Average travel speed and traffic accident risk: Evidence from nationwide data for trunk highways and expressways in Japan

This study empirically analyzed the determinants of traffic accident risk, highlighting the average travel speed, using nationwide data for trunk highways and expressways in Japan. Negative binomial regression models were established using data describing traffic accidents on 86,835 trunk-highway links, 84,091 trunk-highway intersections, and 4322 expressway links occurring from 2019 to 2021. The results indicated that the average travel speed had a significantly negative association with the annual frequency and rates of traffic accidents regardless of the road type, road median presence, or area characteristics. Furthermore, the speed limit was positively associated with the traffic accident risk on trunk highways, while it was negatively associated with the traffic accident risk on expressways. Finally, the presence of road median reduced the traffic accident risk, and more urbanized areas exhibited a higher risk of traffic accidents than less urbanized areas. Policy implications of these findings include the mitigation of trunk-highway traffic congestion to reduce the traffic accident risk and the potential benefit of increasing the speed limit on traffic safety in expressways. • This study empirically analyzed the association of average travel speed with traffic accident risk. • Negative binomial regression models were established using nationwide data for trunk highways and expressways in Japan. • Average travel speed had a significantly negative effect on traffic accident risk regardless of road type. • Speed limit had a significantly positive effect on traffic accident risk in trunk highways but negative in expressways. • Road median reduced traffic accident risk, and urbanized areas exhibited a higher traffic accident risk than other areas.