Housing and sudden infant death syndrome. The New Zealand Cot Death Study Group

AIMS: This paper examined factors relating to the infants' place of domicile to see whether they increased the risk of sudden infant death syndrome (SIDS) beyond social and environmental effects previously published. METHODS: A case control study was undertaken in New Zealand between the years 1987-90. From all sudden infant death syndrome diagnoses over this time, parents of 393 (81%) sudden infant death syndrome infants consented to participate and these derive the cases. Controls were ascertained by randomly sampling 1800 infants from all babies born over 78% of the country. Parents of 1592 (88%) control infants consented to participate in the study. RESULTS: The relative risk of sudden infant death for infants usually residing in houses rented from the government (State houses) was 1.73 (95% CI: 1.13, 2.66) times that of infants with parents owning their house, after adjusting for likely social, economic and environmental confounding factors. However, the type of housing, construction of housing, heating and age of housing was not associated with sudden infant death syndrome. Although house size, measured in terms of bedroom numbers, was similar for sudden infant death syndrome and control infants (chi 2 = 0.40, df = 2, p = 0.82), the number of people normally residing within these houses was different. Sudden infant death syndrome infants' houses were less likely to have two adults and more likely to have more children normally resident. Density calculations (derived by calculating the children and/or adult numbers divided by bedroom numbers) revealed a non significant increase in relative risk, suggesting that housing overcrowding was not associated with sudden infant death syndrome in New Zealand. CONCLUSIONS: Infants domiciled in State houses are more likely to experience sudden infant death syndrome. However, this increased relative risk for sudden infant death syndrome appears to have little to do with the house per se and, perhaps, more to do with socioeconomic characteristics.