First-trimester benzodiazepines were not linked to more malformations

Design

Population-based cohort study in US Medicaid (2000 to 2018) and MarketScan (2003 to 2020) claims data, propensity score weighted

Population

About 4.2 million pregnancies; 51,339 publicly insured and 27,779 commercially insured with first-trimester benzodiazepine exposure

Primary outcome

Any major congenital malformation and selected organ-specific malformations

Effect

Absolute risk 4.0% vs 4.1% (Medicaid) and 4.9% vs 4.8% (MarketScan); pooled adjusted RR 0.99 (95% CI 0.96 to 1.03)

A cohort study used two large US insurance databases covering about 4.2 million pregnancies, of which 51,339 of 2.48 million publicly insured and 27,779 of 1.75 million commercially insured had at least one benzodiazepine fill in the first trimester. Propensity score weighting was used to adjust for confounders, and a relative risk of 1.3 or higher was prespecified as clinically significant.

Any major malformation occurred in 4.0% of exposed against 4.1% of unexposed pregnancies in the Medicaid data and 4.9% against 4.8% in the commercial data. The pooled adjusted relative risk was 0.99 (95% CI 0.96 to 1.03). Organ-specific signals were not consistent across sensitivity analyses, there was no dose-response pattern, and different benzodiazepines gave similar results. Some co-exposure analyses with other psychoactive drugs showed raised risks, without a consistent pattern.

The reassurance applies to major malformations, not to every pregnancy outcome, and it relies on pharmacy fills rather than confirmed use. It supports a calmer conversation when a woman reports benzodiazepine use before she knew she was pregnant, and it does not replace the usual aim of using the lowest effective dose for the shortest time.

• When a woman reports early-pregnancy benzodiazepine use, explain that this large cohort found no overall rise in major malformations
• Review the indication and consider non-drug options and the lowest effective dose rather than stopping abruptly
• Ask about other psychoactive drugs taken at the same time, since co-exposure results were inconsistent and need more study
• The study did not examine late-pregnancy exposure or neonatal withdrawal
• Document the discussion and plan follow-up with obstetric care

Why it matters

Earlier studies conflicted, and fear of malformations often drives abrupt discontinuation or unnecessary anxiety.

Don't overread it

Absence of a malformation signal is not a safety endorsement; it says nothing about other outcomes or later pregnancy exposure.

The statistics, in plain English

A relative risk of 0.99 with a confidence interval of 0.96 to 1.03 means exposed and unexposed pregnancies had almost identical malformation rates, and the data can rule out a rise of the size the authors defined as important (1.3). Observational data cannot remove all confounding, even with weighting, and pharmacy fills do not prove the tablets were taken.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.