Background Fatal single vehicle accidents (SVA) which involved ESC fitted passenger cars were examined in order to disclose the immediate and background risk factors for the accidents as well as the validity of ESC function.

Methods The registry of Finnish Road Accident (FRA) Investigation Teams (RAIT) was analysed and all fatal passenger car accidents during the years 2009–13 (5 years) were included. SVAs (N = 351) were subjected to more detailed analysis. Suicide crashes (N = 13) and disease attacks (N = 102) were recognised but ruled out from the final results because ESC has no saving potential in such accidents.

Results ESC fitted vehicles were involved in SVAs less often (P = 0.001) than their calculated exposure would suggest and they were quite rare. The most common immediate risk factor in ESC SVAs was a steering error (56%, N = 10) but unlike in the non-ESC SVAs it did not lead to loss of control of the vehicle.

The common background risk factors in ESC SVAs were substance abuse (56%, N = 10), tiredness (56%, N = 10) and speeding (56%, N = 10). Tiredness was more common (P = 0.01) than in non-ESC SVAs and in half of the accidents it occurred together with substance abuse.

In lethal ESC SVAs the vehicle experienced multiple collisions (2.4 on average) at relatively high speed (on the average 93 km/h before first impact). This is a challenge to both passive and active safety systems. Also in ESC SVAs the seat belt use was low; nine (47%) of the killed occupants were unbelted. Seat belt use would have saved six (%) of them.

Conclusions Based on this study, ESC operates as designed in SVAs and no operational failures were found. Further research on driver fitness is required to improve active safety of the passenger cars. Vehicle systems should force everyone to use seat belt and the strength of roof structures should be improved.