Impact analysis of road infrastructure and traffic control on severity of pedestrian–vehicle crashes at intersections and non-intersections using bias-reduced logistic regression

Pedestrian safety is essential to realize a walkable city. This study analyzed the impact of road infrastructure and traffic control on severity of pedestrian–vehicle crashes at intersections and non-intersections. We applied a bias-reduced logistic regression analysis using 86,406 pedestrian–vehicle crashes data including 2636 fatal accidents in Japan. The results of the study indicate that medians increased the severity of crashes at intersections; on the other hand, they decreased the severity at non-intersections. In addition, flashing traffic signals increase the severity while stop signs decrease the severity at intersections. •Stop signs, median strips, etc. may reduce pedestrian fatality risk at intersections.•Median strips increase crash severity at intersections and decrease it at non-intersections.•Flashing traffic signals increase severity at intersections while stop signs decrease it.•At non-intersections, right curves are associated with fatality risk increase for pedestrians.