Approaches to optimize patient and family engagement in hospital planning and improvement: Qualitative interviews

Abstract Background Patient engagement (PE) in health‐care planning and improvement is a growing practice. We lack evidence‐based guidance for PE, particularly in hospital settings. This study explored how to optimize PE in hospitals. Methods This study was based on qualitative interviews with individuals in various roles at hospitals with high PE capacity. We asked how patients were engaged, rationale for approaches chosen and solutions for key challenges. We identified themes using content analysis. Results Participants included 40 patient/family advisors, PE managers, clinicians and executives from 9 hospitals (2 < 100 beds, 4 100 + beds, 3 teaching). Hospitals most frequently employed collaboration (standing committees, project teams), followed by blended approaches (collaboration + consultation), and then consultation (surveys, interviews). Those using collaboration emphasized integrating perspectives into decisions; those using consultation emphasized capturing diverse perspectives. Strategies to support engagement included engaging diverse patients, prioritizing what benefits many, matching patients to projects, training patients and health‐care workers, involving a critical volume of patients, requiring at least one patient for quorum, asking involved patients to review outputs, linking PE with the Board of Directors and championing PE by managers, staff and committee/team chairs. Conclusion This research generated insight on concrete approaches and strategies that hospitals can use to optimize PE for planning and improvement. On‐going research is needed to understand how to recruit diverse patients and best balance blended consultation/collaboration approaches. Patient or public contribution Three patient research partners with hospital PE experience informed study objectives and interview questions.


| BACKG ROU N D
Patient (and family) engagement (PE) is defined as patients, families or their representatives, and health professionals working in active partnership at various levels across the health-care systemindividual care, organizational design and governance, and systemlevel policy-making-to improve health and health care. 1 PE in health-care organization planning and improvement (henceforth, hospital PE) is a growing practice that can lead to a range of beneficial impacts such as educational tools for patients, programmes and services tailored to patient needs and preferences, enhanced patient experiences and better clinical outcomes such as reduced admissions. 2,3 However, many barriers can result in token PE, and little or no service improvement. For example, a systematic review (26 studies [2000][2001][2002][2003][2004][2005][2006][2007][2008][2009][2010][2011][2012][2013][2014][2015] found that key barriers resulting in token PE included uncertainty among patients about their role and resistance from clinicians to working with patients. 4 Another systematic review (11 studies 2003-2012) found that patients were typically consulted after decisions had been made, which did not lead to improvements. 5 More recently, a systematic review of 42 reviews (up to 2018) identified numerous organizational barriers of PE: knowledge, attitudes, expectations, communication, financing, resourcing, training, patient/family recruitment and representation, and addressing power dynamics. 6 By synthesizing research to date, these reviews identified gaps in knowledge about how to optimize PE in health-care organization planning and improvement including strategies to capture diverse perspectives and approaches to engage patient (and family) advisors. It is not clear whether more active engagement approaches involving collaboration (patients and providers working together to create solutions) are necessary for all planning and improvement decisions and whether collaboration always leads to improved programmes or services, patient experiences or clinical outcomes compared with less active engagement approaches such as consultation (patient opinions or ideas sought via survey, interview or focus group). 2 It has been proposed that employing a 'mosaic' of engagement approaches is best because it alleviates the expectation that a few select patients can represent the voices of all patients, and that including many voices through different types of engagement allows for a more robust understanding of patient needs and preferences. 7,8 While PE is needed in all sectors, data on approaches in the hospital sector remain limited. [4][5][6] Hospitals provide inpatient, outpatient and emergency services, and account for the largest share of health spending in many countries. 9 In a scoping review, we included only 10 studies published in 2016 or earlier that focused on PE in hospital planning and improvement. 3 Included studies provided little detail about precisely how patients were engaged. For example, a survey of hospital quality managers found that 50% of hospitals engaged patients, and in 65% of those hospitals, patients were members of quality committees, but the survey did not gather specific information such as mode or frequency of patient engagement, what information they contributed, and how it was used and with what impact. 10 Given little evidence-based guidance on how to best translate the patient voice in improving hospital services, experiences and outcomes, the purpose of this study was to generate insight for optimizing PE in hospital planning and improvement. The objective was to explore approaches and strategies used to engage patients in hospitals recognized for PE capacity including infrastructure and activities. Those best practices could be used in future by hospitals to develop their capacity for PE in organizational planning and improvement decisions.

| Approach
We chose a qualitative research design to thoroughly explore PE practices and conducted qualitative interviews with individuals involved in PE at hospitals with high PE activity. 11 We employed a qualitative descriptive approach, which does not test or generate theory, but instead explores views and experiences to identify barriers to, and suggested solutions for improving health services. 12 We complied with standards for reporting qualitative research and enhancing rigour. 13

| Sampling and recruitment
We used purposive sampling to recruit individuals whose PE views and experiences might vary by role (managers responsible for PE, or patients/family or clinicians involved in at least one PE project), type of hospital (<100 beds, 100 + beds, teaching) and health-care region in Ontario, Canada. We also used snowball sampling by first interviewing PE managers, who referred us to patients/family and clinicians. We recruited participants from hospitals with high PE capacity, identified by a survey of hospital PE managers that we had administered in the year prior to these interviews. 15 High PE hospitals were those that featured PE in planning and improvement activities across multiple clinical and corporate departments and employed a variety of engagement approaches. We aimed to recruit 1 PE manager, 2 patient/family and 2 clinicians from 2 hospitals of each type for a minimum total of 30 interviews. We first contacted PE managers by email on 13 January 2020 and closed recruitment on 16 July 2020. Sampling was concurrent with data collection and analysis, and ceased when, through discussion, the research team agreed that thematic saturation was achieved.

| Data collection
We conducted interviews by telephone between 21 January and 16 July 2020. NA (MPH, Research Associate) and ARG (PhD, Senior Scientist/Professor) jointly conducted the first two interviews, independently reviewed transcripts and then met to discuss and refine wording of interview questions. NA subsequently conducted all interviews. Interview guide questions (Data S1) aimed to elicit the rationale for, and barriers of, PE approaches, distinguished according to prior research as involving more intensive (collaboration: joint decision making on project teams or standing committees) versus less intensive engagement (consultation: gathering ideas or feedback using surveys, interviews or focus groups). 2 Questions were reviewed and refined by the research team prior to use. We first asked participants to describe a hospital planning or improvement activity that engaged patients in some way that they were involved in or aware of (reported elsewhere), and refer to that as a reference for

| Data analysis
We employed content analysis to identify themes inductively through constant comparison and used Microsoft Office (Word, Excel) to manage data. 11 NA and ARG independently coded the first two interviews and then discussed coding to develop a preliminary codebook of themes and exemplar quotes (first level coding). NA coded subsequent interviews to expand or merge themes (second level coding). NA and ARG met on two subsequent occasions to review, discuss and refine coding. We tabulated data (themes, quotes) by participant role and hospital type to compare themes. The research team reviewed themes and quotes. We used summary statistics to describe participants and text to describe key themes.

| PE was embedded throughout organizations
Participants said that patients were involved in decision making for all hospital activities.
We always have a patient or two involved in everything that we do (038 exec teaching) We sit on all committees in the hospital (002 patient/family <100) PE was considered important because it allowed health-care workers to see issues with a patient lens, resulting in better understanding of patient needs and preferences, and services reflecting those perspectives.
This was believed to lead to improved patient experiences and outcomes.
PE was therefore described as essential to patient-centred care. We have patient and family advisors embedded in… corporate committees but also at the program level committees (032 exec 100+)

| Project teams
Participants also said that patients were included in project teams with finite timelines formed to address specific initiatives at both the clinical unit/department and corporate levels. These were sometimes referred to as working groups or steering committees.
We had a small working group that consisted of about 6-people including a patient and family advisor that helped drive and steer the organization in terms of developing the strategic plan (004 PE manager teaching) The goal of the stroke council working group was to implement a stroke unit. We had monthly meetings that I was involved in and my role is to bring the patient perspective to these meetings (039 patient/ family teaching)

| Engagement approaches
Participants largely employed either collaboration, consultation or a blended approach to engage patients, and provided rationales for and examples of those approaches ( Table 2). A few participants said that the approach chosen for a PE project would depend on the situation or nature of the project including the issue the project was focusing on, the willingness and commitment from those who would be involved and project time frame. Regardless of which engagement approach was used, participants agreed that in-person interaction was preferred because it established rapport between patients and health-care workers, enabled staff to see patients as real people and nurtured an appreciation for the importance of involving patients in planning and improvement.

| Collaboration
Most projects described by participants involved collaboration approaches. Most commonly, the purpose of collaboration was to partner or collaborate with patients in creating, reviewing or editing documents or resources such as patient information handouts or videos, online educational tools, web sites or procedure consent forms. Collaboration was also employed to discuss and evaluate issues pertaining to planning or improving services or programmes, brainstorm or develop solutions for those issues, and to inform the development of innovative new programmes. Collaboration methods included monthly or bimonthly project team or standing committee meetings held virtually or in-person.
Participants who preferred collaboration offered three reasons. Some said it was the best approach for ensuring that patient perspectives were heard and integrated in decision making.
Having people at the

| Consultation
Consultation approaches were less frequently used than collaboration or blended approaches. The purpose of consultation was to gather feedback on existing or newly implemented programmes, or ideas about how to plan or improve services. Consultation methods included surveys, interviews, focus groups and post-discharge telephone calls. Participants who preferred consultation offered three reasons. Some said consultation was the best approach for reaching many patients.
We attended five or six summer festivals and choosing that style allows us to get many perspectives instead of just one or two. We had thousands and thousands of points of data (004 PE manager teaching) The broader the input, the better it is…it's getting as much information from as many people as possible

| Strategies to optimize engagement
Participants described numerous strategies employed to ensure that multiple and diverse perspectives were sought, heard and integrated in decision making (

| Prioritize what benefits many
Participants said that they reviewed a wide range of patient feedback, but prioritized ideas for planning and improvement based on what was likely to benefit the majority of individuals in the community they served. Complementary to this was the strategy of first using consultation approaches to capture a wide range of ideas from many patients, followed by collaboration approaches involving select patients to prioritize and elaborate on ideas. Both of these approaches may not capture the perspectives of underserved or marginalized community members.

| Match patients to projects
Participants described various ways of allocating patients to projects, but differed in how this was defined. Some participants said it was important to match patient experiences or characteristics to a project, while others said they deployed PFAC members with PE experience and skills to multiple projects, a strategy that might limit diversity.

| Train participants
Once recruited, patients were prepared for PE roles through general orientation and then further education in advance of assuming membership on committees or project teams to provide patients with background on committee or project activities. Some participants also said that health-care workers received training on how to effectively engage with patients.

| Ensure patient perspectives inform decisions
The most commonly mentioned factor supporting PE was organization-wide respect for patient perspectives that had developed over time such that PE was the accepted norm. Participants said their hospitals had developed a philosophical commitment that patient/family advisors are experts on the patient perspective and their perspectives were valued equally to those of health-care workers. Given that 5 hospitals had 3 to 5 years of PE experience, and 4 had 6 plus years, perhaps organizational commitment to PE may be just as or perhaps more important than length of time.
Participants described several additional strategies for ensuring that patient perspectives were heard and informed decisions.
One strategy was to include a critical volume of patients on committees or project teams so that they were not outnumbered by

| Engagement challenges and recommended solutions
When asked what they would do differently in the future to further optimize engagement, participants articulated several recommendations. One recommendation pertained to recruiting a larger pool of patients given the ever-increasing number of planning or improvement projects.
We've had to increase the number of advisors because we're being asked to be involved in many dif- Engage diverse patients Aim for diversity in characteristics We believe in making sure that the most marginalized individuals have represented voices at the table…one of the things in terms of how we've been successful with being able to recruit these types of individuals is that we have a strategy around recruiting for diversity. So we specifically are looking for folks that represent the health disparities in our community and engage them (010 PE manager 100+) Employ various recruitment strategies to achieve diversity A lot of our recruiting often times is by word of mouth and we've tried newspaper ads, the last one we got through Facebook…you know it's those kinds of things; how do you reach the biggest population? (036 patient/family <100) Patient/family advisors were largely retired persons It tends to be the retired community that comes forward to be part of the patient and family advisory committee (027 PE manager <100) Because I'm retired I'm able to give the time to things (023 patient/family 100+) Prioritize what benefits many Chose projects that benefit the majority A lot of suggestions come to PFAC, but if they are more individualized, we try to triage that because it's not about one, it's about everybody. We try to talk about who is our catchment and who are we benefiting (002 patient/family <100) Used perspectives expressed by the majority And then when the structured reviews came back from patients and their families we had to set a priority, we're going to take everything that is said by more than you know 'x' percentage of patients and we're going to use that (012 clinician teaching) Blended approach of consultation then co-design We were really trying to be driven by the data from our survey and post-discharge phone calls, and then validating that with the experiences of our patient and family partners to dig a little bit deeper on some things that would have the biggest impact on improving our results (024 PE manager 100+) Match patients to projects Deploy those with PE experience/skill Sometimes we were selected because of other projects we had worked on. I mean they [PE managers/staff] have a good sense of our skills at this point. There's a large number of patient partners but there seems to be a group that does a lot of different kinds of projects and so they know who's got good analytical skills and good communication skills. So we were sought. We were recruited specifically (014 patient/family teaching) Match patient/ family experience or characteristics to PE project And basically what they ask for generally is people who have had a background as a patient in those areas. So for instance, when things come out in the neuro area or the cancer area I wouldn't apply because my background there as a patient just doesn't exist (039 patient/family teaching) Train participants Train patient/family for role of advisors They are trained during orientation and then have, with the interest in the program, a full day of training and then continued engagement throughout the program (037 clinician < 100) We usually have an education session for about a half an hour before we get into the meeting (036 patient/ family < 100) Train health-care workers on how to collaborate with patient/family advisors The staff and leaders received training on how to effectively engage with patient partners (029 patient/ family 100+)

Ensure patient perspectives inform decisions
Include a critical volume of patient/family advisors About four years ago we established our first patient and family experience steering committee. The initial membership, the staff greatly out-numbered the number of patient family partners. Over the last few years we've decreased the number of staff on the committee, increased the number of patient family partners on the committee (024 PE manager 100+) Quorum requires at least one patient/family advisor There's usually two of us [patient/family advisors on standing committees]. One of the requirements of the [Research Ethics Board] is that to have a proper quorum you need to have one patient/family advisor at the meeting (17 pat 100+) PFAC review of standing committee or project team work I would report back to the Patient and Family Council about what is going on in the General Medicine Quality Committee…and sometimes I'd be seeking out advice (007 patient/family teaching) Patient/family advisor feedback loop When we [patient/family advisors] made those suggestions, they were taken away and then at the next meeting they would hand the draft out and we'd go over it to see which of our suggestions had been included (035 patient/family <100) decisions were not great. I wish there was more opportunity for co-design because that could have saved some problems that we were trying to solve later (014 patient/family teaching)

| D ISCUSS I ON
This study found that hospitals selected for high PE capacity had embedded PE activities broadly, engaging patients throughout the organization in many planning and improvement decisions via multiple structures including standing committees, project teams, and general and unit/department-specific PFACs. Participants most frequently employed collaboration approaches (membership on standing committees/short-term project teams), followed by blended approaches (both collaboration and consultation), and, less frequently, consultation (surveys, interviews). They described a wide range of strategies that supported engagement approaches to ensure that diverse perspectives were sought, heard and integrated in planning and improvement decisions, but faced challenges in achieving this goal.
Prior research that referred to PE largely focused on engaging individuals in their own clinical care or as members of research teams. 16

TA B L E 3 (Continued)
committees to prioritize, and then design or implement those ideas.
Research to date shows that organizational PE generates practical tools (eg patient handouts) and tailors services to patient needs and preferences, but evidence on the link between PE and improved patient experiences or clinical outcomes is limited. 2,3 Therefore, we lack knowledge on whether collaboration, consultation or a blended approach is superior. Rather than a 'mosaic' of approaches, thought by some to be ideal, 7,8 blended approaches were less frequently employed than collaboration, and typically involved consultation to first gather a range of perspectives, and then collaboration to prioritize and elaborate on ideas or feedback. Intuitively, the blended approach makes sense, but is the most time-and work-intensive.
Furthermore, prioritizing issues that are relevant to the majority by a small group of patient/family advisors may not result in services or programme that reflect the needs and preferences of marginalized community members. Thus, further research is needed to establish which approach is best suited to different types of projects, decisions or desired impacts or outcomes.
With respect to strategies that support engagement, this study revealed numerous ways to ensure that planning and improve- benefits the majority of our constituents, then we risk increasing the inequities that currently exist for underserved and marginalized populations. Our participants said there was a need to recruit a larger, more diverse group of patient/family advisors, but did not elaborate on specific strategies for doing so. Some guidance is available from research on involving persons from diverse and hard-to-reach communities in research. 23 Recommendations for recruitment included use of existing networks, consulting with the community, accessing outpatient clinics and using social media; and for supporting engagement included using culturally appropriate communication, building rapport between members, equalizing member roles, establishing trust as the basis for longlasting partnerships and establishing a diverse leadership team.
On-going research is needed to identify the infrastructure and processes hospitals must implement to have the capacity to recruit, prepare and support a large, diverse group of patient/family advisors, or, alternatively, how to efficiently and rapidly do so as the need arises.
Strengths of this research included the use of robust qualitative methods that complied with reporting criteria and standard techniques for ensuring rigour. [11][12][13][14] The research was guided by multiple points of input and review by an interdisciplinary research team that included two patient research partners with hospital PE experience. Furthermore, we interviewed persons with considerable PE expertise and experience from hospitals with proven PE capacity. Participants represented different roles (patient/ family, PE managers, clinicians) and hospital types. We must also acknowledge some limitations. Patient/family advisors were not diverse, consisting largely of retired Caucasian women, and the clinicians we interviewed included only one physician. Patient/ family advisors were largely PFAC members with limited insight on consultation approaches. All participants were affiliated with hospitals in one Canadian province; therefore, findings may not be relevant to hospitals in other countries with differing PE practices or health systems.

| CON CLUS ION
Through interviews with 40 patient/family advisors, PE managers and clinicians at hospitals in which PE had become pervasive, we identified approaches and strategies that could optimize PE in organizational planning and improvement. Hospitals engaged patients via standing committees, project teams and general and unit/ department-specific PFACs. Hospitals primarily used collaboration (membership on committees/teams/PFACs) or blended approaches, which typically involved consultation to first capture a range of ideas/feedback via surveys or interviews with many patients/family, followed by collaboration to prioritize and elaborate on the most promising ideas/feedback. Participants who employed collaboration emphasized the ability to integrate perspectives in planning/improvement decisions, while those who used blended approaches emphasized the ability to capture many diverse perspectives, and then prioritize and further develop those ideas. Fewer participants used only consultation approaches. Given that patient/family advisors were largely retired Caucasian women deployed to many projects or committees, and issues common to the majority of community members were prioritized, the use of collaboration or blended approaches may not lead to facilities or services that reflect the needs and perspectives of underserved community members. Participants described a wide range of strategies that supported engagement approaches to ensure that diverse perspectives were sought, heard and integrated in planning and improvement decisions. Given little evidence-based guidance on how best to engage patients in hospital planning and improvement, this research identified concrete strategies that can be implemented in future by hospitals to enhance their PE capacity.

ACK N OWLED G EM ENTS
We thank patient research partners Laurie Proulx, Julie McIlroy and Craig Lindsay, and Amy Lang (formerly Health Quality Ontario), and Mireille Brosseau (formerly Accreditation Canada) for helping to develop the interview questions.

CO N FLI C T O F I NTE R E S T
None to declare.

AUTH O R S ' CO NTR I B UTI O N
GRB, LM, KS, RU, WW and ARG conceptualized and designed the study. NA and ARG collected and analysed the data. All authors reviewed and interpreted the data. All authors drafted or revised the manuscript, gave final approval of the version to be published and agreed to be accountable for all aspects of the work.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data are included in the manuscript and supplementary files.