21(11):2636-46. 500. Jowett A. A Case for Using Online Discussion Forums in Critical Psychological Research. Qualitative Research in Psychology. 2015;12(3):287-97. 501. Society BP. Ethics Guidelines for Internet-mediated Research. Leicester: British Psychological Society; 2017. 502. Heilferty CM. Ethical considerations in the study of online illness narratives: a qualitative review. J Adv Nurs. 2011;67(5):945-53. 503. Keim-Malpass J, Steeves RH, Kennedy C. Internet ethnography: a review of methodological considerations for studying online illness blogs. Int J Nurs Stud. 2014;51(12):1686-92. 504. Campbell R, Pound P, Morgan M, Daker-White G, Britten N, Pill R, et al. Evaluating meta-ethnography: systematic analysis and synthesis of qualitative research. Health technology assessment (Winchester, England). 2011;15:1-164. 623 505. Dixon-Woods M, Agarwal S, Jones D, Young B, Sutton A. Synthesising qualitative and quantitative evidence: a review of possible methods. J Health Serv Res Policy. 2005;10(1):45-53. 506. Elaine B-P, James T. Methods for the synthesis of qualitative research: a critical review. BMC Medical Research Methodology [Internet]. 2009; 9(1):[59 p.]. 507. Hannes K, Lockwood C. Synthesizing qualitative research : choosing the right approach. Chichester: John Wiley & Sons; 2012. Available from: http://site.ebrary.com/id/10504260 https://search.ebscohost.com/login.aspx?direct=true&scope=site&db=nlebk&db=nlabk&AN=399470 http://www.myilibrary.com?id=329532 https://www.overdrive.com/search?q=BF90A3F8-D872-48CC-9CE9-0E3B765083DC https://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=819198 https://public.ebookcentral.proquest.com/choice/publicfullrecord.aspx?p=4043291 https://doi.org/10.1002/9781119959847 https://onlinelibrary.wiley.com/doi/book/10.1002/9781119959847 http://onlinelibrary.wiley.com/book/10.1002/9781119959847 http://catalogimages.wiley.com/images/db/jimages/9780470656389.jpg http://ovidsp.ovid.com/ovidweb.cgi?T=JS&NEWS=n&CSC=Y&PAGE=booktext&D=books&AN=01745884$&XPATH=/PG(0)&EPUB=Y http://VH7QX3XE2P.search.serialssolutions.com/?V=1.0&L=VH7QX3XE2P&S=JCs&C=TC0000541730&T=marc&tab=BOOKS http://www.myilibrary.com?id=329532&ref=toc http://excerpts.cdn.overdrive.com/FormatType-410/0128-1/BF9/0A3/F8/SynthesisingQualitativeResearchChoosi9781119959823.epub https://samples.overdrive.com/?crid=BF90A3F8-D872-48CC-9CE9-0E3B765083DC&.epub-sample.overdrive.com http://images.contentreserve.com/ImageType-100/0128-1/{BF90A3F8-D872-48CC-9CE9-0E3B765083DC}Img100.jpg. 624 508. Al-Janabi H, Coast J, Flynn TN. What do people value when they provide unpaid care for an older person? A meta-ethnography with interview follow up. 2008;67:111-21. 509. Ingleton C, Morgan J, Hughes P, Noble B, Evans A, Clark D. Carer satisfaction with end-of-life care in Powys, Wales: a cross-sectional survey. Health & Social Care in the Community. 2004;12(1):43-52. 510. Butler AE, Hall H, Copnell B. Bereaved parents' experiences of research participation. BMC Palliat Care. 2018;17(1):122. 511. Hewitt-Taylor J. Use of constant comparative analysis in qualitative research. Nurs Stand. 2001;15(42):39-42. 512. Bowen GA. Naturalistic inquiry and the saturation concept: a research note. Qualitative Research. 2008;8(1):137-52. 513. Mitchell PM, Caskey FJ, Scott J, Sanghera S, Coast J. Response process validity of three patient reported outcome measures for people requiring kidney care: a think-aloud study using the EQ-5D-5L, ICECAP-A and ICECAP-O. BMJ Open. 2020;10(5):e034569. 514. Turner N, Almack K. Recruiting young people to sensitive research: turning the ‘wheels within wheels’. International Journal of Social Research Methodology. 2017;20(5):485-97. 515. Sanden S. A Review of “Researching Young Children's Perspectives: Debating the Ethics and Dilemmas of Educational Research with Children”. The Journal of Educational Research. 2012;105(3):232-3. 516. Kinghorn JCaCBaACaP. Measuring and Valuing Outcomes for Care at the End of Life: The Capability Approach. In: Round J, editor. Care at the End of Life: An Economic Perspective: Springer; 2016. 517. Mullen PM. Public involvement in health care priority setting: an overview of methods for eliciting values. Health Expect. 1999;2(4):222-34. 518. Ryan M, Scott DA, Reeves C, Bate A, van Teijlingen ER, Russell EM, et al. Eliciting public preferences for healthcare: a systematic review of techniques. Health Technol Assess. 2001;5(5):1-186. 519. Mulhern B, Norman R, De Abreu Lourenco R, Malley J, Street D, Viney R. Investigating the relative value of health and social care related quality of life using a discrete choice experiment. Social Science & Medicine. 2019;233:28-37. 520. Coast J, Bailey C, Orlando R, Armour K, Perry R, Jones L, et al. Adaptation, Acceptance and Adaptive Preferences in Health and Capability Well-Being Measurement Amongst Those Approaching End of Life. Patient. 2018;11(5):539-46. 625 521. Oort FJ, Visser MR, Sprangers MA. Formal definitions of measurement bias and explanation bias clarify measurement and conceptual perspectives on response shift. J Clin Epidemiol. 2009;62(11):1126-37. 522. Verma JP, Verma P. Understanding Concepts in Estimating Sample Size in Survey Studies. In: Verma JP, Verma P, editors. Determining Sample Size and Power in Research Studies: A Manual for Researchers. Singapore: Springer Singapore; 2020. p. 29-40. 523. Coast J, Peters TJ, Inglis A. Factors associated with inappropriate emergency hospital admission in the UK. Int J Qual Health Care. 1996;8(1):31-9. 524. Flynn TN, Chan P, Coast J, Peters TJ. Assessing quality of life among British older people using the ICEPOP CAPability (ICECAP-O) measure. Appl Health Econ Health Policy. 2011;9(5):317-29. 525. Sherriff A, Peters TJ, Henderson J, Strachan D, Team aTAS. Risk factor associations with wheezing patterns in children followed longitudinally from birth to 3 ½ years. International Journal of Epidemiology. 2001;30(6):1473-84. 526. Peters TJ, Somerset M, Campbell R, Sharp DJ. Variables associated with attendance at, and the perceived helpfulness of, meetings for people with multiple sclerosis. Health Soc Care Community. 2003;11(1):19-26. 527. Braun V, Clarke V. Using thematic analysis in psychology. Qualitative Research in Psychology. 2006;3(2):77-101. 528. Patterson P, Millar B, Desille N, McDonald F. The unmet needs of emerging adults with a cancer diagnosis: a qualitative study. Cancer Nursing. 2012;35(3):E32-40. 529. Gaab EM, Owens RG, MacLeod RD. The voices of young New Zealanders involved in pediatric palliative care. Journal of Palliative Care. 2013;29(3):186-92. 530. Alkire S. The Capability Approach and Well-Being Measurement for Public Policy. Oxford Poverty & Human Development Initiative (OPHI) Oxford Department of International Development. 2015(OPHI WORKING PAPER NO. 94). 531. Price J, Hurley F, Kiernan G. ‘Managing an unexpected life - a caregiver’s career’: Parents’ experience of caring for their child with a non-malignant life-limiting condition. Journal of Child Health Care. 2022;0(0):13674935221132920. 532. Fisher V, Fraser L, Taylor J. Experiences of fathers of children with a life-limiting condition: a systematic review and qualitative synthesis. BMJ Supportive & Palliative Care. 2023;13(1):15-26. 626 533. Varni JW, Seid M, Rode CA. The PedsQL™: Measurement Model for the Pediatric Quality of Life Inventory. Medical Care. 1999;37(2):126-39. 534. Coast SHPMPKSBKBCBPATPIFJ. A qualitative study to inform the development of icecap capability wellbeing measures for children and young people aged 11-15. Health Economists’ Study Group (HESG); Oxford2023. 535. Hart CS. The capability approach and educational research. In: Caroline Sarojini Hart MB, Bernhard Babic editor. Agency and participation in childhood and youth: International applications of the capability approach in schools and beyond: Bloomsbury Publishing. p. 17-43. 536. Draucker CB, Martsolf DS, Poole C. Developing distress protocols for research on sensitive topics. Arch Psychiatr Nurs. 2009;23(5):343-50. 537. Fox D, Holder J, Netten A. Personal Social Services survey of adult carers in England: 2009–10: survey development project. Canterbury: PSSRU, University of Kent; 2010. 538. Rand S, Malley J, Vadean F, Forder J. Measuring the outcomes of long-term care for unpaid carers: comparing the ASCOT-Carer, Carer Experience Scale and EQ-5D-3 L. Health and Quality of Life Outcomes. 2019;17(1):184. 539. Masha SZ, Lotje B, Yvonne NB, Erica W, Agnes van der H, Liza GGvL, et al. Impact of COVID-19 on care at the end of life during the first months of the pandemic from the perspective of healthcare professionals from different settings: a qualitative interview study (the CO-LIVE study). BMJ Open. 2023;13(3):e063267. 540. Middleton W, Raphael B, Burnett P, Martinek N. A longitudinal study comparing bereavement phenomena in recently bereaved spouses, adult children and parents. Aust N Z J Psychiatry. 1998;32(2):235-41. 541. Sanders CM. A Comparison of Adult Bereavement in the Death of a Spouse, Child, and Parent. OMEGA - Journal of Death and Dying. 1980;10(4):303-22. 542. McIlfatrick S, Slater P, Beck E, Bamidele O, McCloskey S, Carr K, et al. Examining public knowledge, attitudes and perceptions towards palliative care: a mixed method sequential study. BMC Palliat Care. 2021;20(1):44. 543. McIlfatrick S, Hasson F, McLaughlin D, Johnston G, Roulston A, Rutherford L, et al. Public awareness and attitudes toward palliative care in Northern Ireland. BMC Palliat Care. 2013;12(1):34. 544. Wallace J. Public awareness of palliative care: Report of the findings of the first national survey in Scotland into public knowledge and understanding of palliative care. Scottish Partnership for Palliative Care; 2003. 627 545. Cho EY-N, Yu F-Y. A review of measurement tools for child wellbeing. Children and Youth Services Review. 2020;119:105576. 546. Harris N, Blackburn M, Noyes J, Aldridge J, Lapwood S, Dunbar H, et al. Undertaking doctoral research with children and young people with life-limiting or life-threatening conditions. J Adv Nurs. 2019;75(12):3183-5. 547. Engler-Stringer R, Schaefer J, Ridalls T, Muhajarine N. Lessons Learned From a Food Environment Intervention Study: Recruitment and Retention of Participants in Disadvantaged Urban Inner-City Neighborhoods. Health Educ Behav. 2018;45(4):473-9. 548. Ejiogu N, Norbeck JH, Mason MA, Cromwell BC, Zonderman AB, Evans MK. Recruitment and retention strategies for minority or poor clinical research participants: lessons from the Healthy Aging in Neighborhoods of Diversity across the Life Span study. Gerontologist. 2011;51 Suppl 1(Suppl 1):S33-45. 549. Macdonald ME, Chilibeck G, Affleck W, Cadell S. Gender imbalance in pediatric palliative care research samples. Palliative Medicine. 2010;24(4):435-44. 550. Simmons WO, Emanuele R. Male‐female giving differentials: are women more altruistic? Journal of Economic Studies. 2007;34(6):534-50. 551. Zipp JF, Toth J. She said, he said, they said: The impact of spousal presence in survey research. Public opinion quarterly. 2002;66(2):177-208. 552. Jane Ritchie JL. Qualitative Research Practice: A Guide for Social Science Students and Researchers. second ed. London SAGE 2013. 553. Kalton G, Roberts J, Holt D. The Effects of Offering a Middle Response Option with Opinion Questions. Journal of the Royal Statistical Society Series D (The Statistician). 1980;29(1):65-78. 554. Whynes DK, Philips Z, Frew E. Think of a number… any number? Health Economics. 2005;14(11):1191-5. 555. Nancy Devlin DP, Bas Janssen. Analysis of EQ VAS Data Methods for Analysing and Reporting EQ-5D Data: Springer 2020. 556. Together for Short Lives E. Policy and influencing: Affordable and accessible childcare 2015 [Available from: https://www.togetherforshortlives.org.uk/changing-lives/speaking-up-for-children/policy-advocacy/access-to-high-quality-childcare/. 557. England N. NHS long term plan: Chapter 1: A new service model for the 21st century: 5. Local NHS organisations will increasingly focus on population health – moving to Integrated Care Systems everywhere. 2019. 628 558. Excellence NIfC. Guide to the methods of technology appraisal 2013. NICE; 2013. 559. Commission CQ. Experts by Experience: Care Quality Commission 2023 [Available from: https://www.cqc.org.uk/about-us/jobs/experts-experience. 560. Dalziel K, Catchpool M, García-Lorenzo B, Gorostiza I, Norman R, Rivero-Arias O. Feasibility, Validity and Differences in Adolescent and Adult EQ-5D-Y Health State Valuation in Australia and Spain: An Application of Best-Worst Scaling. Pharmacoeconomics. 2020;38(5):499-513. 561. Elder Jr GH. Time, human agency, and social change: Perspectives on the life course. Social psychology quarterly. 1994:4-15. 562. Dixon-Woods M, Bonas S, Booth A, Jones DR, Miller T, Sutton AJ, et al. How can systematic reviews incorporate qualitative research? A critical perspective. Qualitative Research. 2006;6(1):27-44. 563. ISSG Search Filter Resource York (UK) The InterTASC Information Specialists' Sub-Group; [updated updated 20 Feb 2023; cited 2023 07/03/23]. 2006:[Available from: https://sites.google.com/a/york.ac.uk/issg-search-filters-resource/home 564. Conditions AfCwL-toT, Families t, Baum D. A guide to the development of children's palliative care services: Association for Children with Life-threatening or Terminal Conditions and …; 1997. 565. Sackett DL, Wennberg JE. Choosing the best research design for each question. It's time to stop squabbling over the “best” methods. 1997;315(7123):1636. 566. World data bank: High income: World data bank [cited 2023. Available from: https://data.worldbank.org/income-level/high-income. 567. Sasaki H, Bouesseau M-C, Marston J, Mori R. A scoping review of palliative care for children in low- and middle-income countries. BMC Palliat Care. 2017;16(1):60. 568. Cochrane Handbook for Systematic Reviews of Interventions version 6.3: Cochrane; 2022 [Available from: www.training.cochrane.org/handbook. 569. Skills CA, Programme. CASP Qualitative Checklist. 2018. 570. Majid U, Vanstone M. Appraising Qualitative Research for Evidence Syntheses: A Compendium of Quality Appraisal Tools. Qual Health Res. 2018;28(13):2115-31. 571. Patton MQ. Qualitative Research & Evaluation Methods: SAGE Publications; 2002. 629 572. H. M. Selection Procedures, Sampling, Case Construction. . In: Flick U vKE, Steinke I, editors. , editor. A Companion to Qualitative Research London: Sage Publications2004. p. 165-71. 573. Kuper A, Lingard L, Levinson W. Critically appraising qualitative research. BMJ. 2008;337:a1035. |
|---|