Depression, Inflammation, and Physiological Risk in Late Life: A National Longitudinal Study

This nationally representative study queried effects of community dwelling older adults' depression and inflammation at baseline on over-time changes in surrogate markers of their cardiometabolic risk. Data were from the 2005-2006 and 2010-2011 waves of the U.S. National Social Life, Health, and Aging Project. Inflammation was indicated by C-reactive protein and depression by the CES-D scale. Cardiometabolic markers included hemoglobin A1c and systolic BP. Lagged dependent variable models were used to examine effects. In none of the models did Wave 1 depression predict residual change in cardiometabolic states (i.e., Wave 2 values net of Wave 1). In contrast, men's baseline C-reactive protein predicted their Wave 2 hemoglobin A1c (Coeff. = 0.02, p < .05) as well as their systolic BP (Coeff. = 3.22, p < .05). No such effects were found among women. Contrary to a growing clinical literature, depression may not increase cardiometabolic risk among older adults on average. Moderators that may interact with depression to yield such effects in delimited samples remain to be identified. Inflammation, in contrast, does seem linked to increase in physiological risk-but only among men, not women. Clinical research is needed to identify biological factors responsible for this sex difference.