MMR Vaccine Safety and (Non-Existent) Autism Risk

By Dr Megan Yap - PaediatricianGeneral18 Mar 2019

Today, I thought I’d whip up something really quick around an article I posted a link to a couple of weeks back about the MMR vaccination and a study done in Denmark that strongly supports that it does not increase the risk for autism – even in children considered to be susceptible.

Now I have posted in the past other articles, and for those of you who have been following the blog for a while, you would know that I always encourage you to look closely and critically at articles that catch your eye because that “all that glitters is not gold.” What I mean by that is, that not all studies out there are legit.  The strength or weight behind the conclusions drawn from the data collected in the study is heavily dependent on a lot of different factors.  All through my university years, firstly in pharmacy and then in medicine, we were taught the art of “critical appraisal.” This is a process of carefully and systematically looking at different aspects of an article to determine if the “evidence” being presented is trustworthy or not. This process is not easy.  I remember when I first started learning how to do it, I found it really tricky.  There are lots of things you need to look at and question about the study methods, population, population size, how participants were recruited, how the intervention was administered, whether the patients were “blinded” or not, and that is all before you even GET to looking at the statistics (urghhhhhh… statistics – don’t even get me started on statistics!).

So why do we bother doing all of this?

  • Because you can’t believe everything you read!!!

This is especially true on the internet.  It may come as a surprise to you, but some of the “evidence” that people might write about, is not actually evidence at all!  They may well have just made it all up – and in fact, a lot of information out there is just exactly that… made up.

I little while ago I posted just on the FB page for the blog a quick evaluation on an American trial about how electronic device/screen time affects markers of childhood flourishing/development. Today I want to practice what I preach and take a quick look at that recent article quoted in the Brisbane Times about the MMR vaccination. Now I am not going to do a full-blown critical appraisal – that would take me far too long and it is already 9:30pm on Monday night and I have a Uni assignment to get back to.  But let’s just have a quick look and chat about this study and how strong the evidence is that is being spread like wildfire all over the world.

So most of you would be familiar with the story about how a scientific journal called the Lancet, published and then later retracted a paper that theorised a link between the measles, mumps, rubella (MMR) vaccination and autism. The guy who authored the paper was found later to have fabricated (ie made up) the data he published in that study to make it LOOK like there was a link, when actually there wasn’t.  He was caught, charged, deregistered etc, and the paper was retracted, but the damage had been done. Interestingly he was found to have failed to disclose financial interests (e.g., Wakefield had been funded by lawyers who had been engaged by parents in lawsuits against vaccine-producing companies).

Now if people only had known the art of CRITICAL EVALUATION (**Tan-tara!! Tan-tara!!**) they would have seen that the original article was only a case series (and this is weak evidence at best ie low level evidence) with a teeny-tiny sample size (12 patients) that we all know, does NOT accurately reflect a whole population.  But because the paper received widespread publicity and people did not know how to critically evaluate an article, the misinformation spread like wildfire.  In addition, the author and associates were also found guilty of ethical violations and scientific misconduct/misrepresentation – but this admission was NOT widely publicised (face palm).

Okay, so SINCE that dreadful article, a meta-analysis ( a study that examines data from a number of independent studies of the same subject, in order to establish overall trends) was performed in 2014 that looked at 10 studies that specifically looked at the the MMR vaccination and autism and all of them reported no association. Further studies were published in 2015 and those results were consistent with the meta analysis.

But despite all of this, the debate has raged on and people continued to be concerned about this “faux risk” and vaccine acceptance continues to be challenged.

Enter this new study Danish study published in the Annals of Internal Medicine on 5/3/2019. Try this link for the full article:

Let’s talk about it.

For starters, the study sample size is HUGE:  657,461 children born in Denmark in 1999- Dec 2010; 6517 cases – so this is the largest single study to date and gives us really strong evidence that increases in autism risk are highly unlikely.  Interestingly these same authors previously addressed the same issue in another study of 537,303 children and found that in that cohort, MMR vaccination was NOT associated with autistic disorder.

WHY then this study if they had already answered the question?

Because these study authors wanted even more powerful evidence in a non-overlapping cohort of Danish children, with a larger sample size and longer follow up.  They also wanted to address a criticism of their previous paper (and other observational studies) regarding whether or not the MMR vaccination triggered autism in children who may be more susceptible to the condition than other children in the general population (eg those born prematurely, low birthweight, smoking during pregnancy, family history autism + other factors) – and they did this.

The data on MMR vaccination was collected from the Danish Civil Registration System and Danish National Health Service Register. Information on ASD diagnoses was obtained from the Danish Psychiatric Central Register.  This allowed the investigators to look at risk in subgroups of children – like children with a sibling with autism.  The data was stratified and adjusted for confounding variables (obtained through the Danish Medical Birth Registry).  Patient specific information was also used from the Danish National Patient Register and the Danish Medical Birth Registry were also used.  This allowed the authors to also observe no increase in risk of autism in other subgroups of kids including those who received other childhood vaccinations, or during a certain time period after getting a vaccination and to exclude children with syndromes or conditions that increased their risk for autism (if diagnosed before their 1st birthday).

The statistical analysis was involved… I am not going to go into this in detail.

I note that the study was funded by a grant the Novo Nordisk Foundation and the Danish Ministry of Health.  The funder had no role in the study design, data collection, data analysis/interpretation or report writing.  Novo Nordisk is a pharmaceutical company, but does not directly manufacture the MMR or ProQuad vaccination (manufactured by Merck). I tried, but couldn’t find out if Novo Nordisk is a shareholder for Merck.

The loss to follow up was very small (thus acceptable; 0.65%).

The main results between the 2 large studies were similar – which supports their validity. The data strongly supports that MMR vaccination does NOT increase the risk of autism in children. The large number of cases in this study allowed the authors to define subgroups according to environmental and familial (with a positive family history) risk factors for autism – and there was no support for the hypothesis of MMR vaccination triggering autism in susceptible subgroups of children.

A limitation of the study was noted by the study authors themselves and this was that no individual charts were reviewed.  Now, reviewing over 657,000 charts would have delayed the publication date of this study significantly, but it also means that the date of autism diagnosis was used rather than the date of reported onset of symptoms.  This means that when the symptoms of autism were there before a child was vaccinated (and diagnosis happens after vaccination) then this biases the study towards effect – that is, it makes it look like more cases of autism are in vaccinated children. If onset of autism symptoms resulted in avoidance of vaccination, or if symptoms made it more likely a child would be vaccinated, then bias in either direction would be possible. With registry data however, not much could have been done to avoid this.

Anyways, yet again I’ve ended up with a 3 page article, when I meant to only write a short one…

In summary, this article checks out.  The data, the analysis, the sample size, the reproducibility of the result – the results and conclusions are trustworthy.  They are not the HIGHEST level of evidence (ie multiple randomised, double-blinded controlled trials are the best kind of evidence, and a meta analysis of multiple studies of this type would be the bomb), but still, the evidence is good and the strength of this study is undoubtedly in its size.

MMR vaccination does NOT cause an increased risk of autism in children – even in children deemed to be “susceptible.”


If you’ve read this far, I hope you’ve found this article helpful!!!

Till next time, stay well and vaccinate your kids!


xxDr Megs

For more articles from Dr Megan Yap visit her blog – “Dr Megs – Paeds & Feeds” at


We have a number of locations around Brisbane to help make it easier for you to find us!