"Evidence-based"? You keep using that term. I do not think it means what you think it means.

“Evidence-based”? You keep using that term. I do not think it means what you think it means.

So I was checking out my Facebook page when I stumbled across an article, entitled “200 Evidence Based Reasons NOT To Vaccinate” [PDF download] posted by a physician friend. He and I often post medically related articles for a laugh, and every so often, we stumble across something like this, and share it with those whom we know will laugh at the incorrect pile of garbage it truly is. But, of course, I am prejudicing you in advance.

Let us first look at where this article is found. It comes from a website called GreenMedInfo, complete with the tagline of Education Equals Empowerment. While I don’t disagree with that tagline per se, I will also add the caveat that the education needs to be soundly based in reality. It contains such articles with catchy titles like, “What if HPV does not cause Cervical Cancer?,” “Research Confirms Sweating Detoxifies Dangerous Metals, Petrochemicals,” and “The Toxic Terrain of Airplanes: 4 Steps To Travelling Clean.” It is a veritable cornucopia of medical woo, and you too can access it to your hearts content. The founder of the site, Sayer Ji, has been discussed before on this site, regarding Angelina Jolie’s prophylactic mastectomy and the links between diet and cancer, with even broader discussion by a “friend” of the blog. He “founded in 2008 in order to provide the world an open access, evidence-based resource supporting natural and integrative modalities.”

But enough about the site and the author, we all know you are really here to learn why this paper is a joke. Let us start at the very beginning of the paper itself. To begin with, this is basically a meta-study. While there can be many benefits to a meta-study, in this case, all it really does is let you know how much information they have on a particular topic. But of course, in this case, this is where their study immediately begins to fall apart as they do not properly define it. What we call Meta-Data, they call cumulative knowledge. They do claim to have a YouTube video describing how their Cumulative Knowledge works, but the link takes you to a private video. Their written definition is that “Cumulative Knowledge is determined by ascribing a numerical value to the various study types weighted in descending order as follows: (1) Meta-Analysis; (2) Human Study; (3) Human: Case Study; (4) Animal: Transgenic; (5) Animal; (6) In Vitro; (7) Review; and (8) Commentary.” So a listing of 2,123 for the topic of vaccine-induced toxicity where 301 articles are used would likely include all categories. Of course, they never give you a breakdown of which are used. In some cases, such as their Cumulative Knowledge on tumor necrosis factor-α (TNFα) up-regulation, their math doesn’t even ad up. They have a Cumulative Knowledge of 40 with a total of 4 articles. So unless each article is given more than one category, which I guess is possible as they do not rule it out, the total would only be 32. This is, of course the main problem. They do not offer a good definition of how they go about assigning a Cumulative Knowledge Score, so while the numbers look impressive, you are left feeling kind of lost for how they arrived at them. In this case, I am guessing that appearing impressive is what they really wanted, so their goal is probably met.

Now, I am not going to run through all of their articles as I think I would crash SBM’s servers with a piece that long, but I will look at a few of them to show you how disingenuous they truly are in their analysis. The very first article you come across is, in the paper, entitled, “38,787 adverse events including infant death (highest in 1-3 month olds) after vaccination were reported between 1991-1994. (The authors speciously claim SIDS and not vaccination caused these deaths). – GMI Summary” Now, we all know that commentary of a personal nature by the Meta-analyst does not belong in the title of a study, but let us look at what the analysis of the study provides us. They provide us with where the study was published, and an accurate link so we can get to the actual study ourselves. In addition to that, the publication date and the authors are listed, in addition to the study type, which in this case is Meta-Analysis. They then give us the diseases discussed, each followed by a CK (Cumulative Knowledge) number, and an AC (Article Count) number. Finally, they provide us with what they call the Anti Therapeutic [sic] Actions, followed by its CK and AC numbers. Ok, this looks legitimate, but what happens when I click on the link to the actual study?

This is where it all falls apart. While it does take me to the correct Study in PubMed, the titles are different. The actual study title is “Descriptive epidemiology of adverse events after immunization: reports to the Vaccine Adverse Event Reporting System (VAERS), 1991-1994.” The study is looking at incidences of Adverse Events (AE) and causes for them by examining the reports submitted to VAERS. It is true that death is among those AEs, but as the conclusion of the study itself states, “The peak age of deaths at ages 1 to 3 months could be expected on the basis of prior studies showing that sudden infant death syndrome deaths peak at that age, that most deaths in the Vaccine Adverse Event Reporting System are attributed to sudden infant death syndrome, and that sudden infant death syndrome has not been associated with vaccination. The large number of reports and national coverage of the Vaccine Adverse Events Reporting System make it useful for monitoring the safety of vaccine lots and for accumulating case series to detect or better understand adverse events that may occur too rarely to be assessed in clinical trials or in the larger studies that are sometimes carried out by manufacturers after vaccine licensure (phase IV studies).” Suddenly, the additional commentary makes sense. Of course, the study titles do not match, so how do we account for that? In this case, the language was culled from the text of the study itself.

So how did they do on the next study? Again, they give us a title, “DTP or tetanus vaccination increases the risk of allergies and related respiratory symptoms in children and adolescents. – GMI Summary” and they provide us with the PubMed identification number, and authors, and all the same information, so when we click on the link, it brings us to, “Effects of diphtheria-tetanus-pertussis or tetanus vaccination on allergies and allergy-related respiratory symptoms among children and adolescents in the United States.” The study here is slightly more interesting in that it does appear that there is some linkage between DTP and asthma, however, the authors caution against such a general finding in their own conclusion, “DTP or tetanus vaccination appears to increase the risk of allergies and related respiratory symptoms in children and adolescents. Although it is unlikely that these results are entirely because of any sources of bias, the small number of unvaccinated subjects and the study design limit our ability to make firm causal inferences about the true magnitude of effect.” Again, not quite the knockout blow that GreenMedInfo would like.

Perhaps the next “study” is more honest. Yet again, we have a title from the body of the paper itself, “Neonate exposure to thimerosal mercury from hepatitis B vaccines may be a significant problem. –GMI Summary” and again a PubMed identification number is provided, and the authors listed, and so forth. So again, being the intrepid researcher who likes to see the cold, hard, and above all raw data, I click on the link. The link takes us to an study published back in 2009 in the American Journal of Perinatology, entitled “Neonate exposure to thimerosal mercury from hepatitis B vaccines.” The two titles are similar, but there is that addition of “may be a significant problem” in the GreenMedInfo “study” so let’s take a closer look. What this study tells us is that children in less developed countries are the ones mainly at risk as in those countries, the thimerosol-containing vaccines, or TCV’s are more common as they are less expensive. Another thing to look at, and why these TCV’s may have more of an effect on the children is that these are primarily present in the vaccines administered within the first 24 hours after birth in these countries. The issue here, however, isn’t autism as much as it is the levels, or toxicity of the mercury in the neonates. This study does discuss that, but the conclusions it reaches are that, “Although it is much easier to contemplate the incommensurable good achieved by immunization in preventing disease, it was the plausibility of subtle harm caused by the EtHg (ethyl mercury) that moved rich countries to withdraw thimerosal from vaccines for infants and young children.” So again, the “study” got all of its information from the abstract, rather than reading the body text.

Next up we have, “Over 1,000 cases of vaccine-induced thrombocytopenia were reported between 1998-2008. –GMI Summary” and yet again, they give us the PubMed identifier and authors, and all the other information that makes it look nice and correct. The actual study, from the journal Vaccine, is entitled, “Throbocytopenia after vaccination: case reports to the US Vaccine Adverse Event Reporting System, 1998-2008.” Here again, it looks like there might be a grain of truth to this, but this is where those pesky maths come into play. First off, you need to look at the prevalence in each individual vaccine, rather than take them all as a whole. When you begin drilling down into these thrombocytopenia (TP) reports, you see that the hepatitis B vaccine has the most reports at 226. You need to dig down into the discussion of the study to find the reported outcomes. The first sentence there, however, points you in the right direction. “Since acute and chronic TP commonly occurs in the absence of resent immunizations, post-immunization TP may have a purely coincidental relationship with vaccines. Nonetheless, vaccines could conceivably contribute to the development of TP as already demonstrated for measles containing products.” So what about the number, “Over 1,000”? Well, the study did identify 1,510 reports of TP in VAERS, of which 70 were excluded as being linked to other specific causes, such as cancer, meningitis and preceding the vaccination, leaving us with 1,440 cases of TP in that time period. According to VAERS, more than ten million vaccines are given annually to children. This does not include booster and other vaccinations for adults, and specifically looks at children under the age of one. If we use those numbers also, then for this time period, 100 million vaccines were given. Even if we assume all of these 1,440 events occurred in children, the prevalence is 0.00144%.

The final article I will look at was entitled by the “study” as, “Over 600 cases of sudden infant death syndrome following vaccinations were reported from 1990-1997. –GMI Summary” and one more time, a PubMed number along with its authors is provided. This study, properly entitled, “The epidemiology of fatalities reported to the vaccine adverse event reporting system 1990-1997.” comes from the Journal of Pharmacoepidemiology and Drug Safety. The purpose of the study was to examine the fatalities which were reported to the VAERS in its first seven years. There isn’t really much I can say about this. Of the 1,266 fatalities, the study does conclude that over half were classified as due to SIDS. Having said that however, the study also points out that once the “Back to Sleep” program was initiated, the incidence of SIDS decreases. The study itself provides the best conclusion, “These data may support findings of past controlled studies showing that the association between infant vaccination and SIDS is coincidental and not causal.”

It continues in this fashion, culling study “titles” from the text of the reports themselves, and quote mining them for their most extreme statements possible. It even drags up the thimerosal-autism link, which has been debunked numerous times both on this website and others. As a “study” it completely fails. As “200 Evidenced Based Reasons NOT to vaccinate”, it fails. There are no conclusions in the analysis, just the titles of the 200 studies they used, which are themselves inaccurate. It is sad that this is what people think science is today, and further underscores the need for better science-based education in our schools.

Martin A. Lessem is a regulatory attorney with twelve years of working in the pharmaceutical industry. Although currently working for a biologic company, Martin has worked for other innovator companies as well. His current area of expertise is risk management, with a strong focus on opioid risk management, regulatory compliance and the regulatory aspects of clinical trials and pharmaceutical registrations. The views that he expresses in his posts are his own and do not necessarily reflect those of his current or previous companies.




  • A regulatory attorney working as a Director of US Regulatory Affairs in the Pharma Industry with ten years experience in Regulatory Affairs in both Clinical and Post-Marketing areas. Martin has planned US Regulatory Strategies and managing the Regulatory Activities in the US for a multi-national company. Currently he is working on Labeling for Biological products with another multi-national company.

    View all posts

Posted by Martin A. Lessem, J.D.

A regulatory attorney working as a Director of US Regulatory Affairs in the Pharma Industry with ten years experience in Regulatory Affairs in both Clinical and Post-Marketing areas. Martin has planned US Regulatory Strategies and managing the Regulatory Activities in the US for a multi-national company. Currently he is working on Labeling for Biological products with another multi-national company.