OBJECTIVE: The goal was to assess the impact of universal bilirubin screening on severe hyperbilirubinemia and phototherapy use. METHODS: In this retrospective cohort study of 358086 infants of > or =35 weeks and > or =2000 g born between January 1, 1995, and June 30, 2007, we obtained demographic data, bilirubin levels, and codes for inpatient phototherapy from existing databases. We compared the incidence of high total serum bilirubin (TSB) levels and phototherapy before and after implementation of universal screening and examined risk factors for high TSB levels. RESULTS: A total of 38182 infants (10.6%) were born at facilities that had implemented universal bilirubin screening. Compared with infants born at facilities that were not screening, these infants had a 62% lower incidence of TSB levels exceeding the American Academy of Pediatrics exchange guideline (0.17% vs 0.45%; P < .001), received twice the inpatient phototherapy (9.1% vs 4.2%; P < .001), and had slightly longer birth hospitalization lengths of stay (50.9 vs 48.7 hours; P < .001). Of those receiving phototherapy, 56% after initiation of universal screening had TSB levels at which phototherapy was recommended by the guideline, compared with 70% before screening. The adjusted odds ratio for developing TSB levels exceeding the guideline value was 0.28 (95% confidence interval: 0.20-0.40) for those born at a facility using TSB screening and 0.28 (95% confidence interval: 0.19-0.42) for those born at a facility using transcutaneous bilirubin screening. CONCLUSIONS: Universal bilirubin screening was associated with a significantly lower incidence of severe hyperbilirubinemia but also with increased phototherapy use.