Women's reproductive traits and major depressive disorder: A two-sample Mendelian randomization study.

BACKGROUND

Evidence suggested strong associations between women's reproductive factors and major depressive disorder (MDD), but their causalities are unclear.

METHODS

Using female-specific SNPs as genetic instruments obtained from large-scale genome-wide association studies for women's reproductive traits, we designed two-sample univariable and multivariable Mendelian randomization (MR) analysis to evaluate the causal effects of women's reproductive traits on MDD. For both univariable MR (UVMR) and multivariable MR (MVMR), the inverse variance weighting estimates were reported as main results. MR-Egger, weighted median, and generalized summary-data-based MR (GSMR) methods for UVMR, and MVMR-Egger and MVMR-robust methods for MVMR were used as sensitivity analyses. Negative control analyses, MVMR of age at first birth (AFB) and age at first sexual intercourse (AFS) on MDD, and sex-combined genetic variants for AFB and AFS were performed to enhance the robustness of our study.

RESULTS

There was substantial evidence for associations of genetically predicted later age at menarche (AAM) (odds ratio (OR) = 0.97, 95 % confidence interval (CI) = 0.94-0.99, P = 0.007), AFB (OR = 0.91, 95 % CI = 0.86-0.97, P = 0.002) and AFS (OR = 0.70, 95 % CI = 0.60-0.80, P < 0.001) with lower MDD risk in UVMR. After adjustment of BMI and educational attainment using MVMR, we found consistently significant causal effects of AAM (OR = 0.95, 95 % CI = 0.92-0.99, P = 0.006), AFB (OR = 0.88, 95 % CI = 0.84-0.91, P < 0.001) and AFS (OR = 0.71, 95 % CI = 0.64-0.79, P < 0.001) on MDD.

CONCLUSIONS

Our results provide compelling evidence that early AAM, AFB, and AFS are risk factors for MDD. Promoting the cognition of reproductive health care for women may reduce the risk of MDD.