Symptoms and Concerns Among Children and Young People with Life-Limiting and Life-Threatening Conditions: A Systematic Review Highlighting Meaningful Health Outcomes

The design and provision of quality pediatric palliative care should prioritize issues that matter to children and their families for optimal outcomes. This review aims to identify symptoms, concerns and outcomes that matter to children and young people (“young people”) with terminal illnesses and their families. Findings from the systematic review will inform the development of a relevant framework of health outcomes. This is a systematic literature review across multiple databases for identification of eligible primary evidence. Data sources included PsycINFO, MEDLINE, Embase, CINAHL, OpenGrey, and Science Direct Journals, searched from 1 August 2016 to 30 July 2017. The study also incorporates consultations with experts in the field, citation searches via Scopus, and a hand search of reference lists of included studies. Of the 13,567 articles that were evaluated, 81 studies were included. Most of these studies (n = 68) were from high-income countries and foused on young people with cancer (n = 58). A total of 3236 young people, 2103 family carers, 108 families, and 901 healthcare providers were included in the studies. Young people did not contribute to data in 30% of studies. Themes on priority concerns are presented by the following domains and health outcomes: (1) physical (n = 62 studies), e.g., physical symptoms; (2) psychological (n = 65), e.g., worry; (3) psychosocial (n = 31), e.g., relationships; (4) existential (n = 37), e.g., existential loss; and (5) “other” (n = 39), e.g., information access. Burdensome symptoms and concerns affect young people with malignant and nonmalignant conditions and occur across the disease trajectory; pediatric palliative care should not be limited to the end-of-life phase. A child–family-centered framework of health outcomes, spanning the patient, family, and quality of service levels is proposed to inform service development. Future research should address gaps identified across the literature (i.e., the involvement of young people in research, evidence for developing countries, and a focus on nonmalignant conditions.


Background
The number of children and young people ("young people") (aged 0-23 years) living with Life-limiting and Life-threatening Conditions (LLC) is increasing worldwide. [1][2][3] Of the 20.4 million people, annually who need palliative care towards the end of their life, approximately 6% are children. These data exclude factors such as the needs before the end of life and, thus, is an underestimation of the true needs of this group. Recent estimates have set the worldwide number of young people that need palliative care at any point during their disease trajectory at 21 million, with 8 million requiring some form of specialist palliative care. 4 The need for paediatric palliative care has been increasing over time; for example, in the UK, the prevalence of LLC in children increased from 25 to 32 per 10,000 population between 2000-2010. 5 In America, it is estimated that each year, approximately, 500,000 children are diagnosed with LLC and nearly 53,000 die from trauma, congenital conditions, extreme prematurity, and other acquired illness. 6 In low-and middle income countries, the HIV epidemic continues to pose a public health concern, and burden symptoms persist despite the advent of antiretroviral therapy. According to the Joint United Nations Programme, between 2.9-3.5 million children are living with HIV infection, with sub-Saharan Africa shouldering 91% of the global burden. 7 The situation is further exacerbated due to the increasing incidence and prevalence of various types of cancer among young people, with over 80% of deaths occurring in resource-limited settings. 8 This high mortality in resource-limited settings is largely attributed to health system challenges, such as late diagnosis, which limits curative treatment options, 9 and the poor coverage of supportive care services. 10 Besides HIV and cancer, complex chronic, neonatal, and other non-communicable diseases contribute to mortality and morbidity among the children in resource-limited settings. 11 The current delivery of palliative care provision for young people typically runs parallel to existing health care systems, without integration of the existing and speciality services. 4 The provision of quality care to young people with LLC requires the critical establishment of robust evidence on the symptoms and concerns that matter to the patients and their families. 12 Despite the need for this evidence, there is currently limited information on meaningful outcomes for young people with LLC. 12,13 This need for patient-level data is more pressing than ever amidst the worldwide demand to address the absence of person-centred outcome measures in the measurement of the quality of paediatric care. 13,14 Generating population-specific measures of Health-Related Quality of Life (HRQOL) is the key to developing palliative care for young people. Measurement of HRQOL will enable those developing and evaluating services to determine their effectiveness. 15 Furthermore, it can enable improvements in clinical care, research, and informed decision making. 16 Currently, there is no appropriate outcome assessment measure for use in paediatric palliative care. 15 In order to develop outcome assessment measures it is essential to understand the perspectives of the population in which they will be used. 17 At present, there is limited reporting on the needs and experiences of young people with LLC. In those cases where reviews of existing literature have been completed, they have been limited by inclusion of evidence from the North America alone 18 or focused solely on cancer and neuro-disability, neglecting other types of LLC. 18,19 Research with young people with LLC is hampered by several methodological challenges. These include clinical considerations such as participants being very ill, 20 limited access to potential participants, 21 and limited capacity to generate self-reports due to the less developed (or impaired) verbal and cognitive skills of this population. 22 As symptoms and concerns affect children's beliefs, expectations, and perceptions, it becomes important for self-reports from young people to be prioritized wherever possible. 23 The inclusion of the perspectives of young people and their families is critical to ensure that outcome measures are meaningful to them and their families. 24 Regrettably, the level of young people's involvement in research remains limited and self-reports of their outcomes and experiences are not commonly reported. 25 Therefore, innovative and feasible approaches for engagement of young people in research that will shape their care should be prioritized.
This review aims to appraise the global evidence on symptoms and concerns that matter to young people and their families in order to identify meaningful, core person-centred health outcomes in young people with LLC and their families. The objectives of this review are to: i) appraise the methodological quality and extent of research literature, detailing patient, caregiver, family, and health provider reports of symptoms and concerns across disease trajectories for young people living with LLC and their families; ii) identify the gaps that exist in the research literature (e.g. study design, countries, and conditions), and; iii) synthesise reports of symptoms and concerns, using a conceptual framework to identify the domains of importance in the development of outcome assessment measures for young people with LLC. These findings are utilised to discuss the implications for paediatric palliative care service development and outcome measurement.

Methods
This review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). 26 2.

Analysis 2.6 Data extraction
Data from studies that met the inclusion criteria were extracted into Microsoft Excel spreadsheets, which has been piloted by the lead reviewer templates (see Table 2)., Quotes to support reported themes and original author notes have been extracted for qualitative studies. The identified outcome measures have also been extracted into a pre-designed Microsoft Excel template and assessed for quality of measurement properties using the COSMIN checklist. 30

Data synthesis
The data has been synthesised using a systematic review and integrative design. 31,32 Both qualitative and quantitative narrative syntheses approaches have been used. Descriptive statistics have also been used to summarize the studies under selected subheadings such as country, setting, focus, diagnosis, characteristics of respondents, and the main themes identified. Descriptive themes, encompassing the themes or codes of the primary studies, have been developed with attention to similarities and differences across and between studies, and then grouped by phenomenon/themes. Disagreements have been resolved through discussions, guided by references to results, discussion, and conclusions sections of included studies and through consultation of content experts. Data has been assessed for contrasting themes within different diagnostic groups and by developmental age, grouped as follows: (0-5 years, 6-9, 10-14, 15+); it has been based on guidance on feasibility as informed by included studies, best practices 33,34 and expert guidance.
Subsequently, a conceptual framework of domains underlying the concept of health outcomes has been developed. A reference has been made to the WHO definition for paediatric palliative care and the core domains have been mapped out; physical, psychological, social, and spiritual/existential. 35 A fifth domain, "other", was adopted to accommodate any themes and sub-themes that did not seem to fit into the existing four domains. References have been made to WHO definitions for health, 36  3 Results

Study selection process
A total of 13,569 articles were identified, after eliminating the duplicates. These were assessed for eligibility, after which 81 have been included in this review; of these 81, 79 were original studies, published between 1996 and 2017 (see Figure 1).
[Insert Table 2 here] Most of the studies (n=58; 71.6%) involve cancer patients; of these, 46 studies provided details on types of cancer, with treatment status reported in all studies (Table 3).
[Insert Table 3 [20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38]. These studies are mainly descriptive, non-experimental, and cross-sectional, with some of them using convenience samples; one randomized controlled trial is included 40 (see Table 2 for details). The quality of included studies is poor in the following domains: ethics, bias, sampling, and limited articulation of the generalisability of the findings. In most instances, the discussion of ethics has been limited to seeking approval from ethics review boards; assent and consent processes have rarely been explained in detail. It is found that only two papers have mentioned the use of child-appropriate information sheets. 41 Extremely few studies have mentioned the consideration of respondent age in questionnaire development as a way of ensuring age-appropriateness of the questions. [41][42][43][44][45][46][47][48] The process surrounding data collection with young people has rarely been explained in detail. [Insert supplementary results appendix 1 link here] 3.5 Physical (n=62 studies; 77%) 3.5.1 Physical symptoms and concerns Disease and procedure-related pain has emerged as a major concern across the disease trajectory, and during the end of life stage; the pain is often reported as 'intolerable', or 'out of control', with procedure-related pain being associated with fear, anxiety, and suffering. 61,46,[62][63][64][65][66][67][68][69][70][71][72][73][74] For this reason, the need for parental protection and assistance during treatment is stressed upon in cases of young children with cancer. 46,75 Two studies have found such pain to be more prevalent in children with solid tumours. 76,77 Inadequate treatment of pain, side effects of opioids, and treatment of non-pain related symptoms have also been mentioned as concerns by paediatric oncology patients. 61,69 Concept elicitation and illness experience studies highlight pain control as a priority in cancer and non-cancer disease groups. 19,27,58,60,[78][79][80][81] Other commonly reported physical symptoms are lack of energy, nausea, vomiting, dry mouth, weight loss, and drowsiness 63,71,82-84 27,65,71,82,83,85-88 . Symptoms associated with severe distress during cancer treatment include difficulty in swallowing and shortness of breath. 63,89 Symptoms associated with severe distress include shortness of breath, feeding difficulties, fatigue, drowsiness, nausea, loss of motor function, pain, reduced mobility, decreased appetite, respiratory failure, and lack of energy. 46,62,64,65,90,91 Difficulty in breathing/shortness of breath, fatigue, drowsiness, and nausea is a common concern at the end of life, in both cancer and non-cancer patients. 62,64,65,68,76,91,92,90 Providing support with eating, sleeping problems, and minimizing symptom distress are commonly expressed as the end of life care priorities. 60,93,94 In two studies, neurological deterioration, loss of the ability to communicate, and decreased physical activity have been associated with impending death. 56,95 On-going seizures are a major concern among epilepsy patients. 96 Weight loss, fever, mouth sores, stunted growth, diarrhoea, wasting, lymphadenopathy, oral candida, acute malnutrition, pneumonia, and respiratory tract infections are common in paediatric patients with HIV. 87,97,98 One study conducted among HIV positive children, aged 6-15 years, reported a high prevalence of other morbidities such as hearing impairment, visual impairment, gingivitis, speech impairment, and gross motor deficits. 98 The end-stage renal failure patients have also reported physical needs of post-renal transplant adjustment and coping; 99 their dependence on and need for a wheelchair compromises their function and has been associated with negative emotions. 99 Studies that address concerns in neuro-disability and, particularly, the one on the development of the suffering scale in adolescents with cancer, prioritised physical symptoms and associated distress as core domains, which should be included in outcome measures for young people with LLC. 27,58 It is noticed that children tend to use unique language to describe their symptoms and, at times, have difficulty explaining their feelings. 100 45,101 The 'no symptom' syndrome is also reported in one study and has been interpreted as a form of denial, lack of specific symptoms, or unchanged health status. 65 Young people express the need to be normal with full ability to perform age-appropriate functions such as self-care, mobility, and physical activity. 82,83 51,54,95,99,102-107 Symptoms become more of a concern when they lead to physical and mental changes or affect the ability of young people to engage in daily activities. 44 The psychological consequences of living with LLC span a spectrum of sub-domains including emotional, mood-related, cognitive, behavioural change, and isolation. Based on observations, children aged 0-6 years express a need for emotional satisfaction and expression of their own will. 75 Mood-related symptoms include emotional instability 27 , feeling shy about living with the disease, 80 feeling horrible, furious, upset, and disappointed 99 , angry, scared, bored, sad, nervous, and crying. 45,49,50,58,91,104 80,86 Children receiving cancer treatment commonly experience mood swings, depression, anxiety, and lack of concentration. 46,63,71 110 At the end of life, psychosocial symptoms of cancer patients include fear of going to sleep and dying, displaying a distance from family, confusion, anxiety and depression, irritability, inability to laugh or smile, insecurity, and mood swings. 42,46,65,76,82 One study has found emotional concerns to be more dominant in older age groups (12 and above). 83 Palliative care professionals mention the need for alleviation of psychological suffering as an important domain of quality of life in paediatric palliative care. 107 Cognitive changes that have been identified include disturbance of consciousness, declining attention and concentration orientation, social skills, cognition, energy, and drive. 57,76,80,81,84 Other concerns include the need for a sense of self-worth, 111 resilience, coping with illness, accepting the present pain for potential gain in future, and the desire to protect other people from similar illness experiences.
Children have sometimes reported psychological growth; for example, on achieving milestones like treatment completion, they reflect on the whole experience cognitively; "they either felt the same as old-not having noted any changes on outlook to life, completely different, or not normal." 112 Schoolgoing children have also reported poorer performance at school. 58 110 Parents of young people report a common range of behaviours in their children: anti-social, disobedient, unwilling to take medication, creating difficulties in incorporating daily medication into their routine schedules, and pill burden associated distress. 73,99,102,103,[113][114][115] Adolescents (aged 12-18 years) are concerned that isolation or being like a "prisoner" has radically altered their lives and made it alien to them; some have cited memories of fear that could not be forgotten. 46 Hospitalisation, for treatment, is associated with isolation, affecting opportunities for interaction with friends and siblings, and invoking emotions of sadness, and homesickness. 104,109 The impact of living with LLC on social and physical functioning leads to feelings of frustration regarding a strong wish for 'normalcy'. 46,56,103,108 3.7 Psychosocial (n=31 studies; 38%) Young people and their families perceive living with LLC as a stressful life experience, irrespective of the type of diagnosis. Some concerns reveal a social and physical health overlap (e.g. young people with epilepsy who "felt different" due to their need for medication 114 ). For paediatric HIV patients, disease-related features such as skin rash and facial lipodystrophy have been found to be associated with stigma. 116 Paediatric cancer patients are more concerned about hair loss and skin changes, which affect their body image. 95,117 The experiences of becoming the centre of attention to peers, being bullied at school, and isolation, have been prominently expressed. 52,72,102,114,115,117,118 . Young people who have received liver transplants cite the plight of post-treatment features such as large scars, clubbed fingers, and short stature. 72 Hospitalisation is a particularly undesirable experience associated with disruption of school schedule and social interaction, and resulting in isolation. 23,44,49,50,54,58,72,82,99,106,108,117 Young people generally value social relations where they feel comfortable talking, being listened to, share secrets, and are treated with respect. 23,39,60,72,108,114 119 107 Young people, health professionals, and families express the need for children to experience fun, humour, laughter, recreation, and leisure alongside treatment. 23,27,51,60,95,108 107 Family relation concerns have been highlighted in 20 studies. 19,39,41,44,46,52,58,72,73,75,103,109,114,116,[120][121][122][123][124][125][126] Young people value support from their families as it is a structure that helps them feel comfortable and secure. Concerns regarding lack of family support have also been reported; children report discomfort in having to deal with expressions of anger, shock and sorrow from their parents/caregivers, which engenders a sense of being a burden. 46,125,126 Young people are worried about their family carers/parents emotions of fear, hopelessness, depression, and anxiety. 52,72,95,114,116,120 Older children have raised concerns regarding receipt of information about their diagnosis directly from the right people, rather than by overhearing parents and physicians. 103,109,115,127 Some young people prefer to keep their diagnosis a secret from peers for fear of social stigma. 103,109,114,115 In three studies, adolescents have reported concerns relating to sexuality. These concerns include initiating and maintaining romantic relationships, painful sex, and fertility concerns after treatment. 103,109 88 3.8 Existential/spiritual/religious (n=37;46%) The concerns under this category include existential loss, existential vacuum, worry about death, not being at peace, uncertainty arising from inability to anticipate situations, a need to be remembered, hopes, and finding meaning in life in situations, especially when young people feel that their dreams and hopes for the future are being ruined due to terminal illness. 39,54,58,67,78,79,95,108,110,118,128,129 In one study, health professionals have noted that the "life goes on" ideology is important. 107 It is common for the young people, including three-year-olds, to end their narratives with concerns about impending death. 130 Young people and families have also expressed the construct of connection to something larger than the self. These beliefs seem to help them build resilience. 47,92,95 The desire for religious prayers has been mentioned in several instances as a priority. 59,108 The young people have also reported a sense of spiritual growth in maturity and some are thankful for the "gift of life" and wish to protect others from similar experiences. 93,131 3.9 Other concerns (n=39; 48%) Besides physical, psychological, psychosocial, and spiritual/existential concerns, additional pressing problems have also been found; they include communication and information, decision making, and care provision concerns (see Table 4 for details).
[Insert Table 4] 3.10 Sub-group analysis by age group and diagnosis The differences in symptoms and concerns that matter to young people, with respect to age and type of diagnosis, are noted in this study; the results for the same are presented in Table 5. For example, treatment procedural pain and alienation are more dominant in younger children (0-5 years), while an existential loss, self-image, and need for access to information are more dominant in older children (6-9, 10-14, 15+) years. 132 [Insert Table 5] Symptoms and concerns that have been identified for young people with LLC and their families are mapped in a summary diagram, alongside illustrative examples of useful health outcomes, in Figure  2.

Discussion
Through the process of drawing together a comprehensive body of literature across global regions and different conditions, this paper identifies the symptoms and other concerns faced by young people with LLC and their families. Previous reviews have focused solely on cancer and neuro-disability. 18,19,133 This synthesis and presentation of symptoms and concerns across core health domains can be used to guide the development of outcome assessment measures for paediatric palliative care. While the studies are of intermediate methodological quality, it has been possible to extract data on what young people with LLC consider as important, to inform the development of the child/family centred conceptual framework. Studies in this review recruited patients at different stages of the disease trajectory, but multi-dimensional burdensome symptoms and concerns were found across studies. This finding informs debate around the appropriate timing of referral to, and the initiation of, paediatric palliative care (i.e. soon after diagnosis vs. later in the disease trajectory and towards the end of life).
The key message is that, for optimal outcomes, paediatric palliative care should be provided from the time of diagnosis and through to death and bereavement, as is recommended by the World Health Assembly. 35 This approach would align with recent evidence demonstrating the benefits of providing early integrated palliative care in adult populations. 134 The themes concerning symptoms and concerns that have been identified in this review are embodied in illness experience and multidimensionality and are underpinned by the three overlapping domains of child, carer/family, and quality of services. This is in agreement with proposed models of care in palliative care, recommending the incorporation of these components in outcome assessment. 135 Given the task at hand -that of meeting such multi-dimensional concerns -paediatric palliative care models of care may benefit from key elements that have been proposed for person-centred care, which include respect, coordination and integration, physical comfort and emotional support, involvement and support for carers/family, information and education, continuity, and transition. 136 Person-centred care also proposes key activities such as personalised care, self-management support, and shared decision making. 137 This ethos embraces the core child/family concerns that have been identified in this review.
The review findings also demonstrate the intrinsic link between child and carer/family, and care provider interactions, as reflected under the quality of services domain. Positive engagement through information, education, and communication has the potential to enhance child/carer/family selfefficacy and self-management, which can have an impact on outcomes of care. Positive provider interactions may explain the manner in which system process related concerns link to the optimal goals of care and the reasons for their importance. An important consideration for development of services for young people with LLC is the development indicators that can be used to assess the structure, process, and outcomes aspects of health services. This review makes an important contribution by putting forward an evidence-based child/family framework of domains, from which such indicators could be selected. This makes it easier for care providers to gather more information about the relevant domains some of the constructs and symptoms to facilitate prompt action.
The information and communication theme identified in this review warrants further exploration in young people with LLC. Young people are a unique population with varying symptoms and concerns which occur alongside continuing physical, emotional and cognitive development, and a dynamic socio-ecological environment. 33 Indeed, several studies have highlighted the uniqueness of the language that children use to describe their symptoms and concerns. 18,101 Furthermore, paediatric palliative care patients may also have reduced communication and cognitive abilities. 15 This review highlights the ability of young people (6+ years) to self-report on symptoms and health outcomes. As such, young people should be central to and involved in the elicitation of preferences and development of outcome measures. Such an approach should be aligned with simultaneous investment in appropriate information and communication tools and strategies.. It is time to prioritise the provision of self/proxy reporting options for outcome measures in paediatric palliative care to make self-report a preferred option for subjective outcomes, whenever possible. 34,138,139 This review identifies differences in the way health concerns, such as social and psychological wellbeing, are expressed on the basis of developmental age. With age, the cognitive, emotional, and socio-ecological aspects of children undergo change. For example, an advanced understanding of illness emerges in adolescents 140 alongside a shift towards a preference for self-efficacy and shared decisionmaking models. 83 Although the core domains of health remain robust across adult and paediatric populations, the developmental age of young people needs to be considered. Differences across cognitive, emotional and socio-ecological facets render the use of adult-based measures inappropriate, even with adolescents. 138 It is observed that young people have not been interviewed in 30% of the studies, and 35% are mixed samples of young people and proxies, suggesting a low level of involvement of young people in the research that aims to inform the direction of their care. The findings echo previous reports about children with cancer. 25 Involving young people in research that informs their care is the first step to allowing their experiences to update the models of care; this is far from commonplace across the literature. In order to guide best practices on research involving young people, methodological concerns regarding the following need to be addressed: the use of age-appropriate methods of data collection; question wording; duration of interviews; processes of data collection; the manner in which challenging issues of interviewing ill children are dealt with; provision of sufficient details on recruitment strategies, and; informed consent processes . 12,33 The review observes considerable overlap across themes related to the subjective experiences of illness across diagnostic groups, study locations, and age groups of children. For example, there are parallels in themes identified across previous reports in paediatric cancer, 18,133 paediatric HIV, 141 and paediatric neuro-disability. 19 Furthermore, similar indicators have been found that are useful for comparing models of care across different settings. 136 This enhances the feasibility of multi-setting comparisons. 142 It also lends credibility to the use of generic palliative care outcome measures in children with LLC, with minor adaptations wherever necessary, for aspects such as health status, the process of care, or socio-cultural concerns. 143 There are over 300 LLC conditions experienced by young people that may require palliative care; the development of disease-specific outcome measures across all conditions may not be appropriate or feasible, moreover users want fewer tools. 144 145,146 Overlap in the conceptualisation of health outcomes in paediatric palliative care can support the growth of the research field. 138

Strengths and limitations
To reduce bias, the review adopted a broad and comprehensive search strategy across multiple databases, did not limit article inclusion by language, and involved field experts to identify any additional relevant literature. The search has been conducted following PRISMA guidelines. The quality of studies was also assessed, although not used as a basis for article exclusion. This is the first review to comprehensively appraise the state of evidence on symptoms and concerns in young people with a broad range of life-limiting and life-threatening conditions, across the disease trajectory. This is also the most comprehensive framework of meaningful outcomes for young people with LLC. This review has some limitations. Data from a disparate evidence base has been compiled, which utilises a wide range of methods to understand the symptoms and concerns of young people. The variety of approaches meant that it was not possible to assess the extent or magnitude of identified symptoms and concerns among study participants. The inclusion of studies, with both qualitative and quantitative approaches, led the team to adopt narrative methods of synthesis, with efforts made to be transparent about how this was undertaken. Furthermore, some studies did not report the recruitment strategies and as such potential bias could not be assessed, compromising our judgement regarding the methodological quality of the studies included. 147 Many conditions require palliative care and different terminologies are used in different settings and consequently relevant articles may not have been identified. 144 There is a high burden of interacting and multidimensional symptoms and concerns in paediatric palliative care populations. These occur across the disease trajectory, in both malignant and nonmalignant conditions. Therefore, early integration of paediatric palliative care into care plans to address these issues is recommended.
A skilled multi-professional team will be needed to address the symptoms and concerns raised, given they are so wide-ranging. . This study challenges the unidimensional or typical biomedical models of care for children with LLC, which fail to comprehensively address their multi-dimensional symptoms and concerns. This child/family centred framework of child/family domains, grounded in their illness, mirrors the structure, process, and outcomes domains of health service improvement, and can guide the development of appropriate outcome measures to assess existing services and support their development. 15 The measures will inform service audits, research, and evaluations in order to stimulate service development. Developmental age will be important to consider when developing paediatric palliative care outcome measures, with differences across young people identified across the emotional, cognitive, and socio-ecological levels in this review. The developmental age categorizations that have been used to explore these variations in this review were broad and future studies should explore this further, using narrower categories or those that have been recommended for paediatric palliative care. 140 Our findings indicate commonality in the illness experience, suggesting that unified person-centred outcome measures for children across different diseases are feasible. It is instead developmental age which may determine variations in the domains (content) and form of a measure. . 148 Internationally, the state of science remains poor for aspects of care for young people with LLC, including spiritual/existential concerns, patient-reported experiences of care, service delivery, decision-making information, and approaches to communication. 133 Future studies should further explore these areas, alongside addressing gaps in evidence on symptoms and concerns for young people with LLC with non-malignant conditions, those in developing countries and those from different social-cultural settings.
Acknowledgments A note of thanks is extended to Simon Etkind, Cicely Saunders Institute King's College, London (UK), for the useful comments on the framework and manuscript of health outcomes. The authors also extend a special note of appreciation to Melanie Merriman, Moses Bateganya, and Chang Victor for their useful comments on the manuscript.

Contributors' Statements
Eve Namisango conceptualised and designed the study, reviewed literature, abstracted and analysed the data, and drafted the initial manuscript.
Professor Richard Harding, Professor Fliss Murtagh, and Dr Katherine Bristowe reviewed the protocol, data abstraction tools, data analysis framework, and the results.
Professor Irene Higginson and Dr Melanie Abas reviewed the research questions, search strategy, and review findings.
Dr Matthew Allsop reviewed the protocol, abstracted data, and carried out data analysis in association with the lead reviewer.
Professor Julia Downing reviewed the analysis framework and appraised the interpretation of the review findings.
All authors have contributed to the final manuscript.