Previous: Myths About Antibiotics: Healthcare Triage #11
Next: Trans Fats, Sugary Soda, and Effective Regulation



View count:648,246
Last sync:2023-10-28 13:00
Thanks to for supporting this episode of Healthcare Triage. You can get Nate Silver's The Signal and the Noise, or a book of your choice, free at

There is almost no topic in health and health policy that immediately polarizes people more than the idea that vaccines cause autism. Even though the original big paper on this topic came out at the end of the last century, the anger this causes is still raw and potent. But there is a very, very large amount of research showing that vaccines and autism are unrelated. Please, watch the video before you jump down my throat.

Make sure you subscribe above so you don't miss any upcoming episodes!

Here are references for all the studies I talk about:

John Green -- Executive Producer
Stan Muller -- Director, Producer
Aaron Carroll -- Writer
Mark Olsen - Graphics

 1998: The Hypothesis

In 1998 an article was published in the Lancet that followed the cases of twelve children with developmental regression and gastrointestinal symptoms, such as diarrhea or stomach pain. Nine of these children had autism, and eight of the nine had parents who thought the symptoms of their autism had developed after the vaccine for measles, mumps, and rubella (or the MMR vaccine) was administered. 

This was not a randomized controlled trial, nor even a scientific study, it was merely a description of a small group of children. To be honest, it's difficult to imagine that this study could get published in the Lancet today. But based on the beliefs of the parents of those eight children, a frenzy of fear about vaccines and autism has ensued for more than a decade. That's the topic of this week's Healthcare Triage.

[Intro sequence]

Let me start with some caveats: I'm a pediatrician and a health services researcher. I see kids with autism, I treat kids with autism, I've even been part of funded NIH research to improve the ways we screen for and diagnose autism in children.  It's a real condition that is increasing in prevalence and has a significant impact on children and families across the world. I do not deny in any way that we need to do more about autism, but it has nothing to do with vaccines. 

And every dollar that we waste on that topic is a dollar we can't spend on important research or treatment to help children with autism and their families.

The initial article I mentioned in the Lancet was not a study. It had no real statistics and proved no association or causation. But it caused so much concern that led to a whole bunch of real studies to combat it. 

 1999-2004: No Supporting Evidence Found (1:31)

Just one year later in 1999, a study was published in the same journal. No difference was seen in the age of diagnosis for those who did and did not receive vaccines. That meant either there was no association between MMR and autism or that it was too weak an association to be detected even in a larger sample of children.

Some were unimpressed. They still were concerned. In 2001, a study was published in the Journal of the American Medical Association that looked at data on over ten thousand kindergartners born in California from 1980 through 1994.

The incidence of autism over that time increased from 44 per 100,000 births to 208 per 100,000 births. That's a 373% increase. MMR coverage, on the other hand, rose from 72% to 82%, a relative rise of only 14%. It was determined in this study that the relatively small increase in MMR could not possibly be responsible for the huge increase seen in autism.

The next year in 2002, a study was published in the New England Journal of Medicine that followed all children born in Denmark from 1991 through 1998. They obtained data on over 537,000 children born in those years. They could find no association between the development of autism and the age of vaccination, the time since vaccination or even the date of vaccination. That same year, a similar study was published in the journal Pediatrics. It followed over 535,000 children in Finland born between 1982 and 1986. It too could find no association between the MMR vaccine and autism. In 2004 the Lancet published another study, which matched 1294 children with autism against 4469 kids without it. They found no relationship between the MMR vaccine and autism.

 2004: Retraction (3:16)

In 2004, ten of the twelve authors of that first Lancet paper retracted the supposition that the MMR vaccine could cause autism. This kind of thing is unbelievably rare in the medical literature. An eleventh author could not be contacted before the retraction. Only one researcher --the main one, in fact --held firm. For the record, that researcher is no longer licensed to practice medicine in the United Kingdom.

In 2005 a systematic review or 'study of studies' examining the effectiveness and unintended effects of the MMR vaccine was published in the Cochrane Collaboration. They identified 31 studies which met the criteria for their review. After a thorough investigation, even though the MMR could be associated with a number of side effects or other issues, there was no evidence for an association between the vaccine and autism.

And in 2012, they  updated their work. This time the research they found included 5 randomized controlled trials, 1 controlled trial, 27 cohort studies, 17 case control studies, 5 time series trials, 1 case cross-over trial, 2 ecological studies, and 6 self controlled case series studies. All of these together involved about 14,700,000 children. And in all that data, they could find no link between vaccines and autism - because there is no link!

And another study, no matter how many times you ask for one, isn't going to overcome this massive amount of data.

Humans try to make sense of the world by seeing patterns. When they see a disease or condition that tends to appear around the time a child is a year or so old --as autism does --and that is also the age that kids get particular shots, they want to put those things together. Parents watch their kids more carefully after they get shots; sometimes they pick up on symptoms then. But just because two things happen at the same time doesn't mean that one caused the other. This is why we need careful scientific studies. And as I've outlined here, there have been many, many such studies that have failed to find any real evidence to support the hypothesis that vaccines cause autism.

Don't tell me it's the mercury in thimerosal that's to blame. There has been no thimerosal in infant vaccines since 2003, and autism hasn't disappeared in the last decade.

This was all because of one paper, a decade and a half ago, that described the beliefs of the parents of eight children with autism. And that's what makes the next part in this all the more tragic.

 The Tragic Results (5:32)

In 2011, the British Medical Journal released an article which described, in detail, how that 1998 Lancet paper wasn't just junk science, it was a lie. It described how the main author --the only who still supports its findings --changed the records, changed the stories and changed the numbers to create the appearance of an association where none existed. The journalist who wrote the BMJ article tracked down the patients in the study, and showed how none of their stories or information matched up to what was published in the final paper. He found that there were discrepancies as to whether the children actually had regressive autism. He found that there were falsifications and other symptoms that were alleged to cause the autism. He even discovered the dates had been changed. 

Although the Lancet paper alleged that 8 of the 12 patients reported symptoms days after the MMR vaccine, the BMJ investigation confirmed that for almost all of these children, that wasn't the case. And if that wasn't bad enough, it turned out that all of the patients had been recruited by anti-MMR campaigners. The study was also commissioned and paid for by a group that planned litigation against the vaccine manufacturers.

 'A Fraud' (6:38)

The British Medical Journal called the original Lancet paper, 'A fraud'. 

It's easy to become cynical about people's loss of trust and understanding in science, to the point that you think it's unlikely that we will ever be able to convince some people that the MMR vaccine is safe. That's a tragedy in and of itself. It's easy to believe that the perpetrator of this fraud will not suffer the repercussions he deserves. Many still continue to lionize him and believe him to be a victim of some powerful cabal.

It's hard for me to be dispassionate about those who abuse the trust people give physicians. I get even more riled up when someone violates the rules of ethical science. I think it's likely that children have not been given an MMR vaccine because of this fraud. I think it's likely children have gotten sick because of this fraud. I think it's likely some children have died.

I hope we can find some way to change that in the future.

 Closing and Sponsor (7:20)

This episode of Healthcare Triage is supported by, a leading provider of premium digital spoken audio information and entertainment on the internet. allows its users to choose the audio versions of their favorite books with a library of over 150,000 titles. We recommend The Signal and the Noise by Nate Silver, which explains the analysis of data and statistics in an entertaining and informative manner.  It's great; you should go listen to it now.  You can download an audio version of The Signal and the Noise or another of your choice at