Background Medical conditions and medications may be associated with motor vehicle collisions (MVCs), which pose a major public health problem worldwide. Further epidemiological assessment is necessary for certain diseases and medications. Moreover, since disease aetiology and patterns of medication use may differ among ethnicities and healthcare systems, a population-specific approach is necessary. The present epidemiological study is designed to assess the medical conditions and medications associated with the risk of fatal MVCs among at-fault drivers in the Korean population.

Method and design A retrospective cohort will be constructed for individuals who died in MVCs between 2005 and 2014 in the Korean Traffic Accident Analysis System database, which is linked to the Korean National Health Insurance database between 2002 and 2014. In order to compare medical conditions and medication use among drivers who died in a fatal MVC with the general population, standardised prevalence ratios will be calculated. In the culpability study, we will identify conditions and drugs associated with MVCs, comparing drivers with higher levels of responsibility to those with lower levels of responsibility. In the case-crossover study, the transient effects of medical conditions and medications will be examined using a conditional logistic regression model that adjusts for confounders.

Discussion The results of this study will help to characterise the associations of diseases and medications with fatal MVCs in an Asian population, with the goal of informing regulatory and clinical decision-making regarding patients with the relevant conditions and the establishment of strategies for improving traffic safety.