Within a reanalysis of the "Kiel Group Psychotherapy Study", the videorecorded intake interviews of 34 patients were analyzed using the adult attachment prototype rating. Three subgroups, differentiated according to their attachment patterns (avoidant, ambivalent, mixed insecure) differed marginally with respect to their treatment success, with ambivalent patients showing the most positive outcome (escpecially with regard to symptom- and self-system-related outcome criteria). In addition, there were indications for a differential assessment of group related therapeutic factors by the subgroups. Patients with an avoidant attachment pattern evaluated group-related factors such as cohesion and altruism as least helpful. The study is discussed in relation to existing findings on the significance of attachment related variables for the indication and prognosis of psychotherapy.