Priorities of patients, caregivers and health‐care professionals for health research – A systematic review

Abstract Background Based on subjective experience, patients can identify research priorities important for health services research. A systematic method for priority setting has been developed by the James Lind Alliance. Objective This article reviews the literature on the research priorities of patients, caregivers and health‐care professionals and presents the prioritized research themes and prioritization methods used. Search strategy Three electronic databases were searched on 22 May 2018. The search was not limited to any time period or language. Inclusion criteria The included studies reported the identification and prioritization of research priorities involving patients, relatives and caregivers. Each included paper addressed a specific ICD‐coded health problem, and at least one‐third of the sample involved in the prioritization process was affected by the health problem. Data extraction and synthesis The 10 top‐ranked research priorities were included in the thematic analysis. With an inductive approach, a system of identified themes and subthemes was developed from the research priorities. Each research priority was assigned to one research theme. Main results The priority lists of 34 publications involving 331 research priorities were included. Nine main themes represent the content of the research priorities. The most frequently represented main themes are ‘Treatment’, ‘Patients’ and ‘Health condition’. The distribution of the research priorities varied depending on the health conditions and prioritization methods. Discussion and conclusions This review provides a comprehensive overview of the overarching research themes in research priorities of affected individuals. The results can guide future patient‐oriented research.


| INTRODUC TI ON
Currently, a paradigm shift is occurring in health care and related research towards the needs of individuals affected by diseases. 1 So that their needs are recognized, the people affected by diseases are increasingly involved in different stages of the research process and acknowledged as experts with valuable-specific disease experience. [2][3][4] On the one hand, this development stems from the idea that individuals affected by diseases ultimately bear the burden of the diseases and therefore have the right to participate in the process of determining the direction of affiliated research. 5 On the other hand, if research addresses genuine problems in health care more adequately through patient involvement, scientific findings are more likely to be implemented into practice, thus leading to an efficient use of limited funding resources. 6 Developing research agendas are a key strategic factor in this context when they are used for the conception of funding programmes and research projects, as they determine the content alignment of future research. This concept applies in particular to evidence-based medicine since research is a prerequisite for the evidence and thus for the practical use of an intervention. 7 To date, the research priorities for health research have been predominantly determined by researchers and funding bodies. These research priorities are identified from a scientific or commercial point of view, and the needs of the affected people are not necessarily considered. 8 However, according to the outlined paradigm shift, an increasing number of initiatives are being developed to include patients, caregivers and health-care professionals (HCPs) in the process of identifying research priorities. 9 These research agendas have already been addressed in national research funding programmes. 10 The inclusion of patients in the process of identifying research priorities has been explicitly discussed since the 1990s. An important milestone of this development is the establishment of the James Lind Alliance (JLA) in 2004. The JLA aims to identify and prioritize unanswered research priorities for specific health conditions with affected patients, caregivers and practitioners. In addition, the initiative uses extensive networking to ensure that the research agendas that are developed are actually being pursued. 11 The JLA facilitates a specific method in which research agendas are developed by disease-specific priority setting partnerships (PSP); different stakeholders are equally involved in all stages of the prioritization process. The course of a PSP can be described in different stages. From the results, an interim shortlist of research priorities is developed. It is used as the starting position for the third stage of a PSP, where a prioritization workshop is held to develop a ranked top 10 list of disease-specific research priorities. 12 From the results, an interim shortlist of research priorities is developed. It is the basis for the third stage of a PSP, when a prioritization workshop is held. By using an adapted nominal group technique, a ranked top 10 list of disease-specific research priorities is developed. The guidebook presents the methodological basis of the JLA approach. It has been continuously revised in recent years and is now in its eighth version. 13 The JLA approach can be considered well established and is supported by research infrastructure in the United Kingdom. 14 In addition, there are further methods that have been used to identify and prioritize research priorities together with patients or the public, such as focus groups, voting exercises or citizen juries. 15 The structured JLA approach offers transparency and replicability, but also requires resources and supervision. Further methods can easily be tailored for specific groups or settings. 14 Overall, the increase in efforts to improve the quality of health care and health research by involving the affected people in the process of identifying research priorities can be assessed: a comprehensive review of this development through 2008 has been provided by Stewart and colleagues. They identified 27 papers in which patients were actively involved in the identification and prioritization of research priorities. Moreover, their analysis reported a broad variety of disease-related health topics for which patients and clinicians identified research priorities. However, neither the extent of lay involvement nor the contents of the research priorities were analysed. 9 Accordingly, the work did not provide any conclusions about which research priorities were identified by the affected individuals.

Crowe and colleagues approached this question in an analysis of 14
PSPs based on the JLA methodology. Their review focused on research priorities and how they are addressed in on-going research.
Only about 18% of the research priorities concern drugs, vaccines and biologicals, and 23% radiotherapy, surgery and devices, whereas 60% have been linked to 'other intervention' including education and training, service delivery as well as psychological and physical therapies. In contrast, actual research studies mainly focus on drug interventions. 16 On the basis of this background information, we aim to identify overarching research themes identified in prioritization studies with substantial involvement of patients affected by different health conditions and their caregivers.

| ME THODS
We conducted a systematic review guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement 17 to ensure transparent and complete reporting (Appendix S1).
The protocol is not registered.

| Search strategy
The PubMed, SCOPUS and PsycINFO databases were systematically searched on 22 May 2018. The keywords used were as follows: patients, carers, service users, clients, consumers, lay AND priorit* AND research OR James Lind Alliance, James Lind Initiative. The search strategy was adapted for the different databases by applying the respective operators, and it was not limited to any time period or language. The detailed search strategy is provided in Appendix S2.

| Study selection
The study selection process was performed in two steps. First, the titles and abstracts of identified articles were screened. Original research papers that reported the prioritization of at least five research priorities involving patients or caregivers were included.
When only priorities for a specific type of treatment (eg surgical treatments) were ascertained, the respective study was excluded.
Second, full-text articles were obtained for all studies assessed as eligible in the abstract screening process. For the full-text screening process, the following inclusion criteria were added: studies that are published in a peer-reviewed journal, studies that focus on a specific ICD-coded health problem and studies in which the identification and prioritization processes were distinct phases and finally a ranking of the identified research priorities was carried out. To ensure substantial involvement of the affected people, they had to constitute at least one-third of the sample that performed the prioritizing task for study inclusion (for details, see Table 1). Due to a lack of quality assessment tools for the prioritization studies, quality assurance aspects were addressed by the selection criteria. Two researchers (ML, MVB) screened the references and full texts. In the case of a discrepancy, a third reviewer was consulted (ALB).

| Data extraction
One author (ML or MVB) extracted information from the papers into a summary table (Table 2), and a second reviewer (ML or MVB) checked the content for accuracy. This information included data on the author, year of publication, health conditions of the participants, country in which the study was conducted, methods (identification, interim prioritization and prioritization) and the number of research priorities.

| Data analysis
For analysis of the priority lists, we used a descriptive thematic synthesis, considering suggestions for incorporating qualitative evidence into systematic reviews. 18 Researchers (ML, MV) copied all reported research priorities from the included papers into the qualitative data analysis software MAXQDA 2018. 19 The 10 top-ranked priorities of each paper were included in the further analysis. If lists displayed less than 10, all priorities were included. For synthesis of the priority lists, we used a descriptive thematic synthesis, considering the outline for the use in systematic reviews by Dixon-Woods and colleagues. 18 Using an inductive approach, two authors (ML, MV) independently assigned a code to each of the research priorities according to the coding rules (Appendix S3). After one-third of the research priorities were coded, the codes were independently conceptualized in main and subthemes.
The identified themes and subthemes created by ML and MVB were discussed and merged by the three authors (ML, MVB and ALB). ML and MVB coded the remaining research priorities independently using the merged themes and subthemes. Disagreements were again discussed by all three authors, and the identified themes and subthemes were refined in an iterative process. Finally, themes and subthemes were agreed upon by the authors and applied to the data (Appendix S4). The final themes and subthemes were used as the basis for further analyses.

| Search strategy and studies
Across all databases, the search yielded 10,572 citations, of which 8,036 remained after the duplicates were removed. After the title and abstract screening process was performed, 223 publications underwent full-text screening. Of these publications, 34 met the eligibility criteria and were included in the analysis. The most common reasons for exclusion were as follows: articles did not report original research (n = 39), and the study objective did not focus on research priorities for a specific ICD-coded health problem (n = 28) (Figure 1). Papers report original research.

IC 2
Research topics or questions are prioritized.

IC 3
The research objective relates to a specific ICD-coded health problem.

IC 4
Papers do not focus on specific treatment aspects.

IC 5
Sample or separable subsample is affected by the same health problem as a patient or caregiver.

IC 6
The proportion of affected persons in the sample is documented as well as whether they are directly or indirectly affected.

IC 7
Participants are at least 18 years old.

IC 8
Affected people are involved in identification and prioritization of research priorities.

IC 9
The prioritizing sample comprises at least one third affected people.

IC 10
Methodical identification and prioritization of research priorities as well as corresponding analysis.

IC 11
Identification and prioritization take place in two distinct phases.

IC 12
The methodical approach is reproducibly documented.

IC 13
A list of at least five research priorities is provided.

IC 14
Research questions in the lists are ranked involving affected persons.

TA B L E 2 Data extraction table
Authors, year

Number of prioritized research themes
Aliberti

| Summary of the themes in the priority lists
The research priorities were subsumed in nine identified main themes with additional subthemes (Figure 2). The identified themes and subthemes summarized the content of the included priority lists and provided an overview of the research priorities that are frequently prioritized by patients, caregivers and health-care professionals.

| Patients
The

| Health-care professionals (HCPs)
The research theme 'Health care professionals' comprised research priorities (n = 13) that focus on HCPs regarding their attitudes, roles, education, communication and effectiveness. For example, these research priorities prompted research studies that investigate whether HCPs need more training on person-centred skills.

| (Informal) Carers
Research priorities in the research theme '(Informal) Carers' (n = 5) addressed carer needs regarding education and emotional support.
Specifically, the linked research priorities questions prompted research on family-involved care and support by family and friends in situations in which the affected person has communication impairments.

| Health-care system
We

| Diagnosis
The main theme 'Diagnosis' was related to research priorities that

| Health condition
The

| Public awareness and education
Research priorities in the research theme 'Public Awareness and education' (n = 8) indicated a need for research that raised public awareness regarding the patients' situations. Specifically, the linked research priorities prompted research on effective methods of increasing awareness and educational campaigns to improve the situation of the affected individuals and to reduce stigmatization.

| Distribution of themes within priority lists
The priority lists varied in their inclusion of different main themes (Table 3)

| Distribution of themes across the methodological approaches
We identified three different methodological approaches. The JLA method was used in 17 studies. 20 More than four research priorities in two lists related to the main themes 'Treatment' and 'Patients'.

| Distribution of research themes across health conditions
When the priority lists were contrasted according to the health condition that affected the sample, different focus areas were identified. We investigated three different groups of health conditions that varied in the inclusion of main research themes.
Six studies dealt with research priorities for neoplasms (C00-D48). 21,24,28,30,43,48 The main research theme 'Treatment' was included in 100% (n = 6) of these priority lists. 'Patients' was the second most prevalent main research theme, with a prevalence of 83% Research priorities regarding the health condition of the genitourinary system (N00 -N99) were developed in four studies. 27,36,38,47 In 100% of these priority lists, the research theme 'Patients' (n = 4) was addressed as the most prevalent main research theme, fol-

Health conditions
Main research themes JLA JLA adopted other methods

Themes by disease
Mental disorders (n = 5) x 0 20 Genitourinary system (n = 4) x 25 0 Neoplasms (n = 6) x 33 33 Other diseases (n = 19) The JLA and its approach play a pivotal role in the methodology for developing research priorities together with affected individuals.
This idea is reflected in the finding that approximately two-thirds of the included studies were methodologically oriented to the JLA. The importance of the JLA method is also reflected in Yoshida's systematic review of methods used for setting research priorities with and without patient engagement. In this review, the JLA approach was the third most frequently used method. 14 The JLA is an initiative developed by the United Kingdom (UK) and supported by its government. Accordingly, 12 of the 17 included JLA PSPs were conducted in the UK, 4 were conducted in Canada, and one was conducted in Spain. Of the studies that adopted the JLA method, two were conducted in Canada, one was cross-nationally conducted in the UK and Ireland, and one study was conducted in each of the following counties: the UK, Germany and Australia.
These findings show that the JLA approach is being increasingly recognized internationally. This conclusion is especially true for the JLA-oriented studies that do not have a formal affiliation with the organization. An enabler for the implementation in the UK and Canada may be the commitment of both governments to support public involvement in their respective health research systems. 54 The international cooperation and involvement of a JLA supervisor entail increased costs affiliated with an official JLA PSP outside the UK that need to be compensated by research funding programmes.
In addition, the difference in language can be a barrier for the adoption of the JLA approach in non-English-speaking countries.

| Limitations
Our selection criteria excluded studies without substantial involvement of affected individuals. For some health conditions, the symptoms (eg severe cognitive disabilities) limit the patients' ability to participate in the research priority setting process. For example, a JLA PSP on dementia 55 was excluded from this review because too few affected persons were involved.
We attempted to adequately map and compare the relevance For the prioritization studies, no specific quality assessment tool is available. The use of a multitude of different instruments would not have been appropriate given the level of methodological heterogeneity. Quality assurance was therefore based on the selection criteria, which also addressed methodological aspects and their documentation. In addition, only pertinent literature databases were searched, and all included articles were published in peer-reviewed journals.

| CON CLUS IONS
The present review indicates that the relevance of patient involvement in the process of identifying research priorities is increasing. The On the one hand, the review indicates there is variation in priority lists, so disease-specific approaches are needed. On the other hand, the context should be taken into account, and inter-country comparisons may further identify international differences. Finally, in order to encourage patient-centred health research, identified research priorities should be considered and integrated into research activities. Strategies monitoring the uptake of research priorities and reporting their impact should be strengthened.

ACK N O LED G EM ENT
English and German language versions are available (approved text for English language journal). Open access funding enabled and organized by Projekt DEAL. [Correction added on 28 September 2020, after first online publication: Projekt DEAL funding statement has been added.]

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to disclose.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.