Antidepressants in Pregnancy and Neurodevelopmental Risks

Antidepressant use in pregnancy is linked to ADHD and autism, but is this correlation of causation?

STUDY: Chan JKN et al, Lancet Psychiatry 2026

STUDY TYPE: Systematic review and meta-analysis

FUNDING: Independent

Background

Observational studies have found a link between prenatal antidepressant exposure and ADHD and autism spectrum disorder in offspring. The question this meta-analysis tackles: is that association real, or does it reflect the illness driving the prescription, with the known genetic risks that entails?

The Study

• Researchers pooled 37 cohort and case-control studies covering 648,626 antidepressant-exposed and nearly 25 million unexposed pregnancies.
• The primary comparison was offspring of mothers who took antidepressants during pregnancy versus those who did not.
• Key sensitivity analyses compared exposed mothers to those who had discontinued antidepressants before pregnancy, to sibling-matched controls, and to women with untreated psychiatric illness — each designed to isolate the drug’s effect from the underlying disorder.

Results

Before adjusting for counfounders, prenatal antidepressant use was associated with modestly higher rates of ADHD (relative risk 1.35) and autism spectrum disorder (relative risk 1.69). No increased risk appeared for intellectual disabilities, motor disorders, or speech and language disorders.

Those numbers shrank considerably once confounders were addressed. For ADHD, the association became nonsignificant in every sensitivity analysis that controlled for maternal mental illness, shared family genetics, or whether mothers had actually stopped their antidepressants before pregnancy. For autism spectrum disorder, sibling-matched analyses also turned nonsignificant, though some adjusted analyses still showed a weakened but present signal.

Two findings argue against a causal drug effect.

1. Pre-conception antidepressant exposure carried similar risk estimates as prenatal exposure, a timing pattern that doesn’t fit a direct fetal mechanism.
2. Paternal antidepressant use during pregnancy predicted the same elevated ADHD and autism risks as maternal use, despite no fetal drug exposure. The signal follows the family, not the medication.

After thorough confounding adjustment, the only antidepressants that showed a persistent signal were amitriptyline and nortriptyline, both tricyclics.

Practice Implications

1. This is the most rigorous analysis to date, finding no link between the medications and adverse neuropsychiatric outcomes.
2. With the exception of valproate (which is contraindicated), there is always a balance between the risks that untreated illness poses to the pregnancy vs the risks of medications.
3. When discussing risks, remind patients about all the things they can do to improve the child’s health; caffeine, nicotine, alcohol, sleep, healthy diet, prenatal vitamins, and good parenting are critical.