Abstract
OBJECTIVE: We aimed to evaluate whether depression is associated with increased risk of dietary inflammatory index (DII) or energy-adjusted DII (E-DII) and whether the association is partly explained by insulin resistance (IR). METHODS: Base on the National Health and Nutrition Examination Survey (NHANES) 2005-2018. Univariate analyses of continuous and categorical variables were performed using t-test, ANOVA, and χ (2) test, respectively. Logistic regression was used to analyze the relationship between DII or E-DII and depression in three different models. Mediation analysis was used to assess the potential mediation effects of homeostatic model assessment-IR (HOMA-IR). RESULTS: A total of 70,190 participants were included, and the DII score was higher in the depressed group. DII score was related to all participant characteristics except age (p < 0.05). After being included in covariates (Model 3), participants in the highest quartile of DII score have increased odds of depression (OR: 1.82, 95% CI: 1.28-2.58) compared with those in the first quartile of DII score. And, a significant dose-response relationship was found (p-trend <0.05). No interaction between DII and HOMA-IR was observed in terms of the risk of depression, and HOMA-IR did not find to play a mediating role in the association between DII and depression. Similar results were obtained for the association between E-DII and depression. CONCLUSION: Our results suggest that a higher pro-inflammatory diet increases the risk of depression in U.S. adults, while there was no evidence of a multiplicative effect of DII or E-DII and HOMA-IR on disease risk, nor of a mediating effect of HOMA-IR.