The reported prevalence of depression concomitant with Parkinson's disease varies greatly in the literature, which may partly be explained by symptom overlap. To determine the impact of symptom overlap on the prevalence, the authors tested 100 Parkinson's disease patients for major depression (DSM-III-R) with both a standard, inclusive method and a diagnostic-etiologic, exclusive method. The authors found that the prevalence detected with the inclusive method (23%) decreased when the exclusive method was used (13%), which was mainly caused by lower scores on the item "loss of interest." The study's findings give empirical support for the relevance of the new category in DSM-IV "mood disorder due to a general medical condition."