Objective A substantial number of bulimic patients either fail to complete group therapy or miss a large number of sessions. This study examines the differences in eating psychopathology and other clinical features of women who either drop out of group therapy or attend intermittently. Method: Fifteen women who attended group therapy for bulimia completed measures of their eating psychopathology and other characteristics at the outset of the group. The women who completed the group were compared with those who did not, and the proportion of sessions attended was correlated with the women's pretreatment characteristics. Results: In general, poor attendance at group therapy for bulimic disorders appears to have been related to the group not addressing the women's immediate wishes (e.g., weight loss). Good attendance was associated with more severe levels of bulimic pathology (binging and vomiting). Discussion: These findings are related to the small existing literature. Clinicians setting up groups might use these conclusions to indicate who would be most likely to benefit from group therapy for bulimic disorders. © 1996 by John Wiley & Sons, Inc.

Objective Group psychotherapy is a widely used form of treatment for bulimia nervosa, and many open and controlled studies have demonstrated that it is relatively effective. However, it is not clear what factors enhance or reduce that effectiveness. This review considers the potential role of a number of such factors. Method: The literature on the effectiveness of group psychotherapy was selected on the basis of its publication inclusion in peer‐reviewed journals. Unpublished material was not included. Factors considered included theoretical orientation, therapy dose, concurrent treatment, client characteristics, and service characteristics. Results: The research to date shows no obvious advantage attached to any single therapeutic orientation or to the gender of therapists. However, other aspects of group psychotherapy do seem to influence outcome. Better outcomes are associated with longer, more intensively scheduled groups and with the addition of other treatment components (e.g., individual work). There is also evidence that relatively large groups are a viable option. These conclusions are reasonably congruent with clients' own expressed preferences regarding therapy. Discussion: It is suggested that future research should focus on matching the characteristics of individual bulimics to the nature of the group psychotherapy, in order to increase engagement and effectiveness while reducing drop‐out rates. © 1997 John Wiley & Sons, Inc. Int J Eat Dis, 22, 1–13, 1997.