Report: Drowsiness Causes 8 Times More Accidents Than Government Data Suggests

Drowsiness in drivers has been found to cause a larger proportion of crashes than previously thought, according to a new Research Brief by the AAA Foundation for Traffic Safety. The study at hand used the PERCLOS measure, which focuses on the amount of time a person’s eyes are closed, or nearly closed, during a certain interval. 3,593 drivers were monitored over the course of several months between October 2010 and December 2013. Various cameras were installed in their cars, as well as other measurement devices, in order to track how often their eyes closed.

A total of 905 accidents occurred to the participants in the study. Footage was analyzed in either in the final three minutes prior to a crash, or the final minute prior to a crash (if the full three were not usable). According to the report, “the driver was classified as drowsy if the driver’s eyes were closed in 12% or more of the video frames in the three-minute or one-minute period preceding the crash.” The researchers found that drivers were considered drowsy in 9.5% of the crashes, and in 10.8% of high severity crashes that involved “property damage, airbag deployment, rollover, injury.” For younger drivers (under the age of 25), as much as 31% of the crashes were categorized as severe.

These findings, which are the first to use the validated PERCLOS measure to determine drowsiness before a crash, show that far more crashes occur as a result of drowsiness than had previously been reported by the National Highway Traffic Safety Administration (NHTSA). It was previously believed that “driver drowsiness was involved in an estimated 1.4% of all police-reported crashes nationwide, 2.0% of crashes that resulted in injuries and 2.4% of crashes that resulted in a death in years 2011–2015.” It is recommended that adults get at least seven hours of sleep per night. Doing so would help cut down on the number of crashes resulting from drowsiness — a far greater problem than government statistics have indicated in the past.