Two weeks ago over at Mary’s Skepchick Quickies post, one particular comment caught my eye. Pcisek wrote the following:

This has nothing to do with today’s news stories, but I know of no other forum where I can ask this question. My girlfriend has been hearing the claim that “44 women have died from the HPV vaccine”, and is looking for debunking info. Has anyone else encountered this claim, and has anyone debunked it online yet?

As someone who likes to torture myself by regularly reading anti-vaccine blogs, I’ve seen this “44 deaths from the HPV vaccine” stat and others like it mentioned many times. I was pretty sure I knew where the source of this number came from, but did a quick Google search to make sure. One of the first results was the Wikipedia page for the HPV Vaccine which says there were 44 reported deaths following the HPV vaccine as of Sept 1, 2009 and helpfully confirms my suspicions as to the source of that data: the CDC’s Vaccine Adverse Effect Reporting System known as VAERS. In fact, almost any time you hear any stat about the number of people harmed by vaccines, it comes from VAERS, but most people, especially those with an anti-vaccine agenda, misinterpret the numbers that come from this database.

VAERS was created in 1986 as part of the National Childhood Vaccine Injury Act (NCVIA), one of the most important pieces of health legislation ever passed by the U.S. Congress. It saved the vaccine industry in the U.S., created a system that compensates victims who get a legitimate injury from a vaccination, and created VAERS to monitor adverse effects that may be caused by vaccines.

NCVIA created a fund called the Vaccine Injury Compensation Program (VICP) to pay for non-negligent vaccine injuries, paid for by a small tax on each vaccine. Mass immunization for children has been mandated by public schools in the U.S. since the late 70s so it seems fair that the government should compensate victims who have adverse reactions to the vaccines.

VICP has a list of common injuries from vaccines. If your child has an adverse response after receiving a vaccine and this type of injury is listed in the Vaccine Injury Table, then you receive compensation. The family of the victim does not need to prove causation, which can be extremely difficult, only that injury followed the administration of the vaccine.

If the injury is not listed on the Vaccine Injury Table, the case will go to a Vaccine Court, which is presided over by attorneys that specialize in their knowledge of vaccines and vaccine injuries. The Vaccine Court will even pay for the family’s legal representation whether or not they win their case. Unlike a normal US court, which in malpractice cases must prove causation “beyond a reasonable doubt,” (see at the bottom for correction) the vaccine court merely must prove that the “preponderance of evidence” points to causation. In other words, you only need to show an over 50% chance that causation was involved. This is a lower standard than in a typical courtroom, a fact that is widely misunderstood in media reports on Vaccine Court settlements. When the Vaccine Court rules in favor of a family, they are not saying that the child’s injury or illness was caused by the vaccination, only that it plausibly could have been caused by the vaccination and likely was, but not necessarily. This is a fair standard that errs on the side of compensating victims. Keep this in mind whenever you read articles about cases where families were awarded compensation, such as in the case of Hannah Poling. The media will often claim things such as “Court says Vaccines Cause Limbs to Rot and Fall Off,” but in fact it just means that the court thought that it’s plausible that vaccines could cause limbs rotting and it’s likely that the victim got her rotting limb syndrome from a vaccine, so they’re going to assume this is the case and compensate her.

In addition to the Vaccine Court, NCVIA established the VAERS database, which is run by the CDC. This system stores all reported adverse events following the administration of a vaccine in a public database. Doctors and hospitals or even the patient and his/her family can send in a report of any possible side effect following the administration of a vaccine. This data is all public and you can download it yourself on the VAERS website.

Researchers can use this rich data source to look for diseases that seem to correlate with certain vaccines, making vaccines one of the most well-studied pharmaceuticals. Unfortunately, the public data means that non-researchers can use it to create their own “facts” about the dangers of vaccines. Remember, VAERS contains any adverse reaction following a vaccination. That does not mean that the vaccine caused the adverse reaction.

If you are looking for a disease caused by a certain vaccine, you would compare its prevalence in VAERS to its prevalence in the general public or to people who received other types of vaccinations. If it is sufficiently higher in the recently vaccinated group, it suggests that it might be causally linked to the vaccine and you might want to research it in a more serious manner. In this way, the VAERS database helps researchers find possible bad side effects from vaccines, even those that they weren’t specifically looking for or that present only in a small number of individuals. They can then use this data to make vaccines even safer.

However, most of the time when you see data from VAERS mentioned on the Internet, it is presented on its own as a causal vaccine injury. When you hear stats that people get from VAERS like “44 women died from the HPV vaccine,” in actuality it means that 44 women died soon after receiving the vaccine. It’s highly likely that most if not all of these cases are mere coincidences. Statistically, people die all the time and every once in awhile someone will die after receiving the HPV vaccine even from things that have nothing whatsoever to do with the vaccine. Some of these cases may be legitimate negative reactions to a vaccine that caused serious illness or death. However, from the VAERS data alone, it’s impossible to tell which cases were caused by the vaccine and which were coincidental.

Almost any time you hear an anti-vaccine advocate mention stats like “OMGZ YOU GUYS, DID YOU KNOW THAT 1000 PEOPLE WHO GOT THE HPV VACCINE GREW WHISKERS!!1,” remember that it’s highly likely their data is coming straight from VAERS and therefore is just reporting on one event following another without any claims to causality.

Just because data from the Vaccine Court and VAERS is often misused by antivaxxers doesn’t mean they are bad laws. In fact, they both help to keep vaccines safe and affordable and compensate families of children who have bad reactions to their vaccination. It is interesting that the anti-vax advocates seem to believe that the CDC is trying to cover-up bad side effects from vaccines and yet they use public data from the CDC itself in order to prove all these bad things they believe stem from vaccinations. They claim over and over again that the government has not studied vaccines and vaccine injuries at all, when in fact the CDC has a whole infrastructure set-up to monitor adverse effects from vaccines and compensate families of legitimate vaccine injuries. In fact, the National Childhood Vaccine Injury Act (NCVIA) was actually lobbied for by anti-vaccine organizations in the 1980s. We wouldn’t even have this law if it weren’t for them.

Next time I’ll write about the history of how the NCVIA came to be and how anti-vaccine lobbyists, Reagan and a mythical creature called bipartisanship saved the vaccine industry in the US.

Correction: Laurel pointed out in the comments that malpractice cases are tried in a civil court and require only a “preponderance of evidence.” Basic googling shows she is right. Sorry everyone. I am not a lawyer. I took a law class once and got my only “C” in grad school.  What I do know is that up to 2011, 30% of all non-autism related cases in the Vaccine Court awarded compensation, totally $2 billion (Source). By comparison, estimates of how many malpractice cases are typically won are in the range of 15-25% and that’s only for the cases that go to trial. Many don’t even make it that far. (Paul Offit’s book The Cutter Incident claimed 15%, but I don’t own the book and can’t remember what page it is on, so here’s a different source claiming 20%). My original statement was definitely wrong, but I think it is fair to say that patients stand a better chance at getting compensation for their claim in the Vaccine Court rather than in a Civil Court.

Featured Photo by Jamie Bernstein from Flickr

Jamie Bernstein

Jamie Bernstein

Jamie is a data, stats, policy and economics nerd who sometimes pretends she is a photographer. You can usually find her at skeptic events in Chicago or on Twitter or Flickr. She also blogs about music at Notes From Chicago Music Underground.

Previous post

Skepchick Quickies 9.4

Next post

Skepchick Quickies 9.5

18 Comments

  1. Profile photo of moonglaive
    September 5, 2013 at 11:40 am —

    So, we seem to have a problem in our society where people consider courtrooms to be arbiters of the absolute truth. Of course, that’s only when the courts find in favor of what a person believes, otherwise they’ll acknowledge that courtrooms are not peer-reviewed situations blah, blah, blah….

    • Profile photo of criticaldragon1177
      September 5, 2013 at 12:06 pm —

      Moonglaive,

      It also sounds like people find one or two things that seem to back up their belief system, and than ignore everything else, which is something I always believed.

  2. Profile photo of criticaldragon1177
    September 5, 2013 at 12:09 pm —

    Jamie,

    So basically antivaxxers are blaming deaths on vaccines, in-spite of the fact that it many cases those deaths probably have nothing to do with the the vaccines. Ironically, if we were to eliminate vaccines, like they want us to do, it would lead to far more deaths.

    • Profile photo of Jamie Bernstein
      September 5, 2013 at 12:37 pm —

      For many of the top anti-vax writers, they very much understand VAERS and are purposefully misinterpreting the data in order to mislead. In many other cases though, I don’t think it’s on purpose but merely a misunderstanding of what VAERS is and what it means for something to be reported in VAERS. In general when the media reports on it, usually it’s pretty clear that journalists just don’t understand VAERS. Because it comes from a government agency, it’s assumed the data is good quality, but the purpose of VAERS is to have quantity over quality in order to identify areas that might be of concern, not to make actual claims about what types of injuries a person could get from a vaccine. It’s understandable that the general public would misunderstand stats that come from VAERS, but it would be nice if journalists and writers would do some research before reporting on information they get from the database.

  3. Profile photo of Infophile
    September 5, 2013 at 12:16 pm —

    I’ve heard that another problem with VAERS is that there’s little check to ensure that the submission is at all truthful or the slightest bit plausible. I recall a skeptic testing this by posting a claim that a vaccine turned him into the Incredible Hulk. He got a call about it, but they told him that unless he agreed to retract his claim (he did), they’d have to leave it up. This makes it very easy for anti-vaccinationists to game the system if they so wish. Most of them are sincere in their beliefs though, so it’s not too likely. This does tell you how reliable this database should be for making final conclusions, however.

    • Profile photo of Jamie Bernstein
      September 5, 2013 at 12:43 pm —

      Oh, that Incredible Hulk thing sounds really interesting. Let me know if you find a source for it. I’ve heard about people reporting things like broken bones from car crashes that happened post-vaccination, but I couldn’t find a source on it in order to mention it in my post so I just left it out.

      Remember that the anti-vax crowd lobbied for this law specifically so that they could make reports on all sorts of things that they believe are caused by vaccinations. They thought that it would help prove their claims, but instead it did the opposite. VAERS definitely has data quality issues since anyone can send in just about anything. It relies on quantity over quality, which is why its data should never be used to make claims about vaccine side effects, only to look for possible areas to study further. Unfortunately this memo never seemed to of made it to the general public, media, or the anti-vaxxers.

      • Profile photo of medivh
        September 5, 2013 at 1:33 pm —

        James R. Laidler, at the neurodiversity blog, no longer online but archived at http://web.archive.org/web/20100725175227/http://neurodiversity.com/weblog/article/14 said:

        The chief problem with the VAERS data is that reports can be entered by anyone and are not routinely verified. To demonstrate this, a few years ago I entered a report that an influenza vaccine had turned me into The Hulk. The report was accepted and entered into the database.

        Because the reported adverse event was so… unusual, a representative of VAERS contacted me. After a discussion of the VAERS database and its limitations, they asked for my permission to delete the record, which I granted. If I had not agreed, the record would be there still, showing that any claim can become part of the database, no matter how outrageous or improbable.

        • Profile photo of Jamie Bernstein
          September 5, 2013 at 1:50 pm —

          Wow. That is definitely the craziest thing I’ve ever heard being in VAERS. I almost wish he’d left it in so we could use it to refute anti-vaxxers whenever they misuse VAERS data.

          • Profile photo of swivels
            September 7, 2013 at 2:35 am

            Apparently some anti-vaxxers didn’t believe the Hulk story because he had it removed from the database, so a UK-based blogger reported that a vaccine turned his daughter into Wonder Woman: http://leftbrainrightbrain.co.uk/2006/03/14/on-using-vaers/. As far as I know, that one’s still in there.

      • Profile photo of Buzz Parsec
        September 5, 2013 at 2:01 pm —

        VAERS isn’t data; it’s a collection of anecdotes. I just listened to the SGU 5×5 podcast about anecdotes yesterday. Rebecca points out that while anecdotes aren’t data, they can be useful for generating testable hypotheses.

        • Profile photo of Jamie Bernstein
          September 5, 2013 at 5:21 pm —

          I would say that in this case it still counts as a type of data, just a low-quality one. I love the way you called it more of a collection of anecdotes though. It pretty much is just a really large collection of anecdotes. Each one is kind of useless on its own, but the entire database is useful as a whole for at least pointing researchers in the right direction in terms of what they should be looking for.

  4. Profile photo of katiehov
    September 5, 2013 at 1:36 pm —

    Great post! And I bet some people want whiskers.

    • Profile photo of Buzz Parsec
      September 5, 2013 at 2:03 pm —

      I want whiskers! See my highly realistic avatar, to the left. (As a registered Evil Twin, I’m required by the rules of the Guild of Evil Twins to have facial hair.)

  5. Profile photo of Jack99
    September 5, 2013 at 5:44 pm —

    Great post, Jamie. This got me thinking, there must be a concept analogous to background radiation here, yet I’ve never seen it stated explicitly. In other words, even if you were injecting water, or even doing nothing at all, there would still be a certain death rate involved (x deaths per million population per annum).. I am betting that that figure would be around the 1 in a million procedures mark. Perhaps others have some information on this?

  6. Profile photo of pieterb
    September 5, 2013 at 6:52 pm —

    In order to even begin to look at VAERS data, you must acknowledge that you have read and understand a page that begins with this

    When evaluating data from VAERS, it is important to note that for any reported event, no cause-and-effect relationship has been established. Reports of all possible associations between vaccines and adverse events (possible side effects) are filed in VAERS. Therefore, VAERS collects data on any adverse event following vaccination, be it coincidental or truly caused by a vaccine. The report of an adverse event to VAERS is not documentation that a vaccine caused the event.

    and ends with this

    A report to VAERS generally does not prove that the identified vaccine(s) caused the adverse event described.  It only confirms that the reported event occurred sometime after vaccine was given. No proof that the event was caused by the vaccine is required in order for VAERS to accept the report. VAERS accepts all reports without judging whether the event was caused by the vaccine.

  7. Profile photo of pascale68
    September 6, 2013 at 12:18 am —

    Thank you for this. Although it sickens me that this is even necessary.

  8. Profile photo of Laurel Halbany
    September 7, 2013 at 12:18 am —

    “Unlike a normal US court, which in malpractice cases must prove causation “beyond a reasonable doubt,” the vaccine court merely must prove that the “preponderance of evidence” points to causation” – this is completely incorrect.In a CRIMINAL court, where a person is being accused of a crime by the government, the standard is that they must be proven guilty of each element of the crime beyond a reasonable doubt. In CIVIL courts, the standard of proof the plaintiff must show is “preponderance of the evidence”. And yes, malpractice cases are civil cases.

    I appreciate the article and pointing out the lies of anti-vaxxers, but when you get something this basic so badly wrong it casts a great deal of doubt on the article as a whole, which is unfortunate.

    • Profile photo of Jamie Bernstein
      September 10, 2013 at 7:56 pm —

      Thanks Laurel. You are indeed correct. I’m not a lawyer and made some assumptions that were wrong because I clearly don’t understand law. I took a law class once and my professor told me my paper was a policy paper and not a law paper. That was probably a good thing since I was working on a master of public policy and not a law degree at the time. Anyways I updated the post and put a correction at the end. Thanks for pointing this out.

Add Comment Register



Leave a reply