Epi Wonk has completed part II of her deconstruction of the latest abuse of epidemiology and statistics by those pseudoscientists for the mercury militia, Mark and David Geier. (I commented on part I here):
Pretty steep slopes and, therefore, apparently strong associations. But there’s no attempt to control for, or adjust for, the confounding effect of birth cohort. Just one look at Figure 1 (or a basic knowledge about trends in autism) tells you the regression coefficients (slopes) are being driven by increases in autism risk over time. Given the increase in frequency of autism (and other neurodevelopmental disabilities) during time time period, you could do an ecological regression analysis of almost any factor that varied over time and you would find an an association with autism. I would bet that you could enter number of sushi bars per capita into an ecological regression and you’d find an association with autism rates.
I once pointed out this very thing about the way antivaccinationists abuse correlation as causation, in this case starting at 1983 as the baseline, and facetiously said:
A lot of other things have happened since 1983 as well. For example, in the early 1990s, the diagnostic criteria for autism were broadened, and campaigns for greater awareness were begun. Diagnoses of autism in 1983 were made using the DSM-III, where the criteria for an autism diagnosis were much more restrictive than those in the DSM-IV, released in the early 1990s. Moreover, in 1983, categories of Asperger’s and pervasive developmental disorder-not otherwise specified, both of which are lumped into the 1 in 150 figure for 2008, weren’t recognized in the DSM-III. Of course, if I wanted to be snarky (and perish forbid that I would ever be snarky), I could point out that 1981 was the year that the IBM PC was released, followed by the Apple Macintosh in 1984, both of which led to the exponential growth of households owning and using personal computers. That’s it! It must be computer use that led to the increase in autism in the 25 years since 1983! Wait, what about the compact disc? It just so happens that 1983 is the year that the CD was first released in the American market. Ergo, it must be CDs that cause autism.
I can’t wait until part 3.