Raised speed limits, case fatality and road deaths: a six year follow-up using ARIMA models
- 1University of Illinois at Chicago, School of Public Health, Division of Environmental and Occupational Health Sciences, Chicago, Illinois, USA
- 2The Social Policy Research Institute, Skokie, Illinois, USA
- 3University of Miami School of Medicine, Department of Anesthesiology, Center for Patient Safety Miami, Florida, USA
- 4Hebrew University, Unit of Occupational and Environmental Medicine and Center for Injury Prevention, Hadassah School of Public Health and Community Medicine, Jerusalem, Israel
- Correspondence to: Lee Friedman Director, The Social Policy Research Institute, 8423 Monticello Avenue, Skokie, IL 60076;
- Accepted 16 December 2006
Background: In November 1993, the Israeli government increased the speed limit for all vehicles from 90 to 100 km per hour on a total of 115 km of its three major interurban highways.
Design/Setting: We use ARIMA time series intervention models to evaluate the effect of the raise in speed limit on fatalities, serious injuries, and case-fatality for years 1988–1999. Motor vehicle crash data came from the Central Bureau of Statistics of Israel.
Results: Between January 1988 and December 1999, a total of 6029 persons were killed and 45 616 were seriously injured on roads in Israel. For all roads combined, the time series ARIMA model indicated that there were 4.69 more deaths per month (p<0.001), or 347 more than expected in the post-intervention period. Case-fatality rate (CFR) on all roads combined rose significantly (p<0.001). Modified case-fatality rate (CFRS) showed an increase of 2.5 deaths per 100 serious casualties (p<0.001).
Conclusions: The impact of raised speed limits was immediate and sustained. The largest increase in deaths occurred on interurban roads but a spillover effect was observed on urban roads as well. The increases in deaths and case-fatality rates persisted six years after the speed limit change despite major countermeasures and increasing congestion throughout the period of follow-up.
- ACF, autocorrelation function
- AR, autoregressive
- ARIMA, autoregressive integrated moving average
- CBS, Central Bureau of Statistics
- CFR, case-fatality rate
- CFRS, modified case-fatality rate
- INTV, intervention variable
- PACF, partial autocorrelation function
- MA, moving average
Competing interests. None.