Background: Accurate assessment of the natural history of late-life depression requires frequent observation over time. In later life, depressive disorders fulfilling rigorous diagnostic criteria are relatively rare, while subthreshold disorders are common. The primary aim was to study the natural history of late-life depression, systematically comparing those who did with those who did not fulfill rigorous diagnostic criteria. Methods: Within the Longitudinal Aging Study Amsterdam, a large cohort of depressed elderly persons (n=277) was identified and followed up for 6 years, using 14 observations. Depression was measured using self-reports (the Center for Epidemiological Studies Depression Scale) and diagnostic interviews (the Diagnostic Interview Schedule). The natural history was assessed for symptom severity (Center for Epidemiological Studies Depression Scale score), symptom duration, clinical course type, and stability of diagnoses. Results: The average symptom severity remained above the 85th percentile of the population average for 6 years. Symptoms were short-lived in only 14%. There were remissions in 23%, an unfavorable but fluctuating course in 44%, and a severe chronic course in 32% (percentages do not total 100 because of rounding). Comparing the outcome, there was a clear gradient in which those with subthreshold disorders had the best outcome, followed by those with major depressive disorder, dysthymic disorder, and double depression. However, the prognosis of subthreshold disorders was unfavorable in most cases, while this group was at high risk of developing DSM affective disorders. Conclusions: The natural history of late-life depression in the community is poor. DSM affective disorders are relatively rare among elderly persons, but do identify those with the worst prognosis. However, subthreshold depression is serious and chronic in many cases.