Abstract
This review provides a comprehensive analysis of the factors contributing to driver injury severity (DIS) in road traffic accidents, structured around five main domains: driver, vehicle, environmental and temporal, roadway, and accident-specific characteristics. A systematic literature search, guided by PRISMA methodology, identified 133 studies published between 2000 and 2025. This review shows that the most used model in the reviewed studies was the mixed logit model (MLM), applied 22 times, followed by its extension, the MLM-Extension (20 times), and Machine Learning models (15 times). Additionally, the results demonstrate that older people, female drivers and those involved in a head-on collision or rolling over accident have an increased risk of severe injury. For vehicle factors, vehicle age and type are major determinants of injury severity, with older vehicles and two-wheeled vehicles having higher risks. Environmental factors such as night time driving, bad weather and rural roads also significantly affect DIS. The review identifies important gaps in the current literature, which include the absence of integrated research investigations and data availability in the lower-middle-income countries (like India) with its different road infrastructure and peculiar traffic scenarios compared to what is available from high-income countries. Addressing these weaknesses, results from our study highlight the need for regional-relevant research to help us understand how different local patterns are affecting DIS. This review proposes that specific interventions aimed at controlling these determinants would mitigate DIS rates worldwide.