N.G. Sorum, D. Pal
Pages: 89-110
Abstract
Nowadays road safety is one of the biggest public health issues in India. There have been limited efforts to study the injury severity of drivers in road traffic accidents in India, particularly in the North-East region of the country. Therefore, this study aimed to examine the driver, vehicle, temporal, and accident-specific factors that contributed to driver injury severity in one of the North-East states (capital city of Arunachal Pradesh) of India. Police reports of single-vehicle and two-vehicle accidents with fatal and non-fatal driver injuries that occurred during the year 2011–2020 in the capital city of Arunachal Pradesh, India, were used to analyze the driver injury severity. The binary logistic regression (BLR) approach was employed to examine the correlation between influencing factors and driver injury severity. Two sets of models (Model-without-interaction and Model-with-interaction) were developed for the statistical analysis. The results of the Model-without-interaction demonstrated that 14 independent variables (out of 19 variables) were found to be statistically significant to the driver injury severity. This included three age groups (18-24, 25-40, and above 40-year age group), heavy motor vehicle, over-speeding, out-of-control, rushed driving, other cause (like distraction), 6pm-12am time period, side-collision, rear-end collision, fell into gorge, hit-object, and other types of collision. The results of the Model-with-interaction showed that a total of 56 combinations of interactions among independent variables were found to be statistically associated with the driver injury severity. The BLR model used was found to be a capable tool in providing significant interpretations that could be used by decision-makers for road safety improvements in the capital city of the state. Further research can be carried out in other cities of the state and other North-East states with a more detailed data set.
Keywords: road traffic accidents; driver injury severity; binary logistic regression; public health issues; driver accident severity modeling