The authors investigated risk profiles of sudden infant death syndrome (SIDS) as a function of age at death. A case-control study carried out in the Tyrol region of Austria enrolled 99 infants who died of SIDS between 1984 and 1994 and 136 randomly selected controls. Early and late SIDS (< 120 days of age vs. > or = 120 days) were defined according to the clear-cut bimodal age-at-death distribution. Inadequate antenatal care, low parental social and educational level, and the prone sleeping position were risk conditions that applied to both early and late SIDS. A marked seasonal variation (winter preponderance) was the most outstanding feature of late SIDS. A gestational age of < 37 weeks (odds ratio (OR) = 8.4, 95% confidence interval (CI) 2.6-26.0), repeated episodes of apnea (OR = 5.7, 95% CI 1.2-27.0), low birth weight (< 2,500 g) (OR = 3.4, 95% CI 1.1-11.0), a family history of sudden infant death (OR = 2.9, 95% CI 1.1-7.5), and maternal smoking during pregnancy (OR = 2.2, 95% CI 1.0-4.5) were associated with early SIDS. This study identified two distinct subgroups of SIDS infants characterized by different risk conditions and ages at death. These results underline a multiple-cause hypothesis for SIDS etiology which involves a genetic predisposition, immaturity in the first months of life, and environmental factors acting at various ages.