Development of risk prediction models for preterm delivery in a rural setting in Ethiopia

Abstract

Importance: There is a lack of accurate predictive models to identify pregnancies that are at high risk of preterm delivery, especially in resource-limited settings. Objective: To assess whether it is possible to develop highly accurate prognostic predictive models for preterm delivery using data from a resource-limited setting in Ethiopia. Design: Nested cohort study using data from the Birhan Health and Demographic Surveillance System and its associated pregnancy and child cohort. Setting: The study was conducted at the Birhan field site, North Shewa zone, Ethiopia, a platform for community and facility-based research and training, with a focus on maternal and child health. Participants: Women enrolled during pregnancy in the pregnancy and child cohort between December 2018 and March 2020, who are followed-up in home and facility visits beyond 28 weeks of gestation and through delivery. Exposures: A wide range of socio-demographic, economic, clinical, environmental, and pregnancy-related factors were considered as potential predictors for preterm delivery. The selection of predictors was guided by literature review and expert knowledge. Predictors that were not available in the surveillance system and cohort datasets were not considered. Main Outcome Measure: Preterm delivery, a composite indicator defined as any delivery occurring before 37 completed weeks of gestation, regardless of vital status of the fetus or neonate. For time-to-event models, time to delivery in weeks was the main outcome measure. Results: A total of 2493 pregnancies were included. Of those, 138 women were censored due to loss-to-follow-up before delivery. Overall, predictive performance was poor. The area under the receiving operator characteristic curve was highest for the tree ensemble classifier (0.60, 95%CI[0.57, 0.63]). When models were calibrated so that 90% of women who experienced a preterm delivery were classified as high risk, at least 75% of those classified as high risk did not experience the outcome. Conclusions and Relevance: Prediction of preterm delivery remains a major challenge. In resource-limited settings, predicting high-risk deliveries would not only save lives, but also inform resource allocation. It may not be possible to accurately predict risk of preterm delivery without investing in novel technologies to identify genetic factors, immunological biomarkers or the expression of specific proteins.