Sir, - I am genuinely delighted that the vaccination and SIDS debate is being dragged into the scientific arena with Myles Crowe (September 24th) quoting from a 1992 American Journal of Epidemiology article which reviewed the literature relevant to this controversy.
Mr Crowe chides the "medical establishment" for not discussing the conclusions of this study, which he then proceeds to ignore. The article concluded that "most published studies have reported a deficit of sudden infant death syndrome among vaccinees, which may reflect confounding in their study design". I agree wholeheartedly. This conclusion, of fewer SIDS deaths in vaccinated infants, has been seen in most studies. An Australian study found the majority of SIDS deaths occurring in unimmunised infants. A large North American study, of 800 SIDS cases and matched controls, found fewer SIDS cases occurred in immunised infants. A New Zealand study, of 485 SIDS and 1,800 control infants, found immunisation to significantly reduce the risk of SIDS.
Large, SIDS-related databases have also been established in Germany, France, Scotland, England, Norway and Sweden, and in none has immunisation emerged as a risk factor for SIDS, with the majority finding fewer cases of SIDS among immunised infants. In 1992 Ireland established a national register of all infant deaths which also shows no association between immunisation and SIDS, but rather a clear link between socio-economic deprivation (the most likely confounding factor) and the associated low immunisation rates, and SIDS.
Against this background Mr Crowe picked a tiny, old (197283) study of 29 SIDS cases and used a sub-group of nine of these as the entire basis of his argument. This practice of "data dredging" or "torturing data" until some speck of evidence is found to support your bias is pseudo-science of the worst kind. - Yours, etc,
Prof Tom Matthews,
Rotunda Hospital, Dublin 1.