Background & aims: Only a few studies have assessed the association between a proinflammatory diet and the risk of depression in older adults, and they have rendered weak results. The present study analysed the association between the Dietary Inflammatory Index (DII) and incident self-reported diagnosis or symptoms of depression in two cohorts of community-dwelling older adults in Spain. Methods: We used data from the Seniors-ENRICA-I (SE-I) and Seniors-ENRICA-II (SE-II) cohorts. In both cohorts, the baseline DII was calculated from habitual food consumption estimated with a validated computer-based diet history. The incidence of both physician self-reported diagnosis of depression and mild-to-major depressive symptoms (3 on the 10-item Geriatric Depression Scale) was analysed. Logistic regression models were adjusted for the main potential confounders, such as sociodemographics, lifestyles, and comorbidities. The results of both cohorts were pooled using a random effects model. Results: Among the 1627 participants in SE-I (mean age 71.5 ± 5.5 y, 53.1% women) and the 1579 in SE-II (mean age 71.4 ± 4.2, 46.7% women), 86 (5.3%) and 140 (8.9%) incident cases of depression were identified after a mean 3.2-y and 2.3-y follow-up, respectively. The fully adjusted odds ratio (95% confidence interval) of incident depression for the highest (the highest proinflammatory diet) versus the lowest quartile of DII was 2.76 (1.25e6.08, p-for-trend ¼ 0.005) in the SE-I, 1.90 (1.04e3.40, p-for-trend ¼ 0.005) in the SE-II and 2.07 (1.01e3.13) in the pooled cohorts. The results were consistent across strata defined by sex, age, physical activity, loneliness/poor social network, and morbidity. Conclusions: A proinflammatory dietary pattern is associated with depression risk in older adults. Future research should evaluate whether reducing the inflammatory component of diet leads to reduced depression symptoms in this population