That is, of course, the key question that must be answered before one is to draw any final conclusions about the current autism rates in this country being associated – or not associated – with exposure to thimerosal used in childhood vaccines. Yet, defenders of thimerosal – a highly toxic compound that is 50% mercury by weight - conveniently ignore this question.
One such example of this was on “Penn and Teller’s Bullshit!” If there is one good thing I can say about the Penn and Teller episode on vaccinations – it’s that it at least did not run on basic cable for more people to see. The show was so erroneous that even “Science”Blogger David Gorski corrected the hosts on the false assumption that thimerosal was removed in 1999. They ought to claim that to the parents of Hannah Poling, who got 50 mcg of mercury from shots given in 2000 right before she developed full-blown autism.
Of course, Gorski’s “correction” wasn’t any better, claiming it was “the end of 2001/early 2002.” That’s funny, not even the CDC has used that date. The government has provided many conflicting dates for when thimerosal was removed: as early as 1999 and as late as 2005, and individual reports reveal thimerosal-preserved vaccines with expiration dates that stretch even beyond that, not counting flu shots which continue to be preserved in thimerosal.
The specific time, however, repeated by Gorski which I’ve seen circulating around the internet was based on minutes from an ACIP meeting held eight years ago during which a CDC official claimed that in a “convenience sample” of health care providers scattered throughout the country, all but 17% of them were found to have had no thimerosal-preserved vaccines in stock since October 2001. This figure suddenly became widely used five years later after being dug up by tomato writer Arthur Allen to claim this time range was when thimerosal disappeared.
Except, the only problem is that being a convenience sample, this statistic is inherently incapable of being generalized to any population beyond its own by virtue of not being randomized. If doing site visits and lot counts of the vaccine stockpiles used by these particular providers is “convenient ” for public health officials, then so is getting them to switch to thimerosal-free vaccines as soon as possible in the first place – making these numbers uninformative and misleading about the thimerosal content of childhood vaccines overall.
Even worse, these statistics have been applied to a specific state – California – where the reported autism rate has not decreased (although I argued earlier this year that the old California autism rates could potentially correlate with thimerosal exposure HERE). Our current knowledge of autism trends in California, however, is based on information from no later than 2007, after which the entire autism reporting system of California was overhauled. Arthur Allen took the statistics informally obtained by the CDC, and used them to assess thimerosal’s association to autism in California when debating David Kirby, even though they cannot be generalized to nationwide thimerosal content of vaccines, much less to specific states. The convenience sample numbers ultimately raise more questions than they do answers.
Certainly the CDC has a track record of not removing this preservative as early as the agency said it would. At HuffPo, for example, Robert F. Kennedy Jr. discussed how the CDC tacitly declined an offer by GlaxoSmithKline to supply the US market with all thimerosal-free DTaP vaccines, despite the CDC later making false assurances to the public that thimerosal would be removed from all childhood shots as soon as possible.
Of course, the most obvious calculation that the CDC could have easily made was to compare the autism rates in children who received shots before and after October 2001 by the very providers who stopped using thimerosal after this date. That would have given us a clear assessment of thimerosal’s association with autism, rather than falsely extrapolating the convenience sample to all 60,000 pediatricians in the country. Not doing so indicates that such an assessment would have yielded a result less than supportive of the CDC agenda. Instead, we’re left with an unreliable convenience sample estimate obtained by a dishonest government agency.
Jake Crosby is a college student with Asperger Syndrome at Brandeis University who is double majoring in History and Health: Science, Society and Policy. He is a contributing editor to Age of Autism.