Engaging patients with dementia, caregivers, and clinicians to study decisions about driving

Background

While family and clinicians can help older drivers plan for the transition to non-driving, conversations about driving are logistically and emotionally complex. For people with dementia, delaying these conversations can mean they are no longer able to participate in their own decision-making. Our objective was to recruit and sustain two advisory boards of (1) community members and (2) clinicians to engage them in co-designing the approach for a feasibility study to investigate a clinical care pathway that supports earlier care planning discussions about driving for patients with dementia.

Methods

We partnered with the University of Wisconsin Community Academic Aging Research Network (CAARN) to recruit two advisory boards. The community board included Black older adults living in the greater Milwaukee, WI who care for a person with dementia (n = 5) or have dementia (n = 1). The clinical board included clinicians (i.e., MDs, APPs, RNs) with expertise in dementia care across Wisconsin (n = 6). We facilitated four meetings with each board focused on key study activities (i.e., refining research priorities, recruiting participants, meaningful outcomes, sharing results). We presented summaries at subsequent meetings to facilitate convergence between boards.

Results

We engaged the advisory boards to refine our research priorities, recruitment strategy, meaningful outcomes and, ultimately, our plans to share our findings. The advisory boards’ framing of barriers to communication and care planning aligned with existing models (i.e., transtheoretical, social ecological), informing the focus and scope of our intervention co-design approach. Boards also identified key recruitment considerations related to fostering trust, cultural congruence, and resources for clinicians to engage in study recruitment.

Conclusions

Partnering with community and clinical representatives to co-design a clinical care pathway has refined our approach to supporting people with dementia and their caregivers during transitions to non-driving. Insights gained from these advisory boards will increase the applicability of our findings to patients in diverse primary care practice settings. This work represents the early stages of developing relationships with advisory board members and will necessarily require ongoing effort to identify and respond to needs and priorities posed by advisory board members to ensure that our co-design activities are more than confirmatory steps.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40900-025-00820-1.

Plain language summary

For people with dementia, losing the ability to drive can be one of the most difficult life changes—impacting independence, daily routines, and safety. Our goal for this developmental project was to begin building relationships with representatives from community and clinical settings who have experience with discussing driving and dementia. Our long-term objective is to engage community and clinical partners in co-designing meaningful and feasible interventions to support people navigating the transition to non-driving. We worked with existing community engagement groups at the University of Wisconsin to recruit two advisory boards—one with community members with dementia caregiving experience and one with clinical team members. We held four meetings with each advisory board to discuss our research priorities, recruitment plans, what outcomes will be most important to measure in our research, and how we can effectively share what we learn with the community. Working with these community and clinical representatives mattered in ways that were personally profound and useful for ongoing research. As the principal investigator for this research, this co-design process changed my perspective about what matters most for the potential impact of this work. The advisory board members shared that this process changed their perspectives on driving and dementia. The boards offered actionable feedback on improving communication and care planning about stopping driving and study recruitment strategies related to developing trust, cultural alignment, and relevant resources for clinicians.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40900-025-00820-1.

Background

Despite driving being the primary mode of transportation for many older adults [1], older drivers with dementia experience declines in cognitive function that decrease their driving safety and eventually necessitate driving cessation [2]. Family caregivers and clinicians can facilitate planning for driving cessation, but they often delay conversations about driving and dementia because this transition is logistically and emotionally complex [3, 4]. In current practice, many family members identify a car collision or near-miss as primary reasons to start conversations about driving safety (i.e., after serious consequences have occurred) [4]. Similarly, clinicians typically wait to initiate conversations about driving safety until significant concerns about a patient’s health develop [5]. Multiple barriers lead to these delays, including lack of time and training about resources as well as worry that a negative conversation about driving could harm clinician–patient relationships [3, 6].

Older drivers and their caregivers are better able to plan for the transition to non-driving when these conversations begin earlier and occur more than once [7]. By waiting until patients have advanced to a moderate stage of dementia to discuss driving recommendations, clinicians and caregivers miss the opportunity to engage patients in shared decision-making about their future transition to non-driving. The perspectives of people with dementia, family caregivers, and clinicians are vital to developing meaningful and feasible interventions to support people navigating the transition to non-driving.

More than two decades of research have produced multiple driving cessation resources and interventions [8], including evidence-informed websites [9] and evidence-based online decision support tools [10] and coaching programs [11, 12]. However, numerous barriers to implementing evidence-based support for driving cessation remain: limited awareness of the impact of driving cessation on older drivers and caregivers, limited training to provide driving cessation support, reluctance to raise the topic based on negative experiences, and a focus only on the risks of driving without also addressing the risks of driving cessation [13, 14]. Furthermore, racial and ethnic inequities directly influence transitions to non-driving and the ability of older adults to live meaningful and dignified lives after this transition [15]. Research to develop and test driving cessation interventions has not yet investigated whether cultural adaptation and tailoring of these interventions might address inequities. Developing feasible community and clinical interventions to overcome these barriers is imperative. Applying community-engaged research practices will help to increase the relevance and feasibility of interventions early in the design process. Focused inclusion of specific groups under-represented in research such as Black older adults can also help prevent future interventions from perpetuating inequitable outcomes.

Community engagement spans a continuum of activities ranging from outreach for the purpose of informing the community to more meaningful involvement, collaboration, and ultimately shared leadership [16]. Community advisory boards are one commonly used approach in community-engaged research to formalize relationships between academic researchers and key representatives from different community organizations [17]. The relationship between researchers and community advisory boards can take a variety of forms including consulting with a community advisory board regarding individual questions related to the research process or developing longitudinal relationships to co-design a program of research from conceptualization through dissemination of findings [18].

Research co-design or co-production approaches have grown in popularity to overcome barriers to implementing relevant and feasible health care and social interventions. Further work is needed to rigorously define the mechanisms underlying these approaches, but key elements include bringing people together as active and equal partners to develop shared understanding of the problem and to iteratively and creatively develop solutions that meet the needs of the affected community [19, 20]. Co-design approaches have the potential to not only increase the validity of the research, but to yield benefit to participating community members including increased knowledge about and skills for contributing to the research process, as well as a general sense of accomplishment and increased confidence [21].

As the principal investigator for this research and first author of this manuscript, I began this developmental project as a clinician-researcher in my first year as an Assistant Professor. My prior research had been primarily quantitative, conducting secondary analyses of nationally-representative survey data to characterize associations between individual, social, and environmental characteristics associated with the transition to non-driving [22–24]. However, as a practicing primary care clinician, I am deeply motivated to work towards conducting pragmatic research that can meaningfully improve the quality and equity of clinical care that responds to community health needs and leverages community health resources. For this reason, we sought to apply a pragmatic inquiry framework [25] as we take the preliminary steps to develop a study protocol for investigating the feasibility and effectiveness of a clinical care pathway to support drivers with dementia as they navigate the transition to non-driving with their family caregivers and clinicians. Guiding principles of pragmatic inquiry that informed our rationale for and approach to using co-design included an emphasis on how our findings could impact the outcomes of our future research, the inter-relationships of how people experience, know, and act, and on the experiential process of our overall inquiry [26]. With this aim in mind, our objective in this developmental work was to recruit and engage a community and a clinical advisory board to ensure that we included the expertise and lived experience of key representatives in the design of a future implementation study and subsequent research.

Methods

Overview, pre-recruitment, and planning activities

At the outset of this project, the principal investigator, Dr. Hansmann, partnered with two consulting organizations at the University of Wisconsin and continued to work with them throughout the planning and conduct of this developmental project to leverage existing institutional efforts of building and sustaining community-academic partnerships and to ensure rigorous group facilitation and co-design activities throughout our work.

To support recruitment and study design, Dr. Hansmann collaborated with the University of Wisconsin Community Academic Aging Research Network (CAARN) [27, 28]. CAARN has a 15-year history of connecting community and health systems with academic researchers to improve the health of older adults and promote health equity. CAARN’s model of supporting community-academic partnerships allows them to sustain relationships with community organizations beyond the scope of individual grants or research projects, including employing community research associates who work part-time with partner organizations and parti-time with the university. Prior to committing to collaborating on this research project, CAARN had Dr. Hansmann complete an intake process to review her purpose, past experience working with community partners, work plan, long-term goals, and potential benefit of the proposed research, reviewing whether this work and the rationale for seeking community representatives to participate in this work was aligned with the needs and priorities of members of their partner organization networks. Additional questions posed to Dr. Hansmann in the intake process included expectations about how community partners would be engaged in specific research activities (e.g., study design, data collection, dissemination) and how the work might inadvertently worsen inequities. CAARN’s partner organization networks include the UW Center for Community Engagement and Health Partnership, the Wisconsin Alzheimer’s Institute (WAI) Regional Milwaukee Office, and MetaStar, a health care quality improvement organization. We focused recruitment for the older driver community advisory board around Milwaukee, WI based on CAARN’s existing partner relationships and to allow for at least some of those meetings to be held in-person.

Dr. Hansmann began attending community events hosted by the WAI Regional Milwaukee Office several months before recruitment for this developmental project’s advisory boards began. The WAI Regional Milwaukee Office seeks to promote community outreach, professional education, advocacy, service, and research in Milwaukee and the surrounding communities to address current health inequities with a primary focus on Black older adults [29]. Dr. Hansmann has continued to attend these in-person events (e.g., educational sessions hosted in community spaces, dementia choir concerts) about once every other month to develop and sustain connections with community group members.

Dr. Hansmann also collaborated with the Wisconsin Network for Research Support (WINRS) to prepare agendas and facilitation guides for the four meetings we planned to hold with each advisory board over the course of six months. Since 2010, WINRS has offered consultation services to researchers for community engagement activities including facilitating effective advisory board meetings [30]. WINRS is also the home of Community Advisers for Research Design Strategies (CARDS), an advisory board that has consulted on research projects for the past 15 years. We consulted with CARDS to refine the introductory presentation and activities for the first community advisory board meeting. This input from the CARDS board was particularly valuable given its importance in (1) building relationships between advisory board members and the research team and (2) finding a shared agenda. The CARDS board shared feedback about the content and style of our introductory presentation, highlighting areas where slides had too much jargon or lacked clarity.

To further ensure rigor and reproducibility of our work, we conducted this co-design project in accordance with the Guidance for Reporting Involvement of Patients and Public (GRIPP2) Short Form (see Additional File 1) [31]. Advisory board member recruitment and participation in advisory board meetings took place from April 2024 to February 2025. We invited advisory board members to participate in the dissemination of our findings after the final official meeting including the preparation of a poster presentation for the American Geriatrics Society Annual Meeting and the preparation of this manuscript.

Recruitment and formation of the community advisory boards

We used convenience sampling to recruit two parallel advisory boards made up of 5–7 members each: (1) a community advisory board made up of Black older adults (age > = 55 years) living in or near Milwaukee, WI, with experience navigating transitions to non-driving either as a person with dementia or a family caregiver and (2) a clinical advisory board made up of health professionals with significant experience caring for patients with dementia and helping those patients navigate transitions to non-driving. To reflect greater geographic diversity, and because we anticipated virtual meetings would allow for the greatest ease of scheduling for clinical advisory board meetings, we recruited clinical advisory board members from across the state of Wisconsin.

We conducted outreach to community organizations in Milwaukee, WI, with the support of one of CAARN’s community research associates, Darona Woods, who recruited participants through CAARN’s partner organization networks including the WAI Regional Milwaukee Office. Another CAARN community research associate, Mia Croyle, led recruitment for the clinical advisory board, reaching out to clinical practices across Wisconsin through Metastar and other partner organizations. The CAARN community research associates worked directly with individuals interested in participating in the advisory boards to finalize recruitment based on people’s past experiences and their availability to meet. Recruitment drew upon a curated list of individuals with prior involvement in quality improvement programs and existing professional relationships with the recruiter.

Initial meetings and agreements

Our objectives for the first meeting with each group were to welcome and meet the board members, confirm a shared group agreement for each advisory board, and provide an overview of the research project’s background and goals. The primary discussion activity for the first meeting was to engage the group in a preliminary discussion about the problems and priorities for improving the experience of transitions to non-driving for people with dementia, their caregivers, and clinicians.

Ongoing meetings, data collection, and analysis

The research team, including team members from CAARN (D.W. and M.C.) and WINRS (K.P.), met around two times between each round of the advisory board meetings to prepare for subsequent meetings. The general format of the remaining three meetings involved:

1. a welcome, including a brief introduction to the meeting’s agenda and an ice breaker;

2. a brief review of the group agreements;

3. a 10-20-minute presentation of “what we’re learning” (i.e., pragmatic themes we had identified from both advisory boards’ discussions at the previous round of meetings) and “how we’re listening” (i.e., how the research team had begun to incorporate these pragmatic themes into refining the research plan for our future work) to solicit feedback not only on how we had summarized the discussion of each group at the previous meeting but also their reaction to the other board’s feedback; and

4. a group discussion activity. We spent the majority of the meetings on these discussion-based activities to refine our recruitment strategy (Meeting #2) and outcome measures (Meeting (#3) for our future implementation study. The discussion of the final meeting (Meeting #4) focused on sharing and reaching consensus on our findings.

While the meeting topics were consistent between the community and clinical advisory boards, we approached the structure of the meetings slightly different for each board. With the community advisory board, each meeting was 90 min in duration and we held the first two meetings in person at a location hosted by the WAI Regional Milwaukee Office with breakfast provided. Meeting in person and for a longer duration allowed us to spend more time on rapport-building activities. Members of the community advisory board requested that we host the second two meetings virtually so that they would not have to travel during the winter months (January and February) in Wisconsin.

For the clinical advisory board, we held each meeting virtually, as members were living in different regions of the state. We opted to schedule these meetings for 60 min, starting at 7:00a, with a goal of finishing these activities in time for clinical advisory board members to attend to their clinical responsibilities. We facilitated these shorter meetings by spending less time on rapport-building activities (inviting members to answer an ice breaker question in the chat rather than taking turns to answer one-by-one) and spending less time in small group breakout discussions.

Data collection included audio-recording the advisory board meetings and detailed note-taking during meetings by a member of the research team. We also gathered process evaluations by inviting community advisory board members to share “pluses and deltas” (i.e. what went well at this meeting? what would you change at future meetings?) at the end of their two in-person meetings. We invited all advisory board members to complete a process survey after each virtual meeting, which we distributed using Google Forms. In these process surveys, we used a combination of closed- and open-ended questions to gather advisory board members’ perspectives on the meeting content (e.g., relevance, meaningfulness) and facilitation (e.g., effective use of time, flow of topics), and communication (e.g., pre-meeting communication, communication between meetings).

In between meetings, we used a modified approach to thematic analysis to rapidly identify actionable knowledge gathered from meeting notes and process surveys [32]. After each meeting, one team member would review the audio-recording and meeting notes to prepare a detailed written summary. The principal investigator and research team members from CAARN and WINRS reviewed these summaries separately before meeting 1–2 times to discuss pragmatic themes to inform the “What we’re learning” and “How we’re listening” presentations for the next advisory board meeting (described above). For each round of meetings, we held the community advisory board meeting 1–2 weeks before the parallel clinical advisory board meeting. This cadence supported our ability to identify pragmatic themes from the community advisory board discussion to present to the clinical advisory board, which further supported our ability to focus the clinical advisory board discussions and complete those meetings in a shorter (i.e., 60 min) timeframe.

Results

We recruited six Black older adults living in the greater Milwaukee area to participate in the community advisory board, (5 family caregivers and 1 person with dementia) and six clinicians to participate in the clinical advisory board (1 social worker, 1 speech pathologist, 1 nurse, 1 nurse practitioner, and 2 physician assistants). Clinical advisory board members worked in health systems in Madison and Milwaukee as well as Western Wisconsin.

We held our initial meetings with the two advisory boards (community and clinical) in September 2024 and October 2024 respectively. We have summarized key findings from the Advisory Board Meetings in Table 1.

Table 1.

Key findings from the advisory board meetings [Return to results Section]

Community Advisory Board	Clinical Advisory Board
In-person Meeting (September 2024)	Virtual Meeting (October 2024)
Meeting 1: Problems and Priorities
• Barriers to stopping driving include: o being afraid to lose independence; o not being emotionally ready to stop driving; o losing an important form of identification when surrendering your license; o not having good alternatives to driving such as dementia-friendly transportation services	• Barriers for clinicians include: o not enough time; o patients may not be ready to stop driving; o caregivers may not be prepared for the transition
Meeting 2: Recruitment Strategy
In-person Meeting (October 2024)	Virtual Meeting (November 2024)
• To encourage study participation: o trust o clear communication o community outreach (e.g., events, radio programs)	• To encourage participation in study recruitment: o a script and/or brochure
Meeting 3: Meaningful Outcomes
Virtual Meeting (January 2025)	Virtual Meeting (January 2025)
• To evaluate success of the intervention, ask: o about how people feel o if they feel ready for the transition o if people with dementia and their caregivers agree about issues of safety and independence o how they are getting around in their community o if their personal goals are being met	• To evaluate feasibility of the intervention: o it will be important to identify who should follow up after visits o how often clinical team should reach out for updates on driving
Meeting 4: Findings and Consensus
Virtual Meeting (February 2025)	Virtual Meeting (February 2025)
• Continue meeting to sustain connections • Apply for community–university partnership grants • Expressed interest in refining the driving cessation care pathway in the future study	• The length and number of meetings may need to increase to support design refinement • Schedule meetings before clinic (7:00a) and keep them to an hour or less • Increase board membership with the understanding that not every member can attend every meeting • Ensure meeting agendas are well organized

Meeting #1: problems and priorities

In our first meeting with the community advisory board, we began by presenting the background and goals for our program of research as well as presenting our project team, timeline, and how we envisioned the role of the advisory boards in this developmental project. We asked advisory board members to consider what might make it hard for a person with dementia to plan a transition to non-driving with their caregivers and clinicians.

First, we asked the community advisory board members to discuss potential barriers in small groups of 3–4 before coming together as a large group to review the discussion. Community advisory board members shared that drivers with dementia may be afraid to lose their independence or may not be emotionally ready to stop driving. They identified that many people may struggle to get around after stopping driving because they do not have access to good alternatives (e.g., dementia-friendly transportation services). They also raised concerns about the difficulty of losing a driver’s license as an important form of identification and the confusion people with dementia face in understanding the difference between a state ID and a driver’s license.

Next, we presented a summary of three theoretical frameworks that have been used to understand the problems associated with driving cessation to the community advisory board - the theory of planned behavior, the transtheoretical model, and the social ecological model [33]. We summarized these as (1) people see stopping driving as a negative, (2) different people are more less ready to stop driving, and (3) how people make changes to their driving behavior is going to depend on the people around them and the resources in their community.

Finally, we asked the community advisory board members whether any of these three problems seemed to stand out as important or relevant based on their experience. Community advisory board members shared experiences and stories that reinforced all three barriers. They also identified actionable themes that were either new or expanded on the theory-based barriers including that the person with dementia might have distrust of their clinician and/or their caregiver during conversations about stopping driving, a lack of coordination between clinical team members who may give conflicting information to patients and families, a perception that the medical team lacks training or skill to have conversations about driving cessation, and that driving cessation has many disincentives (e.g., increased time to get places, giving up a common form of identification, losing independence).

In our first meeting with the clinical advisory board, we began by presenting an introduction to the background and goals of our research program, followed by a presentation of the three theory-based problems associated with driving cessation and shared examples raised by the community advisory board. We asked the clinical advisory board members to consider what might make it hard for a clinician to support their patients with dementia to make a plan to transition to non-driving and which of these problems are most important to address. Clinical advisory board members identified the biggest challenges as the lack of time during clinical encounters and the readiness of not just the patient but also their caregiver(s) for the driving transition.

Meeting #2: refining the recruitment strategy

In our second round of meetings, we asked both advisory boards to reflect on what would encourage and discourage people with dementia, caregivers, and clinicians from participating in a clinic-based study evaluating the feasibility and effectiveness of a new driving cessation care pathway. The community advisory board identified three themes of trust, communication, and outreach with specific examples of how these could be applied to our future recruitment strategy. Encouraging factors included knowing a trusted individual who is participating in or involved with our research study and having shared identity between study team members and participants where possible. The community advisory board members also suggested preparing eye-catching and clearly written handouts or brochures that were available in the languages spoken by patients at the clinics where we will be recruiting. Additionally, they recommended sharing information about the research study at community events and on radio programs to increase familiarity with the research before we even attempt to recruit potential participants at their clinics. The clinical advisory board added that preparing a script for clinicians to share information about the research opportunity and a brochure to leave with patients would be an encouraging factor for them to let their patients know about the study.

Meeting #3: identifying meaningful outcomes

In our third round of meetings, we asked both advisory boards to share what outcomes we should measure and what questions we should ask to understand the impact of the driving cessation care pathway in our future study. The community advisory board members emphasized opportunities for qualitative data collection, including asking patients and caregivers how they feel during this transition. An important theme raised by the community advisory board was the idea of concordance between patients and caregivers and the concern that patients might prioritize their independence while caregivers prioritize their safety, which leads to disagreements and tension. Other potential outcomes identified by the community advisory board included gathering data on how the person with dementia was getting around in their community and whether the person with dementia and their caregiver felt their goals were being met. The clinical advisory board considered feasibility outcomes for the study, including gathering data to determine the best clinical team member for follow-up with patients and caregivers and how often they should reach out for updates on driving.

Meeting #4: sharing our findings and confirming consensus

In our fourth round of meetings, we engaged the community and clinical advisory boards in different activities to gather consensus on our findings from this co-design process as well as to clarify our best practices for next steps, including sharing our findings with others. Together with the community advisory board members, we prepared a two-page summary of the advisory board findings and findings from interviews we had conducted in parallel with other people with dementia and caregivers to share with a community audience. The members of the community advisory board had expressed an interest in participating in the refinement of the driving cessation care pathway in our future study and hoped to continue meeting to sustain the connections they had formed with each other in the interim, so we also used our fourth meeting to plan opportunities to support their continued meetings, including applying for community–university partnership grants.

In our final meeting with the clinical advisory board, we focused on the optimal format for forming and sustaining an advisory board of clinicians from different health systems. We discussed strategies that had worked well for this advisory board, including scheduling meetings before clinic (7:00a) and keeping the duration to an hour or less. For the next stage of our research, we anticipate intervention co-design activities will require more, and potentially longer, meetings over a shorter period of time, so we discussed strategies to support the function of a clinical advisory board for our future study. Strategies included having a larger advisory board membership with the understanding that not every member would be able to attend every meeting and ensuring that meeting agendas were well organized. The clinical advisory board members expressed appreciation for having a brief amount of time for icebreakers to get to know each other’s personalities and hear about each other’s different and shared experiences. We presented an alternative option of engaging clinicians in individual interviews with the option to complete a second or even third follow-up interview to give feedback on the design of the clinical care pathway. This also prompted the clinical advisory board members to suggest strategies for sharing useful follow-up information via email newsletters or potentially as a short video or podcast-style summary for participating clinicians to listen to on the go.

Ultimately, the clinical advisory board members were most in-favor of a hybrid approach that combined activities with a clinical advisory board and individual interviews with clinicians. Advisory board members felt this approach allowed for both continuity and flexibility by allowing clinicians to participate in sequential interviews, advisory board meetings, or both as their schedules allowed (Fig. 1). The approach refined with the clinical advisory board members distributes the co-design process between a core design team (i.e., the study team and practice-based research network, or PBRN, focus group in Fig. 1) and individual interviews with a broader network of primary care clinicians to increase the applicability and generalizability of potential design solutions. This iterative design process allows for a core, longitudinal group to incorporate design feedback from other clinicians over the course of the co-design process. Overall, the clinical advisory board was highly engaged and interested in sharing experiences and resources – they agreed that if time allowed they could talk longer and that they appreciated the sense of support and camaraderie around the often challenging issue of the transition to non-driving for people with dementia.

Fig. 1.

Proposed approach to merging advisory board and interview data collection in future research (Return to Results)

Process evaluations

Community advisory board members shared multiple positives during the “Plus/Delta” checkout activity at the end of the two in-person meetings. They described enjoying participating in the meetings, feeling a sense of pride participating in the meetings, and appreciating that they were able to hear from other advisory board members and learn about their different experiences with driving and dementia. During this activity they did not suggest any changes to our meeting process. For the remaining two virtual community advisory board meetings, four of the six board members responded to each process survey. In both process surveys, all four respondents reported that they felt their input was valued by the research team. In response to open-ended questions about challenges and opportunities for improvement, they shared a range of feedback from reporting being “completely satisfied with the process,” to suggesting that opening the meeting with an icebreaker used up valuable time that could have been spent on discussion activities.

We invited clinical advisory board members to complete process surveys after each of their four virtual meetings, with response rates ranging from two to four of the six clinical advisory board members. Respondents unanimously reported that meeting activities, including the welcome, researcher presentation, and discussion activities were very useful. In response to open-ended questions, clinical advisory board members shared that they felt there was space to express themselves and ask questions. They reported feeling challenged by the questions but enjoyed thinking through them.

After the completion of the final meetings for both advisory boards, we invited members to share their interest in being involved in future work such as co-presenting the work of these advisory boards for academic settings including papers and conference presentations. Three board members, one from the community advisory board and two from the clinical advisory board, expressed that they would like to be invited to give feedback on future presentations and papers depending on their time and availability.

Discussion

Over the 12-month period of this developmental project, we engaged in important preparatory work to increase the feasibility and ultimately the impact of a program of research to evaluate and implement a clinical care pathway to support people with dementia and their caregivers during complex life transitions like driving cessation. With the goal of incorporating the perspectives of key representatives from the beginning of this work, we recruited and oriented two parallel advisory boards made up of community members and clinical team members who had lived experience with dementia and driving transitions. We worked with these two advisory boards to refine our research priorities, recruitment strategy, meaningful outcomes, and strategies to share our findings. Simultaneously, we began to foster collaborative relationships with advisory board members and invited their feedback to continue to improve engagement and co-design activities in our future work.

We sought to overcome known challenges to developing and sustaining community-academic partnerships by working with an established community organization committed to fostering meaningful and long-term community partnerships that exist beyond the life cycle of an individual researcher’s projects and funding. Sustaining long-term relationships with community partners is consistently a challenge for community-academic partnerships due to the realities of, and structural differences between, community and academic work [34, 35]. At the outset of community-engaged research, academic researchers may not have adequate knowledge of the community to identify key representatives that have a strong knowledge of the population’s health status and trusted roots within the community. Researchers are also often limited by the urgent and time-limited nature of grants, making it difficult to sustain relationships with community representatives between periods of research funding [36]. Turnover and change in structure within community organization staffing can also make it difficult for researchers to stay connected with partners while they work to sustain funding [37]. A strength of our approach to recruiting and engaging a community and clinical advisory board was that we first partnered with the Community Academic Aging Research Network (CAARN) to leverage their ability to sustain relationships over time and have a deep understanding of community dynamics. Building our work with existing institutional infrastructure was both a pragmatic choice and one that helped to circumvent common challenges new researchers face when they attempt to develop community relationships without existing community support.

In our meetings with both the community and clinical advisory boards, board members emphasized the importance of caregivers to the transition to non-driving. From our first meetings, both groups identified that caregivers who were not prepared for the driving transition could be a greater challenge than the person with dementia being unprepared. This was echoed in our discussion about meaningful outcomes for our research with the community advisory board members—board members identified the discordance between a person with dementia’s concern for independence and their caregiver’s concern for safety as one of the greatest sources of distress during this transition. These findings are consistent with previous research about the significant increases in caregiving burden associated with providing transportation support to a family member or friend who stops driving [38]. Notably, previous driving cessation research for older drivers without dementia has focused on establishing the emotional and attitudinal readiness of the patient when tailoring counselling about driving cessation [39, 40]. As with other areas of dementia care research, our findings highlight an important need for caregiver perspectives to be included in assessments and care planning [41, 42].

One limitation in this stage of our community-engaged research is that we as the investigators were still driving the research question and agenda for each community advisory board meeting. We sought to address this by including time to share back what we were learning and how were applying it to our research plans early in each follow up meeting. Going forward, we intend to explore with the members of the advisory boards the extent they wish to be equal partners in this research-to-action process and what resources (e.g., compensation, training) we can offer to support their ideal level of engagement [18]. Without seeking to understand and meet the needs of participating community members, community-engaged research, including the work we have described here, could potentially be described as a confirmatory step for research that is already in progress (including research that has only been conceptualized). To address this potential limitation, we will continue to encourage open, bi-directional communication with members of the advisory boards to inform their future role in this work and the ongoing direction of the research.

Another limitation that impacts the generalizability of our findings is that our community advisory board recruitment relied on convenience sampling, a common challenge for community advisory board formation. Findings from our discussions with these community advisory boards may not be reflective of the perspectives of people with dementia, as all but one member of the community advisory board were caregivers. We also faced common challenges to recruiting clinicians to our clinical advisory boards including clinicians’ time constraints caused by an ever-increasing workload and decreasing workforce that makes it difficult for clinicians to actively participate in research [43, 44]. We sought to overcome these challenges by keeping clinical advisory board meetings shorter and scheduling them earlier, and while there is potential for the combined advisory board-interview approach to engaging clinicians in co-design that we refined with our clinical advisory board members in their final meeting, further testing is needed to determine if this increases clinician participation and representation in research.

Notably, the perspectives of these advisory boards may not be generalizable to other groups that are known to also experience significant challenges during the transition to non-driving such as those living in rural areas where alternative transportation options are scarce or Hispanic older adults who face unique challenges compared to Black older adults. These efforts to engage key representatives in community advisory boards allows for developing relationships and working towards equal partnerships. However, this work also needs to be supported by additional data collection from both the larger community and other important groups experiencing unique challenges by applying more rigorous sampling approaches.

This work represents an important preliminary stage of research to begin engaging key community representatives in refining research priorities and study design for our future investigation of a care pathway to support people with dementia and their caregivers during the transition to non-driving. The insights shared by the advisory board members will increase the relevance of our research to people with dementia, caregivers, and their clinical team members. To deliver on the full potential of this work, we as the investigators will need to continue to develop our relationship with the advisory boards towards a goal of meaningful partnership. These efforts will require that we continue to invest in these relationships beyond the scope of the 12-month pilot award that supported this work and between future grants we are pursuing. By investing the time and effort into these relationships, our hope is to create interventions and implementation strategies that make it possible for those who struggle to navigate complex medical and social systems to get the care they need to live meaningful and dignified lives.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

K.H wrote the main manuscript text, participated in data collection, and data analysis. D.W. and M.C. led participant recruitment and participated in data collection. K.P. facilitated advisory board meetings and participated in data analysis. A.M. conducted an additional literature review and conducted analyses of process survey data for the revised manuscript. All authors contributed to reviewing and editing the manuscript.

Funding

This project was made possible by the UW-Madison Institute for Clinical & Translational Research (ICTR) with support from NIH-NCATS Clinical and Translational Science Award (CTSA) 1UL1TR002373 and funds through a grant from the Wisconsin Partnership Program at the University of Wisconsin School of Medicine and Public Health Program, Wisconsin Partnership Program (WPP 5129).

Data availability

Meeting agendas, meeting notes, process survey, and process survey results are available on request. Please contact the corresponding author, KJ Hansmann, at kellia.hansmann@fammed.wisc.edu.

Declarations

Ethics approval and consent to participate

Prior to conducting the work described in this project we completed the University of Wisconsin’s Institutional Review Board QI/program evaluation self-certification tool finding that the scope of this project did not constitute research requiring IRB review. To ensure ethical treatment of all collaborators and advisory board members, we conducted this project in accordance with the University of Wisconsin’s Responsible Conduct of Research practices.

Consent to publish

All authors have consented to the publication of this manuscript.

Competing interests

The authors declare no competing interests.