Association of depression and antidepressant treatment with pregnancy outcomes in infertility care.

Aims

The associations of depression and antidepressant treatment patterns with reproductive outcomes among women undergoing infertility care remain uncertain. We examined depression history, the timing of antidepressant treatment, and time to first claims-identified pregnancy, and explored antidepressant prescription coverage among treated women.

Methods

This retrospective cohort study used nationwide administrative health-insurance claims data from the Republic of Korea for women receiving infertility care between 2017 and 2023. Overall, 601,664 women aged 20-45 years with an infertility diagnosis were identified. The final analytic cohort included non-depressed controls (group A; n = 8734) and women with depression (group B; n = 4367), stratified as past-treated (B1; n = 2091), ongoing-treated (B2; n = 1336), and treatment initiated after the start of infertility care (B3; n = 940). The primary outcome was time to first claims-identified pregnancy. Group comparisons used adjusted Fine-Gray models with treatment discontinuation as a competing event.

Results

Compared with non-depressed controls, the past-treated group had a higher subdistribution hazard of pregnancy (B1 vs. A: adjusted sHR, 1.10; 95% CI, 1.03-1.18), whereas ongoing-treated and newly treated groups had lower subdistribution hazards (B2 vs. A: 0.53; 95% CI, 0.48-0.58; B3 vs. A: 0.31; 95% CI, 0.28-0.35). Among antidepressant users (B2 and B3), each 10-percentage-point increase in prescription coverage during the observed infertility-treatment period was associated with a higher hazard of pregnancy in a secondary Cox model (adjusted HR, 1.08; 95% CI, 1.06-1.11). Antidepressant-treated pregnancy cohorts had higher odds of spontaneous abortion than non-depressed controls.

Conclusions

Depression history and antidepressant treatment patterns were associated with reproductive outcomes during infertility care. These associations cannot distinguish medication effects from underlying illness burden, confounding by indication, exposure timing, healthcare utilization, or reverse causation. The prescription-coverage association should not be interpreted as evidence that greater medication adherence improves fertility.