The American Medical Association Older Driver Curriculum for Health Professionals: Changes in Trainee Confidence, Attitudes & Practice Behavior

Abstract

Few gerontology and geriatrics professionals receive training in driver fitness evaluation, state reporting of unfit drivers, or transportation mobility planning, yet are often asked to address these concerns in the provision of care to older adults. The American Medical Association (AMA) developed an evidence-based, multi-media Curriculum to promote basic competences. This study evaluated reported changes in practice behaviors three months post-training in 693 professionals trained via the AMA approach. Eight Teaching Teams, designated and trained by AMA staff, offered 22 training sessions across the U.S. in 2006–7. Trainees (67% female; mean age 46) completed a pre-test questionnaire and a post-test administered by mail. Physicians were the largest professional group (32%). While many trainees acknowledged having conversations with patients about driving at pre-test, few endorsed utilizing specific techniques recommended by the AMA prior to this training. The post-test response rate was 34% (n = 235). Significant improvements in reported attitudes, confidence, and practices were found across measured items. In particular, post-test data indicated new adoption of in-office screening techniques, chart documentation of driver safety concerns, and transportation alternative planning strategies. Findings suggest that a well-designed, one-time continuing education intervention can enhance health professional confidence and clinical practice concerning driver fitness evaluation and mobility planning. Targeted dissemination of this Curriculum (in-person and on-line) will allow more to benefit in the future.

Introduction

A growing concern in the U.S. and other developed countries is how to meet the mobility needs of an aging population. The automobile continues to be the primary mode of transportation for most adults, and this is true across the lifespan. As a group, however, older drivers face heightened risk for health and/or functional conditions that may impact on driving ability and/or driving longevity. Questions about medical fitness to drive and mobility alternatives are increasingly part of the doctor-patient dialogue. The American Medical Association (AMA) is committed to educating clinicians concerning evidence-based evaluation and intervention in this important area (Wang & Carr, 2004).

The older segment of the population is growing faster than the population at large, and the same is true for older drivers. There was an 18% increase in the number of older licensed drivers (65+ years) between 1996 and 2006, leading to 30 million older drivers in 2006 as compared to a 13% increase in total licensed drivers (National Highway Traffic Safety Administration, 2008). Older drivers will retain their licenses longer than did those of past generations (Hakamies-Blomqvist, 1994), travel greater distances, and take more driving trips (Rosenbloom, 2000). Medically impaired older drivers have been found to have elevated crash histories when compared to age matched controls (Meuser, Carr, & Ulfarsson, 2009). When older adults are involved in crashes, they are more likely to be seriously injured or die in comparison to middle-aged counterparts (Bédard, Guyatt, Stones, & Hirdes, 2002; Ulfarsson & Mannering, 2004). Physicians and other health professionals are often faced with questions about chronic and/or acute conditions and fitness to drive (Charlton et al., 2004; Dobbs, 2005), yet may lack the knowledge and/or confidence to respond (Meuser, Carr, Berg-Weger, Niewoehner, & Morris, 2006).

Driving cessation or retirement brings concerns about independence and well-being to the forefront (Marottoli et al., 2000). Such concerns are important for individuals and society as a whole, as it has been estimated that most adults living into the 7^th and 8^th decades will outlive their driving life expectancy. The period from driving cessation to death may average 6 years for men and up to 10 years for women (Foley, Heimovitz, Guralnik, & Brock, 2002). Physicians and health professionals who care for older adults (such as social workers, occupational therapists, and other rehabilitation therapists) play key evaluative and supportive roles with respect to driving life expectancy (MacLean, Berg-Weger, Meuser & Carr, 2007).

Stakeholders, notably patients and family members, view physicians as leaders in the evaluative process (Perkinson et al., 2005). A lack of perceived time for driver fitness evaluation remains a barrier for significant involvement (O’Connor, Kapust, & Hollis, 2008). Physicians may also be reticent to address the issue of driving fitness and safety with older patients. Yet, continuing education research supports that when clinicians have improved knowledge and tools they will incorporate this information into their practice (Byszewski et al., 2003; Marottoli, 2000; Meuser et al., 2006). While some physicians suggest that discussions of driving cessation could be harmful to the physician-patient relationship (Shawn, Mitchell, & Gilbert, 1999), many acknowledge the issue of driver safety as germane in the care of older patients and a target for focused education (Kelly, Warke, & Steele, 1999; Marshall & Gilbert, 1999). How to report to state authorities is a particular area of interest (Cable, Reisner, Gerges, & Thirumavalavan, 2000).

The Older Drivers Project of the AMA was created to educate physicians and other health professionals concerning medical fitness related to driving such as safety and cessation. The AMA’s Physicians Guide to Assessing & Counseling Older Drivers (Carr, Schwartzberg, Manning, & Sempek, 2010; Wang & Carr, 2004; Wang, Kosinski, Schwartzberg, & Shanklin, 2003), available in print and electronic formats, provides background information and a detailed rationale for the application of medical principles to the driving task. In-office approaches to screening and assessment receive significant attention, including a suggested group of tests called the Assessment of Driving Related Skills (ADReS Battery). State reporting requirements are also covered. More importantly, the Guide teaches how to frame the broader topic of personal mobility and discuss results in a way that balances patients’ independence needs with larger public safety concerns. If retirement from driving is required, most patients will benefit from this holistic approach.

The Guide may be reviewed by physicians and a quiz is typically taken to earn continuing medical education credits. The Guide also serves as the foundation for a five module, multimedia Curriculum (including slides, video case segments, handouts) for live presentation to physician and other trainee groups by specially designated and trained Teaching Teams (comprised of a physician, rehabilitation professional, and other driving specialist). The Curriculum was updated in 2006 with the addition of new slides and video clips (Meuser et al., 2006) to demonstrate components of the evaluative process, and data were collected from 22 live presentations by different teams in 12 states for this evaluation. The Guide was updated again in 2010 to reflect new information in the literature, current state policies and laws on older drivers, and new organizations and/or educational efforts to assist clinicians in this area (Carr et al., 2010; see www.ama-assn.org/go/olderdrivers).

While the Guide and Curriculum provide an in-depth examination of the aging process and associated co-morbidities with respect to older driver fitness, the Guide underscores that age alone should never be the sole criterion for determining whether someone is a safe driver; how the individual driver/patient functions is the critical issue. The physician is not alone in managing this process. Also emphasized is the importance of a “team approach” involving physicians, nurses, social workers, psychologists, rehabilitation professionals, family members, law enforcement, state officials, and other stakeholders. Each “stakeholder” may contribute to the mobility management process.

The Curriculum was designed to provide the core knowledge and techniques necessary for a clinician to evaluate driver fitness in a typical care encounter, determine if the patient is at heightened risk due to a medical-functional problem, and, if so, to develop a plan for further evaluation (e.g., referral for on-road testing, reporting to state) and/or consideration of non-driving transportation options. We evaluated the AMA Older Driver Curriculum to determine whether there were changes in attitudes, practice, and behavior after 1–2 hour Curriculum interventions with health professionals. We hypothesized that our trainees would report increased confidence, clinical activity concerning driver fitness (questioning patients, documenting findings, encouraging proactive planning), and adoption of specific tools/approaches recommended by the AMA.

Method

This research analyzed data from 22 training sessions presenting the AMA older driver Curriculum to audiences of physicians and other health professionals between October 2006 and July 2007. Presentations were offered in 12 US States¹ and took one of two forms: 1-hour lecture or 2-hour workshop format. The same core topics were covered in each, with the workshop format offering additional opportunities for clinical case review and discussion.

To explore the participants’ attitudinal and behavioral change subsequent to participation in the AMA training, data were collected on participants through the use of two surveys: (1) a pre-test administered in-person immediately before participation in the Curriculum, and (2) a post-test administered by mail three months later, along with an incentive² to encourage returns (see Robertson, Walkom, & McGettigan, 2005).

A total of 693 participants responded to at least one questionnaire. Of those, 52% (361) participated in the one hour lecture format and 48% took part in a two hour workshop; 90% (622) completed the pre-test and 35% (243) completed the 3-month post-test, however, of those only 235 respondents had also completed the pre-test. Behavioral change was explored in this subgroup (n = 235) of individuals who completed both pre and post-tests.

Analytic Approach

Several survey questions measured attitudes, practices, and knowledge before and after Curriculum participation (see items summarized in Table 1). To test if responses to these questions changed in a statistically significant way, the odds ratios of these variables before (pre-test) vs. after (post-test) were calculated. A number of recent training-related studies have used this approach to document attitudinal and behavioral change (Carroll, Vreeman, Buddenbaum & Inui, 2007; McCormick, Tomlinson, Brill-Edwards, & Detsky, 2001). Odds ratios were calculated identically for each question. For example, an estimator for the odds of responding correctly to a knowledge question is the number of participants with a correct answer over those with an incorrect answer (e.g., n_b,correct/n_b,incorrect for the before analysis). The maximum likelihood estimator of the odds ratio before (pre-test) and after (post-test) is then given by Agresti (2007):

$$\widehat{\theta }=\frac{n_{\text{b},\text{correct}}/n_{\text{b},\text{incorrect}}}{n_{\text{a},\text{correct}}/n_{\text{a},\text{incorrect}}}.$$ (1)

Table 1.

Survey Results for the 693 Participants

Variable		Pre-Test Sample	Post-Test Sample	Total Sample
About Program and Participant				693 (100%)
Program Type	1 hour lecture			332 (47.9%)
	2 hour workshop			361 (52.1%)
Completed Both Surveys -	Yes			235 (33.9%)
	No			458 (66.1%)
Completed a Survey	Yes	622 (89.8%)	243 (35.1%)
	No	71 (10.2%)	450 (64.9%)
Reviewed Guide Since -	Yes		54 (22.2%)
	Somewhat		122 (50.2%)
	No		67 (27.6%)
Age		46.1 (13.2)a	47.4 (14.0)
Gender	Male	204 (32.8%)	76 (31.9%)
	Female	417 (67.0%)	162 (68.1%)
	Did not respond	1 (0.2%)	0
Profession	Physician	209 (33.6%)	75 (33.2%)
	Occupational Therapist	125 (20.1%)	65 (28.8%)
	Social Worker	48 (7.7%)	26 (11.5%)
	Counselor/Psychologist	13 (2.1%)	5 (2.2%)
	Physician Assistant	7 (1.1%)	4 (1.8%)
	Physical Therapist/Rehab	49 (7.9%)	19 (8.4%)
	Nurse	107 (17.2%)	21 (9.3%)
	Other Health Professional	7 (1.1%)	2 (0.9%)
	Social Service Professional	10 (1.6%)	4 (1.8%)
	Other	15 (2.4%)	5 (2.2%)
	Did not respond	32 (5.1%)	17 (7.0%)
Years in Profession		17.8 (12.5)	18.8 (13.0)
Percent Patients 65+		61.9 (30.3)	63.2 (30.9)
ATTITUDES
Comfort with knowledge of medical driving impairments	Agree	95 (15.5%)	63 (26.1%)
	Somewhat agree	223 (36.3%)	132 (54.7%)
	Neutral	180 (29.3%)	37 (15.4%)
	Somewhat disagree	95 (15.5%)	6 (2.5%)
	Disagree	22 (3.6%)	3 (1.2%)
Familiar with driving rehabilitation options
	Agree	54 (8.8%)	81 (33.6%)
	Somewhat agree	106 (17.2%)	107 (44.4%)
	Neutral	165 (26.8%)	37 (15.4%)
	Somewhat disagree	175 (28.5%)	12 (5.0%)
	Disagree	115 (18.7%)	4 (1.7%)
Comfortable counseling medically impaired drivers	Agree	60 (9.8%)	65 (27.2%)
	Somewhat agree	156 (25.5%)	104 (43.5%)
	Neutral	191 (31.2%)	48 (20.1%)
	Somewhat disagree	149 (24.4%)	17 (7.1%)
	Disagree	56 (9.2%)	5 (2.1%)
Familiar with state’s licensing reporting and regulations	Agree	58 (9.4%)	61 (25.5%)
	Somewhat agree	92 (15.0%)	94 (39.3%)
	Neutral	173 (28.1%)	51 (21.3%)
	Somewhat disagree	172 (28.0%)	24 (10.0%)
	Disagree	120 (19.5%)	9 (3.8%)
PRACTICE BEHAVIORS
Regularly discuss driving safety with older patients	Agree	77 (12.8%)	65 (28.0%)
	Somewhat agree	133 (22.1%)	69 (29.7%)
	Neutral	136 (22.6%)	61 (26.3%)
	Somewhat disagree	157 (26.1%)	19 (8.2%)
	Disagree	99 (16.5%)	18 (7.8%)
Documented driving abilities in patient’s chart		237 (38.1%)	117 (48.2%)
Incorporated driving questions into assessment		280 (45.0%)	140 (57.6%)
Referred older driver to licensing agency for evaluation		130 (20.9%)	46 (18.9%)
Referred older driver for driving assessment		191 (30.7%)	81 (33.3%)
Worked with older driver and family to implement driving cessation		183 (29.4%)	93 (38.3%)
Used AMA functional assessment of vision, cognition, motor function		35 (5.6%)	41 (16.9%)
Use clock drawing for cognitive screening		292 (47.0%)	113 (46.5%)
Use trail making for cognitive screening		98 (15.8%)	67 (27.6%)
Knowledge
Knowledge test – techniques – Correct		554 (89.1%)	231 (95.1%)
Knowledge test – vision – Correct		415 (66.7%)	169 (69.6%)
Knowledge test – medications – Correct		517 (83.1%)	204 (84.0%)
Training of Others after Participation
Have educated others with the AMA Guide or older driver project materials			26 (10.7%)
Groups trained	Medical students		3 (1.2%)
	Residents		5 (2.1%)
	Physicians		10 (4.1%)
	Other health professionals		21 (8.6%)
Number of sessions taught			1.7 (1.2)
Number of people trained			32.4 (56.0)

^aIndicates average value with standard deviation in parentheses.

This assumes the responses of different participants are not dependent on each other. To explore the statistical significance of the odds ratio, the standard error of the estimator was found with

$$\widehat{\sigma }(ln\widehat{\theta })=\sqrt{\frac{1}{n_{\text{b}}}+\frac{1}{n_{\text{b},\text{correct}}}+\frac{1}{n_{\text{a}}}+\frac{1}{n_{\text{a},\text{correct}}}},$$ (2)

where the natural logarithm of the estimator, ln θ̂, was used since the distribution is skewed to the right. n_b indicates the total number of responses in the before study, n_a for the after study. The 95% confidence interval around the estimator was given by

$$[exp(ln\widehat{\theta }-1,96\widehat{\sigma }),exp(ln\widehat{\theta }+1,96\widehat{\sigma })].$$ (3)

Interpretation is straightforward. If the 95% confidence interval around the estimator does not include 1.0, there is a significant difference in the odds ratio of a correct answer before and after at the 0.05 level of significance. Furthermore, the result should be statistically significantly less than 1.0 to indicate a significant increase in the odds of a positive answer to the question in the post-test (i.e., an improvement).

To explore behavioral change between the pre-test and post-test, the sample of participants that responded to both surveys was used. For each behavioral question, participants who did not indicate engaging in the behavior at pre-test were taken as a subsample. Within this subsample, those that endorsed the behavior at post-test were counted as new adopters.

Results

Sample Characteristics

Table 1 presents the results for both pre and post-test surveys. Participants in the full sample (n = 693) were 67% female, with a mean age of 46 years (range 20–82 years). Many (62%) reported treating patient populations composed primarily of older adults (i.e., half or more of practice). Physicians were represented most (32%), followed by rehabilitation professionals³ (25%), nurses and physician assistants (16%), social workers and psychologists (9%), and other/unknown (18%). Various physician specialties were represented: family/general practice (26%), geriatrics (20%), internal medicine (18%), neurology (8%), and ophthalmology (6%). Students comprised 12% of the total sample, and were categorized with their professional groups for this report. Medical students were coded in the physician group.

The gender distribution (female > male) and high proportion of geriatric specialists suggests that this volunteer sample is not representative of practicing health professionals in general. The gender breakdown is largely due to the influence of two professional groups (social workers, rehabilitation professionals) which were mostly female. The fact that many participants reported treating many older adults was largely a function of marketing (i.e., flyers emphasized the benefits in geriatric care) and self-selection. This sample is representative of those most likely to benefit from and use this training.

Pre-test and three month post-test surveys were available for 235 respondents, indicating an effective response rate of 34%. Comparisons (T-test, Chi Square) were made between respondents that returned the post-test survey and those that did not. Although some variations are suggested in Table 1, these subsamples did not differ statistically in gender proportion, age, physician vs. non-physician status, and percentage of older patients in the practice.

Post-test returns did differ somewhat by professional group. Physicians assistants were proportionately the most likely to respond (4 of 7, 57%), followed by social workers (26 of 48; 54%), occupational therapists (65 of 125; 52%), physical therapists (19 of 49; 39%), counselors and psychologists (5 of 13; 38%), physicians (75 of 209; 36%), and nurses (21 of 107; 19%).

Changes in Attitudes, Practices or Knowledge

Table 2 shows the odds ratio results. All attitude and basic knowledge items (i.e., comfort with knowledge of medical driving impairments, familiarity with driving rehabilitation options, comfort counseling medically impaired drivers, familiarity with state’s licensing reporting and regulations indicated a significant improvement. Three months post-training, respondents reported being more comfortable with their knowledge of driver fitness and more confident in their ability to apply it in the care of patients/clients.

Table 2.

Odds Ratio Results for Changes between the Pre- and Post-Test Responses

Variable		Odds Ratio	Total Sample
Survey Items & Response Options (in survey order)
ATTITUDES
Comfort with knowledge of medical driving impairments	Agree	0.52	Improvement
	Somewhat agree	0.47	Improvement
	Neutral	2.27	Improvement
	Somewhat disagree	7.12	Improvement
	Disagree	2.93NS
Familiar with driving rehabilitation options
	Agree	0.19	Improvement
	Somewhat agree	0.26	Improvement
	Neutral	2.01	Improvement
	Somewhat disagree	7.54	Improvement
	Disagree	13.55	Improvement
Comfortable counseling medically impaired drivers	Agree	0.29	Improvement
	Somewhat agree	0.45	Improvement
	Neutral	1.80	Improvement
	Somewhat disagree	4.19	Improvement
	Disagree	4.71	Improvement
Familiar with state’s licensing reporting and regulations	Agree	0.31	Improvement
	Somewhat agree	0.28	Improvement
	Neutral	1.45	Improvement
	Somewhat disagree	3.49	Improvement
	Disagree	6.22	Improvement
PRACTICE BEHAVIORS
Regularly discuss driving safety with older patients	Agree	0.39	Improvement
	Somewhat agree	0.69	Improvement
	Neutral	0.83NS
	Somewhat disagree	3.98	Improvement
	Disagree	2.37	Improvement
Documented driving abilities in patient’s chart		0.66	Improvement
Incorporated driving questions into assessment		0.60	Improvement
Referred older driver to licensing agency for evaluation		1.13NS
Referred older driver for driving assessment		0.89NS
Worked with older driver and family to implement driving cessation		0.67	Improvement
Used AMA functional assessment of vision, cognition, motor function		0.29	Improvement
Use clock drawing for cognitive screening		1.02NS
Use trail making for cognitive screening		0.49	Improvement
Knowledge
Knowledge test – techniques – Correct		0.42	Improvement
Knowledge test – vision – Correct		0.88NS
Knowledge test – medications – Correct		0.94NS

NS indicates the change is not statistically significant relative to a 95% confidence interval. An odds ratio < 1 indicates a higher ratio in the post-test; conversely, an odds ratio > 1 indicates a lower ratio in the post-test.

Clinician practice behaviors (i.e., regular discussion of driving with patients; documentation of driving abilities in patient’s chart; incorporating driving questions into assessment; working with older driver and family to implement driving cessation; use of the AMA-recommended assessments of vision, cognition, and motor function) all showed significant increases suggesting training-related practice change. Other practices, such as discussion about driving concerns with older patients, referral of older drivers to licensing agency for evaluation, referral of older driver for driving assessment, and use of clock drawing for cognitive screening, did not show a statistically significant change. Many reported having discussed the driving issue with patients regularly before training. The lack of change in referrals and use of the clock drawing test may be explained, in part, by the three-month timeframe for follow-up; the opportunity to make a referral or use a more advanced screening tool may not have presented itself during this relatively short period.

All knowledge questions showed improvement with a larger share of respondents responding correctly in the post-test, however, the only significant difference (p < 0.05) was for a question about appropriate evaluation techniques (see Appendix 1, Pre-Test Survey Part III, Item 1). Respondents were asked which of four approaches would not be appropriate (i.e., Tell the patient’s family to hide the car keys).

Reported Behavioral Change

A majority (174 of 235; 74%) reported referring to the printed Physician’s Guide (i.e., bound paper booklet) during the 3-month post-training period. Use of the Guide was similar across professional groups. To further understand the benefits of training, responses to the five confidence items (see list in Table 3) were summed for the participants responding to both the pre- and post-test to yield pre-test and 3-month post-test composite scores (Pre-test Mean = 15.4, SD = 4.6; Post-test Mean = 19.2, SD 3.8; Post > Pre statistically significant at the p < 0.01 level). While all professional groups gained in reported confidence, the degree varied by professional group and training format (see Figure 1). The 2-hour workshop format yielded the highest gains across groups, with non-significant trends favoring physicians, social workers, and psychologists in the longer format. A similar pattern was found for nurses and physician assistants. Although most reported gains, a small number did not; 9% (2 of 19) of physical therapists and rehabilitation professionals and 13% (10 of 75) of physicians reported less confidence, overall, at the three month post-test.

Table 3.

3-Month Change in Self-Rated Confidence & Experience

Confidence/Experience Item	N	More Confident/Experienced	Unchanged	Less Confident/Experienced
Comfort with personal knowledge of topic	231	45%	46%	9%
Familiarity with driver rehab options	231	75%	27%	8%
Comfort in counseling patients at risk	228	67%	34%	9%
Familiar with state laws and procedures	230	63%	30%	7%
Regularly discuss driving with patients	220	47%	41%	12%

Figure 1.

Mean Change in Confidence Composite Score by Profession & Format

Nine behavioral items were included in pre/post questionnaires, and these are listed in Table 4. Of the 235 respondents to complete both surveys, the participants not indicating a behavior in the pre-test but indicating the behavior in the post-test were counted as new adopters. There were a total of 117 course participants showing change in their practice behavior; half (59 participants) reported change in just one activity during the 3-month post-training period and a quarter (31) reported changes in two. Eight new adopters reported changes in four or more activities; and half of these (4) were physicians.

Table 4.

Percentage of Respondents Endorsing Behavioral Change on 3-Month Post-Test

Behavioral Item	Subsample Size4	# (%) New Adopters
Incorporated driving questions into regular assessment	113	35 (31%)
Discussed driving as a clinical concern with any patient	58	18 (31%)
Specifically documented driving abilities in patient chart	129	32 (25%)
Worked with patient/family on driving retirement plan	154	34 (22%)
Used AMA ADReS Test Battery	204	31 (15%)
Reported patient to State DMV for re-evaluation	177	26 (15%)
Used Clock Drawing Test	118	15 (13%)
Used Trail Making Test	186	22 (12%)
Referred patient for other driving evaluation (e.g., OT)	143	13 (9%)

⁴Subsamples composed of individuals responding no activity concerning the item at pre-test. In other words, these were individuals not engaged in a particular task prior to receiving training, but changed their behavior afterwards.

New adopters came from all professional groups, but especially physical therapy and rehabilitation. Almost two thirds (62%; 12 of 19) of rehabilitation professionals adopted at least one new practice behavior during the 3-month post-training period, and this was true in almost half of the others (42%, 13 of 31, SW/Psych; 48%, 12 of 25, Nurse/PA; 46%, 35 of 75, Physician). Adoption patterns were similar across professional groups in terms of the number of behaviors changed, such that approximately half of each group reported change in just one activity area.

Another measure of educational impact is dissemination to other trainee groups. The Physician’s Guide and Curriculum were developed with secondary training in mind. A total of 26 participants (11% of the post-test sample) reported utilizing the materials to educate others. Of these, the majority (14 of 26, 54%) were rehabilitation professionals. Five physicians reported engaging in secondary training activities.

Effect of Curriculum Format/Duration

Physicians participated in the two presentation formats equally, but this was not the case for other professional groups. Most nurses (83%; 89 of 107) participated in the 1-hour lecture, whereas most physical therapy and rehabilitation professionals (80%; 39 of 49) and social workers and psychologists (79%; 49 of 62) participated in the 2-hour workshops. Some of these differences were likely due to how the programs were marketed, the interest level of different groups, and where the programs were offered. Most 1-hour lectures were held in medical settings, whereas some longer workshops were held in non-medical, conference-oriented settings. In general, more effort and planning were required of participants choosing the workshop format (i.e., more time off work, off-site travel).

Few significant differences were found based on format participation: favorable responses were obtained from both groups across a variety of indicators. Both reported that the programs achieved stated objectives and fostered the learning of new tools and approaches. Significantly more of those attending the longer workshop endorsed an interest in receiving advanced training in the future (70% vs. 59%, p < 0.01). Physicians involved in the one hour lecture were more likely to desire additional training in comparison to their workshop counterparts (68% vs. 57%, p < 0.01). Members of the other groups were not distributed sufficiently to make any meaningful statements in this regard.

Discussion

This evaluation study confirmed that an evidence-based, resource-oriented continuing education intervention can enhance healthcare practice with respect to the evaluation of older driver fitness and safety. Benefits of enhanced confidence and adoption of recommended practice behaviors were found across professional groups at three months post-training, and gains were similar for both training formats (1 hour rounds/lecture, 2 hour workshops). While physicians accounted for the largest trainee subgroup, not infrequently social workers, nurses, rehabilitation professionals and others also participated, speaking to the inherent multidisciplinary nature of driving and mobility evaluation. These findings appear to justify the AMA’s investment in a broad, team-oriented approach to training and practice in this important area.

Personal efficacy beliefs are important when implementing new learning. Respondent reports suggest that the AMA Curriculum was particularly effective in this regard. At the three month post-test, trainees reported greater knowledge and confidence concerning available driver rehabilitation options (75% reported enhancement), strategies to counsel at-risk drivers (67%), and familiarity with state laws (63%). Fewer (45%) reported an increase in overall comfort with the topic, but this may be due to the three-month time interval. Firsthand experience can enhance comfort, and it is possible that some trainees did not yet have many opportunities to exercise their new skills. Longer-term gains have been documented in other similar studies (Meuser et al., 2006).

The data indicate reasonably broad adoption of AMA recommendations. The data show significant new adoption of chart documentation, incorporation of driving questions into the clinical approach, coordination with patient and family to support driving retirement, use of integrative assessment approach (ADReS Battery), and use of the Trail Making Test for screening. The largest proportion of new adoption (31% of post-test respondents) occurred in questioning of patients. The Curriculum can be judged as successful based on this finding, alone, given the importance of initiating a driving and mobility dialogue in the context of medical care. By starting the dialogue, the clinician opens a door to engage the patient, family and other stakeholders in considered evaluation and planning.

While physicians often play key coordinating roles in care of the older driver, other health professionals provide similarly important supportive and ancillary services. Through their training in counseling and resource management skills, social workers, for example, are well-suited to provide mobility counseling and planning assistance to patients and families when driving cessation becomes a possibility. Rehabilitation professionals – those reporting the greatest adoption of new evaluative techniques in this project – may assist patients in the use of adaptive devices or behaviors to keep them optimally mobile, whether behind the wheel or via another means of transportation. Another element of this educational initiative was the opportunity to reach students of medicine and other fields. One in ten of those trained were students with decades of professional practice ahead of them.

While many trainees indicated making changes to their assessment approach on the three month post-test, just a modest number indicated adoption of the full Assessment of Driving Related Skills test battery (ADReS). Since what is considered “assessment” may vary across professional groups, and the influence of personal preference, professional orientation or experience on assessment choices, it is possible that adoption of the ADReS was not a realistic goal for many trainees. Some may have been trained in the use of other, similar screening approaches, and so able to accomplish similar goals without adoption of the full ADReS. Data on use of other approaches were not collected. Limited face time with patients and reimbursement constraints are also potential barriers to adoption of the ADReS. More research is needed to understand these issues.

Anecdotally, a number of participants reported a desire for additional training in the future. At first glance, one might assume that the Curriculum opened a door to new learning and professional competence. Learning some information may have motivated a desire to increase knowledge further in this area. It is notable that many of those wanting more information after the training attended the longer workshop format which included extra case-related discussion. The longer workshop format may have drawn more committed participants from the outset (since they had to make arrangements for time off, travel). These are important issues to consider in the design of future programs, yet changes in attitudes and behaviors were found for participants involved with both presentation formats. Further studies are needed to determine which presentation components and in what format are necessary for continued motivation and learning. The fact that a small proportion (<10%) of our post-test respondents reported less confidence supports this recommendation, as their learning needs may have been better served with a different approach.

This study had a number of limitations. First, while all trainees completed the pre-test questionnaire, only slightly more than a third (235, 38%) completed the three month post-test. Analyses concerning changes in confidence, knowledge and behavior were based on this small subsample. We cannot rule out the impact of selection bias. It is possible that those who responded three months after training were more motivated to change, or simply had more opportunities to use the training with the patients who presented for care. If true, a longer follow-up period may have encouraged more to respond. A prior similar study (Meuser et al., 2006) measured and documented change over 12 months. Still, low response rates are more often than not the norm in health professional surveys, and 38% is within the acceptable range (Cummings, Savitz & Konrad, 2001).

Another limitation was the administration of the educational offerings in a variety of settings by different teaching teams since the AMA Curriculum is not 100% scripted. While trained over two days in how to implement the Curriculum, teaching teams still enjoyed some latitude for tailored presentation of slides and video material. While every effort was made to cover the same core topics in both formats, some variation is likely to have occurred. Teams presenting workshops had more time, and so may have covered more material and addressed more participant questions/needs. Data was not collected on how material was conveyed to participants within and across presentation types. While this discussion is speculative, it should be noted that the complexities of this topic defy full standardization and every trainee group will have different questions and learning needs. It may be more important to offer a choice of lecture and workshop formats in any geographic area, such that trainees may self-select to attend one or the other based on personal preference. This was not the case in the AMA education series, as teams relied on local partners to define which format (lecture vs. workshop, but not both) would be offered in their respective facilities.

This type of educational intervention and evaluative research is resource intensive. Multiple trainers are needed and, ultimately, only those in the room for any given presentation may benefit. Secondary dissemination is a plus when it occurs, but fewer than thirty trainees (11%) reported such secondary efforts. The AMA is developing a web-based system for offering this training in a more standardized and targeted manner. Changing demographics mean that safety concerns for older drivers will increase for years to come, and so ready access to training is a must. Training about driver fitness early in medical education is also important. Medical and other students are a reasonable target group for such training, as they will determine practice priorities in the future.

When surveyed concerning participation in continuing education programming, physicians and nurses alike endorse a strong preference for live, didactic instruction over self-study materials (Goodyear-Smith, Whitehorn, & McCormick, 2003; Meuser, Boise & Morris, 2004; Reddy, Harris, Galle, & Seaquist, 2001). Certain professionals want and need to discuss the gray areas with a live instructor, offer suggestions, and build their own local networks to address the challenges of mobility management. Our data reveal that in-person presentations of 1–2 hours duration can impart sufficient knowledge to impact favorably on clinical practice with respect to the evaluation of driver fitness.

While physicians and other health professionals appreciate live, in-person training opportunities, a benefit of the AMA approach is its flexibility. As noted previously, the Guide is also formatted for self-study and includes a quiz for continuing education purposes. The Guide & Curriculum will be reformatted for module-based on-line training to start in early 2011. On-line training is most accessible, as anyone with a computer and internet connection may log in and participate. A downside of the self-study and on-line approaches, however, is the inability to ask questions in real time and learn from the experiences of other trainees. Since this training has broad applicability in the healthcare of older adults, choices are important to maximize dissemination and adoption of evidence-based strategies.

Further research is needed to determine the relative merit and impact of these three teaching formats: self-study, in-person, and on-line offerings. Our data suggest that certain professional groups will gravitate to one or another. Busy physicians, for example, may prefer on-line training, whereas rehabilitation professionals and social workers may prefer longer, in-person offerings. The self-study and on-line approaches may be least expensive, and so data on attendance preferences would allow for appropriate targeting and resource allocation in the future. The “staying power” of the training in terms of practice change is another issue for longitudinal investigation.

Concluding Thoughts

This study demonstrates that targeted, evidence-based educational programming can enhance the medical care of older patients with respect to driver fitness. Physicians and other health professionals are coming to see driving as a relevant concern in the care encounter. It is important to recognize, however, that this concern applies primarily to a subset of older drivers. Statistics show that the majority of older drivers are safe drivers, and their continued on-road mobility should be encouraged and facilitated. Many older drivers are adept at self-regulating their driving (e.g., daylight hours, lower speeds) to avoid problems. The AMA Curriculum is focused on the minority who are at-risk due to medical conditions that impact directly on the driving task. The core benefit of this training for public health and safety is early detection and intervention.

Too often, in our current system, the medically at-risk driver is not recognized as such until a crash or other on-road incident has occurred (Meuser et al., 2008). The AMA Guide and Curriculum encourage action to detect relevant conditions as they emerge and before driving may be compromised. While the AMA emphasizes the important role of physicians and other health professionals, the responsibility for action extends more broadly, and includes older patients themselves, their family members, social service professionals, driver licensing officials, law enforcement, and other community stakeholders. Communication and collaboration among these stakeholders are essential, as any one may be in a position to detect a problem and initiate action before a crash compromises public safety.

Answering the medical fitness question is an important step in a larger process. When problems necessitate retirement from driving, new mobility needs and challenges appear. This is where the emerging profession of mobility counselor/manager can step in and connect the dots between driving and non-driving mobility options (MacKean et al., 2007). Since we all risk losing the driving privilege if we live long enough, an important challenge for our aging society is building viable alternative transportation options so retirement from driving can become part of an accepted, long-planned process. The AMA Curriculum assists health professionals in tackling these important issues and steps with their patients and families.