Determinants of preterm delivery in low-risk pregnancies. The RADIUS Study Group

From 14,948 low-risk singleton pregnancies, we calculated incidence, risk ratios, and attributable risks for characteristics associated with spontaneous and medically induced preterm delivery. There were 754 women who gave birth prior to 37 weeks of gestation (50.4/1000 deliveries). The greatest fraction of the incidence of prematurity among low-risk pregnancies was due to unknown factors associated with carrying a first live birth, regardless of preterm delivery mechanism (i.e., spontaneous labor, PROM, medical intervention), with population-attributable risk percents (PAR%) ranging from 16.0 to 30.5%. Other than nulliparity, male sex of the fetus accounted for the greatest fraction of spontaneous labor-induced prematurity incidence (PAR% = 13.6%), and maternal age greater than 30 years or a positive urine culture accounted for the greatest fraction of PROM-induced prematurity incidence (PAR% = 7.9 and 6.7, respectively). All other risk factors for either preterm labor or PROM accounted for less than 5% of the incidence. Three characteristics explained a large fraction of medically induced prematurity: women over 150 pounds at the onset of pregnancy (PAR% = 23.8), a > or = 2+ prenatal urine protein (PAR% = 18.7%), and cigarette smoking during the first trimester (PAR% = 8.6). Our results suggest known risk factors may explain only a small fraction of spontaneous preterm delivery incidence in low-risk pregnancies.