Prediction of Fatal Road Traffic Crashes in Iran Using the Box-jenkins Time Series Model

Introduction: Frequency of traffic related mortalities is increasing worldwide. The present study aimed to predict deaths from road traffic accidents for the first time in Iran. Methods: All death statistics from traffic accidents in Iran between March 2004 and March 2011 were available for analysis. The Box-Jenkins time series model was used for trends purposes. Death from traffic accidents were predicted from March 2011 to March 2013 and then compared with the actual frequencies. Results: Overall, 21548 deaths (95% CI: 15426-27669) due to road traffic accidents were predicted for 2011 compared to 2010, with a negative growth of 7.32%. The corresponding frequency was 20404 deaths from road traffic accidents (95% CI: 9914-30893) for 2013 compared to 2011 with a negative growth of 5.31%. An accuracy rate of 93% was found for prediction of fatal traffic accidents compared to the formal reports by the government. Conclusions: The Box-Jenkins time series model is an acceptable method for prediction of road traffic accidents.