I have previously reported on a newish trend in anti-vaccine propaganda, the “vaccinated vs. unvaccinated study” gambit. The anti-vaccine proponents have retreated to this new position after the thimerosal, MMR-autism and other hypotheses of theirs were shown to be worthless. This …..strategy is well known to anyone with a basic understanding of logic as moving the goal post. Superficially, the request does not sound unreasonable. Why not compare the non-vaccinated with the vaccinated and see if there is any increased risks with vaccination?
To be clear, I am not against such research being done, to the limit it can practically be done. There are a few things to keep in mind though. For starters, the problem is presented by the anti-vaccine proponent as a make-it-or-break-it thing. They behave as if the mountain of evidence in support of the use of vaccines is worthless, until and unless such vax vs. unvax study is done. They also act as if the whole vaccination issue hinges upon this one study. However, nothing could be further from the reality. One study doesn’t prove anything, one way or another. We have a ton of science on vaccines, a literal mountain of scientific papers and that will no be overthrown by one study.
Secondly, the unvaccinated population is quite small and dispersed. As we will see, the German Study had over 13,000 participants, and under 100 fell into the unvaccinated category. This hinders considerably the ability to design a study that would include enough non-vaccinated children, and at the same time be able to control for the myriad factors that could affect the results.
Thirdly, this study could not be done prospectively. Current ethical standards of study design would not permit randomly assigning a large number of children not to receive vaccines, when risk of pain, suffering, and even death is so high with vaccine-preventable diseases. That means that the only kind of study we can do is a retroactive study, looking at existing non-vaccinated populations and trying to match the control group as much as possible. That has its own pitfalls, randomization would be impossible, and the point of randomization is precisely to control for all those variables that the scientist didn’t, or can’t, think off on his own. By definition, the non-vaccinated group would be a self-selected group, which may exhibit biases or other behaviors that can affect the results in a non-intuitive way, so that it would be impossible for the researchers to control for said behaviors.
So, we are caught in a Catch 22: we cannot design a prospective, randomized, double blind test due to ethical considerations, and any retrospective studies are inherently limited due to the small numbers of unvaccinated children and the impossibility of randomization. Clearly, any results we may get will need to be analyzed carefully, taking into account all the shortcomings we just went over.
Keeping that in mind, let us look at th German Study.
Vaccination status and health in children and adolescents—findings of the German health interview and examination survey for children and adolescents (KiGGS). Dtsch Arztebl Int 2011; 108(7): 99–104.DOI: 10.3238/arztebl.2011.0099
Summary – The authors utilized data collected through the German Health Interview and Examination Survey for Children and Adolescents (Kinder- und Jugendgesundheits survey, KiGGS) which collected health data on 17,641 children and teenagers aged 0-17 years of age, selected at random in 167 German locations. The data was collected over 3 years, between May 2003 and May 2006. Of these, useful data existed for 13,453 subjects which made up the study population. Any child for which no record existed showing receipt of even a single dose of vaccine was classified as unvaccinated. Any child who had received at least one dose of vaccine was classified as vaccinated. There were a total of 94 unvaccinated children and over 13,000 made up the vaccinated group.
At the time, the recommended German vaccine schedule included the DTaP, MMR, Hib, HepB, and Polio vaccines which were all included in the study. For girls 12-17 the HPV vaccine was also recommended. Lifetime prevalence calculations were done for pertussis, measles, mumps, and rubella (diseases that the vaccines protected against) and also for other infectious diseases outside of the vaccine preventable ones. Atopic (allergies) disease rates were also calculated. Prevalence rates were then compared between vaccinated and unvaccinated children.
Results -The study showed that for pertussis, measles, mumps, and/or rubella, unvaccinated children had on average triple the number of infections when compared with sufficiently vaccinated children. Specifically:
- Pertussis: Unvaccinated 15.8% – Vaccinated 2.3%
- Measles: Unvaccinated 15% – Vaccinated 5.2%
- Mumps: Unvaccinated 9.6% – Vaccinated 3.1%
- Rubella: Unvaccinated 17% – Vaccinated 6.8%
Furthermore, no differences were found in rates of other infectious diseases whose burden was similar on each group, thus providing contradictory evidence to the claim that vaccines “overload” the immune system and make vaccinated children more vulnerable to other diseases. The same held true for medically diagnosed atopic disorders.
Conclusion -The study adds to the existing evidence about vaccine safety and efficacy. As expected, the study showed a significant increase in prevalence in unvaccinated children for the diseases for which vaccines exist. Furthermore, it did not show any increase in susceptibility to other infectious diseases, or atopic diseases, in vaccinated children therefore contradicting the hypothesis that vaccines overwhelm childrens’ immune system.
Nevertheless, the study on its own does not prove that vaccines are safe, but when taken in context with all the other existing scientific evidence, it adds to the safety and efficacy data. However there are certain shortcomings that are unavoidable: firstly, even though the sample size was extremely large, the unvaccinated represented a very small fraction of it, 94 children,which makes it impossible to do any meaningful statistics on things such as autism which has a prevalence rate of roughly 1 in 110. Secondly, it is possible that at least some of the children for whom no record of vaccination existed might have still received a vaccine which went unrecorded. Thirdly, a lot of the data was based on a survey, which relied on parent’s memories, thus it is possible that prevalence rates may have been under, or over, reported. Fourthly, the survey did not ask for date of disease onset, therefore some vaccinated children could have gotten sick before being vaccinated and still be included in the vaccinated prevalence data.
To sum it up, the first vaccinated vs. unvaccinated study’s results are completely aligned with current scientific knowledge on vaccines, they are safe and effective at preventing disease. This study did not, and cannot, say anything about autism though, and I suspect that will be the number one reason used by the anti-vaccine crowd to dismiss this study’s results. After all, they know vaccines cause autism.
The study’s authors concluded as such:
Conclusion: The prevalence of allergic diseases and non-specific infections in children and adolescents was not found to depend on vaccination status.