Background The effect of traffic parameters on road safety has gained considerable attention from researchers. High resolution traffic data extracted from motorways are used widely nowadays to study traffic safety. However, there are no studies exploiting high resolution traffic data in urban areas.

Methods This study uses accident and traffic data from two major arterials of Athens for the period 2006–2011. Traffic data were extracted from the closest upstream as well as the closest downstream detector to the accident location. Five-minute raw traffic data were aggregated hourly to obtain the mean values so that a more mesoscopic approach is achieved. Analysis of variance (ANOVA) was applied in order to highlight potential statistical differences in speed and statistical differences in occupancy by severity level and by accident type, one hour prior to accident occurrence.

Results Overall, several significant statistical differences are observed, either when accident severity or the type of accident is explored. This implies that traffic parameters can significantly influence the severity or the type of the accident.

Conclusions The findings of the study justify the need to further explore the effect of traffic parameters on traffic safety.