Autism and vaccines...never was?

[pallas_marisa]

The last time I wrote about autism and vaccines in this blog, I presumed that there was legitimate scientific debate and their relationship. An article in the recent issue of Scientific American shows that the scientific data never actually was: the article that laid the empirical claim for vaccines causing autism was recently retracted.

Impact

Comments

[mr_sadhead]

Yeah, good luck putting the furies back in the box .. There are still people going around citing that study that showed abortion causes breast cancer, though it's been refuted a dozen times.

[don_negro]

I really wish I could get some better information about this. All I've seen is that the paper was retracted because it didn't go through an IRB and that there was something out of the ordinary in how the subjects were recruited. That might mean there's a problem with the science, or it might mean that this whole thing is political. Let me rephrase that: everything associated with this topic has become political, but I don't care about the politics; I only want to know about the science, and I can't find the information I need to work out my own opinions on the science.

[mechasaprophyte]

Here is a pretty good article,
http://www.psychologytoday.com/blog/child-myths/201002/lancet-takes-it-all-back-vaccines-and-no-autism
with a follow-up here:
http://www.psychologytoday.com/blog/child-myths/201002/the-bicker-wakefield-the-real-reasons-why-lancet-was-right-retract

The upshot being that the way they tried to diagnose symptoms of autism pre- and post-vaccination was really sloppy, and that the sample size of 12 subjects was way too small. (Elsewhere I've read that this latter problem is compounded by the subjects not having been consecutively referred, because it's even more likely that the apparent effect was the result of a sampling error.)