Prenatal Metformin Exposure and Childhood Development at School Age
Antenatal metformin exposure shows no statistically significant association with developmental vulnerabilities in children during their first year of full-time school, according to a cohort study published in JAMA Network Open. Researchers evaluated singleton births in Victoria, Australia, between January 1, 2009, and December 31, 2020—to determine whether crossing the placenta impacts children by ages 4 to 6. Researchers examined anonymized health records to track outcomes across 871,627 total births, identifying 1,095 children with documented in utero exposure.
Data Linkage Across Victorian Registries
The research team linked multiple administrative health databases to evaluate long-term outcomes, adhering to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines. Data sources included the Victorian Perinatal Data Collection, the National Diabetes Service Scheme, the Pharmaceutical Benefits Scheme, the Australian Early Development Census (AEDC), and the Victorian Admitted Episodes Dataset. Metformin exposure required at least one dispensed prescription between the last normal menstrual period and birth. Among the exposed cohort, 64.2% filled their first prescription during the first trimester, 12.9% in the second trimester, and 22.9% in the third trimester.
Assessing Developmental Domains at School Entry
Children underwent teacher-completed assessments via the AEDC during their first year of formal schooling, typically between ages 5 years 1 month and 6 years. Evaluators measured outcomes across five distinct categories: physical health, social competence, emotional maturity, language and cognitive skills, and communication and general knowledge. Investigators screened for variables such as sex, gestational age at birth, mode of delivery, birth weight, first language learned, age at assessment, and special educational needs. Unadjusted figures initially showed developmental vulnerability—defined as scoring below the 10th percentile in at least two domains—in 18.2% of exposed children, compared with 13.9% of unexposed peers.
Adjusting for Maternal and Pregnancy Confounders
When researchers applied inverse probability-weighted regression adjustment to account for background clinical differences, the apparent gap vanished entirely. Mothers who took metformin presented with higher mean body mass indices, higher rates of polycystic ovary syndrome, type 2 diabetes, and gestational diabetes, and more frequent use of assisted reproductive technology. Their infants experienced higher rates of preterm birth, cesarean delivery, and birth weights exceeding the 97th percentile. Once investigators equalized these confounding variables, adjusted analyses revealed no increased risk across any of the five individual AEDC domains, a finding that held steady when restricted to first-trimester prescriptions or pregnancies complicated by diabetes.
While the findings offer reassurance for clinicians and expectant mothers considering the medication, observational limitations persist. The study could not confirm whether patients actually ingested the dispensed pills, nor did it track specific daily dosages or distinct treatment indications. Analyses focusing exclusively on multiple prescriptions or late-pregnancy initiation lacked the statistical power required for fully adjusted models. Researchers emphasize that while neurodevelopmental markers appear secure at school entry, longer-term investigations remain necessary to monitor childhood growth patterns and cardiometabolic health outcomes.
How Many Children Were Analyzed and What Defines Vulnerability?
How many children were included in the final analysis?
Out of 871,627 singleton births recorded in Victoria between 2009 and 2020, researchers analyzed 177,409 children with fully linked developmental outcome data, which included 1,095 children with documented antenatal metformin exposure.
What threshold defines developmental vulnerability in the AEDC assessment?
The Australian Early Development Census classifies a child as developmentally vulnerable if they score below the 10th percentile in at least two of the five evaluated domains during their first year of school.
When during pregnancy were mothers prescribed metformin?
Among the exposed pregnancies evaluated in the study, 64.2% of women received their initial metformin prescription during the first trimester, 12.9% during the second trimester, and 22.9% during the third trimester.
Worth a look