Predictive tools in maternal health: Making futures for patients and populations

Abstract

Researchers develop tools from population data to predict health outcomes for individuals and for groups. Clinicians use some of these predictive tools to guide medical interventions; policymakers use other predictive tools to allocate funds and inform policy decisions. Building on critical metrics scholarship, ethnographic research, and medical experience in Malawi and the United States, I first review how clinicians use three tools—height measures, partographs, and Vaginal Birth After Cesarean calculators—as decision aids in obstetrical care.  Each tool was designed to predict and avert devastating obstetric outcomes. Each has been used globally for millions of women, shaping actual futures by forecasting probable ones. Predictive tools provide important guidance. They also share limitations: they can incorporate feedback loops, naturalize biases, and misdirect attention from alternative possibilities. Those problems may scale up into the increasingly complex global-health projections used by policymakers to guide public-health decisions. Without systematic vigilance, health-related predictive tools may have unhealthy unintended consequences. Extrapolation from what we know about simpler clinical tools provides researchers with vulnerabilities to investigate, and ideas about how we might reimagine population-level predictive tools to build a healthier future.