The bidirectional relationship between depression and type 2 diabetes mellitus (T2DM) has gained increasing attention, yet the strength and heterogeneity of this association remain inadequately evaluated.

This study aimed to quantify the bidirectional association between these conditions and explore key effect modifiers through systematic review and meta-analysis.

We systematically searched PubMed, Web of Science, and other databases (from inception to June 2025) for prospective cohort studies investigating the bidirectional association between depression and T2DM. Random-effects models were used to pool effect estimates. Subgroup analyses and meta-regression were performed to explore sources of heterogeneity.

Eight prospective cohort studies (total n = 173,854; follow-up: 3.2–12.0 years) were included. Meta-analysis showed that: (1) Depression significantly increased the risk of T2DM (RR = 1.38, 95% CI: 1.26–1.50; I2 = 38.5%), with stronger associations observed in older populations (RR = 1.47, 95% CI: 1.35–1.60) and those with longer follow-up duration. (2) In the diabetes-to-depression pathway, pooled analysis showed that T2DM increased depression risk by 35.7% (RR = 1.36, 95% CI: 1.22–1.51, p < 0.001). Age and follow-up duration were identified as major effect modifiers (p for interaction < 0.05). Subgroup analyses identified age, disease severity, and medication use as major effect modifiers.

This meta-analysis reveals a significant bidirectional association between depression and T2DM, with disease severity and treatment status as key effect modifiers. These findings suggest the need for targeted screening strategies and integrated management approaches for both conditions, particularly in high-risk populations.