Background: The relationship between bipolar and unipolar psychotic depression has not been well studied. Therefore, the aim of the present study was to compare bipolar with unipolar psychotic outpatient depression. Methods: Seventy consecutive unipolar (n = 40) and bipolar (n = 30) psychotic depressed outpatients were interviewed with the Structured Clinical Interview for DSM-IV, the Montgomery Asberg Depression Rating Scale, the Global Assessment of Functioning Scale, and the Brief Psychiatric Rating Scale. Results: Of the variables studied (age, duration of illness, severity, recurrences, atypical features, chronicity, gender, comorbidity, hallucinations, delusions), none was significantly different between unipolar and bipolar psychotic patients. Conclusions: Bipolar psychotic depression was similar to unipolar psychotic depression on variables reported in the literature to distinguish bipolar from unipolar disorder. Clinical implications: The findings might suggest, but do not necessarily imply, that psychotic depression might be a distinct clinical entity. Limitations: Single interviewer, nonblind cross-sectional assessment, outpatient sample, sample size. (C) 1999 Elsevier Science B.V. All rights reserved.