Background: Motor vehicle crash (MVC) mortality rates are inversely related to population density. The purpose of this study was to analyze if there is a regional variation in the risk of MVC death after controlling for injury severity.
Methods: The study utilized the Crash Outcome Data Evaluation System (CODES) data set in Nebraska. All fatal or injury-related crashes during a 4-year period (1996 through 1999) were analyzed. Injury Severity Scores (ISSs) were calculated from the CODES listed International Classification of Diseases diagnoses. Logistic regression analysis was performed to analyze the odds ratio for death in three rural county groupings compared with urban locations.
Results: During the 4-year period, 56,727 people were injured and 1,237 were killed in 38,493 MVCs. Of these, 45,222 (78%) records had complete information on variables of interest. In addition, 28,859 (50%) records had enough information to calculate an ISS. A total of 22,181 (39%) records had complete information on the variables of interest and ISSs. After adjusting for the effects of speed limit, age, and alcohol involvement (but not ISS), the odds of death were 1.24 (1.01-1.53) higher in the large, non-adjacent and 1.38 (1.14-1.66) small, non-adjacent rural counties. After adjusting for the effect of ISS, the odds of death were 1.98 (1.18-3.31) higher in the small, non-adjacent rural counties.
Conclusion: After controlling for ISS, the risk of MVC death is nearly twice as high in the most rural counties in Nebraska. This finding suggests that variation in medical care may contribute to this regional variation.