Current Screening and Assesssment Practices

Despite acknowledgment among many researchers that people with dementia should cease driving at some point during the disease, there is no universally accepted test or standard that defines when driving should stop (Lincoln, Radford, Lee, and Reay, 2004). Who should test, how extensive it should be, and what constitutes a valid test are many of the questions that remain without definitive answers. This section reviews current screening practices including the differences in State policies related to driving assessment.

Screening for Driving Safety: What Tests Should Be Used and Who Is Responsible?

Screening Processes

Although research provides insights about why driving is difficult for those with dementia, it has yet to determine what degree of cognitive impairment constitutes an unacceptable risk (Vegega, 1990). As a result, many clinicians have sought methods by which fitness to drive may be determined. Much of the literature surrounding this question focuses on the Mini-Mental State Examination (MMSE) and additional neuropsychological tests to rate cognitive ability and estimate driving fitness from the results (Adler et al., 2000; Lincoln et al., 2004). However, the MMSE was not designed to be, nor has it been validated as, a predictor of driving safety; and indeed studies have shown it to be inconclusive at predicting level of crash risk (Dobbs et al., 1998; Fox, Bowden, Bashford, and Smith, 1997; Lincoln et al., 2004; Shua-Haim and Gross, 1996) and also that it fails to evaluate perception, attention, and motor skills, three areas thought to be essential to competent driving (Reger et al., 2004; Vegega, 1990). Uc et. al (2004), in assessing navigation and safety errors in comparing a sample of 32 people with probable AD with 136 neurologically normal adults, concluded that drivers with AD made more errors than the normal group on a route-following task that placed demands on driver memory, attention, and perception.

Other tests besides the MMSE have been used in assessments of driving fitness. The Clinical Dementia Rating (CDR) scale is used by neurologists to classify Alzheimer’s disease severity. The scale is based on categories that include memory, judgment, problem solving, and personal care. In 2000 the American Academy of Neurology (AAN) released a review recommending that Alzheimer’s patients with an impairment level of CDR 1 should not drive an automobile. In addition, it recommended that individuals with a CDR of 0.5 be referred for a professional driving evaluation as they may pose a serious traffic safety problem (Dubinsky, 2000). It should be noted that the AAN is currently reviewing its guidelines.

The Useful Field of View test (UFOV) has also been used to detect cognitively impaired drivers. This test measures speed of mental processing when the attention of an individual is divided, as it often is during driving. The test was used during the Maryland Pilot Older Driver Study conducted by NHTSA in 2003 as one of a number of tests to determine driving ability (National Older Driver Research and Training Center, 2003). The Gross Impairment Screening tool (GRIMPS), developed by the NHTSA, has also been used for driver evaluation of individuals with dementia. While the CDR, UFOV, and GRIMPS have not yet been validated for this purpose, researchers are using these tools in studies of cognitive and perceptual factors in aging and driving performance. Rinalducci, Mouloua, and Smither (n.d.) observe that changes in older drivers might best be measured using neurological measures and the UFOV.

However, neuropsychological tests have been found to be no more reliable than the MMSE, despite the fact that they can measure specific areas of cognitive ability, including visual-spatial skills, attention, and choice reaction time. Such tests measure the ability to perform nonspecific tasks, while driving involves task-specific processes that are learned through practice (Withaar, Brower, and van Zomeren, 2000). It is thought that only through a combination of tests can driving abilities be evaluated in a valid way (Hunt, 2003; Lincoln et al., 2004; Reger et al., 2004). Cognitive tests alone are not sufficient to determine fitness to drive. It is essential to develop some battery of tests that can distinguish safe drivers, those who are unsafe, and individuals with dementia who need further evaluation. Such a test would reduce the number of individuals needing further evaluation and would result in less cost for screening programs (Lococo and Staplin, 2005a).

Some researchers uphold on-road driving assessments by experienced driving evaluators as the gold standard to evaluate driving abilities (Wang, Kosinski, Schwartzberg, and Shanklin, 2003). The DriveABLE Program, developed by Dr. Allen Dobbs at the University of Alberta, is an example of an evaluation program based on eight years of research on the driving abilities of those with mental impairments (Dobbs et al., 1998). The program consists of two phases, starting with in-office testing of cognitive abilities and proceeding to in-car testing when necessary.

Snellgrove (2005) reports on a new cognitive screening instrument, the Maze Task, developed to assess the competence of drivers with mild cognitive impairment (MCI) or early dementia. In a study of 115 community-dwelling older people, 50 percent of those with MCI failed the task and 75 percent of those with early-stage dementia failed. This task correlates with known measures of attention, visuoconstructional skills, and executive functions of planning and foresight and underscores the concern related to safe driving among people with early-stage dementia or MCI.

Some argue that individuals with dementia who need driving evaluations should be assessed multiple times in multiple settings (Lococo and Staplin, 2005a). This is believed to be necessary to counteract the “good day/bad day” behavior of Alzheimer’s disease; evaluators must assure that competency is constant and not the result of an individual having a good day. Advocates of simulator use believe that the simulator more easily addresses that concern. Driving simulators that score safety error are also believed by some researchers to be effective, and some studies have shown them to correlate directly to driving evaluators’ assessments (Szlyk et al., 2002). Such simulators have multiple advantages in that they require less training for staff and do not put evaluators in cars with potentially dangerous drivers. Concern has been raised, however, about the simulators’ ability to replicate the vehicle environment accurately. They also induce motion sickness in some test-takers.

Who Conducts the Screening

When nothing more than basic cognition tests are done, a physician is usually responsible for the testing. However, physicians generally dislike being designated the “licensing gatekeeper” (Skinner and Stearns, 1999). They also express concern about the potential to disrupt the rapport they have established with patients and cite insufficient time during the office visit to discuss driving (Silverstein and Murtha, 2001). Where no other provisions are in place, responsibility for assessment typically shifts between the Department of Motor Vehicles (DMV) and licensed medical practitioners (Skinner and Stearns, 1999). When asked, individuals with dementia mentioned both themselves and their family members as better evaluators of their driving abilities than their physicians (Adler et al., 1999; Adler and Kuskowski, 2003). Ott et al. (2005) studied a cross-section of 50 drivers with mild dementia and compared clinicians’ safety assessments with those of a professional driving instructor. The researchers noted that while clinicians who were especially trained in dementia assessment were the most accurate predictors among the range of clinicians (62 to 78%), clinicians’ assessments alone was not adequate to determine driving competence with mild dementia. McKenna, Jefferies, Dobson, and Frude (2004) also provide useful insight regarding a cognitive battery to predict who will fail an on-road driving test, demonstrating 100 percent accuracy for subjects less than 69 years old; more research is needed with older age samples, however. Their cognitive assessment tool, the Rookwood Driving Battery, is described in McKenna et al. (2005).

When the decision to stop driving is made, Adler and Kuskowski (2003) found that, among those they surveyed, the physician was most often identified as the decision maker. In September 2003, the American Medical Association (AMA), in conjunction with the National Highway Traffic Safety Administration, published a Physician’s Guide to Assessing and Counseling Older Drivers (see Wang et al., 2003). In this manual, the importance of driver evaluation is discussed, including what the AMA sees as the ethical obligation of physicians to assess drivers for the safety of society. The AMA asserts that, in the case of a known unsafe driver, the threat to the public safety outweighs considerations of the doctor-patient relationship.

The Physician’s Guide recommends two brief tests for conducting a driving evaluation. The Trail-Making Test, Part B (only), and the Clock Drawing Test (CDT) with the Freund Clock Scoring for Driving Competency are considered by the AMA to be useful in identifying people who should be referred to a specialist for more in-depth screening (Wang et al., 2003). Freund et al. (2005) report that the CDT is a reliable, valid, time-effective screening tool for primary-care physicians to use in identifying at-risk drivers in need of further evaluation.

However, primary care physicians may not be the best qualified to evaluate driver safety for individuals with dementia. Brown, Ott et al. (2005) found that general practitioners’ assessments of drivers’ ability matched those of a qualified driving evaluator just 72 percent of the time. The same study showed that only experienced neurologists who were able to conduct full patient evaluations were able to predict driver safety with accuracy comparable to that of the driving evaluator.

When an evaluation calls for an in-depth screening, patients might be referred to a driver rehabilitation specialist (DRS). These are professionals who specialize in assessing driver ability and implementing strategies to increase driver safety. They also make recommendations about when and where individuals should drive, or whether driving should cease altogether (Wang et al., 2003). Such evaluations are most effective when the examiner has the complete medical history of the individual being assessed (Marottoli, 1998), but the history may not be available because of issues related to doctor/patient privilege or the patient’s unwillingness to cooperate. Some occupational therapists also provide similar evaluation and counseling services. The American Occupational Therapy Association (AOTA) recently launched an Older Driver Initiative aimed at increasing the number of occupational therapists (OTs) who offer driving therapy and evaluations. (AOTA [2004] describes the OT’s role in driving and transportation alternatives for older adults.) An initial evaluation session can cost up to $500, however, and Medicare, Medicaid, and private insurance companies often do not cover the cost of these services.

Some States mandate driver screenings, either upon diagnosis and referral by a physician or when the driver reaches a specified age. The effectiveness of such practices remains to be studied extensively in the United States. Currently, responsibility for recognizing driver impairment lies chiefly with the impaired drivers themselves and their family members, a risky situation given that dementia patients often lack the insight to evaluate their own abilities and their caregivers may not understand the implications for community mobility of an Alzheimer’s diagnosis (Silverstein and Murtha, 2001). For example, Adler, Rottunda, and Kuskowski (1999) found that 43 percent of caregivers surveyed believed that the driver with dementia would be able to continue driving throughout the course of the disease. While it is true that many older adults begin to modify their driving as their abilities decline (Brayne et al., 2000), individuals with Alzheimer’s disease are often unable to recognize the loss of their abilities (Molnar, Eby, and Dobbs, 2005; Wild and Cotrell, 2003).