Preterm Birth Prediction Overview

Principles

Definition

Preterm birth prediction is the academic field focused on developing and evaluating methods to assess the likelihood of a pregnant individual delivering before 37 weeks of gestation. This research investigates clinical, demographic, biological, and environmental factors to identify individuals at high risk, enabling targeted surveillance and interventions aimed at preventing preterm delivery and improving perinatal outcomes.

Ontological type

Prediction Models

Risk Factors

Preventive Interventions

Key developments

Key Figures

Biomarker-guided Risk Assessment era

Robert L. Goldenberg [1] is associated with Baylor College of Medicine [3] and Vanderbilt University [4] during the Biomarker-guided Risk Assessment era. He contributed to the preterm prediction study, showing that maternal stress is associated with spontaneous preterm birth at less than thirty-five weeks' gestation [7], underscoring the significance of vaginal infections [9], and analyzing how gestational age and cause of preterm birth influence subsequent obstetric outcomes [8]. Donald McNellis [2] is associated with Harvard University [5] and New Haven Public Schools [6] during this era. His work on the preterm prediction study includes the gestational age and cause of preterm birth on obstetric outcomes [8], the significance of vaginal infections [9], and the development of a clinical risk assessment system [10], advancing biomarker-guided risk assessment and targeted surveillance in this era.

Quantitative Risk Stratification era

Brady E Hamilton [1], and Joyce A Martin [2], were researchers at Johns Hopkins University [3] during the Quantitative Risk Stratification era (2003-2016). Their contributions drew on Births: preliminary data for 2012 [4] and Annual Summary of Vital Statistics: 2010–2011 [5] to advance quantitative, subtype-aware risk stratification and probabilistic risk scores, enabling personalized management in this era. Joyce A Martin [2] advanced epidemiologic risk stratification by integrating vital statistics into probabilistic models, with data drawn from Births: preliminary data for 2012 [4] and Annual Summary of Vital Statistics: 2010–2011 [5]. This work, within the Johns Hopkins University [3] ecosystem, helped shift the field toward dynamic, signal-informed risk assessment for preterm birth by combining clinical measures with population-level data.

Integrated Machine-learning Prediction era

Wanda D. Barfield [1] has been affiliated with Emory University [3] and the University of Maryland, College Park [4] during this era. Her key contributions in this era, based on the 2019 paper Racial/Ethnic Disparities in Pregnancy-Related Deaths — United States, 2007–2016 [6], include identifying racial and ethnic disparities in pregnancy outcomes and underscoring the importance of integrating equity considerations into predictive models for preterm birth. William M. Callaghan [2] has been affiliated with Johns Hopkins University [5] and the University of Maryland, College Park [4] during this era. These contributions, reflected in the 2019 paper Racial/Ethnic Disparities in Pregnancy-Related Deaths — United States, 2007–2016 [6], helped advance equity-focused components of perinatal risk assessment within integrated machine-learning predictions.