OBJECTIVE To determine if preterm infants are at greater risk for sudden infant death syndrome (SIDS) than term infants and to determine if the postconceptional age of SIDS deaths varies by gestational age at birth. METHODS A cohort analysis was conducted using data from the 1987 United States' Birth Cohort Linked Birth/Infant Death Certificate tapes. SIDS was defined as the death of any infant who was > 24 weeks gestation at birth; weighed > 500 g at birth; was assigned an International Classification of Diseases-9th Revision (ICD-9) underlying cause of death of 7980; and had an autopsy. RESULTS The overall SIDS rate using our definition was 1.20 deaths/1000 live births. The SIDS rates by gestational age categories of 24 to 28 weeks, 29 to 32 weeks, 33 to 36 weeks, and 37 or more weeks were 3.52, 3.01, 2.27, and 1.06 deaths/1000 live births, respectively. Because of misclassification of gestational age among the most preterm infants, a restricted analysis was conducted on SIDS victims whose gestational ages fell within cutoff values derived from a methodology that excluded gestational age assessments assumed to be invalid. This subgroup analysis showed a mean (standard deviates) postconceptional age of death for SIDS for infants of 24 to 28 weeks, 29 to 32 weeks, and 33 to 36 weeks gestation to be 45.8 (8.3), 47.3 (8.6), and 48.0 (8.3) weeks, respectively, compared with 52.3 (8.5) weeks for term infants (ANOVA P = .0001). CONCLUSIONS We infer from this analysis that preterm infants are at higher risk for SIDS than term infants, and that the postconceptional age of peak vulnerability for SIDS may differ by 4 to 6 weeks between preterm and term infants.