Background The purpose of the present study was to examine the effectiveness of the injury awareness and prevention programme P.A.R.T.Y. (Prevent Alcohol and Risk-Related Trauma in Youth) in Germany. On a designated P.A R.T.Y. day, school classes spend a day in a trauma hospital experiencing the various wards through which a seriously injured person goes. A further goal of the study was to reveal indications of the programme's mechanism of action by testing theory-based impact models of fear appeals and cognitive beliefs. Methods In a quasi-experimental longitudinal study with three measurement times the participants of 19 P.A.R.T.Y. days (n = 330), as well as pupils who did not attend the programme (n = 244), were interviewed with a standardised questionnaire. They reported risk behaviour, feelings of threat and cognitive beliefs about road traffic. The data were analysed using a meta-analytical approach to estimate an average effect size across the different P.A.R.T.Y. days. Path models were used to identify possible mechanisms of action. Results For most of the parameters, small positive effects could be proven immediately after the P.A.R.T.Y. intervention. However, after four to 5 months only one statistically significant effect was found. Using path analytical models, important predictors for behavioural changes (e.g. self-efficacy) could be identified. But for these predictors no or only short-term effects were observed in the meta-analysis. Conclusions Fear appeals as used primarily in the P.A.R.T.Y. programme appear to cause behavioural changes only to a limited extent and only in the short-term, especially if the strengthening of psychosocial resources is not given sufficient consideration. The participants must also cognitively process the experiences in the hospital. Accordingly, consideration should be given to how the P.A.R.T.Y. program could be adapted to complement the fear appeal with cognitive components.