Using in-depth accident data to identify limitations when applying crash injury risk curves

Abstract

Injury risk curves outline the relationship between speed and the risk of high severity outcomes for certain crash configurations, and hence are a convenient tool for road infrastructure design practitioners when assessing the safety of certain road designs. However, aggregation of data used to create these risk curves can hide important complexities that limit their usefulness. The aim of this study is to contextualise such risk curves with respect to other determining factors of crash injury severity. In-depth crash investigation data from the Initiative for the Global Harmonisation of Accident Data (IGLAD) database is used to compare the predicted risk of high severity outcomes with actual severity outcomes of crashes. The results of this study suggest that the risk of high severity outcomes was either under- or over-predicted for a substantial proportion of crashes within the database.