A few years ago a book said vaccines contained mercury and this caused many children to be autistic. But a recent editorial said that mercury is no longer used in vaccines, and besides it was safe all along anyway. But if it was safe all along, why'd they take it out? Why'd they put it in in the first place? --Darren Provine, via e-mail
I recently saw Robert Kennedy Jr. speaking about the link between autism and vaccines. What's the straight dope here? --Bart Zimmer, via e-mail
While expounding the other day on the lost antiseptic mercurochrome, I mentioned that vaccines once contained an antibacterial and antifungal agent called thimerosal. Keeping vaccines safe is a good thing--nobody wants a repeat of the 1928 Australian case where a dozen kids died from staph-infected diphtheria vaccine drawn from a multidose vial. Unfortunately thimerosal, like mercurochrome, has the drawback of containing mercury, a toxin known to cause neurological disorders. Children are especially vulnerable. In 1999 the American Academy of Pediatrics and the U.S. Public Health Service determined that standard childhood vaccinations could lead to a dangerous accumulation of mercury. They called for thimerosal's elimination from vaccines, and within a few years it was mostly gone.
Thimerosal hasn't totally disappeared. It continues to show up in some contact lens solutions, and as of last fall was still being used in certain vaccines for diseases including tetanus, meningitis, and flu--often ones used in multidose applications, where contamination presumably remains a concern. But according to the FDA, apart from the occasional flu shot, no vaccine routinely recommended for U.S. kids now contains more than a trace amount of the stuff.
Controversy over thimerosal persists, though. One concern is a suspected link, often circulated on parenting and fringe-science Web sites, between vaccines containing the compound and autism in children. Autism diagnosis has become more common lately, and nobody knows why--could just be improved diagnostic methods and such. Most major studies at any rate have failed to show a connection between autism and thimerosal. A 2000 vaccine safety study initially showed an association, but after errors in the analysis were corrected the link vanished. In 2002 a Danish study of more than 500,000 children over age seven found no link between measles, mumps, and rubella (MMR) vaccinations and autism. A 2003 follow-up study focusing on thimerosal-containing vaccines again showed no link, and also showed autism rates increasing despite removal of thimerosal from vaccines. A British study of 5,763 children in 2004 showed only a statistically insignificant link between MMR vaccine and autism; a study of 27,749 children in Montreal found no MMR link to autism or related problems. The World Health Organization reviewed four major studies and found no link between thimerosal from immunizations and neurological disorders in children.
So what's Bobby Jr. talking about? Most people probably connect him with thimerosal thanks to his 2005 report for Rolling Stone and Salon.com entitled "Deadly Immunity," wherein he tells a dark tale of secret conferences, whitewashed data, and ties to prominent Republicans. A focus of his article is a June 2000 conference discussing the thimerosal research of Tom Verstraeten of the Centers for Disease Control. Kennedy claims that though Verstraeten later tweaked his study till the link went away, at the gathering he pretty much blamed thimerosal "for a dramatic increase in autism and a host of other neurological disorders among children"--damning words if that's anything like what the man said. From what I can tell, though, it's not. According to the meeting transcript, Verstraeten actually said, "[W]e don't see much of a trend except for a slight, but not significant, increase for the highest exposure. The overall test for trend is statistically not significant." And the researchers in attendance agreed that there was "no evidence of causality" for autism. Some data pointed to a link with other neurological disorders; the attendees recommended further investigation.
So far, though, evidence for a connection between thimerosal and neuro-logical problems is unpersuasive. A 2001 California study appeared to show a clear link between cumulative thimerosal exposure and autism diagnosis, but a similar comparison using that Danish study data found the opposite. A 2006 study by father and son Mark and David Geier found an association between self-reported adverse effects and timing of thimerosal-containing injections, but it didn't correct for environmental, genetic, social, or other factors. WHO evaluated two U.S. studies alleging a decrease in neurological disorders after the switch to thimerosal-free vaccine but wasn't convinced.
Thimerosal/autism research is complicated by the difficulty of diagnosing autism, especially in very young children; varying amounts of thimerosal used in different countries; and failure to correct for the above-mentioned confounding factors. Since thimerosal appears to be going the way of the mercury fever thermometer, the question may soon be moot--except, of course, to aficionados of the class-action lawsuit.
Art accompanying story in printed newspaper (not available in this archive): illustration by Slug Signorino.