No, the autism prevalence in Denmark did not go down with the removal of thimerosal (again!)

17 Jul

There’s a story that goes around the vaccine/autism groups that, contrary to published reports, the autism rate in Denmark went down after thimerosal was removed from their vaccines. Of course, this is all part of a cover-up. I discussed this previously in No, the autism prevalence in Denmark did not go down with the removal of thimerosal. In that piece, I looked at a number of published studies and found that the autism rate in Denmark has continued to climb post thimerosal. Later, a study came out which in one place listed autism prevalence vs year for Denmark, and I discussed that as Autism, Denmark and again no link with vaccines.

Denmark removed thimerosal from pediatric vaccines in 1992. One of the graphs I created from the published data is here:

See how the prevalence of childhood autism (autistic disorder, dark purple bars) is higher for all birth years post 1992? ASD prevalence (lighter bars) is more complex–it peaks. ASD prevalence is higher for most years after 1992, but there is that peak. At the time I speculated that someone would rework these data and claim “see! there was a drop! Vaccines cause autism!”. It would be very dishonest and misleading, so I’m not surprised to see that the Age of Autism blog did exactly that.. Let me take a second to discuss how this is dishonest. First, the prevalence of childhood autism is always higher post 1992. This is what the Age of Autism and similar sites consider “real autism”, but they just ignore that. Even one of their own commenters asked about this and was ignored. Next, even with the apparent peak in ASD prevalence, the prevalence is higher for kids borh after 1992. If thimerosal was the primary cause of autism (or even a large cause), the prevalence should have dropped immediately after 1992 in both the childhood autism and ASD groups. It didn’t.

Why is there a peak in the ASD prevalence? Most likely years of follow up. Kids with, say, Asperger syndrome are diagnosed later. The data in the graph above were for kids diagnosed by 2010. So the kids in birth year 2004 were only 6 when the study was performed and fewer of them were diagnosed by the time the study was performed.

At age 6, many autistics remain undiagnosed. If you are skeptical that kids are diagnosed late, read on.

Another study has been recently published on autism prevalence in Denmark, The increasing prevalence of reported diagnoses of childhood psychiatric disorders: a descriptive multinational comparison. In this study they compared autism prevalence as well as ADD, Tourette syndrome and OCD in 3 different countries and noted that the trends were similar. The Simons Foundation SFARI blog has covered this in Autism not the only neurodevelopmental disorder on the rise.

Thimerosal was removed in Denmark in 1992. The autism prevalence increased after that. Same for Sweden. I’m not sure when Finland phased out thimerosal, but I suspect it’s about the same time. In which case, again, the rise in prevalence following the removal works against the thimerosal hypothesis.

This all said, what caught my eye was the supplemental material available with the paper. They show autism prevalence for various birth cohorts by age. For example, for kids born in 1990-1992, the authors give the autism prevalence at age 5, 8, 10, 15 and 20.

But, you say, it’s autism prevalence, how can it change with age? Autism is obvious and once the DSM IV was published, everyone has the exact same idea of who is autistic, right? Let’s use these data to see. Autism prevalence is almost always the prevalence of identified autism. Which is to say, it’s a count of who has a diagnosis already. If someone isn’t diagnosed, s/he isn’t counted in these sorts of prevalence reports. So, we can test the idea that (a) autism is obvious and (b) the understanding of what autism is doesn’t change.

So let’s look at data for Danes born in 1990-92. Autism prevalence vs age. If autism is obvious and the definition isn’t changing, this should be a level line. Or, let’s say, after an initial increase as kids are diagnosed, this will be a flat line. Right? Oh, you know with this much of a lead in, this isn’t going to be the case.


The line is not just a guide to the eye. It’s a fit. For some reason, these data are almost a straight line. Autism prevalence increased as the kids aged.

edit to add: in case you prefer bar graphs I’ve added one (for both, click to enlarge).


Consider just the last two points: age 15 and age 20. Autism prevalence went from 73.4 to 109.7 per 10,000. That’s like a 50% increase in 5 years! Is that an epidemic? No. Those autistics were always there. Just uncounted. And they were 15 years old. They were missed.

There are a few ways to look at the increase in autism prevalence in Denmark, the US and elsewhere:

1) The rates reported are accurate counts of how much of our population is autistic and, thus, represent a real increase.
2) There are social influences like shifts in our understanding of what is autism and how well we can diagnose autism and these are behind the increase.
3) Much of the increase is socially driven. It could be all. We can’t rule out that some is a real increase.

It’s not hard to see that (3) fits the data.

People pushing the idea of an autism epidemic don’t even accept that social factors and “real” increases are on the same footing. No, they promote the idea that social influences are a “denial” mechanism. But the fact is, there are data showing that social influences have had a huge impact on autism prevalence. The only data pointing to a “real” increase are those linking increased risk of older parents with increasing age of parents to the autism prevalence (estimated to account for 4-10% of the observed increase in California). This is both a “real” increase and a socially driven increase, by the way.

For what it’s worth, “real” isn’t in “scare quotes”. Some people use “real” to mean “a secular increase in autism prevalence. An actual increase in the fraction of autistics born.” Social factors are real, making real changes in autism prevalence.

Lastly, people will say, “autism is a crisis! If you don’t acknowledge the epidemic you aren’t taking this seriously!” Crisis: a difficult or dangerous situation that needs serious attention. Crisis does not equate to epidemic.

The fact that autistics are being missed *is* a crisis. It is something I take seriously and feel “needs serious attention”. Also, the idea that there could be an increase in autism prevalence is a very serious question getting very serious attention.

Are those promoting the “vaccine epidemic” idea actually treating autism as a crisis? No, they are not. Avoiding the question of undiagnosed autistics is denialism and is hurting our communities.

I know this article is pounding nails into a coffin that is already not only nailed shut, but arc-welded shut and then encased in concrete. Beyond that, the acceptance for the thimerosal causation idea–which was never a majority idea among parents– has waned, dramatically. New parents know their kids were not exposed to thimerosal. That, more than anything, killed the hypothesis within the autism parent community. Parents aren’t chelating their kids any more. The idea is dead. Sure there’s a hard core of believers still talking about it. And, more, they use this idea to scare others about vaccines.

Why talk about it then? For those few people who do get suck into the autism-as-mercury-poisoning world today. It’s still a very damaging notion, leading parents to lead lives of guilt and to become a ready market for the faux-therapies that have been built around the thimerosal idea. They need to know–the idea has no merit, whatsoever. The data are as clear as data can get. Thimerosal in vaccines doesn’t cause autism.

By Matt Carey

15 Responses to “No, the autism prevalence in Denmark did not go down with the removal of thimerosal (again!)”

  1. reissd July 17, 2014 at 07:18 #

    Are you sure parents aren’t chelating their children anymore? I’d expect the chelation sellers to try and sell chelation based on aluminum adjuvants. It’s not as if this is about scientific facts.

    • Sullivan (Matt Carey) July 17, 2014 at 15:38 #

      I recall a study that came out in the past few years that showed fewer than 1% of families are chelating their kids. I’ll see if I can find the link.

  2. Dr Mitzi Waltz July 17, 2014 at 15:25 #

    Another likely reason for the increase is that we are getting better about diagnosing autism in populations with comorbid conditions, including older children and adults. I’m old enough to remember when many clinicians thought autism and Down syndrome couldn’t co-exist–these days it is much more frequently diagnosed in people whose first presenting condition was Down syndrome, global developmental delay, intellectual disability, cerebral palsy etc.
    There’s also the issue that diagnostic instruments are typically written in Engish, and have to go through a bit of a process to be translated and, sometimes, tested for cross-cultural validity before they are widely adopted outside the English-speaking world.
    And speaking of culture, I believe I can dig up an actual citation for this if you want one, but some of the characteristics that might lead to an AS/HFA diagnosis elsewhere are not seen as terribly problematic in Scandinavia. So a rising diagnostic rate may be evidence of cultural change–not necessarily a positive thing if it means the boundaries of “normalcy” are shrinking because of new ways of looking at the same traits!

    • Sullivan (Matt Carey) July 17, 2014 at 15:40 #

      I recall a paper from about 1980 discussing how for a long time people wouldn’t diagnose autism in children with intellectual disability.

  3. BA July 17, 2014 at 17:52 #

    The bars when stacked provide a clearer picture.

    • Sullivan (Matt Carey) July 18, 2014 at 22:55 #

      I’m not sure I understand, but I added a bar graph in addition to the line graph. Is that what you meant?

  4. Science Mom July 19, 2014 at 16:27 #

    Sullivan, Barry is a spammer. In the meantime Barry share the data that demonstrates your product’s safety and efficacy.

    • Sullivan (Matt Carey) July 19, 2014 at 16:53 #

      I let on of “Barry’s” zeolite spam messages go through recently just because I was amused to see Zeolite try to re emerge. It was big for a short while on autism parent forums. No longer.

      I don’t think “Barry” will be back to answer, so I will (even though we both know the answer). There is no data that shows zeolite is safe and/or effective. Which reminds me, they were charging parents to let their kids be part of a zeolite clinical trial a few years ago. Besides the ethical questions of charging to be in a trial, I don’t recall the “study” being published.


  1. No, the autism prevalence in Denmark did not go down with the … | My Autism Site | All About Autism — My Autism Site | All About Autism - July 17, 2014

    […] Read more from the original source: No, the autism prevalence in Denmark did not go down with the … […]

  2. El por qué del 60% del incremento de la prevalencia del autismo en Dinamarca - Autismo Diario - November 7, 2014

    […] infundado. Vemos como por ejemplo en Dinamarca, quien retiró el tiomersal de las vacunas en 1992, siguió aumentando la prevalencia hasta prácticamente el año 2010, donde se […]

  3. Why Does Robert F. Kennedy Jr. Get Brain Science So Wrong? | USA North America News - July 28, 2015

    […] doesn’t read as a refutation of the Madsen study, and the data are elaborated here even more. Thorsen’s involvement in it doesn’t refute it, either. Thorsen, by the […]

  4. Review of Trace Amounts – bad science and conspiracies - September 9, 2015

    […] the removal of thimerosal from vaccines. But that’s not actually correct – as demonstrated by Matt Carey, rates of autism in Denmark did, in fact, go up after the removal of thimerosal from […]

  5. Who Is Poul Thorsen – VAXOPEDIA - October 15, 2017

    […] No, the autism prevalence in Denmark did not go down with the removal of thimerosal (again!) […]

  6. Has the CDC Been Hiding and Destroying Evidence About Vaccine Safety? – VAXOPEDIA - August 8, 2018

    […] No, the autism prevalence in Denmark did not go down with the removal of thimerosal (again!) […]

  7. No, the Recent Danish Study on MMR and Autism Doesn’t Show That Autism Rates Are Lower in Denmark – EpidemioLogical - March 11, 2019

    […] and RFK Jr. could afford some basic epidemiology courses, or to hire an epidemiologist. Then again, they keep getting things wrong… And what self-respecting epidemiologist worth their salt would ever want to be part of […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: