VAERS Misinformation

Updated: Feb 11

Vice.com recently published a scathing article against people misusing VAERS. https://www.vice.com/en/article/qjpmp7/anti-vaxxers-misuse-federal-data-to-falsely-claim-covid-vaccines-are-dangerous


It is full of attacks on "anti-vaxers" and we need to know how to respond to this. I won't bother enumerating them all, because they are obvious when you read the article. I wrote an analysis for a friend and wanted to post it here:


In general, the article's claim is predicated on the fact that it is improper to use aggregated VAERS data to make conclusions about causation, which many people tend to do. People do this because they overestimate the data quality of VAERS reports as well as don't understand that the VAERS program does not follow up with every report with a determination of if the vaccine caused injury or not... they deliberately leave it open-ended, at least in the public data.

That said, the major strength of VAERS is it's qualitative data, not the quantitative. For example, the free form text field in each report that describes in a narrative form what happened offers all kinds of insights that you wouldn't be able to capture with a bunch of yes/no or multiple choice questions. In turn, the MedDra terms that are coded from this freeform text capture a wide array of qualitative data.

So why do I make a difference between qualitative data and quantitative data? The scientific method is basically a 2 part process.

1) Formulate a hypothesis

2) Submit the hypothesis to a test

Scientifically, step 1 is usually done by a qualitative analysis of a case study (like Wakefield's infamous paper). Most people don't realize that Wakefield's original paper was a case study, a qualitative examination of a handful of people in the pursuit of formulating a hypothesis.

Step 2 for Wakefield, would have been to take a hypothesis he formulated from the first study, and perform a controlled experiment with a control group, and an experimental group, and perform a statistical analysis to provide evidence for his hypothesis, or not.

In short, people largely focus on (Step-2) type thinking where they're just looking for answers and not (Step-1) type thinking which is asking the right questions. This is the difference between quantitative and qualitative analysis, respectively.

VAERS is very good at helping us formulate the right questions.

That said, if you critically examine individual reports sometimes reasonable people can make a reasonable conclusion if that person was actually harmed by the vaccine. For instance, if anaphylaxis is mentioned in the report it's highly likely that it was caused by the vaccine, just like "redness at the injection site". It's like, duh. But in some cases it's not so obvious.

As far as the specific claims of the article, it sounds like a lot of straw man arguments and half truths. For instance, the article points out that RFK Jr. says that VAERS is broken, but then calls him a hypocrite because they use it as a source. VAERS is broken in some ways, but if it is understood properly is very valuable.

Next the reference the common claim that only 1% of adverse events are reported, though they don't cite the Lazarus report.

In this case, both sides are wrong because nearly no-one understands the Lazarus report because you need to read it closely to understand what the 1% means.

This part of the article probably has the most truth: "“The best thing you can say about VAERS is that it’s a hypothesis-generating mechanism,” said Paul Offit, ".


Feel free to comment on this post!

15 views0 comments

Recent Posts

See All

Recovering from the Storm

We are still recovering from the recent storms, so I have limited Internet access. I want to let you know that I'm in the process of featuring prominent survivors of vaccine injury, so please let me k

VAERS Underreporting

Harvard Pilgrim published a followup VAERS research study in 2015. It goes into a lot of detail of how doctors vastly underreport vaccine injuries to VAERS. https://pubmed.ncbi.nlm.nih.gov/26060294/ .