21

Got this today on my facebook wall...

Alarming New Study: Unvaccinated Kids are Healthier

The study’s findings really make you wonder. Here’s some of the specifics:

  • Vaccinated children were more than three times as likely to be diagnosed on the Autism Spectrum (OR 4.3)
  • Vaccinated children were 30-fold more likely to be diagnosed with allergic rhinitis (hay fever) than non-vaccinated children
  • Vaccinated children were 22-fold more likely to require an allergy medication than unvaccinated children
  • Vaccinated children had more than quadruple the risk of being diagnosed with a learning disability than unvaccinated children (OR 5.2)
  • Vaccinated children were 300 percent more likely to be diagnosed with Attention Deficit Hyperactivity Disorder than unvaccinated children (OR 4.3)
  • Vaccinated children were 340 percent (OR 4.4) more likely to have been diagnosed with pneumonia than unvaccinated children
  • Vaccinated children were 300 percent more likely to be diagnosed with an ear infection than unvaccinated children (OR 4.0)
  • Vaccinated children were 700 percent more likely to have surgery to insert ear drainage tubes than unvaccinated children (OR 8.01)
  • Vaccinated children were 2.5-fold more likely to be diagnosed with any chronic illness than unvaccinated children

Are unvaccinated Kids are Healthier?

inund8
  • 591
  • 4
  • 20
Matas Vaitkevicius
  • 1,062
  • 2
  • 10
  • 21
  • 9
    Do you happen to know what they mean by "vaccinated" and "unvaccinated"? Because there's a lot of different vaccines and several different suggested regiments. –  May 17 '17 at 02:47
  • 3
    If you go to the article it does make the claims you cite. But the peer reviewer journal article it cites is **only measuring NDD**. The remaining claims are simply made up by Homestead Guru. – DJClayworth May 17 '17 at 03:00
  • 18
    Note that the "study" in question attracted criticism for major design issues (see points raised [here](http://scienceblogs.com/insolence/2016/11/29/antivaccinationists-promote-a-bogus-internet-survey-hilarity-ensues-as-its-retracted/) for example), and was subsequently [retracted](http://retractionwatch.com/2016/11/28/study-linking-vaccines-autism-pulled-frontiers-following-heavy-criticism/) by the journal. And then [retracted again](http://retractionwatch.com/2017/05/08/retracted-vaccine-autism-study-republished/) by another journal. – ff524 May 17 '17 at 04:40
  • 4
    Given the retractions (see comments by @ff524), is this claim even noteable? – DevSolar May 17 '17 at 07:26
  • 2
    @DevSolar considering the blog post provides a copy of the paper in an update even though the original is no longer available, I'd say the retraction doesn't make the claim non-notable. – Andrew Grimm May 17 '17 at 09:47
  • @DJClayworth: Would you like to turn that into an answer that can be voted upon? – Oddthinking May 17 '17 at 13:05
  • @DevSolar: Yes, this appeared on my Facebook wall. People believe it. – Oddthinking May 17 '17 at 13:05
  • 8
    This claim only says that parents that vaccinated their children pay more attention to their kids. – T. Sar May 17 '17 at 13:06
  • 4
    I would go with a lessor "parents that don't vaccinate aren't having doctors help their kids" which is pretty close to a tautology, and seems to be the conclusion Oddthinking found. –  May 17 '17 at 15:26
  • @DevSolar Just because the original source has been retracted doesn't prevent *notability* if people are still repeating the claim. But it does make answering the claim a lot easier (and possible of greater value to the readership). – matt_black May 17 '17 at 16:08
  • I'm not sure simply saying it was retracted fully gets the point across. This was not a scientific study. it was a like-and-share social media survey dressed up as science. If the website Vacines-Are-Evil.net ran a survey on their website for a few months and asked visitors to their website, mostly hardcore-antivaxers, to share the survey with their facebook friends(also mostly hardcore antivaxers) and to fill out the survey asking about the health of their vaccinated and unvaccinated children what would you expect the results to show? – Murphy May 22 '17 at 11:52
  • This article would be far more interesting as the subject of a study of the spread of rumours through social media than the subject it's trying to actually cover! – komodosp May 25 '17 at 08:13
  • Why the using different terms to express the (supposed) numbers - for example, "quadruple" and "300 percent more" are the same thing. – sirjonsnow Jun 12 '17 at 19:59
  • It seems that a [modified 2017 version of the study](http://www.cmsri.org/wp-content/uploads/2017/05/MawsonStudyHealthOutcomes5.8.2017.pdf) discussed here is doing the rounds again, being mentioned on Facebook, pointing to websites like [this](http://www.cmsri.org/wp-content/uploads/2017/05/MawsonStudyHealthOutcomes5.8.2017.pdf), which are also/still misrepresenting the findings. –  Aug 05 '17 at 19:53
  • [Edzard Ernst weighs in.](http://edzardernst.com/2017/08/what-charlatans-dont-tell-you-wcdty/) – Oddthinking Aug 10 '17 at 14:11

2 Answers2

56

Summary: The survey this article was based on was biased, poorly designed and poorly implemented. The conclusions cannot be trusted.

This study was examined by Orac who has been "checking in with and covering periodically ever since its inception in 2012, when antivaxers were fundraising for it."

The study has had a history of being retracted:

I’ve written about this study before. Hilariously, when it was published in its first form, the full study wasn’t published, only the abstract. Then the abstract was, in essence, retracted. Even more hilarious, it was a Frontiers journal, which is an even bigger dis because Frontiers journals are known for tending to be pay-to-publish predatory open access journals. If a Frontiers journal retracts your paper, it’s plenty bad indeed. It turns out that the manuscript had been reviewed by a chiropractor and a peer reviewer without expertise

The next day, Orac returned to announce it had been retracted again!

The first article complains of some minor issues: Apparent bias in the statement of purpose and having antivaccine organisation as the funding source.

He also criticised the misleading description and a biased sampling

Notice how Mawson claims that this is a cross-sectional study, when in reality it’s a survey targeting parents who homeschool. Of course, parents who choose to home school are not like your average parents. There are a lot of confounding factors that go along with home schooling, including the association between home schooling and antivaccine views. This association is very clear in the data, which show that 261 of the 666 subjects were unvaccinated.

He also describes the recruitment techniques as a source of bias too.

no effort was made to construct a representative sample.

In conclusion, he writes:

So what are we to make of the results of this study [...]?

Nothing. The bias and flaws in this study guaranteed no other result, particularly when you consider another confounding factor, namely that the parents of children who are fully vaccinated are very different in their health-seeking behavior than those whose children are unvaccinated. They tend to take their children to visit the doctor more regularly, which means that health disorders their children have are more likely to be diagnosed and treated. They’re also less likely to be seeing naturopaths and other alternative practitioners.

I looked for literature to bolstered Orac's unreferenced claim that "parents of children who are fully vaccinated are very different in their health-seeking behavior than those whose children are unvaccinated." My findings were mixed. I only found minor support for Orac's position.

This paper was useful to me in demonstrating some of the complexity behind the question - there has been a fair amount of research, and it has revealed geographic differenes.

This provided some fairly support for both Orac's statements about visiting doctors and also naturopaths and alternative practitioners:

Inhibitors [for parent vaccinating] included feeling alienated by or unable to trust the pediatrician, having a trusting relationship with an influential homeopath/naturopath or other person who did not believe in vaccinating, worry about permanent side effects, beliefs that vaccine-preventable diseases are not serious, and feeling that since other children are vaccinated their child is not at risk.

However, I don't want to overstate this - one of the "emerging themes" they discovered was:

there is overall trust in the pediatrician but a lack of trust in the information they provided about vaccines.

Community
  • 1
Oddthinking
  • 140,378
  • 46
  • 548
  • 638
  • 23
    This, a thousand times this. It still blows my mind that people still don't get that the number of diagnosis isn't really a indicative of the disease in the population in general. This type of logic is like trying to evade cancer by avoiding doing things that can detect cancer. – T. Sar May 17 '17 at 13:26
  • 1
    " the parents of children who are fully vaccinated are very different in their health-seeking behavior than those whose children are unvaccinated" - citation needed for that specific claim. – user5341 May 17 '17 at 15:35
  • 2
    I find there's two types of parents that don't vaccinate: the careless and the fearful. The fearful are extremely conscientious with their children's health, hence the decision not to vaccinate. I have met many fearful parents, but very few careless ones. If I remember right, the profile of a non vaccinating parent is upper middle class, educated, and very concerned with healthcare. –  May 17 '17 at 19:44
  • 2
    All this is to say they need a citation for "the parents of children who are fully vaccinated are very different in their health-seeking behavior than those whose children are unvaccinated", as @user5341 said, if their conclusion is that non vaccinating children are careless with health decisions. –  May 17 '17 at 19:45
  • How is this up to snuff? The opinions of Orac, fr anyone, are not really a whole lot of proof of anything. 1) Is that not simply how all statements of purpose work, they stare what result they are looking for, their initial hypothesis that they want to prove? 2) Funding is irrelevant, the alternative is funding by some medical group that profits off of vaccines or publicly supports them at least. If funding matters, then no study is safe. – Jonathon May 17 '17 at 20:33
  • 3)Is that not called controlling for variables? Since all the participants are home schooled, then their environment is controlled for. I cannot see any other way of conducting a study on a group that is so predominately homeschooled. Any study like that would just be measuring the difference between home schooled children and publicly schooled children for the most part. – Jonathon May 17 '17 at 20:38
  • 4) Do most studies have such a glut of volunteers that they really get the chance of turning the vast majority down such that they can guarantee a representative sample? Studies I have seen conducted simply asked people to fill out the forms and hoped the sample turned out random enough. Most do not have enough money to take a few people from every city in the world to get a truly diverse and representative sample. but if it was retracted it was retracted, that seems like the main and only thing that needs to be talked about here. – Jonathon May 17 '17 at 20:39
  • @user5341: Thank you for your feedback. Look for an edit shortly. – Oddthinking May 18 '17 at 03:00
  • @JonathonWisnoski: To the degree these are merely opinions, well I could provide evidence that Orac is an expert, but no, you are right. However, to the degree that these are legitimate post-publication peer-review critiques of the claims published in the article, they address the question. – Oddthinking May 18 '17 at 03:04
  • @JonathonWisnoski: Re: 1) - I am tending to agree this isn't as strong a point as it could be. Look for an edit shortly. Re: 2) The funding source isn't all important nor is it irrelevant. It helps us to understand how to focus our skepticism, but shouldn't be used to dismiss evidence entirely. – Oddthinking May 18 '17 at 03:12
  • 3) and 4) are much the same. They took a very biased sample - of an atypical population. They made no attempt to correct the bias. This makes the results largely worthless. If you argue other studies do the same, I would argue they, too, are largely worthless (in terms of getting reliable results - as pilot studies they may have other things to teach.) – Oddthinking May 18 '17 at 03:16
  • @Oddthinking - good edit! One thing I find curious is... there seems to be no study whatsoever mentioned - and given your research patterns, I'm assuming none was found by you - to actually show that unvaccinated children are **less** healthy than vaccinated ones when controlling for variables. Which leads me to suspect that when all said and done, Orac's claim is rather... useless (as in, how parents relationship with pediatrician goes is less relevant than health outcomes). – user5341 May 19 '17 at 01:44
  • @Oddthinking - An interesting side question is, does religious visiting of pediatrician actually cause better health outcomes? Intuitively; considering most visits are to demand antibiotics for a simple viral infection, it may not necessarily be so. – user5341 May 19 '17 at 01:45
  • @user5341: I found lots of papers that show that, in [Country X], increased vaccinations for [Disease Y] lead to lower incidences of children catching [Disease Y]. That's not the same as "overall healthier". – Oddthinking May 19 '17 at 05:55
  • @user5341: I speculate it is like a bathtub curve. After all, you know who else goes to the doctor a lot? Really sick people. – Oddthinking May 19 '17 at 05:56
  • 1
    @Oddthinking - sorry for imprecision, yes, I was specifically thinking "overall healthier". Re: bathtub curve, that's easy(ish) to correct for - you measure willingness to go to a doctor in a hypothetical situation instead of real visits. Less accurate (behavior!=mental model) but removes the problem you mentioned. – user5341 May 19 '17 at 13:46
5

Are unvaccinated children healthier?

Not in the long run.

The quoted study compared the "health" of vaccinated children (in a rather questionable target group and using questionable methods, see Oddthinking's answer and ff524's comment) with the health of unvaccinated children... in a mostly vaccinated population.

It comes to the conclusion that...

...no evidence was found that vaccinated children were more protected against any so-called “vaccine-preventable diseases”. Children in both groups had roughly the same rates of infection from measles, mumps, Hepatitis A and B, influenza, rotavirus, and meningitis (both viral and bacterial).

This claims that being vaccinated has no significant benefit, and that is ludicrous.

Just picking the measles for an example, which are highly contagious. One shot of MMR vaccine is 93% effective, two shots are 97% effective. (Source: CDC)

The effect of vaccination is readily apparent from the significantly reduced number of cases after vaccination started:

Reported cases of measles in the USA, 1950-1993 (CDC)

The fact that the study you quoted did not find "evidence" of significant increases of measles infection among the unvaccinated children is the result of herd immunity, a well-researched and well-understood effect.

If enough members of a population are vaccinated, and (quoting the linked article)...

...it’s worth noting that over 95% of kids in America today are vaccinated...

...then the chances of getting into contact with the relevant pathogen is greatly diminished. Most people who contract the disease recuperate without passing it on, because most people are vaccinated against it.

If the claim is taken at face value, and used to justify not vaccinating children against vaccinable diseases, the percentage of vaccinated people will eventually fall below the threshold at which the vaccinated-against disease will spread (and, exceptional cases nonwithstanding, it will spread among the unvaccinated people mostly).

For measles, the threshold for herd immunity is estimated to be about 83-94%.

And this is not even taking into account the possibility of contracting a vaccine-preventable disease while abroad, or diseases where the pathogen is endemic to the environment (e.g. tetanus, rabies).

Post scriptum, the WHO estimates that measles vaccination alone has saved over 17 million lives since 2000. (WHO)

DevSolar
  • 19,034
  • 8
  • 77
  • 74
  • I totally agree with everything you say, but it doesn't refute the claim in the question. Antivaxers are essentially free-loaders, but that doesn't mean it doesn't work for them... – Benjol May 24 '17 at 11:51
  • @Benjol: Not in the long run. That's what I'm saying -- the sample size is too small. **If** they come into contact with a pathogen, they *will* fall sick, thwarting any benefit not vaccinating might have, real or imaginary. – DevSolar May 24 '17 at 11:53
  • 2
    Yes, my first reaction on seeing the question was "possibly, until they die" :) – Benjol May 24 '17 at 11:56
  • Re "Children in both groups had roughly the same rates of infection...", I strongly suspect that what is seen here is the effect of "herd immunity". That is, when a large enough fraction of the population has immunity to a contagious disease, it tends not to be passed on and become epidemic. See e.g. https://www.vaccines.gov/basics/protection/index.html So the unvaccinated kids that aren't getting those diseases are essentially freeloading off the rest of us. – jamesqf Jun 11 '17 at 17:33
  • @jamesqf: That is the gist of my answer, yes. – DevSolar Jun 11 '17 at 17:35
  • I don't really see the point of refuting the claim that non-vaccinated children are healthier in the long run based on specific details of the study (as the previous answer does) only to have the very next answer headline with the entirely speculative "Not in the long run". Where is your evidence that the two groups are no different "in the long run". I don't see any studies done to same standards as are (and should be) expected of the OP's study having been carried out on the long-term health effects of vaccination/non-vaccination such that we could say anything at all about it at all. – Isaacson Jun 12 '17 at 09:25
  • @Isaacson: Sorry but your comment does not make (grammatical) sense to me. Oddthinking attacks the study mentioned in the claim on its scientific merits and data base. I am attacking the claim on biological / epidemiological grounds. The "not in the long run" is less assertive as I would have liked, but a previous attempt at an answer that made *minute* and *hypothetical* reference to the claim was summarily shot down by the Skeptics community, so I had to write an answer *without* actually making references to the claim. -- Bottom line: In what way do you think my answer should be improved? – DevSolar Jun 12 '17 at 10:05
  • Basically I don't think you can attack the claim on epidemiological grounds, you're only looking at one factor (their chances of contracting the diseases against which they have been vaccinated) and surmising that due to the inevitable breakdown of herd immunity resulting from non-vaccinantion, their health will be worse off in the long run. The study in the OP, however, cites health impacts other than these diseases which your answer does not address. – Isaacson Jun 12 '17 at 10:13
  • Maybe a simpler way of putting it is that the study cited (albeit in the flawed manner described) has already demonstrated, using epidemiology, that unvaccinated children are healthier, by their selected metrics. They have simply failed to demonstrate a causal link. A counter from epidemiology might take a wider selection of health indicators, or a longer time scale (and certainly a wider cohort) to see if that had any impact on the observed effects. I'm not aware of such a study. – Isaacson Jun 12 '17 at 10:17
  • @Isaacson: I don't think the original study has demonstrated anything, not to any scientific standards. I *could* present e.g. the mortality rates and long-term effects of measles, polio etc. as "a wider selection of health indicators", but I feel that it would a) lend undue weight to the "claim" of the antivacs, and b) distract from pointing out the effects of herd immunity, which can be (and has been) *proven*. -- Anyone seriously suggesting that increased cases of hay fever and ADHD offset e.g. being a lifelong invalid due to polio isn't worth the attention. – DevSolar Jun 12 '17 at 10:44
  • The study has demonstrated something, it's demonstrated a link between vaccination and those conditions within that group. Whether that link is causal or coincidental has yet to be shown (except in the case of Autism which has, as a result of the Wakefield scandal been pretty thoroughly shown to be merely coincidental). Your a) and b) do rather play into the hands of irrational anti-vaccers, we should not be concerned about "weight" and "distraction" it makes it sound like you're trying to convince people one way or another rather than just present facts. – Isaacson Jun 12 '17 at 11:02
  • Let us [continue this discussion in chat](http://chat.stackexchange.com/rooms/60304/discussion-between-devsolar-and-isaacson). – DevSolar Jun 12 '17 at 11:03