Freeway speed limits and traffic fatalities in Washington State

Abstract

Background: In 1987 individual states in the USA were allowed to raise speed limits on rural freeways from 55 to 65 mph. Analyses of the impact of the increased speed limits on highway safety have produced conflicting results. Objective: To determine if the 1987 speed limit increase on Washington State's rural freeways affected the incidence of fatal crashes or all crashes on rural freeways, or affected average vehicle speeds or speed variance. Design: An ecological study of crashes and vehicle speeds on Washington State freeways from 1974 through 1994. Results: The incidence of fatal crashes more than doubled after 1987, compared with what would have been expected if there had been no speed limit increase, rate ratio 2.1 (95% confidence interval (CI), 1.6–2.7). This resulted in an excess of 26.4 deaths per year on rural freeways in Washington State. The total crash rate did not change substantially, rate ratio 1.1 (95% CI, 1.0–1.3). Average vehicle speed increased by 5.5 mph. Speed variance was not affected by the speed limit increase. Conclusions: The speed limit increase was associated with a higher fatal crash rate and more deaths on freeways in Washington State.

Introduction

In 1974, a nationwide maximum speed limit of 55 mph was mandated in the USA. Although the reason for this mandate was to conserve oil during the energy crisis of 1973–1974, there was a subsequent decline in freeway deaths per mile driven. In 1987, the 55 mph mandate was relaxed when the Surface Transportation and Uniform Relocation Assistance Act allowed states to raise speed limits to 65 mph on rural freeways. A total of 38 states raised speed limits on rural freeways during 1987, and two more states did so in 1988 (National Highway Traffic Safety Administration, 1992).

Five published studies have used crash data from a single state to examine fatal crash incidence after speed limits were increased in 1987, and all reported an increase in fatalities; the increases ranged from 18% in Alabama to 93% in New Mexico (Gallagher et al., 1989, Brown et al., 1990, Wagenaar et al., 1990, Rock, 1995, Ledolter and Chan, 1996). Eight published studies have used national data, and they have produced conflicting results. McKnight and Klein found a 27% increase in fatal accidents on 65 mph highways, but also reported a 10% increase in fatal accidents on 55 mph highways in states that did not raise the speed limit (McKnight and Klein, 1990). Baum and colleagues reported an increase of 15% in the first year after the speed limit change (Baum et al., 1989), and an increase of 26% in the second year (Baum et al., 1990). Garber and Graham inappropriately used linear regression to analyze state-by-state monthly fatality counts, which are not normally distributed, and concluded that the speed limit increase was associated with more fatalities in some states, fewer fatalities in some states, and had no effect in some states (Garber and Graham, 1990). Lave and Elias used the assumption that total statewide traffic fatalities are a better measure of the effect of the rural freeway speed limit increase than rural freeway fatalities are, and ignored the long-term downward trend in traffic fatalities, to conclude that the higher speed limit resulted in fewer fatalities (Lave and Elias, 1994). The National Highway Traffic Safety Administration reported that the speed limit increase was associated with a 30% increase in rural freeway fatalities in states which adopted the higher speed limit, and was responsible for an increase of 539 fatalities per year nationwide for the period through 1990 (National Highway Traffic Safety Administration, 1992). Chang and Paniati (1990) and Chang et al. (1993) concluded that the increased speed limits did not result in more fatalities, even though their data and analyses appeared to show otherwise.

Some traffic engineers believe that higher speed limits may actually save lives. This belief is supported by views stated in influential traffic engineering reference works; that the 85th percentile of vehicle speeds is a good speed at which to set the speed limit, that speed variance (the difference between the average and 85th percentile speeds) is more important than speed limits in determining crash rates, and that speed variance may decrease if speed limits are raised to the 85th percentile (Warren, 1982, Pline et al., 1992). Despite the importance of these views in determining speed limits, there has been little analysis in the literature on the effect of the speed limit increase on average vehicle speeds or speed variance.

The specific aims of this study were to estimate the effect of the speed limit increase on the incidence of fatal crashes and total crashes on rural freeways, on average vehicle speeds, and on speed variance.