OBJECTIVE: The need to recognize infants that are at high risk for developing signifi-cant jaundice is apparent in the era of routine early discharge. The aim of the present study was to prospectively determine the ability to predict severe hyperbilirubinemia in term healthy newborns (defined as total serum bilirubin of > 10.0 mg/dl at day 2, > 14.0 mg/dl at day 3, and > 17.0 mg/dl at days 4 and 5 of life). DESIGN: Prospective study of 1177 healthy term newborns. SETTING: Two university-affiliated community hospitals in Jerusalem. RESULTS: Using a multiple logistic regression analysis, neonatal jaundice was best predicted (p < 0.0001) by day 1 serum bilirubin (adjusted odds ratio of 3.1 [per mg/dl] [95% confidence limits of 2.4to 4.1]) and byachange in serum bilirubin from the first to the second day of life (2.4 [pec mg/dl] [1.9 to 3.0]). Maternal blood type 0 (2.9 [1.5 to 5.8]), age (1.1 [per year] [1.0 to 1.2]), schooling (0.8 [peryear] [0.7 to 0.9]), and full breastfeeding (0.4 [0.2 to 0.9]) were also associated with jaundice (p < 0.005). Other factors considered in the regression model but not found to be significantly related to jaundice included maternal ethnic origin, smoking, hypertension, diabetes mellitus, intranatal administration of oxytocin, meperidine, anesthesia, premature rupture of the membranes, parity, newborn sex, birth weight, gestational age, presentation, Apgar scores, blood type, hematocrit, cephalohematoma, and history of jaundice in other siblings. A model for predicting neonatal jaundice based on the above factors had a sensitivity of 81.8%, a specificity of 82.9%, a false positive rate of 80.2%, and a false negative rate of 1.1%. CONCLUSION: Individual risk assessment on discharge in association with day 1 total serum bilirubin is of value in identifying infants at greater risk for neonatal jaundice.