Impact of drivers' attributes on children injury severities in traffic crashes

Abstract

Every year, a substantial number of children sustain injuries and fatalities in motor vehicle crashes in Wyoming. Understanding the factors contributing to child injury is crucial for the development of appropriate mitigation measures that aid in alleviating the severity of such injuries. In this study, a hierarchical Bayesian binary logit regression model was developed to investigate the factors that contribute to children's injuries resulting from crashes while accounting for possible intra-class correlation effects (those of unobserved factors common to children involved in the same crash). A strong correlation among crashes justified the use of the hierarchical Bayesian logit model. As per the modeling results, the children's ages, safety restraint types, vehicle types, drivers' ages, alcohol/drug involvement, drivers' seat belt use habits, drivers' actions, manners of collision and environmental conditions contributed to child injury risk. The child's age was found to be inversely related to the risk of injury. Similarly, among safety restraint types, rear-facing car seats and forward-facing car seats were found to reduce injury likelihoods in crashes. When it comes to the drivers' characteristics, the probability of incurring injuries among the child population increased in the presence of young, unbuckled and impaired drivers. Furthermore, improper driving actions, such as running off the road, raised the risk of incurring injuries to children. The findings of this study may be beneficial to authorities regarding developing and implementing road safety programs aimed at ameliorating child injury concerns.

Keywords

• Child injury
• Bayesian inference
• Hierarchical logit model
• Intra-class correlations
• Unobserved heterogeneity effects