|
Abstract
|
Background
Speed limit reductions are known to decrease road risks where they are introduced since speed influences collision and injury severity. However, studies on the impact of speed limit reductions on collisions are limited since the 1990s, especially on roads that are not highways or freeways.
Objective
To evaluate the effect of speed limit reductions on collisions involving killed or seriously injured (KSI) persons in Quebec, Canada using a controlled before-and-after methodology where changes are not introduced all at once.
Methods
A database including all major provincial road segments (n = 3457) was built to reflect ad hoc speed changes introduced between 2006 and 2013 throughout the entire territory and on varying types of rural and periurban roads. The number of KSI three years prior and after speed limit changes were calculated for 20,437 segment-years with or without speed changes (control group). It was then used to model the effect of the speed reduction on segments with initial speeds of 70, 80 or 90 km/h and speed reductions of 10 or 20-40 km/h.
Main findings
Using a zero inflated negative binomial and logistic regressions (at least 1 KSI vs. none), adjusting for decreasing KSI trends over time, traffic counts and segment length, we found a clear downward trend of KSI on all road segments, and a higher decrease where initial speed (90 km/h) were higher and speed reduction (20-40 km/h) were more important. Also, segments with sharp turns or with at least one intersection were more likely to have more KSI while divided two-lane roads were less likely to.
Conclusion
Our results add to the literature on the impact of speed reduction on major roads, helping provincial transportation managers to adopt evidence-based best practices.
Language: en |