An empirically based conceptual framework for fostering meaningful patient engagement in research

Abstract Background Patient engagement in research (PEIR) is promoted to improve the relevance and quality of health research, but has little conceptualization derived from empirical data. Objective To address this issue, we sought to develop an empirically based conceptual framework for meaningful PEIR founded on a patient perspective. Methods We conducted a qualitative secondary analysis of in‐depth interviews with 18 patient research partners from a research centre‐affiliated patient advisory board. Data analysis involved three phases: identifying the themes, developing a framework and confirming the framework. We coded and organized the data, and abstracted, illustrated, described and explored the emergent themes using thematic analysis. Directed content analysis was conducted to derive concepts from 18 publications related to PEIR to supplement, confirm or refute, and extend the emergent conceptual framework. The framework was reviewed by four patient research partners on our research team. Results Participants’ experiences of working with researchers were generally positive. Eight themes emerged: procedural requirements, convenience, contributions, support, team interaction, research environment, feel valued and benefits. These themes were interconnected and formed a conceptual framework to explain the phenomenon of meaningful PEIR from a patient perspective. This framework, the PEIR Framework, was endorsed by the patient research partners on our team. Conclusions The PEIR Framework provides guidance on aspects of PEIR to address for meaningful PEIR. It could be particularly useful when patient‐researcher partnerships are led by researchers with little experience of engaging patients in research.


| INTRODUCTION
Patient engagement in research (PEIR) involves patients or their surrogates undertaking roles beyond those of traditional study participants along the continuum of the research process, including knowledge translation. This practice enables patients to contribute their perspectives to research. [1][2][3][4][5] Integrating a patient perspective is meant to influence the creation, dissemination and use of evidence-based knowledge to reflect and meet the health-care needs and preferences of patients. 6 PEIR could therefore potentially increase the quality and appropriateness of research and ultimately improve health-care services. 7 The term patient and public involvement, commonly used in the UK, 6 is a synonym of PEIR.
The increased popularity of PEIR in the last 15 years coincides with a greater awareness among funding agencies that patients are key stakeholders of health research. 6,[8][9][10] Its impacts have been categorized within three-value systems that cover the "moral, ethical and political concerns," "consequences" and "process" of PEIR. 7 Emerging evidence supports the theory that engaging patients in research can improve the quality and relevance of research. 11,12 Unfortunately, there are no controlled evaluations of methods for PEIR. 12,13 This is possibly because there is limited conceptualization or theoretical development that would underlie evaluation tools, which has hindered its evaluation. 12,14 Recent systematic reviews on the engagement of patients as research partners have highlighted that the practice was generally tokenistic, poorly reported and under-reported. 12,15 The literature provides several key attributes of PEIR as guiding principles and components of models. 1,6,9,10,16 The guiding principles for fostering meaningful engagement of patients include involving patients early (contributing to research questions and design), inclusiveness, co-learning, co-building of knowledge and providing support. 6,9,10 Two models, one in arthritis and the other in breast cancer, have been identified in the earlier literature in this field, 1,16 both of which were developed by teams of patients and researchers using their own experiences of engaging in research. 1,16 Neither of these models, however, was originally developed through empirical studies. Using an empirical approach can advance the conceptualization of PEIR by systematically collected data to build on anecdotal knowledge.
While we found no studies explicitly exploring this phenomenon, "meaningful PEIR" does appear frequently in the health-care literature. 9,10,17,18 For example, besides attributes already mentioned, the National Health Council asserts that meaningful PEIR should benefit both patient partners and researchers and be an informative and constructive process. 18 Patient partners are the patients and informal caregivers, such as their family members and friends, who engage to provide a patient perspective in health research. One empirical study developed a conceptual framework that outlines the broader phenomenon of successful interprofessional collaborations in health-care research but does not have a complete focus on the role of patients. 19 The value of a conceptual framework is that it links "concepts that together provide a comprehensive understanding of a phenomenon" (p. 57). 20 This study aimed to develop a conceptual framework for meaningful PEIR from a patient perspective. The practical value of this framework would include the use of its components to guide the planning, implementing and evaluating of PEIR. We formed a researcherinitiated collaboration between four experienced patient partners, all of whom are living with arthritis (AMH, AMM, KE and SM), and three health-care researchers (CBH, CLB and LCL). AMH is also an experienced knowledge broker and a registered physiotherapist.

| METHODS
We conducted a qualitative secondary analysis of interview data from a study involving patient partners using a thematic analysis approach. The patient partners on our team were female adults from

| Data analysis
Data analysis was guided by Attride-Stirling's six-step thematic networks analytical technique to code and organize the data, and abstract, illustrate, describe and explore the emergent themes. 31 This technique allowed us to focus on identifying concepts that form unique and interrelated themes of meaningful PEIR. 31 Because the patient partners in the current project were APAB members, to protect the confidentiality of their APAB peers they did not review the full transcripts but were presented with critical excerpts from the interviews so they could question and confirm themes and contribute to the framework development. NVivo software (version 11, QSR International Pty Ltd, Burlington, MA) was used to manage the data.
The framework development involved three phases ( Figure 1): 1. Identification of themes.

2.
Development of the conceptual framework.

3.
Confirmation of the conceptual framework.

| Phase 1: Identification of themes
This phase consisted of three steps: developing and applying a coding framework, abstracting the basic themes and abstracting the or- the coding process more manageable. 31,32 Intercoder reliability for each of these eight codes was conducted by the two coders using the approach outlined by Campbell et al. 33 A new transcript was then coded, and proportional agreement between the coders was calculated for each of the eight codes. Negotiated agreement was calculated when the 80% agreement threshold set for adequate intercoder reliability was not reached. 33 These negotiations provided mutual understanding of each code by refining the code's definition and recoding disputed statements. This assessment was repeated with another new transcript to ensure acceptable levels of agreement. The reliable coding framework was then applied to all 18 transcripts by both coders, and any newly emergent codes were iteratively applied to all the transcripts.
After we incorporated feedback on our initial coding framework from LCL and one interviewer, other team members reviewed the refined coding framework via email. Discussions in a subsequent research team meeting led to further refinement of the codes and their definitions. The patient partners then reviewed the final coding framework to ensure it was comprehensive and reflected their experience as patient partners. 32 Next, basic themes were created as statements by abstracting the central ideas in the portions of transcripts corresponding to lowerlevel codes. The basic themes were then arranged into clusters of similar ideas and abstracted as organizing themes. The central idea across the organizing themes was abstracted into a global theme. The basic themes were the lowest order themes derived from the interview data, while organizing and global themes were successively higherorder themes formed by clustering and abstracting the key concepts. 31 The themes were then diagrammed as a thematic network, with the basic themes omitted to streamline the diagram.

| Phase 2: Development of the conceptual framework
This phase involved two steps. CBH explored and interpreted the relationships among the organizing themes by comparing interview data and descriptions across the themes and assessing the relative importance, if any, of the themes. 32 Our patient partners were then provided with several quotes from the transcripts and selected salient quotes corresponding to each organizing theme. CBH used these salient quotes to summarize the organizing themes and then compared the emergent conceptual framework with the existing guiding principles, frameworks, models and recommendations related to PEIR.
This step allowed for broadening the interpretation of the conceptual framework beyond our interview data. 31

| Phase 3: Confirmation of the conceptual framework
In this phase, all team members critically reviewed the framework.
We refined the theme descriptions and documented our findings.
Successive drafts of this paper were reviewed by all patient partners and researchers to ensure agreement on the presentation of the conceptual framework.

| RESULTS
The  Table 2 and Supplementary 2).

| Procedural requirements
"Procedural requirements" refers to the procedural details involved in managing the inclusion of patient partners in a research project to ensure their experiences are rewarding and productive. Participants wanted to contribute their experiences, knowledge and skills to projects that they found interesting, and wanted to be informed about the expectations associated with their roles as patient partners. One participant shared her thoughts on prerequisites for patient partners: members enabled participants to undertake data collection and analysis and become more fully engaged in the research process.
They wanted to be able to express their views without unnecessary restrictions. Some participants also spoke about wanting to be reimbursed and compensated (eg honoraria) for their contributions.
Compensation would demonstrate that they are valued. Challenges relating to the scope of patient partners' roles were also addressed.
One participant highlighted differences in the roles of patient partners and researchers by noting the importance of researchers' roles in maintaining the scientific quality of the research.
Analysis of the key publications corroborated these views.
Specifically, publications including guiding principles on PEIR projects and systematic reviews indicated that projects should clearly outline roles, goals and expectations for team members at the outset. 6,13,15,16,24,25 Furthermore, the publications recommended that patient partners should engage continuously from a project's inception. This practice appeared to help patient partners become gradually more engaged and comfortable with the research process and environment. 30 Two publications supported the use of plain language and minimization of undefined acronyms and technical terms. 24,29 Several publications also indicated that research projects should involve a diversity of patient perspectives, and that more than one patient partner should be invited to each project so that they could learn from and support each other. 13 The sentiments of participants concurred with the key publications, which support scheduling meetings at times convenient to patient partners and ensuring sufficient time for contributing, given their health and practical needs. 13,23,29,30 This might require research leaders to understand and take into account patients' health-related and practical restrictions to remove barriers and ease regulations to facilitate contributions from patient partners. 16

| Contributions
This theme pertains to the roles of and tasks assumed by patient partners. Participants expressed wanting to contribute their perspectives and experiences to research:

One way that I could help is to be involved in a research and impart my experiences and express my ideas and opinions to the researchers and hope that they listen and take into account what I went through as a legitimate contribution to their research. (Jan)
Participants' contributions to research reflected the many dimensions of their lives and sometimes extended beyond sharing their health and health-care experiences to include their professional and personal skills. For example, several participants spoke about assisting in the development of plain language summaries with researchers to disseminate research findings, which required critical thinking and communication skills. Some participants found constructive feedback important because sometimes they initially felt uncertain about their ability to contribute.
The concept of contributions was also addressed in the key publications. Specifically, one discussed the time required for patient partners to be involved in research activities and noted that there was no guarantee of a successful partnership or proper compensation. 26 They corroborate the importance of giving patient partners constructive feedback on their contributions. 6,10,25,30

| Support
Participants indicated that they received valuable financial and skills/ instructional support. They valued financial support that covered their engagement-related expenses, such as attending meetings. Skills/ instructional support provided training to understand the language and processes of research. One participant shared her view on receiving training on data analysis:  The key publications corroborated that participants wanted a reciprocal relationship with other team members (researchers, research staff and patient partner) 16 and valued positive interactions that establish mutual respect, trust and moral support. 9,15,26 The importance of in-person meetings was substantiated by the key publications. 23,30 For example, Forsythe et al 23

| Research environment
This theme conveyed participants' emphasis on the importance of having a positive and inclusive organizational/team culture that allows patient partners to feel comfortable and accepted as equal team members working together. Inclusivity is conveyed, for example, by providing equipment to enable patients to connect remotely and avoid commuting, and routinely inviting comments from all team members via phone or video, or in person. A positive and welcoming research environment fosters a feeling of connection with no emphasis on maintaining power or hierarchical differences among members: This theme was a common concept in the key publications. 9,16,25,30 For example, one guideline recommended, "Do something with the input given by the partners. Listening only is not enough." 30 Three of the key publications also recommended compensating patient partners as a means of acknowledging them. 6,25,30

| Benefits
It was important to participants that they benefitted from their engage-

It's [a] very fulfilling feeling. (Jan)
The key publications highlighted that patient partners engaged when benefits outweighed the risks and that patient partners should learn how to engage in research. 6,16,19,24,29,30 Also, Ahmed et al 24 highlighted the role of learning in building capacity among researcher or patient communities: "Investigators and the community partner to learn from each other and share expertise and knowledge" (p. 1382).

| DISCUSSION
This study revealed eight themes that underscore what patient partners valued as important to PEIR. This study is among the first to use empirical data from a sample of patient partners to provide a conceptualization of meaningful PEIR. The results expand on current understanding in the literature by providing a conceptual framework based on a patient perspective.
The PEIR Framework extends beyond the interviews by incorporating, streamlining and orienting published attributes of PEIR towards a patient perspective. 1,4,6,[9][10][11]13,15,16,19,[23][24][25][26][27][28][29][30] 16 The FIRST model instead emphasized identifying projects that would benefit from patient partners' contributions and indicated that when patient partners learn, the project ultimately benefits. 16 The summary of existing guiding principles of patient engagement by Kirwan et al 13 also does not highlight benefits to patients beyond co-learning. Soever et al 19 reported that learning from research team members was a reason for their participants' engagement in research, but did not include it in their "conceptual framework for interrelational health-care research." To our knowledge, the PEIR Framework is the first framework for PEIR to use empirical data derived exclusively from a patient perspec-  19 The idea of "common interest(s)" is reflected in our procedural requirements for having the research topic and patients' interests aligned, and patients having the capacity to contribute.
Both frameworks present comparable ideas. However, a key difference is that the patient perspective is a defining feature of the PEIR Framework.
The PEIR Framework therefore orientates the reader to what is important to patient partners for meaningful PEIR to be achieved.

| Practical value
The PEIR Framework (Table 2) could be applied by using its eight themes to guide the development of an engagement plan. The specific elements of the themes can be co-constructed and agreed on by the research team members. These elements would then guide PEIR and be updated when necessary. The elements could form items to evaluate the extent to which meaningful PEIR was accomplished.

| Limitations
Participants from the interviews lacked diversity in gender, education and diagnosis (all had arthritis as a primary diagnosis), and did not include informal caregivers (eg family members and friends) of patients.
Our participants' views primarily reflected the context of engagement of patients through an APAB and research project teams led or co-led by researchers at a single research centre for arthritis research. This framework therefore might not reflect other contexts of PEIR, such as patient-directed research. The PEIR Framework does, however, reflect the rich diversity of experiences and views of our participants who had experiences with varying degrees of PEIR, and so could be relevant to other patient communities as many of the concepts were also reported in the key publications.

| Next steps
We speculate that there are both common and unique elements of meaningful PEIR across demographic characteristics of patient partners. The PEIR Framework could be studied with a demographically diverse sample, to further elucidate its general structure and nuances across patient communities, and to determine elements of each theme that are applicable for planning, implementing and evaluating PEIR in various contexts. For example, the level and type of support needed might vary by research experience or education level of patient partners. Further, the PEIR Framework is currently being used to develop an outcome measure for this phenomenon.

| CONCLUSIONS
This novel conceptual framework can be used to guide teams in developing strategies for engaging patients in research activities.
Our findings specifically offer an orientation on what is important to patient partners, including deriving benefits and feeling valued.
The PEIR Framework could be particularly useful when patientresearcher partnerships are led by researchers with little experience of engaging patients in research, but it needs further validation.