Comorbidity is a well-established phenomenon in depressive disorders, and it is widely agreed that the majority of depressive disorders examined in both primary care and the general population are not "pure." This article reviews comorbidity findings in general population and primary care surveys. The implications of comorbid depressive disorders are discussed in terms of their presentation and recognition in primary care, patterns of course and outcome, and associated impairments and disabilities. Data from a World Health Organization study, conducted in primary health clinics in 15 countries, demonstrate that 62% of all depressive cases also suffer from at least one other current mental disorder and that the primary reason for patient visits is rarely of a psychological nature, with the majority of attendees complaining primarily of somatic symptoms (41%), pains (37%), and fatigue and sleep problems (12%). Similar results are presented from 2 recent large representative population surveys, the National Comorbidity Survey and the Early Developmental Stages of Psychopathology Study. The implications of such comorbidity are viewed from various perspectives: the form of depressive presentations in primary care, the recognition of depression in health care, and health services utilization, impairments and disabilities, and course and outcome.