The treatment of adolescents with psychiatric disturbance remains a highly unsuccessful area. Empirical research is limited and treatment decisions are often based on theoretical bias rather than proven results. This study reports on the results of a multivariate analysis of pre- and post-individual and family measures on patients in a community based day and evening programme. The programme utilized a broad range of treatment modalities including group and family therapy and chemotherapy. An accredited educational programme was also provided to the patients attending the day programme. The results of this study showed that adolescents were referred to either the day or evening programme components based on level of acting out behaviour rather than on DSM III diagnosis. Adolescents reported significant positive changes at post testing in all areas, but this improvement was not endorsed by parents. Although therapists perceived considerable improvement in their families after treatment, this improvement did not show up on the family assessment measure chosen. The findings of the study endorse the need for further empirical research in adolescent treatment and underscores the fallacy of relying on anecdotal evidence to report treatment results. Questions are raised as to the operational usefulness of the present DSM III criteria in making treatment decisions and the validity of associating individual dysfunction with family dysfunction. Some question as to the sensitivity of the family measure used was also raised. The study outlined strengths and weaknesses in the present treatment programme and underscored the importance of further evaluative research, if the treatment of adolescents with psychiatric disturbance is to become successful.