Tuesday, August 15, 2006

Vaccinations and autism? My analysis.

Alana sent me an e-mail today. At a meeting recently they discussed the potential link between autism and vaccines. I hadn't heard about this link, but apparently this link has been around for a few years.

Vaccines, up until 1999, regularly included thimerosal. This is a chemical that is similar in composition to ethyl mercury. You know, mercury, that really nasty heavy metal that causes neurological disorders. It was included in vaccines in trace amounts in the 1930s because it breaks down cell walls in bacteria. This was in response to some nasty outbreaks where bottles of multiple doses of vaccine were contaminated with bacteria, resulting in the illness — even death — of many children. The solution, of course, is to bottle the vaccine in single dose bottles. This was either a) too expensive, or b) ate into profits too much, depending on who you talk to.

In the 1990s the number of people diagnosed with autism increased significantly. Some have suggested that this is at least in part due to a change in diagnosis such that there are more, younger children diagnosed as autistic. On the flip side, others have suggested that there are environmental factors involved.

A theory emerged that thimerosal is responsible. The number of vaccinations given to children went up in the 1970s and 1980s. More vaccinations were given more quickly. At the same time the mumps, measles and rubella (MMR) vaccine was given more regularly, and this is essentially three vaccines in one. The theory points to the increased use of vaccines, particularly MMR, giving larger doses of thimerosal to children, resulting in increased numbers of children becoming autistic.

The fear of thimerosal resulted in the substance from being removed from vaccinations in the U.S. in 1999, though stockpiles were not destroyed so it was still in vaccinations as late as 2003. The only vaccines that regularly includes thimerosal are influenza vaccines. The proper, safe method of vaccinating people is by using single dose vials, which is just a good idea for several reasons.

The theory that thimerosal causes autism has grabbed a number of people, to the point where many people believe it is true. Here is a link to OpEdNews.com and an article that declares that the U.S. government has announced thimerosal is linked to autism. The article is written by Evelyn Pringle, who looks for corruption in government. The article is found at:

The government report is not a scientific report, but a paper released by the House Committee on Government Reform.

After I read the article, my BS detector went off. It's not written in neutral tones, it's obviously an opinion piece (and described as such). There were some things that didn't sit right. So, in an entirely scientific study (*chuckle*) I used my Google-fu to search the net for more information.

I found out from the CDC that the only pre-school vaccines with thimerosal are some influenza vaccines. According to the FDA, thimerosal free flu vaccines are available, but in limited quantities.

The CDC web page is here:

The FDA commissioned a study on thimerosal and autism. They determined that a link between thimerosal and autism was biologically plausible. However, their 2004 report suggested that the link was unlikely.

The FDA site, with the information about their report, is here:

The CDC information on the report is here:

Pringle's article implied that the CDC and the FDA were wrong because of a conflict of interest. Whether or not that was the case, surely I'd find some information on the studies they cited. Well, I did. One study is from Denmark. Here is the Denmark study's results, which strongly suggest there is _no_ link between autism and thimerosal.


I read on Wikipedia (stop snickering! It has some good information!) that there was a Japanese study done in the 1990s when they discontinued MMR vaccines. They found that the incidence of autism went up, not down, after stopping the vaccines. The Bandolier site also has this information:


Finally, here's a report from a Canadian study that found that once MMR with thimerosal was stopped in Canada the rate of autism did not decline (and actually increased):


Pringle's site allows people to leave comments. An engineer left a good comment with an opposing view to Pringle's story. All other comments posted shot down this engineer's post, suggesting that the belief in a vaccine/autism link has reached religious proportions with some people. One poster pointed out that the Canadian study was biased because the researcher involved in the study has also been a featured "expert" on the lack of a link between autism and thimerosal. The poster thought that this was a conflict of interest. Of course it could be that he did the research, believes there is no link, and now goes around talking about the lack of a link because he finds the talk about a link disturbing.

I mentioned above that I had some problems with Pringle's story. She says that thimerosal was found in four samples taken to a lab, proving that the big drug companies are using the stuff when they said they didn't. No mention was made of the actual vaccine tested, or of the date the vaccine was produced. We don't know if the tested vaccines were old, or if they were flu vaccines. This makes it extremely hard to do the kind of searching I was able to do from the FDA and CDC information.

I did discover comments about Doctor's Data, the lab cited in the op ed piece. The lab makes a living from finding heavy metals in samples. In other words, they aren't exactly neutral and the testing was not done as a double-blind test. Doctor's Data is also advertised as doing hair sample analysis. Several sites mentioned that hair sample analysis has been essentially discredited. This makes you wonder about the veracity of a lab that uses such tests.

This is information about hair samples from Quackwatch:

I wonder how the poster to Pringle's site feels about Doctor's Data's conflict of interest, given the poster's belief that the Canadian study was biased.

The test was commissioned by Health Advocacy in the Public Interest, a "watchdog" group whose web site says, "Health Advocacy in the Public Interest (HAPI) wishes to serve as a public forum, fund scientific research, and serve as an educational resource of scientific evidence that supports informed medical decisions for children." From another part of their site, "Sarnoff began working with Dawn Winkler, co-founder and Executive Director, in 2001 when Winkler's group, California Vaccine Awareness, was fighting legislation that would have mandated hepatitis A and Prevnar vaccines for entry to kindergarten. They were successful and decided to join forces and become HAPI."

It is quite clear from the web site that HAPI believes that vaccinations have increased the rate of autism. They want to fund a study of vaccinated children versus unvaccinated children to determine the health risks. This isn't a bad thing. More research is important. What is troubling is that this organization is preaching scientific research on one hand while denouncing the FDA's and the CDC's research on the other due to "bias". They seem to be equally biased on the opposite side of the argument.

There web site is found here:

There long term goals can be found here:

Interestingly, I found The Millenium Project, a skeptic site that is out to expose charlatans, frauds, etc. whether it be from pseudoscience, quackery, or multi-level marketing schemes. The author — Peter Bowditch, a computer consultant — got into an e-mail battle with Dawn Winkler of HAPI, and his exchanges are on the site. Apparently Winkler lost a child. She believes it was from a vaccine.

He... doesn't suffer fools gladly. The messages back and forth between him and Winkler are fascinating, even if he does bait her more than I thought he should. He's got a grievance against the anti-vaccination crowd (whom I didn't even realize existed!), and he's not afraid to fight them.

The exchange is here:

The Millenium Project's home page is here:

Doctor's Data is a questionable lab. I'd want to see the samples taken to a neutral lab, like at a university, before I'd trust them. Of course Pringle has to "prove" that drug companies still use thimerosal, otherwise if it was the main cause of autism you'd have seen a drop in autism since the use of thimerosal was stopped. There has been no drop. One of the problems with autism diagnosis is that the guidelines have changed over the years, including the age at which autism is detected. Here's a link with a study that showed the rate of autism leveled off in the 1990s:


They list in that article a rate of 19 per 10,000. That seems low compared to other estimates. I did find this article that said rates were 16.8 per 10,000 for classic autism and 62.6 per 10,000 for autism spectrum disorder in Britain:


What's interesting about those numbers is that they were published in 2002. This site, about autism in the U.S., has the rate at 1 in 175, or 57 per 10,000. That's close to 62.6 per 10,000, but this study was published this year, back in May. If autism is, indeed, spiraling upwards you'd expect a dramatic increase in four years. All the sites I found quote pretty much the same numbers for the past six to eight years in spite of many of them talking about the "drastic rise" in autism rates.


Someone in the chain of e-mails Alana sent said that there was talk of a study on Amish people in Pennsylvania. Apparently there is no autism among the Amish (though I'd have to see a citation on that before I believed it). I suspect that the Amish population study will prove inconclusive. The big problem is that the Amish are a small subculture who marry within their own group. They don't get a huge influx of people. Genetic issues could make autism less prone in the Amish regardless of whether drugs or the environment are being tested. This is why the Danish test is more statistically significant. It covers a large heterogeneous population. This long, technical paper suggests that there is a link between autism and genes:


It sounds like it's a good idea to stay away from thimerosal, seeing as how it's closely linked to ethyl mercury (even though methyl mercury, an entirely different molecule, is the really dangerous molecule) just to be on the safe side, particularly since single doses are available. I personally don't think there is a connection between vaccines and autism. There's no way you can prove that there is _no_ connection, since you logically can't prove a negative. You can only prove that there is a connection, and so far the evidence isn't supporting it.

In her e-mail, Alana mentioned the risk of not giving a child a vaccination and the serious illnesses that result. And, of course, she hit the nail on the head. Children have far more to fear from the measles or mumps than from autism. In countries where measles is not vaccinated against, for instance, the death rate due to measles complications is between 1% and 5%, with it going up as high as 25% when malnutrition is involved. Sixty per 10,000 is 0.6%, so measles alone could be 2 to 10 times as bad as autism. That's comparing the death rate of a single childhood disease to the rate of autism.



Michael Skeet said...

The vaccination-autism thing is fascinating, in a depressing sort of way. I think it demonstrates the human inability (or at least reluctance) to see beyond the immediate and short-term.

The reason so many people campaign against vaccination is precisely because vaccination has been so successful: they don't see, at close hand, the dangerous effects of the childhood diseases vaccination is intended to prevent. Autism, on the other hand, isn't (yet) preventable, so people see its effects in their own lives; even with relatively low rates of occurrance, I suspect that most of us know someone with an autistic child.

Where the anti-vaccination argument becomes pernicious is when those who oppose the practice become free-riders: they insist on being able to keep vaccines away from their own children, but depend on vaccination being sufficiently prevalent to suppress diseases that might otherwise harm their kids.

Since the only subject more likely to result in hostility than politics is other peoples' children, it's not necessarily productive to make this argument to parents who oppose vaccination...

Allan Goodall said...

This being the U.S., I suspect there's a religious aspect to this, too. It's comforting, in a strange sort of way, to have someone to blame. "God didn't punish my child (and me) with autism!" (And, since "God" created everything this couldn't be the result of random genetic mutation, or genetic inheritance.) "No, someone did this, and they will pay!"

Even if you're not religious, it's a human trait to want to blame someone. It's more satisfying to scream at a major drug company instead of just screaming at the universe.

One thing I didn't mention: Dawn Winkler, of HAPI, is running for Governor of Colorado as a member of the Libertarian Party.

You make a good point about the anti-vaccination argument. HAPI and other groups want studies done between vaccinated kids and non-vaccinated kids. Of course they intend to do this in the United States, where — as you point out — the non-vaccinated kids get a free ride. They aren't doing this study in the Third World. The WHO web site I mention was from the Third World. If you let all kids in the U.S. skip vaccinations you would get mortality rates of 1% to 5% for measles alone. It may be safe enough for a few kids to skip vaccinations. Extend that to the population at large and... well, there'd soon be a howling demand for vaccinations.