Frequently Asked Questions

Stimulant use during pregnancy was associated with significantly higher risk for most adverse obstetric outcomes compared with no ADHD medication exposure, although most effects were small to moderate. In adjusted analyses using propensity score–based inverse probability of treatment weighting, increased-risk odds ratios ranged from 1.21 for large for gestational age to 1.85 for gestational hypertension and placenta previa. The study also found increased risk of preeclampsia/eclampsia, intrauterine growth restriction, preterm delivery, and spontaneous abortion, no significant effect for gestational diabetes, and lower odds of placental abruption versus no medication.

No clear safety advantage for nonstimulant medication was identified. Compared with no medication, nonstimulant exposure was associated with significantly higher risk for most obstetric outcomes, with adjusted odds ratios ranging from 1.12 for gestational diabetes and placenta previa to 5.70 for placental abruption. In direct stimulant-versus-nonstimulant comparisons, stimulant exposure had higher odds of placenta previa and large for gestational age, but lower odds of gestational diabetes, placental abruption, intrauterine growth restriction, and preterm delivery; there were no significant differences in gestational hypertension, preeclampsia/eclampsia, or spontaneous abortion.

No. In this study, stimulant use during pregnancy did not have a significant effect on gestational diabetes risk compared with the no-medication group. By contrast, nonstimulant use was associated with higher risk of gestational diabetes than no medication and also higher risk than stimulant exposure in the head-to-head comparison.

Placental abruption and preterm delivery were the most notable higher-risk signals for nonstimulant exposure. Compared with no medication, nonstimulant-associated increased-risk odds ratios ranged from 1.12 for gestational diabetes and placenta previa up to 5.70 for placental abruption, with effects described as ranging from small to large. The authors also reported increased risk for gestational hypertension, preeclampsia/eclampsia, intrauterine growth restriction, and spontaneous abortion.

In direct comparisons, stimulant exposure had higher odds of placenta previa and large for gestational age but lower odds of placental abruption, intrauterine growth restriction, and preterm delivery than nonstimulant exposure. Significant odds ratios ranged from 0.40 for lower odds of placental abruption with stimulants to 2.30 for higher odds of placenta previa. This means the relative risk pattern differed by outcome rather than favoring one medication class across all obstetric endpoints.

The study included 894,986 women aged 18–45 years with birth delivery records between January 2010 and December 2023. Of these, 3,465 were classified in the stimulant group and 575 in the nonstimulant group. Stimulant exposure required at least 2 electronic health record medication entries during pregnancy for a stimulant and no nonstimulant records, while nonstimulant exposure required at least 2 records for a nonstimulant and no stimulant records; the comparison group had no records for either medication type during pregnancy.

The investigators used propensity score–based inverse probability of treatment weighting followed by weighted logistic regression to reduce imbalance in measured baseline characteristics across exposure groups. The weighting model included advanced maternal age, race/ethnicity, smoking/drinking status, obesity/overweight, hypertension, depression, anxiety, bipolar disorder, ADHD, and substance use disorder. This approach was intended to reduce confounding from measured differences, but it cannot eliminate bias from unmeasured factors.

This was an observational electronic health record study, so the findings are signal-detection evidence rather than definitive causal safety estimates. The authors noted possible inaccuracy or incompleteness in EHR data, residual confounding from unmeasured factors, lack of adjustment for concurrent medication exposures, and substantial variability in timing, duration, and adherence of medication exposure during pregnancy. They also could not determine specific prescribing indications for many nonstimulant users, did not compare treated versus untreated women with ADHD specifically, and did not analyze risks for individual stimulant formulations or individual nonstimulant agents.

The study suggests that both stimulant and nonstimulant ADHD medication use during pregnancy were associated with elevated obstetric risk after accounting for measured demographic, psychiatric, and medical differences, and nonstimulants were not clearly safer overall. The authors emphasize individualized risk-benefit decisions, shared decision-making, and monitoring across gestation, including treatment changes, symptoms, blood pressure, and fetal growth. They also note that most associations were modest in magnitude and should not be interpreted as proof of causation.