There is substantial evidence that lack of sleep is a significant factor in motor vehicle crashes experienced by teenage drivers. This paper examines the hypothesis that a later high school start time may reduce crash rates by reducing the interference of school start time with the sleep needs of adolescents. The authors collected data from two jurisdictions — Forsyth County, North Carolina and Fayette County, Kentucky — that changed to a substantially later high school start time. The authors examined whether this change was accompanied by a drop in crash rates among 16- and 17-year-old drivers. Monthly time series were compiled corresponding to the overall rate of crashes on school days, adjusted for changes in the 16- and 17-year-old population. An intervention time series analysis was applied to determine whether the change in school start time was accompanied by a downward shift in the level of the crash rate series. To control for possible confounding factors, comparable statistics were also compiled, using the same analyses, for several other counties where there was no change in school start times. For Forsyth County, NC, there was a decrease in crash rates corresponding to the change in school start time, though the statistical significance of the effect is only moderate (one-sided p-value = .04). No corresponding effect was observed for three comparable counties of North Carolina. For Fayette County, KY, there was no evidence of a statistically significant change in crash rates. Because of anomalous data in the one available comparable county, The authors were unable to derive meaningful results for a comparison county in Kentucky. The authors conclude that there is mild evidence that the change in school start times in Forsyth County, NC had a beneficial effect, but there is no corresponding evidence for Fayette County, KY. (Author/publisher)
Abstract