OBJECTIVE To describe the epidemiology of neonatal jaundice at the University Hospital of the West Indies (UHWI). METHODS A retrospective review of all neonates at the UHWI with clinically significant jaundice between January 1, 2006 and June 30, 2007 was performed. Demographic, clinical and laboratory data were collected. Descriptive analyses were performed. RESULTS The incidence of clinically significant neonatal jaundice at the UHWI was 4.6% for the study period. There were 103 male (61%) and 67 (39%) female infants. The aetiology ofjaundice in the infant was attributed to ABO incompatibility in 59 (35%), infection in 30 (18%), prematurity in 19 (11%), G6PD deficiency in 8 (5%), Rhesus incompatibility in 6 (3.5%) and no cause was identified in 16 (9%) infants. There was a low incidence (26%) ofscreening for G6PD deficiency although it was the most common aetiology for infants presenting from home. Nine (5%) neonates required exchange blood transfusion. Infants admitted from home had a significantly higher mean total bilirubin value at presentation, a significantly higher mean peak bilirubin level andpresented significantly later than those who were admitted from the postnatal ward (p < 0.001). One patient was discharged with a diagnosis of bilirubin encephalopathy but defaulted from follow-up. Two neonates died but from causes unrelated to neonatal jaundice. Sixty-two patients (37%) were followed-up post discharge; 50% had hearing tests done, all tests were normal. Sixty-one (98%) infants had normal development at the time of the study; one patient had impaired motor development but this infant also had a myelomeningocoele. CONCLUSIONS To further reduce morbidity associated with neonatal jaundice at the UHWI, there should be increased screening for G6PD deficiency; current systems in place for follow-up and monitoring of infants discharged from hospital prior to 72 hours must also be expanded and strengthened.