Elsevier

Sleep Medicine

Volume 2, Issue 1, January 2001, Pages 31-36
Sleep Medicine

Original article
Fatal accidents following changes in daylight savings time: the American experience

https://doi.org/10.1016/S1389-9457(00)00032-0Get rights and content

Abstract

Objective: This study examines specific hypotheses that both sleep loss and behavioral changes occurring with the time shifts for Daylight Savings Time (DST) significantly effect the number of fatal traffic accidents in the United States of America.

Background: It has been reported that there is a significant increase in the number of automobile accidents in the spring shift to DST due to the loss of 1 h of sleep. But the extra hour gained at night with the shift from DST in the fall has been variably reported to be associated with increases and decreases in the number of automobile accidents which may reflect either behavioral anticipation with an extended late night prior to the change or the benefit of extra sleep after the change.

Methods: Data from 21 years of United States’ fatal automobile accidents were gathered. The mean number of accidents on the days at the time of the shifts (Saturday, Sunday and Monday) was compared to the average of the corresponding mean number of accidents on the matching day of the weeks preceding and following the shift. This was repeated for each DST shift. The number of accidents for a particular shift was also correlated with the year of the accidents.

Results: There was a significant increase in accidents for the Monday immediately following the spring shift to DST (t=1.92, P=0.034). There was also a significant increase in number of accidents on the Sunday of the fall shift from DST (P<0.002). No significant changes were observed for the other days. A significant negative correlation with the year was found between the number of accidents on the Saturdays and Sundays but not Mondays.

Conclusions: The sleep deprivation on the Monday following shift to DST in the spring results in a small increase in fatal accidents. The behavioral adaptation anticipating the longer day on Sunday of the shift from DST in the fall leads to an increased number of accidents suggesting an increase in late night (early Sunday morning) driving when traffic related fatalities are high possibly related to alcohol consumption and driving while sleepy. Public health educators should probably consider issuing warnings both about the effects of sleep loss in the spring shift and possible behaviors such as staying out later, particularly when consuming alcohol in the fall shift. Sleep clinicians should be aware that health consequences from forced changes in the circadian patterns resulting from DST come not only from physiological adjustments but also from behavioral responses to forced circadian changes.

Introduction

The 1-h adjustments required for Daylight Savings Time (DST) can be seen as producing two very different effects: (1) physiological adjustment to an obligatory 1-h phase advance or delay of the circadian rhythm of sleep and (2) behavioral adjustments anticipating the changes in time available on Saturday night into early Sunday morning. In relation to the first effect, it has been shown that significant disruption in sleep patterns occurs due to DST and persists up to 5 days after a change to or from DST [1]. The resulting change in sleepiness seems likely to alter the risk of automobile accidents. On one hand, the loss of 1 h with the change to DST in the spring leads to a loss of sleep while the body's circadian system adjusts to the phase advance with a possible increase in risk of accidents. The effects will be more pronounced on Monday after the shift when, unlike Sunday, most people cannot ‘sleep-in’ late and must rely upon a successful 1-h phase advance to ensure adequate sleep. On the other hand, the gain of 1 h with the change from DST in the fall permits a 1-h gain in sleep time while the body adjusts to the sleep delay, but this should occur mainly for people who are normally sleep deprived by forced awakenings on workday. This might, for these sleep deprived individuals, reduce sleepiness and therefore somewhat reduce the risk of accidents particularly on Monday after the shift from DST. Physiological adjustment is also generally considered easier for phase delay than for phase advance. Thus the physiological effects expected from DST changes would primarily be decreased sleep time and possible increased accidents principally on the Monday after the spring change to DST.

Sleep related behaviors may, however, also be altered on the night of the DST changes. In particular, in the fall the addition of an hour in the early morning may encourage those who normally stay out late into Sunday morning to stay out even later. They would then drive home 1-h later and presumably less alert than usual thereby increasing risks of accidents on the Sunday morning of the change, particularly if alcohol is involved [2]. The loss of an hour in the spring might encourage some to come home earlier than usual thereby reducing the risk of accidents on Saturday night and Sunday morning. Given the problems of adjusting to an earlier sleep time, it, however, seems unlikely that this behavioral strategy would be either successful or particularly attractive for the spring. In contrast, given the ease of adjusting to a longer day the behavioral strategy of staying out later may be a successful adaptation to the change in the fall. This behavioral adaptation may be particularly appealing to those who already prefer to stay out late that night.

Thus these two hypothesized effects lead to somewhat opposite predicted effects on accident rates but mostly on different days. For the spring change to DST, physiological effects could lead to an increased risk of accidents principally on Monday and less so on Sunday, while, in contrast, behavioral adaptation may lead to some minor decreased risk only on Sunday with no effect on Monday. For the fall change from DST, physiological effects might produce some small decreased accident risk mainly on the Monday after the change and less so on Sunday, while, in contrast, behavioral adaptation may significantly increase risk of accidents only on Sunday with no effect on Monday. No changes in accident risk should occur for the Saturday before either DST change.

In fact, Coren has reported for Canadian traffic accidents over a two year period, that the Monday after the changes showed both a significant increase following the spring change to DST, and a significant decrease following the fall change from DST [3]. Monk similarly reported an increase in traffic accidents after the spring change to DST when changes in level of lighting were controlled by excluding accidents occurring at dawn or dusk [4]. But the fall accident rates for the Monday after the fall change from DST show in various studies both increases, no change and decreases. These prior studies covered only a few years, with one exception did not use data from the United States, and failed to examine all 3 days around the change: Saturday, Sunday and Monday. Thus they could not test for the hypothesized sleep-related behavioral changes on Saturday night and Sunday morning also leading to changes in accidents. An update on this issue using a larger database and more complete analyses of the relevant days to tests for effects of both physiological and behavioral changes is needed.

Section snippets

Database

The United States National Highway Transportation Safety Administration maintains records of all fatal automobile accidents for each day of the year, dating back to 1975. Data for a 21-year period from 1975 to 1995 were obtained for the Sunday corresponding to the change, the preceding Saturday, and the following Monday. Since accidents vary considerably for each day of the week, comparison data were obtained from the corresponding Saturdays, Sundays and Mondays in the week before and also the

Results

For the spring change to DST, the number of fatal accidents on Monday showed a significant increase from an average of 78.2 from the weeks before and after to an average of 83.5 on the day of the shift (t=1.92, P=0.034) (See Fig. 1), but there was no significant change for fatal accidents on Sunday (means: weeks before and after=113.4, week of change=118.1, t=1.49, P=0.14) or Saturday (means: weeks before and after=145.3, week of change=141.9, t=0.813, P=0.21).

For the fall change from DST, the

Discussion

This is the largest data set ever used to address the issue of the effects of DST changes on accidents and the only data set to look at the Sunday data when sleep-wake behavioral changes may also effect alertness and accidents and to use Saturday as a comparison control. Like the preceding studies, a small significant effect was found for increased accidents in the spring, when the change to DST in the spring tends to shorten sleep times. As expected for this physiological change, the effect

References (7)

  • T. Monk et al.

    Adjusting to the changes to and from daylight saving time

    Science

    (1976)
  • T. Roehrs et al.

    Sleepiness and ethanol effects on simulated driving

    Alcohol Clin Exp Res

    (1994)
  • S. Coren

    Daylight savings time and traffic accidents

    N Engl J Med

    (1996)
There are more references available in the full text version of this article.

Cited by (67)

  • Daylight saving time and the incidence of thrombolysis to treat acute ischemic stroke

    2020, Revue Neurologique
    Citation Excerpt :

    It has always been evident that it associates with an interference with biorhythm twice a year; the debate rather focuses on the extent, duration, and consequences of this interference. Some of the studies compare the effects of spring and autumn transitions [4]. DST abruptly alters the “social clock”, which cannot be immediately followed by the body, as the adaptation necessitates hormonal changes.

View all citing articles on Scopus
View full text