Clinically diagnosed Axis II comorbidity and the short term outcome of CBT for Axis I Disorders

Earlier studies found that the effectiveness of CBT for anxiety disorders and depression is not negatively affected by the presence of co-morbid axis II disorders (for reviews, see Mulder, 2002; Dreessen & Arntz, 1998). That is to say, patients with co-morbid axis II disorders tend to have more severe complaints post-therapy, but they already have complaints before therapy and the decrease in symptoms is not affected by axis II disorders.In these studies the presence of axis II disorders was assessed using structured interviews. Compared to the more usual open clinical interview, structured interviews tend to result in larger proportions of the patients receiving a diagnosis of one or more personality disorders. Possibly the inclusion of many relatively mild cases obscured a negative effect of personality disorders and such a negative effect might materialize if axis II disorders were assessed with a open clinical interview.In the present study, data were analyzed from 421 patients, 289 of whom received a diagnosis of one or more personality disorders. Axis I diagnoses were obsessive–compulsive disorder (n = 165), panic disorder with agoraphobia (n = 54), major depression (n = 40), eating disorders (n = 42) or other disorders (n = 120) and CBT was delivered on an in-patient basis.When all groups were collapsed and a general measure of axis I problems was taken, the pattern was similar to the pattern from studies using structured interviews: patients with axis II problems had higher axis I problems both before and after treatment, but the decrease was parallel. When analyzed separately patients with axis II disorders did not score higher than patients without axis II disorders, while improvement in the various axis I groups was not affected by the presence of personality disorders. In the depressed subgroup a trend was observed for patients with axis II disorders to improve less.Even when personality disorders are diagnosed using an open clinical interview, their presence is largely irrelevant to the reduction of axis I problems after CBT. Copyright © 2006 John Wiley & Sons, Ltd