Child/youth, family and public engagement in paediatric services in high‐income countries: A systematic scoping review

Abstract Background Patient and public engagement in paediatric health‐care decision making is under‐researched, and there is a lack of systematically reviewed literature in this area. Objective To examine the extent, range and nature of published research investigating the engagement of children/youth, families and the public in paediatric service improvement, to summarize key aspects of the research identified and to identify gaps to help inform future research needs. Methods Literature was sought in MEDLINE, EMBASE, PsycINFO and CINAHL. Eligible articles presented research focused on patient, family and public engagement strategies in the paediatric health‐care setting. Two reviewers extracted and charted data and analysed findings using a descriptive numerical summary analysis and a thematic analysis. Results From 4331 articles, 21 were eligible. Most were from the United States. The majority of studies were undertaken in hospital settings and used quantitative methods. Various patient and public engagement strategies/interventions were examined, including shared decision‐making tools, questionnaires, youth councils/family advisory groups, patient portals and online networks. Most of the studies examined child/youth/parent satisfaction, with fewer investigating treatment outcomes or service improvement. The majority of studies investigated an engagement strategy at the ‘individual treatment’ level of engagement. Regarding the continuum of engagement, most of the studies were at either the ‘consultation’ or ‘involvement’ stage. Conclusion Future research needs to focus on the investigation of engagement strategies delivered in primary care, and the use of more qualitative and mixed methods approaches is recommended. There is a gap in the area of engagement strategies directed towards ‘service design and resources’ and ‘macro/policy’ levels.


| INTRODUC TI ON
There has been increasing recognition of the need for patient and public engagement in shaping health services. [1][2][3] The rationale behind enhancing patient and public engagement is to improve the quality of services, improve health service accountability and responsiveness, and improve health outcomes. 1,4,5 Patient and public engagement is complex. This complexity is evident given the existence of different definitions, theories, strategies of engagement and differences in how engagement strategies are applied in different health-care settings. [6][7][8] Barello  which methods or strategies are the most effective. 4,9 There are different theoretical models of engagement, most of which specify levels and domains of engagement. Arnstein's 10 ladder of participation is among the most popular models of participation (the term engagement is used interchangeably with participation in this review). Arnstein 10 proposed an engagement model with a ladder of 'Eight Rungs' with ascending levels of engagement. The first two rungs, 'Manipulation' and 'Therapy', are considered non-participation as they relate to persuading citizens to follow existing plans or diverting citizens from important issues. 10 The next set of rungs are 'Informing', 'Consulting' and 'Placation', and are seen to represent tokenistic forms of participation. 10 The higher rungs are 'Partnership', 'Delegated Power' and 'Citizen Control', which are intended to represent true engagement. 10 Although a popular model, criticisms have been made of the linear conceptualization of Arnstein's model. [11][12][13] There are a number of other models for engagement proposed by different authors, 6,[14][15][16] all informed by Arnstein's earlier work.
Recent models describe engagement as a 'continuum' offering a set of choices rather than citizen control as the only ideal goal. 9 Charles and DeMaio 14 describe a multidimensional framework based on decision-making domains, role perspectives and levels of engagement in health-care decisions. The first dimension refers to types of healthcare decision-making contexts or domains, ranging from individual-level treatment to service delivery and organization, and finally to broad system-level decision-making contexts. 14 The second dimension focuses on two alternative role perspectives participants can adopt in health-care decision making: as users of health services or with a public policy perspective. The third dimension depicts levels of engagement in health-care decision making, with Arnstein's ladder simplified into three categories: consultation, partnership and lay control. 14 There is growing recognition of the importance of children's rights, and their need to participate, including health-care decision making. 17 Studies suggest that children can provide valuable perspectives on their experience of local health services. 18 Similarly, families (and/or legal guardians) are involved in their child's care and in turn are able to share their knowledge with professionals, and their participation can provide a more holistic picture of the child and the way services are delivered. 19 Engagement in paediatric health-care decision making can be especially complex due to the interaction between three types of participants: children, their parents and health professionals. 20 Although a number of systematic and narrative reviews have been published in the area regarding various aspects of patient and public engagement more broadly, 17,18,21,22 there is an absence of consolidated and systematically reviewed literature examining patient and public engagement in paediatric health services specifically.
According to Anderson et al, 23 'scoping studies are concerned with contextualizing knowledge in terms of identifying the current state of understanding; identifying the sorts of things we know and do not know'. Scoping studies are particularly valuable in areas of emerging interest, such as patient and public health-care engagement. 24 The goals of this scoping review are to examine the extent, range and nature of research investigating strategies to engage children/youth, families and the public in paediatric service improvement, to summarize key aspects of the research identified and to identify gaps to help inform future research needs. 25 Specifically, this review aims to: (a) identify peer-reviewed literature presenting empirical research investigating strategies to promote engagement in paediatric services; (b) describe the range, strengths and weaknesses of engagement strategies empirically examined in the identified publications; and (c) to identify any gaps in knowledge about patient and public participation in paediatric services to help inform future research needs.

| ME THODS
We used a framework recommended for conducting scoping studies. 24,25 This involved the following stages: identifying the research questions; identifying relevant studies; study selection; charting the data; collating, summarizing and reporting results; and consultation.
Additionally, we followed the PRISMA Extension for Scoping Reviews (PRISMA-ScR) checklist. 26

| Identifying the research questions
To capture a broad range of literature, but at the same time to ensure a manageable volume of literature, we developed research questions which were broad with 'a clearly articulated scope of inquiry.' 24 Refined by using the Population, Intervention, Comparison, Outcome, Context (PICOC) question formulation process advocated by the Center for Evidence-Based Management 27 (see Supinfo S1), the following review questions were used: • How have patients, families and the public been engaged in paediatric service improvement activities in high-income countries?
More specifically, this review aimed to answer: • What is the extent, range and nature of research related to patient, family and public engagement in paediatric services?
• What are the patient, family and public engagement strategies implemented in paediatric services?
• What are the experiences/outcomes/impact of patient, family and public engagement strategies in paediatric services?

| Identifying relevant studies
Four electronic databases (MEDLINE, EMBASE, PsycINFO and CINAHL) were used to capture literature focused on engagement in paediatric services. Different search terms (keywords or phrases) related to the concepts of: (a) patient, family and public engagement, and (b) paediatric health-care settings were used; then, search strings/terms were developed and used by combining the terms according to 'Boolean logic'. 28 The complete list of the terms used to search the four databases (identical or slightly variant versions employed in all databases) is included in Supinfo S2. Where possible, Medical Subject Headings (MeSH) terms were used while searching the databases. However, available MeSH terms were insufficient, so free-text words were also used.

| Inclusion criteria
Guided by the review questions, we included research papers published in peer-reviewed journals which presented research focused on patient, family and public engagement strategies in the paediatric health-care setting including the implementation, experience, effectiveness, process or outcomes. Only papers written in the English language and published between January 1990 and 6 December 2017 were included. Additional inclusion criteria included studies conducted in high-income countries as defined by the World Bank. 29 We excluded studies not focused on paediatric services. Review articles, discussion papers, think pieces, commentaries and editorials were also excluded. Papers focused on patient, family and public engagement in research projects, paediatric patient satisfaction papers (unless explicitly used by a service to enhance engagement), and papers not available in full text were also excluded.

| Study selection
Titles and abstracts of all articles found were uploaded into EndNote to form a database. All the duplicates were then removed. Following this, articles were uploaded into Rayyan QCRI, a web/mobile application that assists with screening, to conduct the remaining review process. 30 After a review of all titles and abstracts by one reviewer (GG) using the inclusion/exclusion criteria, full-text versions of all potentially eligible peer-reviewed papers were accessed. Full-text articles were independently screened by two reviewers (GG and AR).
Any disagreement between them over the eligibility of particular studies was resolved through discussion with a third reviewer (SD).

| Data charting
Two reviewers (GG and AR) independently extracted data using a standardized data charting form, and discrepancies were identified and resolved through discussion with third reviewer SD. Extracted information included author(s); title and aims of the study; study methodology (design, methods of data collection and analysis procedure); setting; study population; details of the engagement strategies; and main findings and outcome measures.

| Collating, summarizing and reporting findings
A descriptive numerical summary analysis and a qualitative thematic analysis of the papers eligible for inclusion were conducted. 24 Information for each study regarding aims, setting, design, sample, engagement strategies (types of interventions and/or activities undertaken), primary outcome(s) and main findings was summarized narratively to describe the characteristics of included studies (Table 1). In order to explore the range of engagement strategies implemented to date, studies were further examined and classified using qualitative thematic analysis 24 according to the levels and continuum of engagement ( Table 2; GG and SD). This framework has been adapted from the work of Charles and DeMaio, 14 Ocloo & Matthews, 16 Arnstein 10 and Carman et al. 32 Levels of engagement utilized in each study were also classified according to whether engagement strategies were targeted at the individual/treatment level, the service design and resource level, or the macro/policy level. For the continuum of engagement, 'consultation', 'involvement' and 'partnership and shared leadership' were used (see Supinfo S3 for operational definitions of these terms). Finally, the settings in which engagement occurred were identified, including whether strategies were implemented in hospitals, primary care settings or the community. This was done to detect gaps in the evidence base regarding types of engagement strategies investigated, levels of engagement targeted and settings in which engagement strategies have been delivered, providing clear opportunities for future research.

| Consultation
Preliminary findings of the review were shared with key stakeholders, including academics and end users (such as policy and health managers) at a University of Otago symposium to increase the relevance of the findings. 33 The findings were well-received by the audience, and their recommendation to include additional references in the discussion of this manuscript was adopted. To test whether parents exposed to a communication skills intervention would participate in a pre-surgical consultation more than parents in a control group To determine whether parents who deliver asthma treatments in a paediatric emergency department report better adherence to inhaler use at home compared with parents whose children undergo standard care

| Literature search
A total of 4331 citations were initially obtained. After restricting the search to the availability of an abstract and to English language sources, 3492 citations remained, of which 827 were removed as they were duplicates. After title and abstract review of the remaining 2665 papers, 275 papers remained. Of these, 21 met the inclusion criteria after fulltext reading and data extraction (Figure 1). We summarized an overview of the principal findings for each paper (Table 1). When relevant, we cited either the paper ID from Table 1 or cited the paper directly.

| Continuum of engagement
The majority of studies included in this review (n = 17) investigated an engagement strategy at the 'individual treatment' level of engagement ( Table 2). Few studies (n = 4; 5, 6, 8 and 15) were found involved in activities, ranging from developing patient bedside booklets to advising on menus and websites. 38 Other examples include involvement in activities such as brochure development, and obtaining patients' and parents' perspectives on cleanliness and privacy of services. 39,40 Regarding the continuum of engagement, most of the studies (n = 20) were categorized to the 'consultation' or 'involvement' stages. Only one study 41 was identified as existing at the 'partnership and shared leadership' stage of the continuum ( Articles included for analysis: 21 Included the involvement stage on the continuum (1, 3, 4, 5, 6, 9, 10, 11, 13, 18, 20 and 21). For example, Benjamin et al 36 studied family-initiated dialogue about medications and health-care team responses during family-centred rounds (FCRs) to understand the potential for FCR to foster safe medication. We categorized the engagement process in this study as involvement as this was looking for child/parent questions and/or discussion in videoed FCRs leading to a change in treatment for the patient. Many other studies showed progress to achieve greater engagement is slow and patchy, and often concentrated at the lower levels of the continuum. 16,45,46 Furthermore, others have found professionals and organizations can, in practice, control the decision making in many contexts. 6 However, the goal is not always to move to a higher level of the continuum. There is wide variation in preferences for participation in decision making which can be determined by capabilities, resources, types of services/cares and the needs of the particular decision-making situation. 6,16,32,47 Hence, it is important to consider the preferences and willingness of patients and the public to participate, with efforts directed away from tokenistic strategies towards partnership approaches to participation. Our review suggests studies also tended to explore strategies that were implemented in the hospital context. There appears to be an absence of studies that aim to promote children/youth and family engagement in other healthcare contexts, for example primary care.
The majority of included studies employed a quantitative design; only a few studies used mixed methods to evaluate engagement strategies. While past reviews have found research in patient and public engagement to be dominated by qualitative methods, 4 fewer qualitative studies in our review might be due to our inclusion criteria. One of our inclusion criteria was that the papers reported can also influence what outcomes/impacts are sought. Engagement efforts can be directed towards achieving intrinsic goals (engagement for the democratic process, accountability, etc) rather than focusing on specific health outcomes. 42 The gap in evidence of engagement in health service development and improved health outcomes needs to be addressed, especially for communities who may not be as well-engaged in existing health services.
The results of our scoping review are subject to certain limitations. First, the emphasis of the scoping review on peer-reviewed publications only may have missed patient and public engagement studies reported in grey literature. It would be worthwhile to further investigate the grey literature, given its potential to identify additional engagement strategies employed in the paediatric context. Secondly, there was limited discussion about the continuum of engagement found in the identified papers. Hence, our categorization of the studies across the continuum needs to be treated with caution. However, the research team made genuine efforts to analyse the studies as accurately as possible. We used clearly defined guidelines to categorize the studies along the continuum, and the analysis was conducted by two researchers independently. Any disagreement between them regarding the eligibility of particular studies was resolved through discussion with the research team. In the context of the apparent literature gap between the availability of theoretical literature regarding depth or continuum of engagement and the application of such a framework in empirical studies, our attempt to categorize studies across the continuum is a strength of the review.

| CON CLUS ION
This review found that literature related to patient and public engagement in paediatric services has mostly emerged since the year 2000, and with a hospital and quantitative focus. Studies tended to investigate an engagement strategy at the 'individual treatment' level, and most studies were at 'consultation' or 'involvement' stages of the continuum. We suggest future studies investigating engagement strategies be undertaken in primary care and that qualitative and mixed methods approaches may help strengthen our understanding of engagement mechanisms. Additional studies are needed to investigate the implementation and effectiveness of patient, family and public engagement strategies in the following areas: codesign of paediatric services; development of policies related to paediatric services; feedback mechanisms to paediatric services; and approaches to improving the experiences of children and their families. Furthermore, the use of theoretical frameworks to measure the engagement process is recommended, with a focus on service design, resources and macro/policy levels.

CO N FLI C T O F I NTE R E S T
None of the authors declared any conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analysed in this study.