Section 7:
Diseases of the Nervous System

7.3 Sleep Disorders

Sleep Apnea and Other Measures of Performance (continued)

George et al. (1996) compared the performance of sleep apnea patients with that of age-matched controls on a divided-attention driving task (DADT). Tracking errors and visual search measures (correct responses and reaction time) were measured. There were significant differences between the two groups on tracking errors, with the sleep apnea patients exhibiting a three-fold increase in errors. Although the results from the visual search measures were statistically significant, there was considerable overlap in the measures between the two groups (see Table 20 below).

Barbé et al. (1998), in a retrospective controlled study, compared the performance of sleep apnea patients with age- and sex-matched controls on a vigilance task (PVT: Psychometer Vigilance Test) and on Steer Clear. As can be seen in Table 20, individuals with sleep apnea performed significantly worse than controls (with the exception of the reaction fatigue measure). It is important to note, however, that despite the fact that the differences between the two groups were statistically significant, those differences are unlikely to be clinically meaningful because of the degree of overlap in the measures between the two groups. Importantly, the authors found no relationship between patient measures on Steer Clear and the vigilance task (PVT) or between Steer Clear measures and crash data from insurance companies.

In summary, although individuals with sleep apnea perform significantly worse than controls on laboratory based tests, it is unclear how those findings translate to real world driving performance. For example, Findley et al. (1991) found higher crash rates in individuals with either narcolepsy or sleep apnea who performed very poorly on Steer Clear compared to sleep apnea and narcoleptic individuals performing either poorly or normally on Steer Clear. However, sample sizes for each of the categories were small. Barbé et al. (1998), using a larger sample size, found no relationship between a number of vigilance measures and crash rates. In both studies, the time periods for crash rate measurement were not assessed as a function of sleep apnea diagnosis. Future studies with larger sample sizes and more clearly defined diagnosis-crash parameters are needed.

Table 20 Summary of Studies Examining the Relationship Between Measures of Disease Severity and Crashes and/or Simulator Measures

Relationship Between Measures of Disease Severity and Crashes and/or Simulator Measure

A handful of studies have examined the relationship between measures of disease severity and crashes, and/or driving simulator measures in individuals with sleep apnea. The results of those studies are summarized in Table 20.

Researchers that have examined the relationship between the AHI (apnea/hypopexmia index) and performance have found little, if any, relationship between AHI and Steer Clear scores (Flemons, Remmers, and Whitelaw, 1993), tracking errors and visual search measures (George, Boudreau, and Smiley, 1996), or crashes (Barbé et al., 1998). Measures of daytime somnolence also have been found to be unrelated to crash rates (Aldrich, 1989; Barbé et al., 1998; Flemons et al., 1993). Other measures of disease severity (e.g., O2 desaturation) or clinical measures (e.g., anxiety, depression, etc.) also have been found to be unrelated to crash data or simulator measures (Barbé et al., 1998; Flemons et al., 1993). Taken together, the results reveal that measures commonly used to measure disease severity in sleep apnea are not very useful in discriminating between individuals who are likely to perform poorly on laboratory based measures putatively related to driving performance or who are at-risk for crashes.