The Prospective Association Between Inflammation and Depressive Symptoms in Type 2 Diabetes Stratified by Sex

Abstract

OBJECTIVE We tested whether inflammation is associated with worsening depressive symptoms in type 2 diabetes and examined whether sex moderated this association. RESEARCH DESIGN AND METHODS In a prospective cohort study of people with newly diagnosed type 2 diabetes, we measured depressive symptoms over a 2-year follow-up using the Patient Health Questionnaire-9 (PHQ-9). The independent variable was a composite inflammation burden score at diagnosis of diabetes, derived from hs-CRP, white cell count, interleukin (IL)-1β, IL-1 receptor antagonist, monocyte chemotactic protein-1, and vascular endothelial growth factor concentrations. General linear models assessed 1) the association between overall inflammation burden and estimated marginal mean PHQ-9 score (ln transformed) at 2 years and 2) whether sex interacted with elevated inflammation burden (above-median score) in predicting change in PHQ-9 score. Models were adjusted for age, ethnicity, BMI, blood pressure, cholesterol, HbA1c, antidepressants, anti-inflammatory medications, and baseline ln PHQ-9 score. RESULTS Of 1,174 people with complete inflammation data, mean (SD) age was 56.7 (11.0) years and 46.1% were of nonwhite ethnicity and 44.1% female. After full adjustment, inflammation burden was not associated with worsening ln PHQ-9 score (P = 0.65). However, female sex interacted with elevated inflammation in predicting higher 2-year ln PHQ-9 score (β = 0.32, P = 0.005), showing that the difference by inflammation burden in females was 0.32 larger than in males. In post hoc comparisons, ln PHQ-9 score was higher in females than males with elevated inflammation (P = 0.003) but not with low inflammation (P = 0.34) burden. CONCLUSIONS In type 2 diabetes, female sex confers specific vulnerability to the effects of inflammation on depressive symptoms.