Sieber, Thyne. "Real and Artifactual Components of Trends in SIDS by Race and SES, 1979-1987." PSC Research Report No. 89-149. July 1989.
The frequency of Sudden Infant Death Syndrome (SIDS) as a cause of postneonatal infant mortality differs among racial and socioeconomic groups. However, it is not clear whether these group differences reflect real differences in the incidence of SIDS, or whether they are in part due to differences in the clinical practice of assigning cause of death.
There is a debate as to whether SIDS can be unambiguously diagnosed, even through autopsy, or whether it remains largely a diagnosis of exclusion. There is evidence that decedent characteristics and the training of the medical personnel assigning cause of death can affect the attributed cause. It would not be surprising if patient and parental characteristics affected the diagnosis of SIDS, for which there are few unambiguous indicators.
This paper investigates the issue of artifactual versus real differences in the level and trend in SIDS mortality by group characteristics. The data examined are linked files of birth and infant deaths for New York state (excluding New York City) for 1979- 1987.
Linear probability models are estimated looking at the proportion of postneonatal deaths that are diagnosed as Sudden Infant Death Syndrome. Predictor variables include race, maternal age, mother's education, birthweight, and year of death.
Although the results presented here are preliminary, they offer some support for the idea that diagnosis of cause of death is not determined in the same way for all groups in society. These findings point to the need for caution in interpreting differentials in rates and trends of SIDS over time, particularly among different racial and socioeconomic groups.