Background: Previous studies have reported conflicting results concerning the association between several inflammatory markers and depression. The association between inflammation and depression may depend on the presence of specific chronic diseases or be relevant in specific sub-groups of depressed patients only. Objective: To assess associations between inflammatory markers and depression in older people, taking account of confounding and effect-modifying factors. Method: Population-based study of 1285 participants of the Longitudinal Aging Study Amsterdam, aged 65 and over. Plasma concentrations of Interleukin-6 (IL-6) and C-reactive protein (CRP) were measured. Major depression (first- or recurrent episode) and sub-threshold depression were assessed. Associations were adjusted for confounding variables. Associations with inflammatory markers were further studied with regard to severity and duration of depression, and with regard to specific depressive symptoms. Results: High levels of IL-6 (above 5 pg/mL) were associated with major depression (odds ratio 2.49 (1.07-5.80), both in recurrent and first episodes. No significant effect of either one of the markers on specific symptom dimensions of depression was found. Mildly elevated plasma levels of CRP (above 3.2 mg/L) were associated with higher CES-D scores, but not after correction for the confounding effect of age and chronic diseases. Limitations: The cross-sectional design limits conclusions regarding causality. Conclusions: A high plasma level of IL-6, but not CRP, is associated with an increased prevalence of major depression in older people, independent of age, chronic diseases, cognitive functioning and anti-depressants. Present results suggest new directions for clinical research into the prevention of physical consequences of depression. © 2007 Elsevier B.V. All rights reserved.