Objective. Better understanding of the relationship between pain and depression in older adults in the community is of particular importance considering the high prevalence of both conditions in these adults. In the present study, the longitudinal relationship between pain and depression in older adults was examined, thereby taking into account the role of physical disability and the possibly modifying effect of sex and age. Methods. The study is based on a sample which at the outset consisted of 325 non-depressed and 327 depressed persons (55-85) drawn from a larger random community-based sample in the Netherlands. Depression (CES-D) and pain (subscale of the Nottingham Health Profile) were measured at eight successive waves over 3 years. Results. Pain was very persistent over time as was to a lesser extent depression. The prognosis of comorbid pain and depression was poor. In longitudinal analyses (Generalized Estimating Equations), pain and depression were strongly associated. At the symptom level, the pain-depression relationship was found to be stronger in men than in women. There was no effect of age on the pain-depression relationship. No support was found for the hypothesis that the pain-depression relationship is mediated by disability. Conclusion. The persistent nature of pain and to a lesser extent depression and the intimate and probably reciprocal association between them stress the need for adequate treatment strategies.