Objective: We aimed to establish a comprehensive digital phenotype for postpartum hemorrhage (PPH). Current guidelines rely primarily on estimates of blood loss, which can be inaccurate and biased, and ignore a suite of complementary information readily available in electronic medical records (EMR). Inaccurate and incomplete phenotyping contribute to ongoing challenges to track PPH outcomes, develop more accurate risk assessments, and identify novel interventions.
Methods: We constructed a cohort of 71,944 deliveries from the Mount Sinai Health System, 2011-2019. Estimates of postpartum blood loss, shifts in hematocrit intra- and postpartum, administration of uterotonics, surgical treatments, and associated diagnostic codes were combined to identify PPH retrospectively. All clinical features were extracted from structured EMR data and mapped to common data models for maximum interoperability across hospitals. Blinded chart review was done on a randomly selected subset of cases and controls for validation and performance was compared to alternate PPH phenotypes.
Results: We identified 6,639 cases (9% prevalence) using our phenotype - more than three times as many as using blood loss alone (N=1,747), supporting the need to incorporate other diagnostic and treatment data. Blinded chart review revealed our phenotype had 96% sensitivity, 89% precision, 77% specificity, and 89% accuracy to detect PPH. Alternate phenotypes were less accurate, including a common blood loss-based definition (67%) and a previously published digital phenotype (74%).
Conclusion: We have developed a scalable, accurate, and valid digital phenotype that may be of significant use for tracking outcomes and ongoing clinical research to deliver better preventative interventions for PPH.