Monday, December 05, 2005

A Modest Prediction (Or Two)

Kev, over at Left Brain/Right Brain has done some excellent work on the prediction of David Kirby concerning a deadline for when the rates of autism should start to go down due to the phasing out of thimerosal in 2001 (The CDC first made this suggestion in 1999). Please check out his post at http://www.kevinleitch.co.uk/wp/index.php?p=310

And will happen if the actual rates don't go down in 2007?
I apologize for the guessing games that such a thought exercise involves, but I think a number of predictions are not "out of order" and can be based on what is seen so far. 2007 (As I remembered quite clearly) was not the first deadline year offered by Mr. Kirby and anecdotally, some of these predictions are not really all that new.

The behavior analytic Law of Effect states: The best predictor of future behavior is past behavior.

So, in consideration to that, here are my predictions of how advocates of the autism epidemic will respond in 2007 to a "no decrease" in autism rate according to the DDS; in order of likelihood.

I am not trying to be humorous or sarcastic (or mean) by presenting these. These are legitimate predictions that can be proven right or wrong.

1. Ignore it.
2. Invoke the possibility that the vaccines caused autism in a very small (compared to the whole) subset of children. The subset is too small to be noticed in the data.
2. Call it a statistical blip
3. Say that it will take more time to see the effects
4. Invoke other things that could take of the slack e.g. "flu shot, dental fillings, out gassing of coal run power plants"
5. Present a new theory that involves accumulated mercury in parents slowing having been built up and now being manifested irregardless of the removal of thimerosal causing vaccines.
6. Accuse the DDS of altering the data sets due to pressure from the vaccine manufacturers.
7. Take a new interest on how the DDS really isn't epidemiology and was never supposed to have been used that way.
8. Take a new interest in the fact that correlation isn’t causation.