(A-354) Methods for Estimating Background Rates of Obstetric and Adverse Events of Special Interest During Pregnancy: A Multi-Site Analysis of Over 3 Million Annual Births

Background: Understanding background rates of adverse events of special interest (AESI) during pregnancy is essential for pharmacoepidemiologic research and regulatory decision-making. In addition to AESI measured in the general population, vaccine safety surveillance during pregnancy requires monitoring of other events unique to pregnancy, including obstetric, fetal, and neonatal complications. Establishing background rates for these events is particularly valuable before the introduction of new maternal vaccines, such as RSV and Group B Streptococcus, as it supports real-time pharmacovigilance in pregnant women.

Objectives: To develop and implement a standardized approach for estimating background rates of obstetric outcomes and AESI during pregnancy using multi-site healthcare databases.

Methods: The Global Vaccine Data Network (GVDN) is an international collaboration enhancing vaccine safety surveillance through standardized data collection and analysis. GVDN sites (n=16; Indonesia [1 site], United States [2 sites], France, Australia [2 provinces], Republic of Korea, China [1 site], Canada [1 province], New Zealand, Argentina [1 province], Denmark, VAC4EU [3 sites], and Taiwan) use national, regional, and hospital healthcare databases to estimate background rates using a retrospective cohort approach from 2017-2023. Pregnancies and incident cases of obstetric, fetal, and neonatal outcomes, and AESI during pregnancy are identified using ICD-10 codes in available healthcare databases.

Results: The total cohort includes data from over 3 million births per year across all sites. Estimating background rates consistently across countries and healthcare settings presents challenges. Variations in data availability, clinical coding, and data collection are addressed using a standardized analytical framework and data harmonization to enhance comparability. Denominator choices depend on data availability and time requirements. “Pregnancy-months” can serve as a denominator that accounts for time-at-risk, with incidence rates expressed as events per 10,000 pregnancy-months. Alternatively, pregnancies or live births be used, offering rapid accessibility through vital statistics. Sensitivity analyses assess the impact of case definitions, data source variations, and population characteristics on estimates.

Conclusions: This study establishes a rigorous methodology for estimating obstetric and AESI background rates during pregnancy, supporting rapid vaccine safety assessments following the introduction of new maternal vaccines.