A few points on what will probably be my last thimerosal-related post for a while.
1) Notwithstanding my P.P.S. below, I'm not going to do a major discussion of the merits of the existing studies on the link between thimerosal and autism. Even though I think the existing studies may be less flawed than Dwight Meredith or Wampum do, the scope of the disagreement is ultimately marginal: I don't think the existing evidence provides conclusive disproof of a link, and they do not argue that the existing evidence provides conclusive proof of a link.
2) Professor Mark Kleiman is threatening to dethrone Prof. Reynolds as Most-Favored Professor of Blissful Knowledge. (All that's lacking is the ability to generate a few thousand more hits with each link, a spot on his permalinks and a 50% rightward correction in political views, and the deal is sealed.).
Regarding the screwups identified in the Tax Notes piece noted below, Kleiman notes:

This is exactly the reason that complicated legislation shouldn't be pushed through without hearings. And these two mistakes suggest that Dr. Frist, the sponsor of the original amendment, Dick Armey, who has sorta-kinda acknowledged that he was responsible for pasting it into the Homeland Security bill, and the still-anonymous-but-widely-believed-to-be-Mitch-Daniels White House gnome who passed the word to Armey all failed to perform due diligence. (Unless, that is, it was their intention to deprive the families of the right to sue and "accidentally" leave them with no recourse whatever.) If you're going to short-cut the process, you ought to be certain you've got the substance right.

There's nothing that keeps the new Congress from fixing these problems. But the Lilly lobbyists, and their friends on the Hill, including the new Majority Leader, now have the huge advantage of the status quo. And Frist may not regard himself as being bound by the promises to undo the damage that Trent Lott made (and immediately started to back away from) to hold on to the Republican moderate votes he needed to pass the bill last session. The families and their friends are now in the position of begging the other side for whatever concessions it might deign to offer. Not a pretty picture.

I agree with most of Kleiman's first paragraph. I'm less cynical than Kleiman about the likelihood that the errors will go uncorrected and even more doubtful that the result was some sort of conspiracy to leave affected families no recourse. Bill Frist's original bill (PDF link) on the subject contained the appropriate conforming amendments to the Internal Revenue Code (click here and here). With that precedent, it does not seem likely (to me, at least) that passage of the appropriate conforming amendments would meet with much difficulty on the merits. It's further argument against sloppy, slapdash lawmaking, but not proof of some conspiracy to wholly eliminate families' recourse.

3) Does anyone know how long of a lag there is in the collection and reporting of figures on autism incidence? We have had almost three years of largely thimerosal-free vaccinations being administered to children in the U.S., and I would expect that the resulting data on autism incidence should begin showing up soon. But I don't know for sure how long it takes for the information to be collected and reported. If any of my new readers knows the answer, please let me know.

UPDATE: Dwight Meredith writes in response to the above question:

While we do not know how long it will take for the “natural experiment” to help us know whether or not thimerosal used in infant vaccines is is related to autism, we suspect that it will be a while. Although thimerosal was removed from infant vaccines in 1999, there are a number of reasons why the question will not be immediately answered.
First, autistic behaviors often do not present themselves until 15-24 months of age. Accurate diagnosis can take additional time once autisic behaviors are present.
Secondly, although thimerosal was removed from infant vaccines in 1999, as we understand it, the vaccines were not recalled and we do not know how many kids were vaccinated from multiple dose vials containing thimerosal after 1999.
Third, the best studies of incidence, (like the UC-Davis study) deal with school system and government program data. It takes a while for changes in incidence to filter upwards to those programs and it takes a while to collect and report the data. The U.C.-Davis study looked at data from 1987 through 1998. The final report was issued this year.
The collection and analysis of the data is tricky. States that have good autism programs are likely to attract families with autistic kids. Thus, the studies have to control for migration. They also have to control for severity of the problems so as to avoid reporting an expansion of the spectrum as an increase in incidence. Those factors add to the lag time.
We do not know when the results of the “natural experiment” will be available. Anecdotal evidence is likely to surface soon. Definitive data may take a while longer.

Meredith also has some other observations on the nature of autism that are worth checking out.

My post below has drawn a lot of attention, and many good comments, all of which I appreciate. Most notably, Dwight Meredith has responded in both the comments section and in a post on his own blog. Mr. Meredith's posts are models of how to rigorously argue a viewpoint on an emotional topic without allowing (justifiable) emotion to usurp reason. I regret not having followed his blog earlier, and a place will be made for his site in my permalinks in their next reorganization (which will occur sometime prior to the second Bill Frist administration).
As a tribute to Mr. Meredith's skills, I will not allow my own reasoning to unjustifiably succumb to his. Specifically, I want to address a few of the points he made in his post.

Mr. Meredith argues that I am "quick to declare that no link between thimerosal and autism has been shown and that there is no such link." I don't agree, at least with the second part. I do not think that the current evidence shows a convincing link between thimerosal and autism, but that does not mean that no such link exists. It just means it hasn't been shown yet. I'm glad that several studies are currently being done to explore the issue further.
Mr. Meredith then reads my post as implying that "the entire issue is driven by greedy lawyers eager to bankrupt Eli Lilly." Not the entire issue, but part of it. More specifically, that is an excellent summary of why I think the recent amendment to the Homeland Security bill was substantively appropriate in its own right. I don't argue that the lawyers created the concerns regarding thimerosal, though I wouldn't be surprised if they try to fan the flames of such concerns even if convincing scientific evidence emerges against such a link.
Mr. Meredith also refers me to Wampum, another interesting read. She notes that, in apparent contradiction to the CDC's officially sanguine view of thimerosal, the Bush administration has agreed to release certain documents in connection with the issue, including internal CDC documents which purportedly state that "mercury in children's vaccines is a potential source of neurological damage in children including ADD/ADHD, speech and language delays and other neurological disorders including autism."
We should all look forward to the revelation of whatever evidence is found in those documents. We should note, though, that the CDC documents referred to in the release are apparently the ones obtained by Safe Minds, an autism activist group, via the Freedom of Information Act. The New York Times Magazine article about the thimerosal issue notes in an aside that those CDC documents were "far from conclusive evidence" of a link. (Again - it's not proof of no link; just not clear proof of a link (assuming the Times' report is accurate).)
Finally, Mr. Meredith concludes:
We have been fighting battles in the area of autism for some time now. So far, money and politics has always trumped science and kids. For once, we should forget about politics, blame, potential liabilities, campaign contributions and other factors. Put all that stuff aside. Forget which side wins and which side loses. Sides suck. We should simply provide the scientists with all available information and plenty of money so that they can save our children.
I couldn't agree more.
And I assume one of my correspondents agrees as well. He identifies himself as a "parent of an autistic child and an active fundraiser for biomedical research into autism." He e-mailed me to describe "how angry I was that this group of trial lawyers was so active at trying to suck resources away from the research and into their pockets in the name of a phony connection between thimerosal and autism. Frankly, I wish whoever added that provision to the Homeland Security bill would come forward so I could express my appreciation."
Unfortunately, resources are never infinite. We should certainly pursue whatever leads appear promising, but not be afraid to reallocate scarce resources if those leads fail to deliver on their promise.
P.S. A tax lawyer friend e-mailed me an interesting article about the amendment that appeared in Tax Notes. Unfortunately, the article is only available by subscription.
The author is openly hostile to the amendment on both procedural and substantive grounds. Nothing new there. But then he adds a kicker:

But a funny thing happened on the way to the forum. Our lawmakers forgot a lot of law. Without amending the Internal Revenue Code, it is not possible for the vaccine-injury amendments to tap revenues from the fund to serve their intended purposes. Internal Revenue Code section 9510 establishes and provides the rules for operation of the Vaccine Injury Compensation Trust Fund. Sections 4131 and 4132 describe the 75-cents-per-dose excise tax that feeds the fund.
Specifically, there appear to be at least three drafting problems. First, section 9510(c)(1)(A) requires that any money disbursed from the trust fund must be in accordance with the Public Health Service Act as in effect on October 18, 2000. Because the Homeland Security Act did not amend that section, no funds can be distributed for the thimerosal cases or any cases newly covered by the amendments made by the Homeland Security Act.
Second, under section 9510(b)(3) if any amounts in the trust fund are used for ineligible purposes, the Treasury Department must cease paying funds (from excise tax revenues) into the trust fund. That is a poison pill. If any funds were paid to thimerosal victims, transfers for all vaccine-injury cases would be cut off.
Third, the Homeland Security Act does not alter the definition of taxable vaccine in code section 4132(a). That does not affect the disbursement of funds under section 9510, but it does raise the question of linking trust fund costs with trust fund benefits. If the fund's purposes are expanded to include cases against manufacturers of vaccine components, should the trust fund tax base be expanded to include separate taxation of components? (If so, that could be problematic given that the current taxable unit for the excise tax is "per dose." How would a shipment of vaccine preservative be taxed?) Or could it be ! argued that components are already effectively taxed under the current law's levy on the final product?
If Congress gets around to making code changes to amend the VICP as intended by the Homeland Security Act, it will have to consider those issues.

Oops! As the author concludes: "[T]his is a great case study in how Washington works . . . and how it doesn't." (Ellipses in original.)

P.P.S. I hope to respond to Meredith's and Wampum's criticisms of the existing studies soon.