What You Want is What You Get
“Wanting” an outcome is what starts a lot of pseudoscience. The Holmes, et al paper is a great example of what happens when you “want” a certain result. Certain that mercury was the cause of autism, the authors took their very screwy results and spun a “Just So” story that, as it turned out, “Just Isn’t So”.
This is not always a conscious distortion of reality. Too often, it is the absolute certainty that their hypothesis is right that leads otherwise rational scientists over the edge and into the abyss. Just ask Stanley Pons and Martin Fleischman, “discoverers” of cold fusion . Neither of them was a “crackpot”, but they let their belief in their own hypothesis blind them to the flaws (major flaws) in the data.
Which leads us to the question of mercury and autism.
Let’s go ‘way back to the beginning of the hypothesis. Some folks thought that the apparent rise in autism prevalence (the “autism epidemic”) starting around 1985 could be due to the increase in thimerosal-containing vaccines that happened at about the same time. The timing wasn’t particularly close, but it wasn’t the worst hypothesis ever written. The known neurotoxicity of mercury made it biologically plausible (but didn’t prove that thimerosal could cause autism).
In real science, a hypothesis needs to explain the data – all the data – or it needs to be revised (or replaced). Things started to go wrong with the mercury-autism hypothesis when the Madsen, et al study failed to show a drop in Danish autism prevalence after Denmark removed thimerosal from its vaccines. There were some methodological problems with the study (which were spelled out by the authors), but it certainly raised a lot of doubt.
Before too long, more studies came out showing the lack of association between thimerosal dose and autism prevalence (Verstraeten, et al and Andrews, et al; Fombonne, et al). Rather than modifying (or abandoning) the mercury-causes-autism hypothesis, its proponents concentrated on attacking the motivations and ethics of the researchers.
A few studies using the thoroughly discredited VAERS database (see Goodman and Nordin) attempted to refute the better-designed studies, but were generally disregarded, except by those who were desperately trying to keep the mercury-causes-autism hypothesis alive. A report of declining autism prevalence (Geier and Geier) was not only poorly done but, as later data revealed, wrong.
Here is the crux of the matter: the autism prevalence data from the California Department of Developmental Services (CDDS) and the United States Department of Education (USDE) have not shown a decline in autism prevalence. That is not an issue in question – it is simple fact. Whether or not this data is valid (and there is some doubt about that, see Shattuck, Newschaffer and Laidler), it was what the original mercury-causes-autism hypothesis was based on.
The reason that the prevalence of autism is so critical to the mercury-causes-autism hypothesis is, of course, because thimerosal was removed from children’s vaccines in the US sometime between 2000 and 2001 (depending on which source you use). No matter when it was finally completely removed, the thimerosal dose received by children in their vaccines has [a] not risen since 1999 and [b] has significantly fallen since 2000.
Thus, even if thimerosal had remained in all children’s vaccines at its 1999 concentration – which it hasn’t – the autism prevalence should have reached a plateau by now. And it hasn’t.
Let me repeat that, for emphasis.
Even if the total amount of thimerosal that a child received in vaccines had remained at the 1999 level, the prevalence of autism would have reached a plateau by now if thimerosal was a major cause of autism.
Given that the thimerosal dose received by children born after 2001 (to allow an overly generous time for all the “on the shelf” doses to be used) has been significantly reduced, we should have seen a decline in autism prevalence by now, if thimerosal caused a significant fraction of autism cases. And we haven’t – decidedly not.
Now, the mercury-causes-autism proponents have been forced to resort to a variety of “bait and switch” tactics to keep their supporters’ eyes off of the poverty of their hypothesis. Mercury from power plants, China and crematoria has been invoked – despite the fact that mercury deposition rates have been declining since 1961 (Roos-Barraclough, et al ).
They have also proposed mercury in dental amalgams, Rho-Gam shots and flu shots as potential sources of mercury.
The basic problem with all this “bait and switch” is that they are acting as though the core assertion of their hypothesis – that mercury can cause autism – has been proven, which it decidedly has not!
The failure of US, UK, Danish, Swedish and Canadian autism prevalence to fall following the removal of thimerosal from children’s vaccines has not supported their claim – in fact, it has weakened it. In addition, showing that mercury can cause autoimmune disorders, “oxidative stress” and neuronal injury is not the same as showing that it can cause autism.
Pretty much everybody knows that mercury is not a good thing for you. This is not the question. The question at hand is whether it can cause autism. And the answer to that question – at least so far – has been “No.”
So, asking if I’m in favor of exposing kids to mercury or if I think that mercury is a good thing for kids is simply a ploy to shift attention away from the glaring absence of data supporting the mercury-causes-autism hypothesis.
And pointing out that flu shots have mercury, or that there is mercury in our food, water and air is just more of the same. The onus is on those who propose that mercury causes autism to bring sufficient data (data, not testimonials, stories about “recovered” children or wild fantasies about governmental conspiracies) to the table to show why their hypothesis is a better explanation of how the universe (and neurobiology) works than the “null hypothesis” (that mercury doesn't cause autism).
A few ground rules:
As a personal pre-condition (a promise made to myself) of returning to ‘blogging, I have elected to moderate all comments. I do this to prevent the sort of free-for-all insult-fests that are all too common from certain individuals (who need not be named).
I expect people to behave themselves in a civilized fashion. Do not bring fights from other ‘blogs into this one simply because you have been banned somewhere else. Fore Sam has already gotten himself banned for a day for doing just that. And because he has already used up his lifetime allotment of second chances. Fore Sam, consider yourself on permanent probation – you’ve earned it.
If it upsets you that your commentary is not appreciated here, then you are free to set up your own ‘blog and rant about my arbitrary justice to your heart’s content. You are free to say what you want here, as long as you remain within the bounds of civilized discourse.
Finally, do not be surprised or upset if your assertions ("It's right because I say it's right! Are you callin' me a liar!?!"}, anecdotes ("It worked for me!"), unsupported hypotheses ("I don't have any data! I don't have time to get data - I'm busy saving lives!") and conspiracies ("It's a guv'mint plot!") fail to convince. That's just the way science works.