Little is known about factors that may predict the response of dysthymia or other forms of chronic depression to treatment with antidepressant medication. We investigated several sociodemographic and clinical variables for their relationship to the acute treatment response to desipramine in subjects with DSM-III-R dysthymia. Subjects with DSM-III-R dysthymia were treated with desipramine in an open fashion for 10 weeks. Various clinical and sociodemographic variables were assessed at baseline. Ratings of depressive symptoms and severity and determination of categorical outcome were done during treatment. Baseline extended family adjustment as measured by the Social Adjustment Scale Self-Report was significantly better in the responders compared with the nonresponders (2.1 +/- 0.5 vs. 2.6 +/- 0.8; t = 2.84, df = 52.11, p = 0.006). There was a trend (p = 0.06) for overall baseline social impairment (SAS-SR) to be higher in nonresponders versus responders. Older age was a significant predictor of higher depressive severity (Cornell Dysthymia Rating Scale) and global impairment in the last week of the study. No other variable significantly distinguished responders from non-responders. Although few of the variables that were examined were found to affect acute treatment response, better extended family adjustment as reported on the SAS-SR was a significant predictor of good acute treatment response to desipramine in patients with dysthymia.