Abstract
In the United States, the future group of older adults, the baby boomers, will be healthy, well educated, largely female, and living in the community (Dellinger, 2012). In 2005, delegates of the White House Conference on Aging voted transportation as one of the top three issues for older adults. The delegates recognized that community mobility, which for most people means driving a privately owned motor vehicle, underlies the means to remain independent in one’s home and community. Currently, most older adults are licensed to drive and will make the majority of their trips in a private vehicle. Considering the lifestyle trends in the growing aging population, older adults will continue to be dependent on the private car and continue to drive or risk becoming isolated (Rosenbloom, 2012). For the practice of occupational therapy, this dependence means that as an instrumental activity of daily living (IADL), community mobility, including driving, is a critical area of occupation that should be addressed with each client. In fact, in a recent study (Dickerson, Reistetter, & Gaudy, 2012), driving was identified as the IADL most important to older adults in a rehabilitation program. Of the 30 participants, 27 indicated that resuming driving was their primary goal in recovery; the remaining 3 had previously ceased driving.
Currently, a driver rehabilitation specialist (DRS), a professional with specialized skills and knowledge, typically assesses a driver if a question arises as to his or her fitness to drive. Many DRSs gain experience, pass an exam, and achieve the title of certified driver rehabilitation specialist (CDRS). Although most CDRSs are occupational therapists, it is not a requirement for taking the exam or for being a DRS. To determine fitness to drive, a DRS or CDRS usually does a comprehensive driving evaluation consisting of a clinical portion and then a behind-the-wheel (BTW) assessment with a specialized vehicle that includes a passenger brake and extra mirrors and may include other adaptations for physical accommodations. Typically, the clinical component, which covers the assessment of the visual–perceptual, cognitive, and physical skills needed for driving, may take 1 hr or more.
At this time, no single assessment tool or set of assessment tools is considered a scientifically valid predictor of motor vehicle crashes (Bédard, Weaver, Darzins, & Porter, 2008; Classen, Dickerson, & Justiss, 2012; Hogan, 2005; Marshall, 2008). Many researchers have argued that the BTW assessment is the gold standard for final determination of pass or fail (Langford et al., 2008; Wheatley & Di Stefano, 2008); however, little consistency exists in what defines a BTW assessment (Classen et al., 2012; Korner-Bitensky, Bitensky, Sofer, Man-Son-Hing, & Gelinas, 2006). Additionally, no standardization exists in the assessment tools that DRSs select to use with clients. Thus, the purpose of this study was to describe the current evaluation practice and assessment tools used by DRSs.
In 2006, results from a similar study were published. Korner-Bitensky et al. (2006) surveyed 114 clinicians who attended the 2006 national meeting of the Association for Driver Rehabilitation Specialists (ADED). The key components of the clinical part of the assessment included vision, visual–perceptual, cognitive, motor function, and sensory function. They found that computer-based evaluations (e.g., Useful Field of View
This study was initiated in an effort to update the evaluations and establish the most frequently used assessments. An additional objective was to explore how DRSs use assessment results to make the recommendation for fitness to drive. Moreover, in recognizing that different assessments might be used with clients with various diagnoses, respondents were specifically asked about which assessments tools they used with client groups with various functional deficits.
Method
Research Design
This study used a descriptive, nonexperimental design with a survey format, approved by the University and Medical Center Institutional Review Board of East Carolina University. A letter described the study and told respondents that consent was granted when they responded to the survey.
Participants
The participants consisted of DRSs initially identified from the ADED’s database and the American Occupational Therapy Association’s (AOTA’s) Driver Safety Web site (www.aota.org/older-driver). Participants were recruited through the databases by means of a request to all practitioners who evaluated clients for driving fitness (inclusion criteria). Practitioners who were in the databases but who did not evaluate clients were excluded.
Instrument
The self-administered survey consisted of a list of all known driving assessments, demographics, and descriptive questions about the driving evaluation process. The initial list of assessments was developed from an extensive review of the literature identifying any clinical screening or assessment tool used to evaluate driving knowledge or skills. The comprehensive list was developed through a survey on frequency of use and sent to four experienced CDRSs. On the basis of their feedback, the survey was redesigned to address not only whether the tool was used but also with what type of client, and additional tools were added when identified as missing.
Screening and Assessment Tools.
A final list of 114 specific assessment tools was identified. The tools were grouped by the kind of assessment: ocular movement, visual skills, visual perception, cognitive and memory, physical skills, driving skills, and off or on road. For each tool, the respondent checked a box to indicate whether the tool was used with all clients or with the following specific diagnostic groups: neurological, spinal cord, dementia, orthopedic, developmental disability, or new driver. The category of never use the tool or not applicable was also used. For the BTW questions, respondents were asked to identify the average length of time taken for off-road and on-road assessments.
Decision Process.
To explore which tools practitioners valued most, a section on the survey asked respondents to list the top five assessments tools, including the BTW assessment, that they typically use to make a final recommendation for driving. After listing their five top tools, respondents were then asked to assign a percentage, to total 100, to each of the five tools and two additional categories, all other assessments and other factors/information.
Demographics.
Demographics included the respondents’ age, experience, and degree. Six questions asked the respondent to identify referral sources, types of setting, methods of communication (written or verbal), number and type of staff, reimbursement sources, and who contributed to the decision process. An open-ended question asked for other relevant information.
Data Collection
Data were collected in three ways. First, after eliminating all duplicates, 600 paper surveys were mailed to all specialists identified from the two databases in 2009 along with a self-addressed, postage-paid return envelope. With feedback from some respondents, the survey was replicated as a Web-based survey and e-mailed to all e-mail addresses in the same databases in 2010, with instructions to complete the survey only if the DRS had not completed the paper survey. At the 2010 ADED annual conference, initial results were presented and additional DRSs completed the paper survey or the Web-based survey. The survey was anticipated to take about 30–60 min to complete depending on the variety of clients seen and number of tools used by the DRS.
Data Analysis
Frequency distributions and means were calculated using SPSS Version 20 (IBM Corporation, Armonk, NY). Descriptive questions were summarized for similar and contrasting content.
Results
Demographics
Two hundred twenty-seven respondents—approximately one-third of the estimated 600 driving evaluators in the United States—completed paper or Web-based surveys.
The 227 respondents were 41 men (18%), 167 women (74%), and 18 (8%) who did not indicate their gender. The mean age was 44 yr (standard deviation = 8.39, range = 27–62 yr). Most were occupational therapy practitioners (n = 184; 81%), with 8 (4%) having another degree and 35 (15%) not listing their professional degree. Just more than half (n = 126; 56%) had a baccalaureate degree; 53 (23%) had a master’s degree, 7 (3%) had a doctoral degree, 7 (3%) had an associate’s degree, and 1 had a high school diploma (0.4%). Education was unknown for 33 (15%). (Percentages do not total 100% due to rounding.) The years involved with driving rehabilitation ranged from 1 to 40 with a mean of 11 (standard deviation = 8.0); 84.5% of the respondents had >3 yr of experience and 54.5% had >10 yr experience. Thirty-nine percent (n = 89) were CDRSs.
Table 1 lists the practice settings of the respondents (n = 186) who completed this section of the questionnaire, differentiating among those who did both clinical and BTW assessment, clinical only, or BTW only. For private practice with both clinical and BTW (n = 39), number of staff involved in delivery of services ranged from 1 to 8. Fourteen respondents reported working in practice on their own, and 12 respondents indicated they worked with 2–3 others; 9 respondents worked in settings with ≥4 in a private practice (4 did not indicate practice setting). The range for number of personnel in hospitals was similar, with 1–9 staff assigned to driving. Fifty-nine respondents indicated they worked in hospitals, with at least 1 respondent assigned to driving full-time and 56 respondents in part-time positions. Seventy-eight respondents (full or part-time) indicated they worked with 2–3 people assigned to driving, and 18 respondents indicated they worked with 4 or more colleagues in hospital settings.
Practice Settings of Respondents
Note. N = 186. BTW = behind the wheel; OT = occupational therapy.
Of the 186 respondents who answered questions about their practice setting, 53 (29%) were assigned to driving rehabilitation only, 28 (15%) spent more than half their time in delivering driving services, 35 (19%) split their time equally between driving and other services, and 65 (36%) spent less than half their time in delivery of driving services.
Screening and Assessment Tools
Table 2 provides the number and percentage of all respondents who used each assessment tool, how many respondents used the tool with all clients, and how many used the tool with select clients. Table 3 lists the tools that were used selectively by ≥19% of the respondents in the specific diagnostic categories of neurological impairment, dementia, spinal cord injury, developmental disabilities, orthopedic, and new drivers. Table 4 documents the decision-making process by listing the assessments identified by the respondents as their top five assessments in making a fitness-to-drive recommendation.
Assessments Most Frequently Used by Driver Rehabilitation Specialists
Note. N = 227. MMSE = Mini-Mental State Examination (Folstein, Folstein, & McHugh, 1975); MVPT = Motor-Free Visual Perception Test; ROM = range of motion; SBT = Short Blessed Test; — = not applicable.
Trail Making Test. bOptec is a visual screening tester from Stereo Optical Company, Inc. (Chicago).
Most Frequent Assessments Used Selectively for Specific Diagnoses (N = 227)
Note. N = 227. MVPT = Motor-Free Visual Perceptual Test.
Trail Making Test.
Assessments Most Frequently Selected by DRSs as Top 5 Used to Make a Driving Recommendation
Note. N = 181. MMSE = Mini-Mental State Examination; SD = standard deviation; UFOV = Useful Field of View.
Trail Making Test.
Behind the Wheel
All 227 respondents addressed the question of frequency and length of time for the off- and on-road assessments. One hundred five (46%) of the respondents took all clients BTW in an off-road venue. Twelve (5%) did this with select clients. The time taken for the off-road assessment ranged from 5 min to 4 hr with a mean of 44 min (standard deviation = 39.8, mode = 10 min, median = 30 min). A total of 170 respondents (75%) reported doing an on-road assessment with all clients, and 13 (6%) reported doing so with select clients. The length of the on-road assessment ranged from 5 min to 6 hr (mean = 70 min, standard deviation = 40, median = 60 min).
Technology-Based Tools
Fewer than a third of DRSs reported using computer-based technology. Seventy-two respondents (32%) indicated they used the UFOV test. Forty-two (19%) indicated they used it with all clients in terms of divided and selected attention (Subtests I and II), and 44 respondents (19%) used Subtest III or processing speed with all clients. Twenty-six respondents (12%) used the UFOV with select diagnostic categories for processing speed, and 28 (12%) used the divided and selected attention with select diagnoses: neurological (n = 27; 12%), dementia (n = 21; 9%), developmental disabilities (n = 14; 6%), and new drivers (n = 8; 4%).
Only 25 respondents (11%) indicated they used driving simulators with a range of different systems. Nine (4%) of the respondents use the Elemental driving simulator (Life Sciences Associates, NY) with clients with neurological or dementia as a diagnostic category. Other systems included the Drive-Able (DriveABLE Assessment Centres, Inc., Edmonton, Alberta, Canada; n = 2; 1%), Drivers Assessment System (manufacturer not known; n = 4; 2%), STISIM (Systems Technology, Inc. Hawthorne, CA; n = 5; 2%), and Doran (Doran Precision Systems, Inc., NY; n = 5; 2%).
Decision Making
When questioned who determined the outcome of the driving evaluation (n = 182), 60 respondents (33%) indicated that the person who performed the clinical and BTW evaluation made the decision. Forty-seven (26%) respondents indicated that the physician made the outcome decision either directly or by recommendation to the state driver licensing agency. Fifty-one (28%) respondents indicated that the decision was a collaborative one between the evaluator and physician. Fewer respondents indicated that the decision was made by the medical board of the state driver licensing agency (n = 5; 2%), clinical therapist (n = 7; 3%), on-road evaluator (n = 9; 4%), and client (n = 2; 0.8%), and 7 (3%) indicated the clinical therapist and BTW evaluator came to a decision that was communicated to the physician.
Communication and Referrals
To answer how the evaluation results were communicated to the referral sources, 105 of 143 respondents (73%) indicated they used written reports, 82 (57%) indicated that they discussed the results in addition to the written report, and 2 (1%) indicated they discussed the results and did not indicate any written documentation. Referral sources (n = 178) were primarily physicians (n = 145; 82%), the state vocational rehabilitation agency (n = 68; 38%), the DMV (n = 52; 29%), other therapists (n = 23; 13%), families or self (n = 21; 12%), and others (n = 31; 17%), which included vendors (e.g., vehicle modifiers), Department of Veterans Affairs, senior or retirement centers, case managers, and schools.
Discussion
Results from this research replicated some of Korner-Bitensky et al.’s (2006) findings. First, the key components identified as vision, visual–perceptual, cognitive, motor, and sensory functions continue to be the key components assessed by this group of DRSs. The assessments identified in this study as commonly used, the Trails A, Trails B, and MVPT, remained remarkably similar in percentage (76%, 84%, and 66%, respectively) to Korner-Bitensky et al.’s previous study. Only use of the brake reaction timer decreased, from 73% in 2003 to only 49% in this study. This decreased use is interesting, given the face validity of the brake reaction timer. DRSs may not be using simple brake reaction timers because evidence has demonstrated that it is not predictive (Classen et al., 2012), or the norms previously used may have been for a machine no longer manufactured (i.e., American Automobile Association brake reaction machine), and the void has yet to be filled. As with the earlier study, only a minority of specialists are using computerized assessments, specifically the UFOV (i.e., n = 72; 32%). However, this number is much higher than the 5 of 114 respondents in Korner-Bitensky’s (2006) study . Finally, as in 2006, standardized BTW evaluations are rarely used, and on-road testing continues to vary significantly.
The results of this study offer a closer examination of the use of assessment tools. Interestingly, practitioners appear to use a majority of assessment tools with all clients rather than to choose select assessments for specific diagnoses. Among the assessment tools listed in Table 1, the most common assessment is visual acuity. Although visual acuity is not as critical to driving safety unless it is <20/70 (Edwards et al., 2008; Margolis et al., 2002; McCarthy & Mann, 2006; Stav, Justiss, McCarthy, Mann, & Lanford, 2008) because of most states’ driver licensing agency guidelines, it is no surprise that most driving specialists test visual acuity to ensure clients meet the state’s minimal guidelines. However, with evidence emerging that contrast sensitivity is linked to driving outcomes measures (Bowers, Peli, Elgin, McGwin, & Owsley, 2005; Janke, 2001; McCarthy & Mann, 2006; Owsley & McGwinn, 1999; Owsley et al., 2002; Stav et al., 2008), it is surprising that fewer than half of the respondents screened for contrast sensitivity.
Because most driving evaluators are occupational therapists, evaluation of range of motion, muscle strength, and muscle tone is completed for almost all clients, because they are basic components of occupational therapy practice. By contrast, fewer numbers of cognitive assessments are consistently used by driving evaluators. Although the most frequently mentioned tools have good evidence linking their results to driving outcomes, for example, the Trail Making Test (Carr, Barco, Wallendorf, Snellgrove, & Ott, 2011; Keay et al., 2009; Munro et al., 2010), MVPT (Edwards, Bart, O’Connor, & Cissell, 2010; Korner-Bitensky et al., 2000; Oswanski et al., 2007), and clock drawing (Carr et al., 2011; Mathias & Lucas, 2009; McCarthy & Mann, 2006; Oswanski et al., 2007), almost half of the evaluators used other assessments, such as road signs, color perception (Owsley, Ball, Sloane, Roenker, & Bruni, 1991; Owsley & McGwin, 1999), and rules of the road (Stav et al., 2008), that have no evidence for predicting fitness to drive. This result may be because driving evaluators use standard protocols from manufacturers (e.g., visual slides from Optec Functional Vision Analyzer) or maintain the belief that using more assessment tools will result in a better outcome. Although that latter hypothesis may hold some truth, evaluation protocols warrant close examination in the present climate of cost containment and need for evidence.
Although the ability to examine driving specialists’ decision process is limited in a survey, the results are thought provoking. At least 40 different assessments were listed as the top 5 assessments for making a fitness-to-drive decision, illustrating the diversity of assessments across settings. However, as expected, the BTW component was by far the primary factor in making a decision. When the DRSs were asked to specify what percentage each of the assessments contributed to the recommendation decision, the range of percent responses ranged from 5% to 100% for the BTW, illustrating the diverse use of BTW assessments and the complexity of decision making. Although only 4 respondents indicated that their decision was based 100% (i.e., only) on the BTW assessment, 23 other DRS indicated that their decision was 95%, very close to being based only on the BTW assessment. With many driving evaluators using the BTW assessment as the gold standard (i.e., making the decision of fitness to drive on the basis of driving performance), it raises the question of how much clinical assessment is needed before the BTW component. If one considers that most driving evaluators conduct many assessments with each client, it is not surprising that driving programs are not profitable and needed services are not provided in a timely or cost-effective manner. These results underscore the need for continued research on and clear evidence for specific assessment tools and application of those tools to specific populations.
With the exception of the fine motor, balance, foot tap, or brake reaction tools, few of the assessment tools are performance based, and these few are narrowly focused on one factor in the driving paradigm (i.e., bottom-up assessments). Moreover, all the frequently used cognitive tests are paper-and-pencil or verbal-answer assessments rather than performance based. Because the practice of occupational therapy emphasizes the analysis of occupational performance (AOTA, 2008), investigating methods of using more performance-based assessments to evaluate fitness to drive seems essential. An example of such a study is the performance-based Assessment of Motor and Process Skills (Fisher, 2006) and driving outcomes decisions (Dickerson, Reistetter, Schold Davis, & Monahan, 2011; Dickerson, Reistetter, & Trujillo, 2010), but much more work needs to be done in this area.
In terms of performance-based evaluation, the use of the interactive driving simulator is clearly emerging as an option for assessment and training in driving. Early studies have demonstrated the effective use of driving simulators (Bédard, Parkkari, Weaver, Riendeau, & Dahlquist, 2010; Shechtman, Classen, Awadzi, & Mann, 2009). However, as evidenced in this study, few DRSs are using driving simulators, probably because of several factors including cost; insufficient evidence; the impact of simulator sickness, especially for older adults; and the fact that driving simulator platforms need much more work to be “clinic ready” for therapists.
The information gathered in this survey describes the practitioners who choose to specialize in driver rehabilitation. Most of the respondents were occupational therapists, with a higher percentage of male practitioners (18%) than in other areas of practice. A majority were older, experienced therapists (i.e., 54.5% had >10 yr experience) educated at the baccalaureate level. Most respondents worked in hospitals, rehabilitation centers, or outpatient clinics; 21% of therapists were in private practice. Only 30.5% of the DRSs indicated that they only did driving rehabilitation, whereas most respondents indicated they participated in other occupational therapy practice areas. In making the fitness-to-drive decision, practitioners appeared to collaborate, especially if different practitioners did the clinical and BTW portions; 26% indicating that physicians made the final decision either directly or by recommendation. The collaboration between specialists and physicians is one that needs further exploration, development, and education because a physician’s clinical examination of an older adult is not sufficient to identify increased crash risk (Johansson et al., 1996; Molnar, Byszewski, Marshall, & Man-Son-Hing, 2005), and physicians have indicated that they often do not have sufficient education and training in evaluating driver fitness (Marshall & Gilbert, 1999) and they are often unaware of DRSs’ services.
With the growing importance of mobility to the baby boomers, especially in maintaining a personal vehicle (D’Ambrosio, Coughlin, Pratt, & Mohyde, 2012), the number of DRSs using the current methods of evaluating fitness to drive will not meet the growing demand for services. As the number of older drivers continues to rise (32 million older drivers were licensed in 2008, a 20% increase since 1999; National Center for Statistics and Analysis, 2009), the results of this research suggest that the need for action is urgent. Work needs to be done to translate knowledge between research and practice. Although significant work has been done to support the use of evidence-based assessment tools, this study suggests that assessment tools may not be selected on the basis of evidence. More than half of the respondents in this study had >10 yr experience; if these experienced therapists are not using research to select assessment tools, there is cause for alarm. However, it is possible that the diversity may reflect the client-centered, individualized selection of assessment tools. More research is needed to explore how experienced clinicians are using their clinical expertise in conjunction with evidence-based assessments and the implications for assessment and intervention outcomes.
Implications for Occupational Therapy Practice
The results of this study suggest several actions for occupational therapy practice in this area:
Guidelines for BTW assessments need to be developed, as do guidelines and evidence for which assessments are most appropriate for specific diagnoses.
Continued work is needed to find the appropriate place for driving simulation and ensure that practitioners are trained to use this highly technical piece of equipment, especially because of the significant side effect of simulator sickness, which occurs frequently with older adults.
Research findings need to be clarified and the work translated to practice to find what is economically efficient for driver screening versus assessment.
Finally, the exploration of partnering general practice occupational therapists and specialists to provide more effective, efficient, and client-centered driving rehabilitation assessment and services is essential.
Fortunately, significant projects are underway that may assist in meeting some of these needs. Consensus statements from an expert panel specifically addressing occupational therapy and driver rehabilitation have been established and are being promoted (Schold Davis & Dickerson, 2012). Genesis Rehabilitation Services, in collaboration with AOTA, has initiated a project to increase the knowledge and skills of its general practice occupational therapists regarding driving and community mobility, making it a systemwide goal. The preliminary analysis is showing promising results (Dickerson, Schold Davis, & Chew, 2011; Schold Davis, Chew, & Dickerson, 2013). Finally, the Canadian Research Initiative for Vehicular Safety in the Elderly (CanDrive), a collaborative and interdisciplinary research project dedicated to improving the health, mobility, and quality of life of older Canadians, has partnered with occupational therapists to develop a valid, easy-to-use screening tool to allow clinicians to assess older adults’ medical fitness (Kennedy, 2010). All of these efforts and more will, it is hoped, provide further opportunities to develop the evidence needed to provide driver rehabilitation services for a growing older population that will want and need to drive to maintain their independence.
Limitations and Future Research
As with all descriptive research, respondents did not provide the reasons for using specific assessments. Although this study goes further than previous research in specifying the types of diagnoses, information regarding why a particular assessment might be used over another is limited. Further exploration is needed to gather this information. Another limitation might be the two different formats (i.e., paper-based survey vs. Web-based survey). However, the Web-based survey replicated the paper-based survey, and care was taken to put the results together for analysis. In fact, the two formats likely increased the diversity of respondents. Finally, respondents did comment that the survey was lengthy, which may have discouraged some busy practitioners from completing it. Nevertheless, one-third of the population of DRSs completed the survey, which gives greater weight to the observations gathered from the data.
The next step in expanding the information from this study would be to explore why specific assessment tools are used with specific client groups. Understanding the clinical reasoning process for determining fitness to drive by expert clinicians would elucidate the process for all practitioners and further this area of practice. The information gained would also assist in determining which tools are most effective, efficient, and appropriate for individual clients. Understanding the specific value of the BTW assessment may also assist with gaining third-party payment for this type of assessment. Finally, specific research is needed on how occupational therapy assessments of function for IADL can be used as screening tools for making an appropriate referral for driver rehabilitation services.
Summary
In summary, this article describes the practice of DRSs, including which assessments are typically used for determining fitness to drive, which assessments are used most frequently with specific types of clients, and how the specialists use specific assessments to make recommendations. When the current results were compared with those of a past survey, the data suggest that current driver rehabilitation practice has not changed significantly, and the issues needing resolution in 2006 continue to be problematic. Current projects will meet some of these goals, but support of continued research and education is essential to meet the demands of our aging population.
Footnotes
Acknowledgment
I thank all the DRSs who contributed to this study by completing the survey.