Baseline adverse birth outcomes in a prospective pregnancy cohort in Uganda: Evidence to inform maternal vaccine readiness.

BACKGROUND: Maternal vaccines have the potential to substantially reduce neonatal morbidity and mortality; however, in low- and middle-income countries (LMICs), their evaluation and implementation are hindered by the scarcity of high-quality baseline data on pregnancy and neonatal outcomes. Harmonised outcome definitions, such as those developed by the Brighton Collaboration's Global Alignment of Immunisation Safety Assessment in Pregnancy (GAIA), are essential for interpreting vaccine safety signals and supporting equitable maternal immunisation programmes. METHODS: PREPARE was a prospective pregnancy registry conducted at a tertiary maternity facility in Kampala, Uganda. Pregnant women were enrolled in the first or second trimester and followed through delivery and until nine months postpartum. Maternal, fetal, and neonatal outcomes were actively ascertained using GAIA case definitions, with independent review and expert adjudication to determine diagnostic certainty. RESULTS: Between September 2020 and March 2023, 3423 women were enrolled; 98.5% were followed to delivery and 95.7% retained through nine months postpartum. The stillbirth incidence was 26.3 per 1000 total births, exceeding national estimates. Preterm birth occurred in 110.9 per 1000 births, and neonatal mortality was 22.5 per 1000 livebirths, with substantially higher incidence among preterm infants. Neonatal infections were commonly identified through active surveillance. Reported rates of most neonatal outcomes were lower among births occurring outside the study site, consistent with under-ascertainment in routine settings. Conversely, through application of GAIA criteria, overdiagnosis of some intrapartum and neonatal conditions in routine clinical records was identified. CONCLUSIONS: Prospective pregnancy registries using standardised GAIA definitions can generate robust, context-specific background rates and reveal important limitations of routine health data systems. These findings highlight the need for prospective surveillance, strengthened diagnostic capacity, and adaptation of GAIA case definitions for reliable use with electronic health record data to support the evaluation of maternal vaccines and to enhance maternal immunisation readiness in LMICs.