Article Text

Download PDFPDF

Research priority setting in childhood chronic disease: a systematic review
  1. Harrison Lindsay Odgers1,2,
  2. Allison Tong1,2,
  3. Pamela Lopez-Vargas2,3,
  4. Andrew Davidson4,5,6,
  5. Adam Jaffe7,8,
  6. Anne McKenzie9,
  7. Ross Pinkerton10,
  8. Melissa Wake4,6,11,
  9. Peter Richmond12,13,
  10. Sally Crowe14,
  11. Patrina Ha Yuen Caldwell2,15,16,
  12. Sophie Hill17,
  13. Jennifer Couper18,19,
  14. Suzy Haddad20,
  15. Behrouz Kassai2,21,
  16. Jonathan C Craig1,2
  1. 1 Sydney School of Public Health, University of Sydney, Sydney, New South Wales, Australia
  2. 2 Centre for Kidney Research, The Children’s Hospital at Westmead, Westmead, New South Wales, Australia
  3. 3 Kid’s Research Institute, The Children’s Hospital at Westmead, Sydney, New South Wales, Australia
  4. 4 Department of Paediatrics, University of Melbourne, Melbourne, Victoria, France
  5. 5 Department of Anaesthesiology, Royal Children’s Hospital, Melbourne, Victoria, Australia
  6. 6 Murdoch Children’s Research Institute, Parkville, Victoria, Australia
  7. 7 Department of Respiratory Medicine, The Sydney Children’s Hospital Network, Sydney, New South Wales, Australia
  8. 8 Discipline of Paediatrics, The University of New South Wales, Sydney, New South Wales, Australia
  9. 9 Western Australian Health Translation Network, The University of Western Australia, Perth, Western Australia, Australia
  10. 10 Department of Oncology, Lady Cilento Children’s Hospital, South Brisbane, Queensland, Australia
  11. 11 Department of Paediatrics, The Liggins Institute, The University of Auckland, Auckland, New Zealand
  12. 12 Division of Paediatrics, School of Medicine, The University of Western Australia, Perth, Western Australia, Australia
  13. 13 Departments of General Paediatrics and Immunology, Princess Margaret Hospital for Children, Perth, Western Australia, Australia
  14. 14 Crowe Associated Ltd, Oxon, UK
  15. 15 Discipline of Paediatrics and Child Health, University of Sydney, Sydney, New South Wales, Australia
  16. 16 Department of Nephrology, The Children’s Hospital at Westmead, Sydney, New South Wales, Australia
  17. 17 Center for Health Communication and Participation, La Trobe University, Melbourne, Victoria, Australia
  18. 18 Department of Endocrinology and Diabetes, Women’s and Children’s Hospital, Adelaide, South Australia, Australia
  19. 19 Robinson Research Institute, University of Adelaide, Adelaide, South Australia, Australia
  20. 20 Patient and Carer Representative, Sydney, New South Wales, Australia
  21. 21 Centre d’Investigation Clinique de Lyon, Lyon, France
  1. Correspondence to Dr Harrison Lindsay Odgers, Centre for Kidney Research, The Children’s Hospital at Westmead, Sydney, NSW 2145, Australia; hodg7590{at}


Objective To evaluate research priority setting approaches in childhood chronic diseases and to describe the priorities of stakeholders including patients, caregivers/families and health professionals.

Design We conducted a systematic review of MEDLINE, Embase, PsycINFO and CINAHL from inception to 16 October 2016. Studies that elicited stakeholder priorities for paediatric chronic disease research were eligible for inclusion. Data on the prioritisation process were extracted using an appraisal checklist. Generated priorities were collated into common topic areas.

Results We identified 83 studies (n=15 722). Twenty (24%) studies involved parents/caregivers and four (5%) children. The top three health areas were cancer (11%), neurology (8%) and endocrine/metabolism (8%). Priority topic areas were treatment (78%), disease trajectory (48%), quality of life/psychosocial impact (48%), disease onset/prevention (43%), knowledge/self-management (33%), prevalence (30%), diagnostic methods (28%), access to healthcare (25%) and transition to adulthood (12%). The methods included workshops, Delphi techniques, surveys and focus groups/interviews. Specific methods for collecting and prioritising research topics were described in only 60% of studies. Most reviewed studies were conducted in high-income nations.

Conclusions Research priority setting activities in paediatric chronic disease cover many discipline areas and have elicited a broad range of topics. However, child/caregiver involvement is uncommon, and the methods often lack clarity. A systematic and explicit process that involves patients and families in partnership may help to inform a more patient and family-relevant research agenda in paediatric chronic disease.

  • outcomes research
  • patient perspective
  • qualitative research
  • stakeholder engagement
  • chronic disease

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What is already known on this topic?

  • The research agenda has been traditionally driven by expert opinion, potentially generating priorities that do not encompass patient concerns.

  • More inclusive research priority setting processes have the potential to generate a more relevant research agenda and thus reduce waste in research.

What this study adds?

  • This study systematically demonstrates shortcomings with the quality of priority setting processes in paediatric chronic disease. This helps guide future processes.

  • This study provides a significant summary of generated research priorities across paediatric chronic disease, providing insight into where the agenda has been driven.


The global expenditure on biomedical research exceeds US$265 billion per annum.1 However, up to 85% of this investment is wasted due to problems across all stages of the research process, beginning with how research topics are selected.2 Despite the well-documented mismatch in priorities between patients and researchers, the selection of research topics appears to be predominantly researcher-driven with little input from patients and their families, the ultimate users of research and so important research needs may be overlooked.3 4 Concurrently, a lack of transparency in the methods used to select research topics have raised concerns about the legitimacy of decisions regarding research investment, particularly as most research is publicly funded.5–7

Priority setting is highly relevant for paediatric research. Children have long since been regarded as ‘therapeutic orphans’,8–10 but in the past two decades, policy initiatives worldwide have driven efforts to support and increase the amount of research involving children.11–16 Paediatric trials are complex and costly due to additional ethical requirements and concerns among paediatricians and caregivers about the vulnerability of children.17–19 Transparent and inclusive research priority setting is needed to ensure that resources are directed towards research that is relevant to children, families and their clinicians. The James Lind Alliance (JLA) has established mechanisms for such research priority setting partnerships,20 and there have been several initiatives conducted in paediatric conditions including neurodisability, eczema and cleft lip and palate.21–23

Childhood chronic disease imposes an often indefinite and overwhelming burden on children and their families. Children with chronic conditions may face prognostic uncertainty, poor quality of life (QoL) and adverse psychosocial outcomes, as well as delayed growth and development.24–28 This study aims to evaluate the process of research priority activities in paediatric chronic disease and to describe the range of research priorities identified. Understanding the current landscape of research priority setting may guide efforts to better align the research agenda in paediatric chronic disease with the priorities of patients, families and health professionals involved in their care.

Materials and methods

Selection criteria

We included studies that elicited stakeholder priorities for research in childhood (aged under 18 years) chronic disease. Stakeholders were defined as patients, caregivers (eg, parents), healthcare providers, policy makers or researchers. Consensus methods (such as Delphi technique), workshops, conferences, surveys or qualitative studies were eligible. We excluded non-primary research articles (review, editorials and guidelines), conference abstracts, epidemiological studies, policy analyses and articles published in non-peer reviewed journals. No language restrictions were applied.

Data sources and searches

We used Medical Subject Headings terms and text words for paediatrics (ie, children, adolescents, parents and family) and priority setting to search MEDLINE (1946–), EMBASE (1980–), PsycINFO (1806–) and CINAHL (1961–) up to 16 October 2016 (online supplementary file 1). We also searched Google Scholar, PubMed and the JLA website. References lists of included articles were reviewed. Two authors (HLO/PL-V) screened abstracts and discarded studies that were not eligible, then assessed the remaining full texts for eligibility.

Supplemental material

Appraisal of transparency of reporting

There are no standard or universally accepted guidelines for appraising the reporting of research priority setting; however, some good practice principles have been proposed.6 29–31 From existing tools we extracted items to develop a comprehensive 32-item checklist to assess the transparency of reporting of research priority setting (online supplementary file 2). This checklist addressed: context and scope, governance, inclusion of stakeholders, identification and collection of topics/questions, prioritisation of research topics/questions, output, evaluation, dissemination and translation, and funding and conflict of interest. Two reviewers (HLO and BK) independently appraised the included studies, and a third author (PL-V) adjudicated discrepancies.

Supplemental material


We conducted a descriptive textual synthesis of the research priorities identified in the primary studies. Each article was imported into HyperRESEARCH V.3.7.3 (ResearchWare Inc: 2009 V.3.0.3, Randolph, Massachusetts, USA) software to facilitate coding and retrieval of textual data. One author (HLO) inductively identified and coded all the research priorities and classified and summarised them descriptively according to broad research topics. The research topics were cross-tabulated against the chronic condition/health area for comparison. Two other reviewers (PL-V and BK) read the articles to ensure that all relevant data were extracted and coded into the appropriate research topics.


Literature search and study characteristics

We included 83 studies involving 15 722 participants (figure 1, online supplementary file 3), of which 17 (20%) studies did not report the number of participants. Of the 18 (22%) studies that involved patients, 13 included adult patients and only four included patients aged below 18 years (table 1). Four (5%) studies did not define patient age. Twenty (24%) studies included caregivers or family members. The number of children and adolescents/young adults in the included studies ranged from 1 to 844. The number of parents in studies ranged from 3 to 876. All studies involved health professionals. Thirty-eight (46%) studies included physicians and 50 (60%) included researchers. The range of methods included workshops/conferences (59%), Delphi surveys (28%), surveys (25%) and qualitative studies (4%). The majority of studies were conducted in the USA (47%), UK (13%) and Australia (10%). Eighteen (19%) studies were multinational. Forty-seven (52%) studies were published since 2010.

Supplemental material

Figure 1

Search results. aEditorials, reviews, letters and guidelines. bPrevalence and incidence, clinical trials and cohort studies. cCost-effectiveness and economic impact studies. dReviews of healthcare policy. eAssessment of health systems. fNumber of participants calculated using only the first-round participants for any included Delphi surveys.

Table 1

Characteristics of included studies (n=83)*

Transparency of reporting

The transparency of reporting was variable across the studies, ranging from 4 to 29 items (of a possible 32 items, table 2). All 83 studies defined the health area or focus. The methods for collecting the initial research topics were described in 50 (60%) studies and 50 (60%) studies outlined the process of prioritisation. The participant characteristics were reported in 73 (88%) studies, and 17 (20%) considered implementation of the research priorities.

Table 2

Comprehensiveness of reporting (n=83)*


The most frequently included health areas were cancer (11%), neurology (8%), endocrinology and metabolism (8%), respiratory (7%), development and behaviour (7%), mental health (6%), multiple-condition focus (6%), palliative care (6%), dermatology (4%) and HIV/AIDS (4%) (figure 2). Multiple less frequently identified areas were included and are described in table 1 and online supplementary file 3. Eight per cent of studies had no specific disease focus.

Figure 2

Matrix of research priorities in paediatric chronic diseases. Horizontal axis indicates health area, and vertical axis indicates research topic. Circle size indicates the number of studies representing the matrix point. Note: the health area category ‘Other’ (n=27) is not included. QoL, quality of life.

The research priority topics covered disease onset and prevention (43%), prevalence (30%), disease trajectory (48%), diagnostic methods (28%), treatment (78%), knowledge and self-management (33%), QoL and psychosocial impact (48%), access to healthcare (25%) and transition to adulthood (12%) (figure 2). The majority of studies presented the research priorities as ranked lists or embedded within the text, with most framed as statements or research questions. There was variability in the level of detail provided on the research priorities in terms of population, intervention, comparator and outcomes, and scope. The prioritised topics are described below with most relevant to multiple disease areas.

Research topics

Disease onset and primary prevention (36 studies)


Research priorities included identifying the genetic,32–47 epigenetic32 48 and environmental32 34 36–38 41 42 44 46 49 causes of disease. Genome-wide assessment to identify candidate causative genes was mentioned,35 46 as were gene–environment interactions.36 37 46

Risk factors

Stakeholders prioritised identifying genetic and environmental risk factors for disease onset.21 32 34 35 37–39 41 42 46 50–55 Family structures,37 diet,21 35 37child abuse,52 comorbid medical conditions (including mental illness)36 38 50 52 and prenatal risk factors were specifically mentioned for investigation.32 37 49 56 Understanding the disparities of risk factors among ethnic minority groups was suggested.37 49–51 57 58

Primary prevention

Primary prevention of paediatric disease was a priority,21 37 38 43 44 48 49 51 52 55 58–67 including specific interventions targeting parents such as prenatal care, community based-education and interventions via social media.44 49 58–61

Prevalence (25 studies)

Prevalence of conditions

Determining the prevalence of conditions was identified as a priority,32 34 35 42 44 45 50 61 62 66 68–72 including variance by geographic location, ethnicity and gender.35 73

Prevalence of symptoms and comorbidities

The prevalence of symptoms and comorbidities was a research priority.34 35 38 44 48–50 54 56 63 69 73–76 Investigating the prevalence of comorbid mental health conditions,35 38 49 75 77 cognitive impairment34 50 and obesity were specifically prioritised.48

Disease trajectory (40 studies)

Monitoring of disease

Monitoring chronic disease-related complications were regarded as a research priority,32–35 40 44 45 48 50 61 65 68 71 74 78–82 and the development of biomarkers were commonly suggested to this end.32 33 35 50 65 68 71 80 81 Research on the best methods to monitor attention deficit hyperactivity disorder and autism spectrum disorder (ASD) were prioritised.48 79

Prevention of progression

Prevention of disease progression and secondary complications were identified as research priorities in many studies.23 32 34–37 42 45 46 48 50 52 56 59 61 63–65 74 75 78 80 83–85 Identifying factors that influence disease progression were prioritised,35 48 50 56 65 74 78 as was understanding the processes that underlie secondary complications.32 34 36 42 59 65 74 86 87 Examples included exploring the reasons for secondary complications of ASD: ‘Why do autistic people appear to be more at risk from some medical conditions than non-autistic people?’.42 Interventions to prevent progression and complications of disease were also suggested.23 32 34–36 45 63 64 75 83

Long-term outcomes

Identifying the long-term medical outcomes of paediatric chronic disease were proposed as research priorities,15 32 33 35 37 38 41 43 46 48–50 59–61 63 65 74 75 81 88 including physical growth,35 59 60 74 and cognition.35 46 50 74

Diagnostic methods (23 studies)

Improved diagnosis

Optimising diagnostic tests in terms of developing tools, biomarkers and training of healthcare providers were suggested.32 37 38 40–42 44–49 57 61 63 67 68 71 74 75 79 81 87 89 In paediatric epilepsy, investigating the impact of healthcare provider competency on diagnosis was prioritised.89

Early detection

Research into the effect of early detection on outcomes including family coping, healthcare decision making, health outcomes and cost-effectiveness was prioritised.38 44 46 49 61 71 89 Two studies underlined the need for more research about screening for ASD,44 48 and one study prioritised research into health system-related factors that could improve the early detection of developmental disabilities.44

Treatment (65 studies)

Medical therapy

Medical therapy research priorities were primarily focused on determining treatment efficacy21 23 32 35 47 48 50 53 63 64 71 73–75 77–79 81 82 84 86 87 90–98 and the safety and long-term effects of medications.21 23 32 39 55 71 74 75 77–79 83 90–94 96 99 For example, ‘What are the adverse effects associated with long-term use of short- and long-acting bronchodilators; inhaled and oral steroids; and combination and additive therapies in children?’.83 Cost-effectiveness was prioritised in three studies.14 43 73 Pharmacokinetic, pharmacodynamic and pharmacogenetic research topics were also identified.32 34 48 74 81 93

Non-medical therapy

Non-medical therapies including cognitive therapy, psychotherapy and family therapy were prioritised.23 35 38 48 57 70 90 91 100–102 Investigating the impact of social and communication interventions on psychosocial outcomes was prioritised,23 35 48 53 56 96 100 including the effect of peer and parental support.53 96 100 Additional interventions included diet,75 83 90 91 exercise,48 101 alternative therapies,75 83 90 91 and self-treatment using technology.43 44 48 75 84 96 100

Treatment adherence

Priorities focused on mediators of adherence,35 48 58 65 74 87 88 92 103 including community and family influence on non-adherence.48 58 Strategies to improve adherence were also prioritised.35 58 65 74 75 82 89 104

Models of care

Research into the effect of inpatient versus outpatient care on safety, cost effectiveness and impact on family were prioritised.58 84 92 102 Identifying the differences between specialist, non-specialist, nurse-led and multidisciplinary care models was suggested,60 78 90 as were methods to effectively integrate paediatric palliative care into mainstream services.70 76 105

Knowledge and self-management (27 studies)


Exploring knowledge of diseases and their management was prioritised by patients, their parents and healthcare providers.48 56 58 69 85 89 103 In one palliative care study, the top priority was ‘Children’s understanding of death and dying’,69while an epilepsy study prioritised ‘What do parents and providers understand about the seriousness of seizures and the need to control them as soon as possible?’.89 Education strategies to improve patient,43 52 67 78 83 92 103 healthcare provider45 46 73 83 105 and parent45 46 52 56 64 82 85 89 102 105 knowledge were prioritised.

Communication and decision making

Exploring communication between healthcare providers, parents and patients was prioritised,43 69 70 85 89 105 including specifically exploring communication at diagnosis.85 89 Priorities to develop effective communication strategies were described.43 69 70 105 Evaluating the roles of patients and parents in healthcare decision making was prioritised,44 58 105 including developing decision aids to help parents make decisions for or with their child.105

Patient preferences and self-management

Research on patient healthcare preferences was prioritised.16 55 69 70 76 103 The perspectives of family members and healthcare providers were also considered relevant,69 70 for example, ‘What matters most for patients and parents receiving paediatric palliative services?’.76 Identifying factors for successful self-management programmes was also regarded as important.43 78 83 88 99

QoL and psychosocial impact (40 studies)

Outcomes and mediators

Stakeholders prioritised research to explore the effect of disease on patient QoL and psychosocial outcomes.15 32 37 46–48 50 54 56 61 64 66 73–75 77 78 84 92 102 106 107 Social outcomes,37 44 46 48 50 106 107 including self-esteem,106 and the impact on education and employment were mentioned,15 61 64 75 77 84 for example, prioritising research into ‘social and emotional development in severely visually impaired children’.50 Identifying mediators of QoL and psychosocial outcomes were prioritised,35 49 56 64 65 67 100 as was developing tools to accurately measure QoL and psychosocial functioning.35 45 49 65–67 74 100 108

Burden and bereavement

Priorities included identifying the needs of families/caregivers,15 58 64 70 76 84 109 healthcare workers70 109 and siblings of affected children.70 Interventions to provide bereavement care and support to families and siblings were prioritised,58 60 64 76 85 92 105 for example, ‘Evaluate interventions designed to help family members cope with the treatment process and its outcomes’.92

Access to healthcare (20 studies)

Stakeholders prioritised identifying general barriers and facilitators to healthcare access.15 16 35 46 49 51 52 60 66 67 69 73 75 89 105 Specific effects of insurance status48 49 51 60 and geographical distance were mentioned.15 46 The need to identify at which level of the clinical care pathway barriers were occurring (eg, diagnosis, accessing treatment and continuing care) was prioritised,73 89 105 as was identification of factors that influence appropriate referral to medical services.66 70 89

Transition to adulthood (10 studies)

Research priorities on transition to adulthood were identified.48 52 60 66 70 86 88 103 104 107 Mediators of successful transition to adult healthcare services,48 52 66 70 103 as well as interventions to facilitate this transition,70 103 were mentioned. Mediators60 104 107 and interventions to facilitate transition to adult society was also identified.52 103 104 For example, ‘What skills and knowledge do adolescents need to learn in order to better manage their chronic illness and enable successful transition to adult health care and adulthood?’.103


Research priority setting for chronic conditions affecting children has been conducted across a wide range of healthcare areas. Across the studies, the most frequently identified topic was treatment, which appeared in 65 (78%) studies and in all healthcare areas. Access to healthcare (24%) and transition to adulthood and adult healthcare services (12%) were less frequent topics. Despite the range of research priorities identified, the methodology was generally inadequately described, with about half describing participant selection, approach to collecting, categorising, refining and prioritising/achieving consensus on research priorities. Over half of the studies were conducted since 2010, indicating a recent substantial increase in the number of research priority setting initiatives in paediatric chronic disease.

Patient and family engagement in research priority setting was rare. Only about one-quarter of studies reported parental/caregiver involvement, and only 5% involved children directly. Two studies involving paediatric patients used strategies established by the JLA, which is an independent organisation that facilitates priority setting partnerships between clinicians, patients and carers, to identify research priorities using consensus-based methods.20 Both studies suggested that professionals prioritise pharmacological therapy and basic science topics, while paediatric patients and their caregivers prioritised mostly psychosocial health and non-pharmacological treatment, for example, education, diet and lifestyle changes.21 90 We observed a similar trend across all included studies; for example, a study on epilepsy involving only researchers predominantly prioritised aetiology, risk factors and medical therapies,34while another epilepsy-focused study that included parents generated priorities relating to education, patient/provider/community communication, treatment adherence and access to healthcare.89 These results suggest that patient involvement in research priority setting is important in generating a research agenda that encompasses the full spectrum of issues that affect patients.

The infrequent involvement of children may be due to age-specific barriers and challenges, and efforts to develop engaging and developmentally appropriate strategies that empower children to articulate their priorities for research are lacking.110 111 Studies in this review that included children/adolescent perspectives employed surveys that were deliberately created in plain English, non-technical versions in order to avoid confusing patients with research jargon, but most studies still reported receiving a large number of responses that were out-of-scope of the survey questions.21 67 83 90 Promotion of these priority setting partnerships was achieved through a mixture of patient advocacy groups, traditional media and social media networks (eg, Facebook).21 67 83 90 Social media groups were specifically used to target teenagers due to their common use in this population and potential for a ‘snowballing-effect’ of participation.21 67While engagement of paediatric patients may not be straightforward, studies included in this review indicate it is possible to integrate paediatric perspectives into research priority setting and generate meaningful results.21 67 83 90 Furthermore, participation in setting the research agenda can likely benefit the children themselves through promotion of communication skills and empowering children to be involved in their healthcare.6 111–113 More recently, there appears to be increasing efforts to involve children and adolescents in research priority setting in childhood chronic disease.114–116

The lack of clarity identified in priority setting processes raises concerns over the legitimacy and relevance of identified priorities. Consequently, this may undermine implementation efforts.5–7 29 Some studies described the strategies for implementing and measuring the impact of the research priorities identified, and this included liaising with key stakeholders, disseminating priorities through key organisations, conducting patient-centred care and translational research, monitoring the impact of priorities on grant applications and grants awarded and informing allocation of funding. We suggest that more use could be made of good practice guidelines in research and reporting of priority setting such as those proposed by the JLA and the WHO.6 29–31 117 However, we note that the JLA approach may not fit all settings, the methods may not be well suited to children and has a narrower focus (eg, does not include prevention). Nonetheless, increasing engagement of paediatric patients in future partnerships is essential in generating research priorities indicative of the needs and concerns of the end-users of the research.20

This systematic review presents an extensive summary of research priority setting in paediatric chronic disease; however, there are potential limitations. We did not include grey literature, and the majority of studies were conducted in high-income, English-speaking countries. Seven studies were published before the year 2000, and some of the research priorities may no longer be applicable.

Research priority setting in paediatric chronic disease has generated a broad range of priorities shared across multiple conditions. However, the involvement of children and parents/caregivers appears to be limited, and reporting of the priority setting process is incomplete at best. We recommend the use of the Guidance for Reporting Involvement of Patients and the Public checklist, which was developed to help improve the quality, consistency and transparency of reporting patient and public involvement in research.118 Research priority setting partnerships need to be more transparent and involve patients and parents/caregivers to help ensure that resources are directed towards the shared priorities of children with chronic disease, their families and clinicians involved in their care.



  • Contributors HLO conceptualised the study, carried out the data collection and analysis, coding of data, drafted the initial manuscript, drafted the manuscript and approved the final manuscript as submitted. AT and PL-V conceptualised and designed the study, contributed to data collection and analysis, reviewed and revised the manuscript and approved the final manuscript as submitted. AD, AJ, AM, RP, PR, SC, PHYC and JC contributed to study design, data analysis, reviewed and revised the manuscript and approved the final manuscript as submitted. MW contributed to study design, reviewed and revised the manuscript and approved the final manuscript as submitted. BK contributed to study design, data collection and analysis, reviewed and revised the manuscript and approved the final manuscript submitted. JCC conceptualised and designed the study, contributed to the analysis, reviewed and revised the manuscript and approved the final manuscript as submitted. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

  • Funding HLO received a scholarship from the Better Treatment for Kids Network funded by NSW Health. AT is supported by the National Health and Medical Research Council Fellowship (1106716). Professor Wake was supported by NHMRC Senior Research Fellowship and Cure Kids New Zealand. Research at the Murdoch Children’s Research Institute research is supported by the Victorian Government’s Operational Infrastructure Program.

  • Competing interests None declared.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.