Modeling non-fatal road crash injuries for Pakistan using aggregate data

Road crash injuries have emerged as a major health problem and have posed serious social and economic challenges around the globe. Traffic safety can be improved only through adequate safety measures if the underlying factors are well understood. Factors affecting crashes in developing countries, like Pakistan, are less studied in the literature. Also, the road crash injuries data reporting and recording systems in these countries are not well established. As such, this study by using the data from multiple sources including World Health Organization (WHO), International Road Federation (IRF) and World Bank (WB) compares the reported non-fatal road crash injuries of Pakistan using two different approaches. First, the road crash injury rates were compared with different groups of countries around the globe using two different indicators: (1) injuries per hundred thousand population and (2) injuries per thousand registered vehicles. Results indicated lower road crash injury rates Pakistan relative to other countries. Using the same indicators as response variables, linear regression models were estimated using Ordinary Least Square (OLS) regression. The total number of registered vehicles, maximum speed on rural roads, enforcement level of the seatbelt law, income level, and safety audit of new roads were found as significant explanatory variables. The average values of these variables were compared with those in the country. Finally, using the number of injuries as the response variables, count data models were developed and the number of road crash injuries for Pakistan were estimated and compared with the reported number of road crash injuries. The estimated road crash injuries were 4.5 times higher than the number of reported injuries.