Section 14: The Aging Driver

14.4 Dementia

A potentially high-risk group of older drivers are those with a dementing illness (Dobbs, 1997; Dubinsky,Williamson, Gray, and Glatt, 1992; Gilley, Wilson, Bennett, et al., 1991; Kaszniak, Keyl, and Albert, 1991).Prevalence estimates from the Canadian Study on Health and Aging (1994) suggest that eight percent of all Canadians aged 65 and older met the criteria for dementia, increasing to a staggering 34.5 percent for those age 85 and older. Alzheimer's disease, the most common form of dementia, currently has a prevalence rate of 5.1 percent overall or 161 cases per one hundred thousand with a projected prevalence rate of 509 cases per 100,000 by the year 2031 (Canadian Study on Health and Aging, 1994).

Alzheimer's disease (AD) is a progressive, degenerative brain disorder characterized by impairments in multiple cognitive functions. The earliest cognitive symptoms include difficulties in recent memory, word finding, confrontation naming, orientation, and concentration. Slowed rates of information processing, attentional deficits, disturbances in executive functions, impairments in language, perception, and praxis characterize the cognitive changes in later stages. The impairments in cognitive functioning increasingly interfere with social and occupational functioning, including the driving abilities of affected patients. Because of the progressive nature of the disease, at some point in the course of their illness, all individuals with a progressive dementia will become incapable of driving safely and will eventually stop driving. However, many patients continue to drive after the onset of their illness. An early study of retirement community residents indicated that 31 percent of those driving were suffering from a dementia (Waller, 1967), and 28 percent of patients (192) referred to a dementia clinic were active drivers (Odenheimer, 1993). Moreover, research suggests that many patients continue to drive for as long as four years following initial diagnosis (Friedland et al., 1988, Gilley et al., 1991; Lucas-Blaustein et al., 1988).

The problems associated with driving in this population are receiving increasing attention, and research findings using crash data suggest there is cause for concern. Results from one of the earliest studies examining the crash risks associated with dementia were reported by Waller (1967) who compared the driving records of 83 normal older drivers to those of 82 older drivers described as 'senile', 80 drivers with cardiovascular disease, and 199 drivers diagnosed with dementia and cardiovascular disease. The comparisons revealed crash rates of 12.1, 19.3, 14.7, and 36.2 crashes per million miles driven for the four groups, respectively. More than twenty years later, Friedland et al. (1988) compared the driving history of 30 patients with dementia of the Alzheimer's type (DAT) and 20 healthy age-matched controls. Results from this investigation revealed that individuals with DAT were nearly five times more likely to have had a crash than healthy elderly controls. Thirty percent of the dementia patients studied by Lucas-Blaustein et al. (1988) had at least one crash and another 11 percent were reported to have caused a crash since the onset of the disease.

Recent research corroborates these early findings of increased rates of crashes for dementia patients compared to age-relevant population rates and/or elderly controls. As shown in Table 35, the most well documented finding from retrospective studies is that patients with dementia have crash rates that far exceed those of non-dementing seniors (Cooper et al., 1993; Drachman and Swearer, 1993; Dubinsky et al., 1992; Friedland et al., 1988; Gilley et al., 1991; Lucas-Blaustein et al., 1988; O'Neill et al., 1992; Tuokko et al., 1995). Despite the variation in the retrospective methodology, the majority of the evidence provides a clear indication that, as a group, patients with dementia who continue to drive pose a considerable public safety risk. It is interesting to note that all studies find a small but significant subset of dementia patients competent to drive, a point that will be addressed shortly.

In addition to retrospective surveys and driving record examinations, a number of studies have examined the driving ability of patients with dementia using on-road assessments (Cushman, 1992; Dobbs, 1997; Fitten et al., 1995; Hunt et al., 1993; Kapust et al., 1992; Odenheimer et al., 1994; Shemon and Christensen, 1991). As evidenced from the results in Table 35, dementia patients, in general, perform less well than their counterparts without dementia on tests of on-road performance.

Of particular note to this discussion is the use of the Mini Mental Status Exam [MMSE] (Folstein, Folstein, and McHugh, 1975) as a predictor of driving competency in individuals with a dementia. The MMSE is a short screening instrument of cognitive status consisting of questions and tasks designed to assess orientation to time and place, registration of verbal information, attention and calculation, recall, language and visual construction. Importantly, a number of organizations, including medical ones, have suggested the use of the MMSE to screen drivers. In fact, according to results from Miller and Morley (1993), the majority of physicians surveyed felt the MMSE was the best available mental status examination for fitness-to-drive evaluation. However, the evidence to date indicates that the MMSE is of questionable utility for identifying individual driving competency.

A number of retrospective studies compared the MMSE scores of patients with dementia, who reported involvement in at least one crash, with patients who were not involved in car crashes. In many cases, the difference in MMSE scores between the two groups was less than one point (Friedland et al., 1988; Gilley et al., 1991; Lucas-Blaustein et al., 1988). No differences in MMSE scores between patients with diminished driving ability and patients with preserved driving ability have been reported by O'Neill et al. (1992). Using a global severity score (rather than MMSE), Drachman and Swearer (1993) found no difference between patients who had a collision and those who had not.

The data from studies that have examined on-road performance provide a somewhat different picture, with reliable correlations reported between MMSE scores and on-road performance. Odenheimer et al. (1992) reported a substantial correlation (r = .72) between the MMSE score and their in-traffic driving score. Similarly, Hunt et al., (1993) found that a combined score from the Clinical Dementia Rating scale was related to driving outcome on a road test (pass/fail) using Kendall's Tau Coefficient, a measure of association between ranked data ( T = .50). Fitten et al. (1995) reported that drive scores from their specially designed road test were strongly correlated with a transformed MMSE score (r = -.63). However, in the Fitten et al. study, mental status scores did not correlate with drive scores at the upper end of the MMSE scale. Although the correlations reported above show a clear relationship between mental status as measured by the MMSE and driving performance over a group of individuals, they do not indicate that the MMSE score is a sufficient predictor of the driving performance of individuals.

Table 35  Summary of the Research Literature on Driving and Dementia