Evaluating the effects of a simulator-based training on knowledge, attitudes and driving profiles of German ambulance drivers

•Holistic training evaluation on all four Kirkpatrick evaluation levels.•Evaluation of simulator based training specifically designed for ambulance drivers.•Results show difficulty of attitude modification and behavioral change in traffic.•Training positively influenced knowledge, risk sensitivity and reduced speed.•Reduced speed due to training did not prolong driving times to operation sites. Accident risk is increased for emergency responders driving with warning lights and sirens compared to other road users’ driving. Currently no standards for education of ambulance drivers exist. Research shows that high order understanding trainings focusing on insight to avoid critical driving situations might be more helpful than trainings focusing on car handling. The present controlled intervention study evaluates a one-day simulator-based high order training program specifically designed for ambulance drivers. In a longitudinal design with three measurement times multiple methods were used to evaluate the training holistically targeting the levels of reaction to training, learning, behavior and results of training. Questionnaire, knowledge test and driving profile data were analyzed with repeated measures analysis of variance controlling for age and sex. Data of two intervention groups and one control-waiting group was collected between 2014 and 2017 in two German federal states. 183 German paramedics (age: M = 33.1, SD = 9.4, 21.9 % female) participated in the study. 147 participants (80.3 %) completed post-training tests, and 30 participants (16.4 %) completed follow-up measurements six months after training. Participants’ reaction to training was positive directly after the training, and dropped slightly over time. Intervention group participants gained traffic-relevant knowledge compared to control group participants. Risk sensitivity of regular driving situations was the only attitude variable positively affected by training. This effect was not sustained six months after training. Training led to a decrease of average and maximum speed in short- as well as long-term measurements but did not affect drivers’ acceleration. Although speed was lower in post-tests, emergency response times did not differ. The simulator-based training for paramedics has small but notable effects on drivers’ knowledge, attitudes and real driving behavior. Although only very few measured variables showed positive training effects, no negative training effects were found. Speed was