The inflammatory marker C‐reactive protein (CRP) is associated with depression. We examined the directional relations between CRP and symptoms of depression among older adults.

The sample consisted of 3397 participants from the English Longitudinal Study of Ageing, a prospective study of community‐dwelling older adults. CRP and depressive symptoms were measured at baseline and follow‐up. A high CRP level was dichotomized as >3 mg/L. Elevated depressive symptomatology was defined as ≥4 using the 8‐item Center for Epidemiologic Studies Depression Scale. Logistic regressions computed the association between high CRP levels at baseline with elevated depressive symptoms at follow‐up, and vice versa.

After adjusting for baseline depressive symptoms, baseline high CRP levels were associated with subsequent elevated symptoms of depression (OR = 1.49; 95% CI, 1.19–1.88). This relationship was no longer significant after simultaneous adjustments for metabolic and health variables. In the other direction, after adjusting for baseline CRP levels, baseline elevated depressive symptoms was not associated with subsequent high CRP levels (OR = 1.12; 95% CI, 0.88–1.42).

High CRP levels at baseline are related to elevated depressive symptomatology at follow‐up due to clinical factors. No association was found in the opposite direction. Copyright © 2014 John Wiley & Sons, Ltd.