Sudden infant death syndrome in Indigenous and non-Indigenous infants in north Queensland, 1990-1998

Objective: To compare the epidemiology of sudden infant death syndrome (SIDS) in Indigenous and non-Indigenous infants in north Queensland, and to assess the quality of data recorded for SIDS deaths. Methods: Records were obtained for SIDS cases from all coronial courts in north Queensland from 1990 to 1998. Demographic characteristics, ethnicity, age at death, sleeping and feeding patterns, smoking incidences and autopsy findings were compared. Incidences, medians and univariate associations were generated where appropriate. Results: There were 83 248 live births for the 9-year period; 71 389 non-Indigenous and 11 859 Indigenous births. There were 69 SIDS deaths (0.83 per 1000 live births). Overall, recording of demographic and death scene data was poor. Thirty-eight autopsies (55%) were performed by specialist pathologists. There were 22 (32%) non-Indigenous and 22 (32%) Indigenous SIDS deaths (25 ethnicity unknown), giving an estimated relative risk of 2.82 (95% CI 2, 4). Median age at death was 13.1 weeks (range 1-83 weeks) with 14% of deaths occurring in the neonatal period for both groups. Sleeping position was not recorded in 42% of cases and co-sleeping was not recorded in 27% of cases. Bed sharing was more common amongst Indigenous infants. Fifty-two per cent of SIDS cases occurred in the wet season and 48% in the dry season. Conclusions: Data recorded for SIDS deaths in north Queensland are poor, preventing specific conclusions concerning SIDS risk factors. However, SIDS rates may be up to three-fold higher in the Indigenous population. A uniform system of post-mortem and death scene data reporting is needed urgently.