Objective  To investigate the association between processed food pattern and the risk of depression during pregnancy and the role of inflammation-related diseases on the association.

  Methods  Totally 6 313 pregnant women at 18 – 27 weeks of gestation were recruited in clinics of 3 hospitals in Hefei city, Anhui province from March 2015 through June 2020. An interview was conducted at the enrollment and two follow-up surveys were carried out at 32 – 36 weeks of gestation and at the delivery. A self-designed questionnaire, Edinburgh Postnatal Depression Scale (EPDS), and Food Frequency Questionnaire (FFQ) were used to collect information on demographics, frequency of food consumption, and diseases/depression. Exploratory factor analysis was adopted to discriminate dietary patterns and binary logistic regression models were used to analyze associations of processed food and inflammation-related diseases with depression risk in third trimester. Results Of all the pregnant women, 13.4% (n = 844) were assessed as having severe depression at the third trimester. The pregnant women with high scores ( ≥ 75th percentile) for processed diet pattern had an increased risk of gingival bleeding, oral ulcers, vaginitis and other inflammation-related diseases (all P < 0.050). The inflammation-related disease score was positively correlated with serum leukocyte count (r = 0.066, P < 0.001). The results of logistic regression analysis showed that (1) for the pregnant women with low scores (< 75th percentile) of processed diet pattern, those suffering from inflammation-related diseases were at a significantly increased risk of mild depression (odds ratio OR = 1.343, 95% confidence interval 95% CI: 1.174 – 1.536) and severe depression (OR = 1.275, 95% CI: 1.062 – 1.531) in the third trimester of pregnancy compare with the pregnant women without the diseases; (2) the pregnant women with high processed diet pattern scores but without inflammation-related diseases were at an increased risk of mild depression in the third trimester (OR = 1.417, 95% CI: 1.156 – 1.736), while no significantly increased risk of severe depression was identified (P = 0.141); (3) and the pregnant women with both high processed diet pattern scores and inflammation - related diseases were at a significantly increased risk of mild depression (OR = 1.778, 95% CI: 1.490 – 2.123) and severe depression (OR = 2.209, 95% CI: 1.766 – 2.763). Mediation analysis revealed that inflammation - related disease exerted a mediation effect on the correlation between processed diet and depression in the third trimester and the mediation effect accounted for 7.54% of the total effect. Conclusion The pregnant women with both processed diet pattern and inflammation - related disease in the second trimester could have a significantly increased risk of depression in the third trimester and processed diet may increase depression risk through promoting the incidence of inflammation-related diseases.