OBJECTIVES: To estimate the risk of neonatal outcomes from patterns of prenatal antidepressant use. METHODS: From the OptumLabs Data Warehouse, 226 932 singleton deliveries were identified. Antidepressant claims with coverage between the last menstrual period and 35 weeks' gestation were converted to fluoxetine equivalents, and a longitudinal cluster analysis was performed. Outcomes included major cardiac malformations (11.7 of 1000 births), preterm birth (75.7 of 1000 births), and newborn respiratory distress (54.2 of 1000 births). The lowest trajectory was the primary reference group, and depression and anxiety with no antidepressant claims served as secondary reference groups. RESULTS: From 15 041 (6.6%) pregnancies exposed to an antidepressant, use patterns were best described as (1) low use ( ∼ 75 mg/day). Moderate sustained use increased the risk of major cardiac malformations, although results included the null when compared with depression or anxiety reference groups. Moderate sustained (adjusted risk ratio [RR] 1.31; 95% confidence interval [CI] 1.16-1.49) and high sustained (adjusted RR 1.78; 95% CI 1.48-2.14) trajectories were associated with an increased risk of preterm birth. All 4 trajectories increased the risk of neonatal respiratory distress in a dose-response fashion (adjusted RRs 1.36 [95% CI 1.20-1.50] to 2.23 [95% CI 1.83-2.77]). CONCLUSIONS: Although findings support continuation of the lowest effective dose to treat depression or anxiety, which benefits the mother, they also highlight an increased risk for newborn respiratory distress in all groups and preterm birth at moderate to high sustained doses. ∼ 10 mg/day) with first-trimester reduction, (2) low sustained use ( ∼ 20 mg/day), (3) moderate use ( ∼ 40 mg/day) with first-trimester reduction, (4) moderate sustained use ( ∼ 40 mg/day), and (5) high sustained use (