Attention-deficit hyperactivity disorder medication use in pregnancy and risk of miscarriage

Background An increasing number of women of childbearing age are treated for attention-deficit hyperactivity disorder (ADHD). Limited evidence exists on risk of pregnancy loss associated with ADHD medication use in early pregnancy. Aims To assess whether ADHD medication use during pregnancy is associated with increased risk of miscarriage. Method We conducted a nationwide, register-based, case–control study, using linked Norwegian data from Medical Birth Registry of Norway, Norwegian Patient Registry, Norwegian Control and Payment of Health Reimbursements Database and Norwegian Prescription Database. Among pregnant women with ADHD, those with miscarriage ( n = 2993 cases) were matched with up to four live births ( n = 10 305 controls) by maternal age and year of conception. ADHD medication exposure during pregnancy was defined as any use (one or more filled prescriptions) and categorised into tertiles of total defined daily doses (DDDs) as a proxy for dose. The main outcome was miscarriage (pregnancy loss before 20 weeks). Conditional logistic regression was used to estimate adjusted odds ratios (aORs) with 95% confidence intervals, adjusting for psychiatric comorbidities, psychotropic and teratogenic medications, and maternal age at conception. Results Of 13 298 pregnancies, 1389 (10.5%) were exposed to ADHD medications. Any ADHD medication use was associated with increased miscarriage risk (aOR 1.60, 95% CI 1.41–1.83). Methylphenidate (aOR 1.55, 95% CI 1.35–1.79), lisdexamfetamine (aOR 1.81, 95% CI 1.06–3.10) and atomoxetine (aOR 2.34, 95% CI 1.41–3.89) were associated with increased risks. Higher levels of medication exposure, categorised by DDD tertiles, were associated with increased odds of miscarriage, increasing from 1.14 (95% CI 0.91–1.42) for the lowest tertile to 2.11 (95% CI 1.71–2.60) for the highest. Conclusions ADHD medication use during pregnancy is associated with increased miscarriage risk. However, filled prescriptions may not reflect actual use. Further research is needed to clarify these associations and refine risk estimates.