The dexamethasone suppression test (DST) was administered after baseline cortisol measurements in 20 patients (10 males, 10 females) who met DSM-III criteria for obsessive-compulsive disorder (OCD). Six patients (30%) showed an abnormal escape from dexamethasone suppression. DST suppressors vs. DST nonsuppressors showed no differences on age, rate of secondary depressive disorders, or scores on the Hamilton Rating Scale for Depression, the Minnesota Multiphasic Personality Inventory D scale, or OCD rating scales. Surprisingly, there was a trend for suppressors to have a stronger family history of depressive disorders, and for nonsuppressors to include an excess of male subjects. Moreover, there was a significant correlation between levels of cortisol before and after DST. In five of six nonsuppressors, both depressive symptoms and obsessive-compulsive behaviors showed a diminution in response to antidepressant therapy combined, in one case, with intensive behavior therapy. The relationships between OCD and endogenous depression, as well as the specificity of the DST, are discussed.