Road traffic injuries are a serious public health problem especially in developing countries such as Ghana, West Africa. Data from developing countries are difficult to obtain, as a result, most countermeasures used to reduce crash severity are based on data from developed countries. There is, therefore, the need to examine which factors possibly would contribute to different crash severity levels using data obtained in Ghana. A binary logistic regression technique is used to model crash injury severity (severe injury and minor injury). The identified factors are compared to factors identified in developed countries. The study data set contain a total of 3 years of crash data (2014-2016) for the whole country. The results identified single vehicle crashes, the time period exclusive of the morning peak, nighttime (lights off and no lights), curved and/inclined road, highly potholed paved and earth road, poor and overgrown shoulder, no median, road segment, right angle, head-on and crashes involving pedestrians as positively associated with severe injury crashes. Some of the identified factors are consistent with factors found in developed countries whiles others are in contrast with factors that have been found in developed countries. This emphasizes the imperativeness that direction for policies and countermeasures should be based on data obtained in Ghana. The research results should provide a direction to policymakers and engineers in the road agencies for establishing effective policies and countermeasures.

CICTP