The present study was designed to investigate the interaction between therapeutic effects and side-effects in fluspirilene treatment of generalised anxiety disorders (GAD). 205 outpatients received in an open trial 1.5 mg fluspirilene per week for 6 weeks. Confirming previous reports of our study group fluspirilene demonstrated marked efficacy and was generally well tolerated. However, the main finding of our study is that side effects, although rare in fluspirilene treatment, predict an unfavourable clinical outcome. This phenomenon is integrated in a psychological model implying that when GAD patients are aware of side-effects this induces anxiety. Again anxiety gives rise to somatic symptoms, both interacting in a forward loop. In conclusion, our data suggest that in the sense of Heinrich (1988) fluspirilene treatment of GAD should be guided by paying attention to possible side effects.