System facilitators and barriers to discussing older driver safety in primary care settings

Abstract

Background

Primary care physicians play a leading role in counseling older drivers, but discussions often do not occur until safety concerns arise. Prior work suggests that routine questioning about driving might facilitate these difficult conversations.

Objective

To explore system-level factors affecting driving discussions in primary care settings, in order to inform the design and implementation of a program supporting routine conversations.

Methods

This qualitative descriptive study used iterative interviews with providers (physicians, nurses, medical assistants, social workers, and administrative staff) working at two clinics (one geriatric, one general internal medicine) at a tertiary-care teaching hospital. General inductive techniques in transcript analysis were used to identify stakeholder-perceived system-level barriers and facilitators to routine conversations with older drivers.

Results

From fifteen interviews, four themes emerged: (1) complexity of defined provider roles within primary care setting (which can both support team work and hamper efficiency); (2) inadequate resources to support providers (including clinical prompts, local guides, and access to social workers and driving specialists); (3) gaps in education of providers and patients about discussing driving; and (4) suggested models to enhance provider conversations with older drivers (including following successful examples and using defined pathways integrated into the electronic medical record). A fifth theme was that participants characterized their experiences in terms of current and ideal states.

Conclusions

Physicians have been tasked with assessing older driver safety and guiding older patients through the process of “driving retirement.” Attention to system-level factors such as provider roles, resources, and training can support them in this process.

INTRODUCTION

Healthcare providers have been identified as playing a central role in older driver safety^1-6 because they are trusted by patients and families⁷ and have a responsibility to public safety.¹ Compared to licensing authority staff, physicians may better understand a patient’s medications and cognitive, sensory or physical impairments, all of which can impair driving ability.^8,9 The difficulty lies in estimating crash risk for an individual and balancing this against the negative effects of “driving retirement.” While fatal crash rates rise at age 75,¹⁰ driving is so tied to mobility and independence that driving cessation can increase morbidity and mortality.^8,11-15

Unfortunately, conversations about driving are generally not routine and occur only when there are safety concerns.^4,16,17 Explanations include: physician discomfort, fear of patient alienation, and inadequate training;^18,19 competing priorities with limited time;⁴ driver assessment tools that may be too long for practical routine use;^1,20 fear of liability;⁵ and inadequate options for testing or alternative transportation.^3,21 Regulations concerning physician assessment or reporting of fitness-to-drive vary widely among states;²² in Colorado, for example, physicians are able, but not mandated, to report potentially unsafe drivers and they receive liability coverage for any reports make in good faith (but non-physicians are not similarly covered).²³ These policies may affect conversations with patients, as some physicians may be reluctant to bring up the topic of driving if they are required to report to licensing bureaus or if they are unsure of regulations.

System-level changes may address some of these barriers, perhaps through a “tiered” approach of routine, brief questioning followed by referral of screen-positive patients for specialist evaluation.^3,24,25 An ongoing challenge is the lack of a “gold standard” brief assessment for older driver ability, but prior work suggests that routine conversations may ease the emotional impact of eventual driving retirement. In addition, in the absence of a “gold standard” test, physician guidance may be even more important in helping an older adult decide when to retire from driving. While there are gaps in provider knowledge concerning driver assessment and referral,¹⁹ there is evidence that provider training can affect confidence, attitudes, knowledge and behaviors concerning driving discussions.^17,18,26,27 Through interviews with healthcare providers and older drivers, we previously developed preliminary frameworks of provider-, patient- and system-level factors influencing conversations between clinicians and older drivers.^3,4 These frameworks identified both barriers (e.g., time constraints) and facilitators (e.g., openness to preventive health guidance) to conversations, along with specific recommendations about how to incorporate driving discussions into primary healthcare (e.g., through annual questioning). Questions remain, however, how best to implement a tiered older driver assessment program.

In this study, we sought to explore key stakeholder-perceived system-level factors in greater depth by using qualitative methods with a new, diverse group of healthcare providers. We hoped that an improved understanding of environmental barriers and facilitators to discussions about driving within the clinical microsystem would inform future implementation of a program for routine discussions with older drivers.

METHODS

Study Population

Eligible participants were current English-speaking staff at one of two hospital-based outpatient clinics (one geriatric, one general internal medicine). Eligible staff included physicians, physician assistants, nurses, medical assistants (MA), social workers, and front-desk administrative staff. We recruited participants via flyers in staff areas and emails from clinic directors; we also used snowball sampling to recruit staff identified by participants as likely to have unique perspectives or key influences in the clinical setting. One-on-one interviews lasted 30 to 60 minutes in private spaces on hospital property. Participants provided written informed consent prior to the interview and each received a $15 gift card. The Colorado Multiple Institutional Review Board approved the study.

Study Design and Procedures

We used semi-structured interviews in a qualitative descriptive study design²⁸ to explore various aspects of factors affecting conversations between older drivers and primary care clinic staff. The interview guide included open-ended questions developed from prior research findings and our current research question of the barriers and facilitators to conversations with older drivers in primary care settings (Table 1; Appendix). The focus of the current analysis is the portion of interviews directed towards system-level factors. Following an iterative approach,^29,30 we analyzed material after each interview and adapted subsequent interviews accordingly to generate a deeper, richer understanding. This included presenting the preliminary themes to interviewees in the later interviews for feedback and refinement.³⁰ Interviews were conducted by MEB from February to June, 2014, and all sessions were digitally recorded and then transcribed verbatim by a professional transcriptionist.

Table 1.

Interview domains and probes

View of own professional role in working with older drivers Description of experience with an older patient who has some impairments that may affect driving, including: ○ Specific steps of what provider did (e.g., discussions, assessments or referrals) ○ Who provided assistance ○ Driver’s response View of concept of “tiered older driver assessment” in primary care settings (including whether it would work, anticipated reaction from patients and providers) Recommendations on how to implement “tiered older driver assessment” in a way that would make it more likely for both providers and drivers to support it

Analysis

We used a team-based approach of deductive and inductive analytic techniques at the micro- and macro-levels of text data. This approach draws on social constructionist epistemology and was guided by the six steps outlined by Thomas.^31,32 Analyzed material included interview transcripts and notes from interviews and research team meetings, and codes were both deductive (predetermined from interview guide) and inductive (arising from participants’ responses). Two investigators (MEB & JJ) independently reviewed the interview transcripts and conducted line-by-line readings for pertinent codes, supported by Atlas.ti v7.1.5 for data management. Through regular discussions and an iterative team-consensus approach,^30,31 we grouped codes into broader categories and then identified core themes arising from the final set of code categories. In a second level of deductive analysis, we applied the social ecological model (SEM) to the codes and themes to allow a comprehensive matrix for analysis and understanding. As a clinical microsystem, the primary care clinic is situated within a broader public health context. The SEM directed our analysis to pay attention to the various systems levels within and beyond the clinic itself at which arising themes and factors could be clustered for targeted future intervention. These included the levels of individual, unit, organization, community and policy influence.

RESULTS

We conducted 15 interviews, yielding 241 pages of transcripts and 18 pages of notes. Characteristics of participants are shown in Table 2. When no new themes were emerging (the point of “thematic saturation”), we stopped recruitment.^30,32 Four core themes emerged related to systems-level factors affecting primary care provider interactions with older drivers (Figure): (1) complexity of defined provider roles within primary care setting; (2) inadequate resources to support providers; (3) gaps in education of providers and patients about discussion driving; and (4) suggested models to enhance provider conversations with older drivers. A fifth theme was that participants characterized their experiences in both the current and future (ideal) states. Representative quotes, organized by theme, are shown in Table 3.

Table 2.

Characteristics of participants (n=15)

Characteristic	N (%)
Position Physician Nurse Medical assistant Social worker Front-desk (registration) staff	10 (67) 2 (13) 1 (6.7) 1 (6.7) 1 (6.7)
Affiliation Geriatric clinic General Internal Medicine clinic	7 (47) 8 (53)
Female	13 (87)
Age in years (median, range)	40 (32-68)
Years working in healthcare (median, range)	12 (4-43)

Figure.

Emergent systems-level themes (white box) affecting primary care environment for discussions between providers and older drivers, shown in context of larger framework of clinician-driver interactions concerning constructive conversations about driving.⁴

Table 3.

Representative quotes, by emergent themes.

	State (Theme 5)
	Current	Ideal
Complexity of defined provider roles within primary care setting (Theme 1)	If there is not a family concern, am I actually going to bring [driving] up independently? I don't think that is necessarily my role. In an ideal world, it would probably be, but not with the resources that we have in general medicine right now. (Physician)	I think if we ever redesign primary care where it would not be my job to do all of these things, then I think we could expand those roles and if there was a way to get paid for that, or get people's time paid for… But right now, primary care has reached its limit, I think, on all the other things it can be doing. (Physician)
Inadequate resources to support providers (Theme 2)	I run 30 min. behind every half-day. And I do get frustrated with talks that come and say I have to do more. (Physician) Just add one more thing to their already …I mean it is beneficial, but would [doctors]…(laughing) Would they be willing to do it on every visit? (MA)	Thinking about primary care though, we have pamphlets and resources, but it's knowing how to access those resources. It's like we almost need a Clearing House of Primary Care Resources that everybody just uses. Where you can go and you can get who are the local therapists? What do I do for spousal abuse? … You know, just these things that come up rarely, but are super important and you just have like one place… and driving is one of those. (Physician)
Gaps in education of providers and patients about discussion driving (Theme 3)	I just Googled it. So once they told me they couldn't schedule a road test with the DMV, I was like, well, what else do I do? I talked to some colleagues. I got blank stares! (Physician) You mentioned the AMA Guide. I've never heard of or used the AMA guide. (Physician)	Helping us know what's out there and evidenced- based would go a long way. Like what do we know is evidenced-based for when we need older drivers to do a behind the wheel test? … I have colleagues who go “I always say they have to take the behind the wheel test because I just feel like, how do I know they are a good driver, just because I know that they are healthy?” And then other colleagues are like, “well, if I think they are doing ok, and no one has told me different, I'll maybe say they are ok.” (Physician)
Suggested models to enhance provider conversations with older drivers (Theme 4)	And I have done a lot better job at Advanced Directives actually. That's a good example. Since I've been using the Medicare Smart Phrase [in the electronic medical record]. Just because it is in there. So as I get to it, I discuss it. (Physician)	I think that if we had a simple screening tool to identify drivers at high risk and then, once we do that, knowing exactly what resources to give them, so it is literally, “here are the resources, this is what you need to do.” And not only that, then being able to follow up on it. Some way… like the loop gets closed. (Physician)

Complexity of defined provider roles within primary care setting

Providers in both clinics recognized that they had a role in discussing driving with older patients, although it was not a role they all enjoyed. As one physician said: “I don't know if I would say hesitant… I would say dread maybe. More like I know it's important to do. I know we have to do it, but we don't like it.” Concerning their current roles as defined within the primary care system, participants identified physicians as leaders for discussions about driving, with support from other staff. For nurses, this was partly due to the nature of their interactions with patients, which was primarily over the phone rather than in person. One clinic employed a social worker, who was recognized by providers in both clinics as a key resource. She also saw the high value in her involvement in driving discussions with older adults, as she felt she was most informed about options for driving assessment and alternative transportation. Providers also mentioned relying on specialists, including occupational therapists (OTs) at a driving program, for guidance, and noted that conversations with these specialists were particularly helpful.

In considering prior findings that older drivers and their family members look to physicians for support, one physician questioned the appropriateness of this relationship. She speculated whether it arose not from trust but rather from inadequate information and blanket recommendations to ‘ask a doctor’ about any concerns. “I think one of the reasons they say their doctors is, they trust their doctor and they think that [doctors] know more than they do, about this. … Because we are supposed to have magic? … It's just a default way of ‘we don't know, go talk to somebody else’.” This same physician argued that the department of motor vehicles (DMV) or licensing authority should take the lead: “It is a privilege to have a driver's license, so why doesn't DMV do the testing?” However, other physicians expressed skepticism and a lack of trust in the DMV; said: “What I've heard from preceptors is that sometime the DMV will ignore or not process your paperwork.”

Providers noted constraints in their current roles related to institutional policies, including a circumscribed “scope of practice” for MAs that limited the ways MAs might be able to help providers in screening, assessing, or counseling older drivers. Some of these difficulties related to working in a teaching hospital; as one physician explained, “It's because [MAs] are employed by one entity, where I am employed by one entity and it is very hard to make the two meet. I was in private practice before. I can tell you it is nothing like private practice.” Especially as related to current time constraints, many providers had suggestions about how to enhance patient care and efficiency through adjusted models of teamwork; one physician said: “If a nurse could ask the question and then flag the doctor to specifically ask the question later, I think that would be great.”

Inadequate resources to support providers

Providers identified a number of critical resources to support them in talking with older drivers. All mentioned time constraints, given the multiple competing priorities providers faced during short visits, sometimes with inadequate access to social workers and other support staff. Concerning the feasibility of routinely asking about driving, a physician said:

“We don't have a social worker here. We are so busy. Over-extended. They want us to do more and more and more. … And we primary care doctors, we just don't have time. And if we had a mechanism that we could immediately go to. A person we could immediately call to help us, absolutely! And if we could do it in a non-threatening way to patients, yes.”

Time was a particular concern related to driving, given the complexity of the issue; as one physician noted:

“I think it's tricky, because on the one hand it is obviously a health issue, trying to keep people safe if they are not able to drive safely. On the other hand, it is hard for us within the clinic to know whether or not what family members or what the patients might be saying is true. … I don't think the systems that are in place for us to do whatever the right thing is for a particular patient.”

Suggested resources included physical brochures or guides to help older drivers with getting a driving evaluation or finding alternative transportation. As the social worker pointed out: “You really do have to offer [transportation] alternatives. And you have to offer services or resources so that people can still get where they need and want to go.” Unfortunately, many providers felt these resources were not easily accessible and in fact were less accessible since the institution adopted an electronic medical record (EMR) system (EPIC). A physician commented: “We used to have a file full of basically anything you could imagine. But now since things have gone paperless, and everything is in EPIC, those have been taken from the rooms. … I just don’t know where they have taken them.”

Identified resources also included referral options for specialist testing by trained driving evaluators. However, finding these referrals could be difficult, and the institution did not have its own driving program. Many providers mentioned a driving evaluation program available to veterans at the nearby Veteran’s Administration (VA) hospital, but they were otherwise unsure of local options. As one physician said: “I know [a driving rehabilitation program] exists and you know, I could do a Google search if I really needed it.” Many physicians did know, however, that these driving programs are often not reimbursed by insurance: “A lot of people probably can't afford it.”

After the first interviews revealed the need for a list of local resources, our research team developed and gave one to participants. We did not ask these interviewees to do anything in particular with the list, though we gave them permission to share it with other providers and we gave the list to each subsequent participant. Interestingly, the list was quickly and widely distributed among the clinics. In one of the later interviews, a clinic nurse was well-informed about local options and mentioned a resource list: “[Interviewer]: Do you know where the list came from? [Participant]: Um…maybe you? I don't know. … Yeah. I mean it's really great information. I think it's something we lack in the Clinic. … I actually gave that information to both of the other [clinic] teams, as a resource.” Additional ideas from participants included making “smart phrases” in the EMR to provide decision support to providers and to allow them easily to give resources to patients; our research team also developed and distributed this with initial positive feedback.

Gaps in education of providers and patients about discussion driving

Participants felt they had not received adequate training on the topic of older driver assessment and counseling; as one physician said: “I think what would help me would be to just even know what the rules and regulations are. I don't think we have been educated very well.” Physicians with fellowship training in geriatrics felt more comfortable than those without. In two cases this was due to having completed an elective rotation working with an OT at a driving rehabilitation program. Others noted how geriatrics training had changed their perspective on medicine in general, including how to balance the risks of driving with the risks of driving retirement. As one physician said of the difficulty of this balance: “One of the lessons in geriatrics is trying to maintain independence. Trying in a patient-centered sort of way. That's what good geriatrics is. Respecting the individuals even as they age. Respecting their freedom, their autonomy, and now you are telling them, ‘don't drive’?” In explaining his discomfort in discussing driving with patients, the same physician continued:

“I'm boarded in internal medicine, geriatrics and palliative care. And I feel really comfortable having the ‘end of life’ discussion. … It is also less contentious than driving. When I say driving is one of my least favorite issues in medicine, that is not hyperbole. It truly is one of my least favorite things. Because it is so sad.”

Most participants reported that any education about driving they had received had been “on the job” from supervisors or colleagues rather than from formal didactics, and some physicians recommended driving be covered in residency training. Another physician suggested training non-physicians to become experts in assessing and discussing driving: “It would be pretty cool, if within Patient Centered Medical Home, you could know that person X went to the day-long training and therefore we could have them do all our driving screening, counseling after the provider.”

Participants also brought up the value of the public about the topic of driving safety and driving retirement, including through media campaigns to encourage older adults to think about their driving and their future plans. Others expressed skepticism about a media approach; as one physician said: “I'm not sure how people would respond to a [public service announcement]. I mean driving in the United States is like a right, like the right to bear arms! It is going to be like the driver's version of the NRA.” Providers were enthusiastic about the idea of patient education at a clinic or hospital setting. They suggested educational discussions with groups of older adults to encourage peer learning and peer support, similar to current models of group visits for discussing advance directives; one physician commented: “Just seeing you are not alone and other people are struggling with this might help you be introspective around the fact that this is part of life.”

Suggested models to enhance provider conversations with older drivers

In discussing ways to enhance discussions with older drivers, all participants brought up analogies to other kinds of clinical programs. A common reference was to the clinics’ current practice of screening all patients for depression. They described that program as working well because of its defined process, as MAs completed brief screening and then flagged screen-positive patients for physicians. The EMR included templates for physicians with prompted medications and referrals, and the nurses received automatic reminders to call patients back two weeks after the visit. Participants also attributed the program’s success to the way it was implemented with staff input and training. As one physician said concerning any screening: “I think it is important to think of who the stakeholders are and how we are going to offset the burden of doing any additional screening or intervention, with lining it up with the existing initiatives and requirements.”

Considering routine screening or assessment of older drivers, participants agreed that an initial question about current driving status might be best, as many providers admitted they were not certain whether some patients still drove or not. A physician explained: Knowing for sure if they drive would be important to know, if I need to make sure they are still safe to drive. And we don't screen people for whether they are driving.” Participants also thought the basic question of “Do you drive?” would not be threatening to patients, and they expressed concern over not wanting to alienate patients over the sensitive topic of driving, especially in a medical culture that emphasizes “customer service.”

A few providers used the term “plant the seed” in reference to bringing up the topic of driving. One physician reported she currently asks basic questions about driving:

“I am using it to plant the seed of preventative health. That eventually, this is something that I want as your provider to normalize and talk with you about. So years down the road, whether that is two or five or however many years, this is something that it is ok for me, as your doctor, for me to check in with you about.”

Providers also liked the idea of including a screening question on driving status in the EMR to prompt physicians to ask the question; one physician referenced the current system at the nearby VA hospital: “It is in the template. And I try to like slip it in there somewhere. Like, ‘Who do you live with? How do you get around? Do you drive or take the bus?” Many participants mentioned the annual Medicare Wellness Visit as an ideal time to include such a question, since the visit is focused on preventive health. Although the clinics’ current EMR template for that visit did not reference driving, one physician pointed out that driving is included as a possible topic in the Medicare guide.

However, providers identified concerns about the next steps after assessing driving status, including feeling limited in the resources they could offer potentially at-risk drivers. One physician emphasized: “I think the worst thing we could do would be to screen for stuff that we are not following through with.” Participants liked the idea of a tiered process, with embedded decision support in the EMR, to prompt physicians to ask standardized assessment questions and refer drivers to other resources based on their responses.

Current versus ideal state of practice

A fifth theme was that providers often expressed their views in terms of the current and ideal state of practice, with reference to the current gaps between the two. Views on how to bridge the gap and move towards an ideal state of practice were varied. Some providers advocated action on the part of providers; as one physician explained:

“The best thing would be just to get them to ask about it. And then they would start discovering things that would make them uncomfortable. And then they would start wondering what to do next. But if they even just asked about it and saw that as part of what they did to care for an older adult, that would be a big step forward.”

Another physician from the same clinic disagreed, however, and emphasized the need for improved systems to support physician action: “I don't think just asking a random question is going to really bring up the conversation. I think that more it would be helpful if we figured out a way to make it less scary for those doctors to bring it up. And that would be, I would think, to have a mechanism in place that would be non-threatening.” Suggestions to move towards a more ideal state of practice also related to education of the public, such as the idea of campaigns featuring positive images of older adults who had retired from driving. A nurse suggested:

“some positive role models of people that could be brought in somehow, saying you know, ‘I saw it was my time to hand the keys over. But this is what I'm doing and it's working great.’ So you know, there might be some opportunities for people to see like life can go on. Even if you don't have access to your car.”

Providers also spoke of necessary changes at the level of policies, including in the realm of provider reimbursement and the structure of primary care medicine.

DISCUSSION

Physicians have been tasked with assessing and counseling older drivers,^1,33 but there has been inadequate attention on how to adopt currently available guidelines and curricula. This study provides novel information concerning system-level barriers and facilitators to routine conversations about driving in primary care settings. Defined provider roles within a clinic can impede conversations by overloading particular providers, but they can also facilitate efficient teamwork, empower staff, and enhance satisfaction of both patients and providers.³⁴ Resources and education of providers and patients are also important; some require national policies or programs while others may be easily addressed at the local level. Finally, routine driving conversations could be facilitated through integration into existing models and by following examples of other successful clinical pathways. With the aging of the population, issues of older driver safety will grow in importance, especially given the fact that most older adults will eventually “retire” from driving and require alternative transportation.^12,33 System-level changes will be critical to support clinicians in their work to balance the safety and independence of older adults.

For routine, tiered assessment of older drivers to function smoothly in clinical practice, providers need access to key internal and external resources. Ideally, these should include: clinical prompts for providers with questions or screening tools; printable hand-outs concerning safe driving, testing options and alternative transportation; and referral systems for driver evaluations and reporting to licensing authorities. Many of these could be embedded in EMRs, although physical signs or brochures in clinical spaces may be easier for some providers to use and may raise awareness among waiting patients. Resources for providers, patients and family members are currently available from a number of national organizations,^35-37 but our findings suggest that they have not been well-linked to practicing clinicians. Some of these barriers might be relatively easy to overcome, as shown by our experience developing and distributing a list of local resources. At a national level, models from Australia and Canada – where frequent updates to driving guidelines are published in print and online^38,39—might be useful but will require leadership from national organizations. Other resource barriers include reimbursement for on-road evaluations by driving rehabilitation specialists,²¹ as well as the time constraints faced by primary care providers in fee-for-service environment. Addressing these barriers will require continued discussions with policy makers, insurance executives and other stakeholders;²⁵ options include insurance coverage or incentives for evaluations, restructuring primary care processes to optimize team efficiency and maximize physician time with patients, and alternative payment structures for primary care practices.

Our findings, in line with those from prior studies,^17-19,40,41 suggest that primary care providers – even some of those with geriatrics fellowship training—may not have received adequate training on the topic of older driver assessment and counseling. This is despite the inclusion of driving in competencies for internal medicine, family medicine, and geriatrics.⁴² Efforts to educate providers on older driver safety using the American Medical Association (AMA) Physician’s Guide to Assessing and Counseling Older Drivers ¹ have shown local success, at least over short time periods.^17,18,40 Given that such trainings can be resource-intensive, the AMA developed a web-based physician module to facilitate dissemination to a broader physician audience.⁴³ The module was effective²⁶ but is no longer supported by the AMA,⁴⁴ although it may be available through the American Geriatric Society in the future. The AMA Guide itself remains available online.¹ Educational efforts should focus not just on primary care physicians but also on older drivers and the range of providers who care for them,¹⁸ including nurses, occupational therapists, social workers, specialist physicians, and providers working across the spectrum of settings. The American Occupational Therapy Association has a current program aimed at encouraging generalist OTs to become engaged in older driver screening and assessment;⁴⁵ similar national initiatives are needed for other stakeholder groups.

A strong emergent theme in this study was providers’ comfort with and desire for clear pathways with assigned responsibilities and easily available resources. This is in line with prior work identifying a multi-disciplinary approach as critical in implementation of clinical pathways.³⁴ Concerning older driver screening and counseling, participants spoke of the usefulness—but often unachievable reality—of a team-based approach to the issue. Participants identified the Medicare wellness visits as an ideal time to incorporate questions about driving; the “Health Risk Assessment” of these visits should include instrumental activities of daily living, which include “mode of transportation.”⁴⁶ For facilities using electronic medical records, inclusion of driving status as an optional question for the Medicare wellness visits may prompt providers to bring up and “normalize” the topic.⁴

Limitations

A study limitation is that participants worked at clinics at a large, urban teaching hospital, so their perspectives may differ from those in other settings. Independent clinics, those in rural areas, or those affiliated with private (non-teaching) institutions may have different barriers and facilitators to routine conversations about driving, and future work should expand into these settings. The exploratory nature of the topic, with a goal of depth rather than breadth, supported the moderate sample size, but generalizability may be limited. Recruitment of certain providers (including medical assistants and nurses) was limited by the clinics’ staffing structure and small pool of eligible providers, so future research could target these groups. Participation was voluntary, so respondents may have been particularly supportive of driving discussions; at the same time, we used snowball sampling to target specific individuals and some participants expressed divergent views. In addition, our iterative study design allowed us to explore new information in subsequent interviews and to ask participants their views on the emergent themes.

CONCLUSIONS

Healthcare providers have important roles in assessing and counseling older adults about safe driving. Routine conversations about driving are unlikely to occur, however, without a supportive clinical environment. Providers’ defined roles should empower them and optimize team function. Necessary resources include those within a clinic, referral systems for specialist evaluations, and alternative transportation. Education of the public and the spectrum of providers who care for older drivers remains critical. Finally, implementation of older driver assessment programs in clinical practice should incorporate stakeholder input, follow successful models, integrate into current workflows, and include a robust evaluation of the process. Increased attention to dissemination and implementation issues, including the use of translational research approaches, in the field of older driver safety has the potential to enhance program effectiveness and spread and, ultimately, prevent injuries.

What is already known on this topic.

• Driving can be important for older adult mobility and independence, but some older drivers face elevated injury and death rates.

• Healthcare providers play an important role in talking with older adults about the risks of benefits of driving, but these discussions are difficult.

• A “tiered” approach could involve routine questioning about driving, followed by further evaluation for higher-risk older drivers.

What this study adds.

• Qualitative methods allow a deeper understanding of the system-level factors affecting implementation of an older driver safety program in primary care settings.

• Barriers to address include inadequate access to resources and inadequate education of providers and the public.

• An older driver safety program in primary care settings should be modeled after other successful programs