MMR

[zotz]

I've also been reading a reanalysis of the famous Danish paper that holed the MMR/autism hypothesis below the waterline. There was somebody on the Today program burbling on about how this reanalysis resurrected the anti-vaccine view (and saying, among other things, that epidemiology wasn't the way to tell if there was a link . . . . er, Hello? Earth calling fuckwit?), so obviously I've found a copy and I've been reading it. I'm quite puzzled. Not only does it not prove that epidemiology doesn't work on these problems (it is, after all, an epidemiology paper), it seems to be quite unclear about what it actually does hope to show. What it seems to show is that autism cases were already rising before formal changes in official definitions were made in about 1993. This, I think, is not news. Oddly, although MMR was licenced in Denmark in 1987, that isn't the boundary between the periods compared. I could drone on for ages about this, but I'll spare you. Executive summary: a very unsatisfying paper that doesn't convincingly say anything about its data.

Goldman et al, Journal of American Physicians and Surgeons 9(3) 70-75