Epidemiologist Research Centre - CHU Sainte-Justine Montreal, Canada
Background: Post-market surveillance concerning medication use in pregnancy is critical and having access to national cohorts to answer emergent safety and effectiveness questions of treatments is essential for decision-making. The AMerican PREGNANcy-Mother-Child Cohort (AM-PREGNANT) aims to be an additional data source that will increase the generability of performing cross-country studies on perinatal pharmacoepidemiology.
Objectives: To present AM-PREGNANT, and the baseline prevalence of the main perinatal outcomes and medication use during pregnancy
Methods: We used the IBM® MarketScan® Research Databases to build the AM-PREGNANT cohort. The linked mother-child cohort was done by identifying all pregnancies in persons 12 to 55 years of age from 01/01/2003-12/31/2021, and linking these to their family unique identifier by calendar year. The method used to create the cohort was based on prior literature for creating US pregnant-child cohorts. The AM-PREGNANT creation did not impose restrictions on duration of eligibility coverage. We further performed the harmonization with the Canadian Mother-Child Cohort (CAMCCO) definitions, restricting the cohort to persons between 15 to 45 years and adapted definitions for gestational age considering the US’ realities and health data captures. We assessed the following baseline indicators: i) Prematurity defined as gestational age, < 37 weeks; ii) Major malformations (according to EUROCAT classification using ICD-9 and ICD-10 codes); iii) Medication exposures (using the National Drug Code). We defined exposure by trimester: first (anytime between the first 14 weeks of gestation, from the first day of the last menstrual period); second (15–26 weeks gestation); and the third (from 26th weeks gestation until end of pregnancy) trimester. We presented the results as absolute numbers and overall frequencies.
Results: We identified 7,967,552 pregnancies from 6,079,885 persons (medium age 30.2 years; standard deviation: 5.7). Prematurity prevalence was 11% among deliveries, and frequency of major malformations recorded among live births was 8%. Among all malformations, the most prevalent were related to the circulatory system (38%), followed by musculoskeletal system (27%). Overall, 56% of pregnancies were exposed to at least one medication in the first trimester, decreasing to 41% and 48% in the second and third trimesters, respectively.
Conclusions: These are the first results from the AM-PREGNANT cohort. Results from cohort creation were similar to previous studies published using these data, which is reassuring. However, prevalence of major congenital malformations was higher than national estimates, which could partly be explained by the fact that no restrictions were applied on duration of eligibility coverage.
