Section 7:
Diseases of the Nervous System

7.3 Sleep Disorders

Sleep Apnea and Driving Literature Review

A number of studies (see Table 18) have investigated the relationship between obstructive sleep apnea and motor vehicle crashes with results revealing a two- to three-fold increase in crashes in individuals with sleep apnea compared to controls.

There are a number of methodological limitations to those studies. The sample size in one half of the studies (Findley et al., 1988; 1989; George, Nickersen, Hanly,Millar, and Kryger, 1987) is small. In those studies with a larger sample size, the data are based on retrospective self-reports (Aldrich, 1989; Gonzalez-Rothi, Foresman,and Block, 1988, but see Barbé et al., 1998). In one study (George et al., 1987), the diagnosis of sleep apnea was not confirmed by polysonography in seven of the study participants. According to the authors, if any of these seven patients had a condition such as narcolepsy or idiopathic hypersomnia, their results would lose significance. In the most recent study (Barbé et al., 1998), the sample size is relatively large, with reports of crashes based on insurance company crash data. As noted by the authors, insurance company crash data are an objective source of data (although under-reporting may be a problem). However, in this study, data regarding motor vehicle crashes were included for a three-year period preceding the diagnosis of sleep apnea. Thus, crashes unrelated to sleep apnea may have been included in their statistics.

An additional limitation in studies investigating the relationship between sleep apnea and crashes is the lack of uniform diagnostic criteria. For example, in three of the studies listed in Table 18 (Findley et al., 1989; George et al., 1987; Gonzalez-Rothi et al., 1988), diagnostic criteria for sleep apnea are not specified. Barbé et al. (1998) defined sleep apnea as greater than 20 apnea/hypopnea episodes per hour of sleep as measured by polysomnography. Individuals included in the study by Aldrich (1989) all underwent polysomnography, with a respiratory disturbance index calculated based on the number of apneas or hypopneas per hour of sleep. Participants were then classified into four diagnostic groups, ranging from sleep apnea to other disorders of excessive daytime sleepiness. However, the criteria used for group classification were unspecified. Lastly, subjects diagnosed as having sleep apnea in the study by Findley et al. (1988) met the criteria of at least five obstructive sleep apneas or hypopneas per hour of sleep, which resulted in a drop in baseline oxyhemoglobin saturation of four percentage points.

Table 18  Summary of Studies Examining the Crash Rates of Drivers with Obstructive Sleep Apnea

* Classified according to severity of nocturnal hypoxemia associated with apnea

Finally, in the majority of studies, degree of driving exposure is not taken into account. Therefore, it is not clear if the elevated crash rates of sleep apnea individuals represent an overestimation or underestimation of risk compared to controls. In the one study where driving exposure is considered (Barbé et al., 1998), individuals with sleep apnea reported driving more kilometers per year (~27,000) than controls (~16,000). The reason for the significantly higher exposure per year in individuals with sleep apnea compared to controls is unclear. The patients and controls were matched for sex and age. The authors note that a selection bias is unlikely to account for the findings as an equal percentage of patients and controls came from areas surrounding the hospital. One possible explanation may be in terms of occupational differences. Although data on occupation were gathered in the Barbé study, those data were not reported. It may be that occupational differences account for the higher driving exposure of the sleep apnea patients. The findings from the Barbé et al. study are noteworthy in that the magnitude of the increased crash rates of sleep apnea patients persisted after controlling for number of kilometers driven per year.

The studies reported above suggest that individuals with sleep apnea are at increased risk of crashes. However, more research is needed using standardized diagnostic criteria, larger sample sizes, driving exposure, and objective measures of driving performance (e.g., crash data, on-road performance) gathered from time of diagnosis onward.