Preterm birth is the second leading cause of neonatal mortality in the United States (1) (second only to birth defects), and preterm labor is the cause of most preterm births (2). Neonatal intensive care has improved the survival rate for babies at the cusp of viability, but it also has increased the proportion of survivors with disabilities (3). The incidence of multiple births also has increased along with the associated risk of preterm delivery (4). Interventions to delay preterm delivery in these settings have not shown conclusive effectiveness. Because the morbidity of babies born after 34–35 weeks of gestation has diminished, most efforts to identify preterm deliveries have focused on deliveries before this age. This document describes the various methods proposed for predicting preterm birth and the evidence for their roles in clinical practice.