Driver risk factors for sleep-related crashes

Abstract

A population-based case-control study was carried out to examine driver risk factors for sleep-related motor vehicle crashes. Cases included 312 drivers involved in recent North Carolina crashes and identified on police reports as asleep at the time of the crash and 155 drivers identified as fatigued. Controls were 529 drivers also involved in recent crashes but not identified as asleep or fatigued, and 407 drivers not involved in recent crashes. All drivers were contacted for brief telephone interviews. Results showed that drivers in sleep-related crashes were more likely to work multiple jobs, night shifts, or other unusual work schedules. They averaged fewer hours sleep per night, reported poorer quality sleep, were less likely to feel they got enough sleep, were sleepier during the day, drove more often late at night, and had more prior instances of drowsy driving. Compared to drivers in non-sleep-related crashes, they had been driving for longer times, been awake more hours, slept fewer hours the night before, and were more likely to have used soporific medications. Knowledge of specific risk factors for sleep-related crashes is an important first step in reducing the thousands of deaths and injuries each year in the US attributed to drowsy driving.

Introduction

Drowsy driving is a key factor in an estimated 76,000–100,000 crashes occurring each year in the United States, resulting in 1500 deaths and thousands of injuries (Knipling and Wang, 1995, Wang et al., 1996). The actual toll may be considerably higher, since drowsiness also contributes to crashes by making drivers less attentive and by slowing their reactions and impairing their judgment (Lyznicki et al., 1998, Leger, 1995). As many as 1 million crashes are attributed to driver inattention, and drowsiness may be a hidden factor in many of these crashes.

Populations that have been shown to be at higher risk for involvement in sleep-related crashes include young persons, especially young males (Wang et al., 1996, Pack et al., 1995, Horne and Reyner, 1995); persons with undiagnosed or untreated sleep disorders (Findley et al., 1988, Findley et al., 1989, Aldrich, 1989, Cohen et al., 1992); drivers who have taken soporific medications such as benzodiazepine anxiolytics or sedating antihistamines (Ray et al., 1992, Ceutel, 1995); and night or rotating shift workers (Dalziel and Job, 1997, Marcus and Loughlin, 1996, Gold et al., 1992). Commercial vehicle operators are also at increased risk for drowsy driving and sleep-related crash involvement due to such factors as extended driving times, irregular work and sleep schedules, higher frequency of nighttime driving, and inadequate sleep (McCartt et al., 1997, McCartt et al., 2000, Abrams et al., 1997, Wylie et al., 1996).

For anyone who is already drowsy, consumption of alcohol can pose a special risk. Research has shown that alcohol and sleep loss interact synergistically to increase levels of sleepiness (Zwyghuizen-Doorenbos et al., 1988, Lumley et al., 1987). Studies conducted using driving simulators show significant decrements in performance when low doses of alcohol are given to sleep-deprived subjects (Roehrs et al., 1994). Circadian factors have also been shown to play a role in drowsy driving crashes. Our bodies are programmed for sleep during our nighttime sleep period and again 12 h later, between 2 and 4 in the afternoon. This cyclic pattern of wakefulness has been demonstrated in time of day plots for drowsy driving crashes as well as in other arenas where sustained vigilance is important for safety (Knipling and Wang, 1995, Wang et al., 1996, Pack et al., 1995, Folkard, 1997).

Prior studies of risk factors for sleep-related crashes have primarily involved either analysis of motor vehicle crash data or population-based surveys. Analysis of crash data can tell us who is most likely to be involved in a drowsy driving crash (young persons, males), when and where these crashes are most likely to occur (late night, rural roadways, etc.), how they occur (run-off-road, rear-end, etc.), and the severity of injuries that result. For example, analysis of North Carolina crash data has revealed that 55% of drivers who fell asleep at the wheel and crashed were age 25 or younger, and 75% were males. Three-fourths of the crashes in which the driver fell asleep involved a vehicle departing the roadway, and nearly two-thirds occurred at speeds greater than 50 miles per hour (Pack et al., 1995).

Surveys of the general driving population provide evidence of the prevalence of drowsy driving and drowsy driving crashes. In a 1994 survey of licensed drivers in New York State, 55% said that they had driven drowsy in the past year, and 28% said that they had fallen asleep at the wheel at least once during their driving careers; 3% had fallen asleep at the wheel and crashed, and an additional 2% had crashed when driving drowsy (McCartt et al., 1996). In UK, 29% of respondents to a mail survey reported that they “had felt close to falling asleep while driving” in the past year (Maycock, 1997), and in Norway, 1 in 12 drivers reported that they had fallen asleep while driving during the past year, with 4% of these episodes resulting in a crash (Sagberg, 1999).

When used separately, crash data and driver survey data can provide only limited information about the characteristics of drivers who are involved in sleep-related crashes or who report driving while drowsy. The present study combines motor vehicle crash and driver interview data to examine specific driver risk factors for sleep-related motor vehicle crashes in the general population of drivers. The study uses a case-control study design, where cases are drivers involved in recent sleep-related crashes, and controls are drivers involved in non-sleep-related crashes or not involved in recent crashes at all. The results of this controlled study offer new evidence for helping to reduce drowsy driving crashes and injuries.