0192 A Re-appraisal Of The Link Between Daylight Saving Time And Traffic Accidents In The US. (12th April 2019)

• Record Type: Journal Article
• Authors: Fritz, Josef; Wright, Kenneth P; Vetter, Celine
• Is Part Of: Sleep. Volume 42(2019)Supplement 1
• Page Start: A78
• Page End: A79
• Publication Date: 2019-04-12
• Subjects: Sleep -- Physiological aspects -- Periodicals; Sleep disorders -- Periodicals; Sommeil -- Aspect physiologique -- Périodiques; Sommeil, Troubles du -- Périodiques; Sleep disorders; Sleep -- Physiological aspects; Sleep -- physiological aspects; Sleep Wake Disorders; Psychophysiology; Electronic journals; Periodicals; 616.8498
• DOI: 10.1093/sleep/zsz067.191
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: To date, evidence addressing the effects of Daylight Saving Time (DST) on traffic accidents is inconsistent and often limited by geographical heterogeneity in accident occurrence and follow-up duration. The DST spring transition is thought to acutely increase traffic accident rates by inducing modest levels of circadian disruption and ~1h of sleep deprivation. Our goal was to re-examine the link between spring DST and traffic accident rates, considering changes in DST timing, potential differences in fatal/total accident rates, both nation-wide and within two urban settings with distinct meteorological profiles. Methods: We analyzed three US traffic accident data sources: (i) the Fatality Analysis Reporting System (FARS) database of the National Highway Traffic Safety Administration (fatal accidents since 1975); (ii) all police-recorded fatal and nonfatal accidents in Denver (2012-2018); (iii) all police-recorded fatal and nonfatal accidents in Seattle (2007-2018). We compared the number of accidents on Saturday, Sunday and Monday of the DST change weekend with corresponding Saturdays, Sundays and Mondays in the week before and after the DST spring transition, using two-tailed paired t-tests and mixed models accounting for potential accident rate changes over time. Additional analyses considered time-of-day of accident occurrence. Results: Data on 1, 874, 672 accidents were available. We did not observe differences in fatal/non-fatal accident rates associated with the DST spring transition in neither of the datasets (p>0.10). Changes in timing of DST did not result in shifts of accident rates (p>0.10). Accident fatality, time-of-day, or geographical location did not influence the association between DST and accident rates. We replicated prior findings restricting FARS analyses to 1975-1995 (Varghuese & Allen, 2001), showing higher accident rates associated with the DST spring transition. Conclusion: In this large study of traffic accidents, we did not observe evidence for an association between the spring DST transition and accident rates, despite being able to replicate prior positive findings in a smaller dataset. To better appreciate the potential burden of DST on human health, further, large-scale studies are necessary to detect this presumably small effect. Support (If Any): N/A