Authors' conclusions are at odds with investigators'

EDITOR--In responding to Jefferson's editorial about vaccination and its adverse effects Classen and Classen say that present immunisation schedules may exacerbate the development of type 1 diabetes and that early immunisation with a cocktail of vaccines may prevent the development of diabetes. 1 2

The authors cite their unpublished reanalysis of the Finnish trial of Haemophilus influenzae type b vaccine.3 Their conclusions are at odds with the analyses and conclusions of the investigator (JT) presented at a National Institutes of Health workshop and were published recently.4 The Finnish investigators' review of the data could not attribute the risk of type 1 diabetes in Finnish children to differences in the timing of childhood vaccination in the vaccine trial. The incidence of type 1 diabetes in Finland has risen over several decades, and the risk seems to have been almost linear since the early 1950s. Given the disparity of conclusions, it was suggested that Classen and Classen's analysis be sent to an independent statistician. We are not aware that this has been done.

The authors quote the aim of the National Institutes of Health workshop as being “to discuss our data” when in fact the title was “an evaluation of the possible role of vaccines and infectious diseases in insulin dependent (type 1) diabetes mellitus.” Two independent reviews of available data specific to vaccines were presented by the Cochrane reviewers and the Johns Hopkins Vaccine Safety Institute; neither found an association such as that reported, and both indicated concerns about methodological issues in statistical analysis and the design and conduct of …