Gottlieb · BMC medicine 2018 · prospective cohort study · n=3201

Sleep deficiency and motor vehicle crash risk in the general population: a prospective cohort study.

Cited 159 times in the scientific literature.

Level 3 - non-randomized controlled study

Prospective cohort study

PubMed 29554902 · doi:10.1186/s12916-018-1025-7 · record verified 2026-08-28

What was done

This prospective observational cohort study nested within the Sleep Heart Health Study evaluated 3,201 community-dwelling adults (1,745 men and 1,456 women aged 40–89 years). Sleep apnea was measured at baseline by home polysomnography, and questionnaires assessed usual sleep duration and daytime sleepiness. A follow-up questionnaire 2 years later ascertained driving habits and motor vehicle crashes occurring in the preceding year. Logistic regression adjusted for relevant covariates was used to analyze crash risk, and population-attributable fractions were estimated.

What was found

Among 3,201 evaluable participants, 222 (6.9%) reported at least one motor vehicle crash during the prior year. Crash risk was significantly associated with higher apnea-hypopnea index (p < 0.01), fewer hours of sleep (p = 0.04), and excessive daytime sleepiness (p < 0.01). Severe sleep apnea was associated with a 123% increased crash risk compared to no sleep apnea. Sleeping 6 hours per night was associated with a 33% increased crash risk compared to sleeping 7 or 8 hours per night. These associations persisted in participants who did not report excessive sleepiness. The population-attributable fraction was 10% for sleep apnea and 9% for sleep duration less than 7 hours.

Why it matters

This study demonstrates in a general community sample that both sleep apnea and short sleep duration substantially elevate motor vehicle crash risk. Notably, these risks persist even in drivers who do not report feeling excessively sleepy.

Limits

Motor vehicle crashes, sleep duration, and excessive daytime sleepiness were all self-reported rather than objectively verified through driving records or actigraphy. The study population was restricted to adults aged 40 to 89 years, limiting generalizability to younger drivers.

Cited by