The purpose of the present study is to examine the effectiveness of CBT on a sample of no selected, pharmacologically treatment-resistant OCD patients and to find the predictors of successful treatment in these conditions. The therapy was conducted in a naturalistic setting and systematic CBT steps were adapted to each patient. Pharmacologic treatment underwent no or minimal changes during the trial period. Outcome measures included the Y-BOCS, CGI, BDI, BAI, Somatoform Dissociation Questionaire, and Dissociative Experience Scale. The primary outcome measure for response was a rating of 25% improvement in Y-BOCS. There was border for remission -12 points or less on the same scale, and 1-2 points in CGI-S.

As far 47 patients completed the trial. All patients finished minimum of 5 weeks of CBT and showed statistically significant improvement on all outcome measures. At the end of therapy 64.1% of patients were improved. 50% of patients reach remission according Y-BOCS and 40.4% according CGI. The main predictors of good therapeutic outcome were lower scores in Y-BOCS, good insight, high resistance against symptoms, low level of dissociation, and aggressive character of obsessions. The negative predictors were control/symmetry obsession and compulsions, and obsessive slowness and ambivalence.

CBT could be effectively used for medication-resistant OCD patients. There are some factors, like intensity of OCD symptoms, level of insight, resistance against symptoms, and level of dissociation, which could predict outcome of the therapy.