The prevalence and nature of stopped on-the-road driving tests and the relationship with objective performance impairment.

The on-the-road driving test in normal traffic is applied to examine the impact of drugs on driving performance. Although participants are accompanied by a licensed driving instructor, under Dutch law, the driver is primarily responsible for safe driving and is not permitted to continue driving when it is judged that the drug compromises safety. This review examined the prevalence and nature of stopped driving tests, and the relationship with standard deviation of lateral position (SDLP), i.e. the weaving of the car¥. A literature search was conducted to gather all publications on clinical trials that applied the on-the-road driving test, examining the effects of central nervous system drugs such as anxiolytics, antidepressants, antihistamines, analgesics, and hypnotics. A total of 47 papers reported on 50 Dutch clinical trials in which 1059 subjects participated (903 healthy volunteers and 156 patients). A total of 7232 driving tests were performed; 5050 after drug treatment and 2042 after placebo. 3.1% of all driving tests were terminated before completion: 4.1% after drug treatment, and 0.7% after placebo. The decision to stop a driving test was 3-4 times more often made by the driving instructor than the subject. The most common reasons for stopping were the driver feeling tired or sleepy, or the driving instructor noticing signs of drowsiness and performance impairment. Although SDLP values of stopped driving tests are sometimes high, there is no clear relationship between SDLP (changes from placebo) and the decision to stop a driving test. Based on 8 studies that reported exact data, 39.6% of stopped drivers had a lower and 60.4% had a higher SDLP than 35ácm, i.e. the cut-off point of safe driving. This confirms that perception of the driver as well as judgment by the instructor of driving to be "unsafe" differs between individuals. Driving tests are sometimes stopped after drug treatment or placebo. The decision to stop driving is not a good correlate of objective performance. (A) Reprinted with permission from Elsevier.