Design: Retrospective cohort, US Medicaid claims 2016-2021
Population: 21,923 people with epilepsy prescribed enzyme-susceptible hormonal contraception
Primary outcome: ≥14 days of overlap with an enzyme-inducing antiseizure drug
Effect: 32.6% overlapped; median 218 days

This retrospective study used US Medicaid claims from 2016 to 2021 for 110,976 people with epilepsy of childbearing potential (median age 24). It looked at those prescribed hormonal contraception whose effect is reduced by liver enzyme induction. Published in Neurology on 7 October 2026.

Of 21,923 such patients, 32.6% had at least 14 days of overlap with an enzyme-inducing antiseizure drug, and the median overlap was 218 days. Overlap was more likely with intellectual disability (adjusted OR 2.33), drug-resistant epilepsy (1.73) and age over 35 (1.45).

The interaction is well known, yet a third of exposed patients had it for months. An unplanned pregnancy on an antiseizure drug carries risks for both mother and baby.

Claims data cannot show adherence, counselling or barrier use, so not every overlap meant unprotected sex. The message holds: every prescriber of either drug owns this check.

Ask every woman of childbearing potential on antiseizure drugs which contraception she uses
Recognise enzyme inducers: carbamazepine, phenytoin, phenobarbital, primidone, oxcarbazepine, and topiramate at higher doses
Offer a copper or levonorgestrel IUD, or medroxyprogesterone injection, which inducers do not undermine
Do not rely on combined pills, progestogen-only pills or implants alongside an inducer
Take particular care in patients with intellectual disability or drug-resistant epilepsy

Why it matters

A well-known interaction is still being missed in a third of the patients it affects.

The statistics, in plain English

The odds ratios show which groups were more often exposed, not why. An adjusted OR of 2.33 means the odds of overlap were more than twice as high in people with intellectual disability, after accounting for other characteristics.