Background Road traffic injuries claim are among the leading causes of preventable death, claiming around 7000 lives every year. Accordingly, injuries to the head and neck are the main causes of death and disability.

Objective This aimed to identify the determinants of prolonged length of stay in the hospital among victims of road traffic injury.

Methods A retrospective cohort study was conducted. Socio-demographics, accident details (e.g. time of injury, type of victim, other party to the accident, time to admission, intoxication, and helmet use), the Glasgow Coma Score upon admission, head injuries, torso injuries, injuries to the limbs, and length of stay) were abstracted from the patient charts and cleaned accordingly. The predictor model for the determinants of prolonged length of stay in the hospital was built.

Results There were 427 road traffic injury victims that were admitted during the time period considered. The distribution of the time to admission differed between those who have a short length of stay and long stay differed significantly (p<0.01). Glasgow coma score, type of victim, mortality, age, and sex were found to be important determinants of prolonged hospital stay. Those who have a low Glasgow Coma Score (i.e. more severe cases) were found to be 2.77 (95% CI: 1.13–6.91) times more likely to stay longer in the hospital as compared to those who have high Score. There is strong evidence for this association. Passengers are 78% (95% CI: 0.06–0.84) less likely to stay longer in the Hospital as compared to drivers. There is also strong evidence for this association.

Conclusion Based on the predictive model from 427 traffic injury patients, determinants of prolonged hospital stay were identified. Consequent policies and programs should address this.