Antidepressant exposure patterns during pregnancy and risk of adverse newborn outcomes.

Studies on antidepressant use during pregnancy often rely on a simplified exposure classification. We introduce a novel data-driven method to identify complex antidepressant exposure patterns during pregnancy and estimates the risk of 24-hour hospitalization, congenital heart defects (CHD) and persistent pulmonary hypertension of the newborn (PPHN). We constructed a nationwide cohort of all newborns born to women who filled at least one antidepressant prescription 24 weeks before pregnancy in Denmark during 1997-2016 using national registries. Antidepressant exposure patterns were identified by hierarchical cluster analysis based on number of antidepressants, Anatomical Therapeutic Chemical code and timing similarities. Risk ratios were estimated using the cumulative incidences in the inverse probability treatment weighted population. The study included 33,776 newborns. Discontinuing antidepressant use during pregnancy was associated with significant decreased risk of hospitalization compared to continued exposure to Citalopram throughout pregnancy. Moreover, discontinuing monotherapy was associated with significant decreased risk of PPHN compared to continued exposure to Citalopram. No risk ratios were statistically significant for the risk of CHD. Our study suggests a potential decreased risk of 24-hour hospitalization for any reason and PPHN if mothers discontinue antidepressant use during pregnancy as compared with continuous Citalopram use.