OUR WORLD IN DATA is a unique and rich online data resource that has been tracking COVID-19 cases throughout the pandemic.

They recently added vaccination data to their data. Well, sort of.

Here, OWID provides a chart of SOME countries’ vaccination rates.

Why one cannot access the vaccination uptake data from all countries is unclear. In fact, no one can access the vaccination rate data from any country via their interface or by downloading the datafile. When one downloads the data for their chart, or does an individual country query, one finds that the vaccination rate data are not present.

So, using the numbers from their chart (not scaled for population size), I typed in the vaccination uptake data and then, using their data on the number of new cases reported (9.9.2021 entries), I could plot the number of new cases by vaccination uptake. Plotted on a log scale with a necessary fixed zero intercept model, that looks like this:

Wait, you say. There is a third variable that could explain this completely you say. Population size. Right. So, when we correct for population size by calculating the number of vaccines per capita and number of new cases per capita, we get a plot that looks like this (plotted on linear scale; Y-axis = New Cases per Capita, X-axis = Vaccines Per Capita).

The coefficient (slope = 0.0003) is even larger, as is the R-squared value. The data suggest that for every vaccination given per person in a country, we get an INCREASE of 3/1000ths of a new case of COVID-19 per person in the country. Another way to look at it, for every 3000 people who are vaccinated per person in a country, we get an extra case of COVID-19 per person in the country. Wait, what? Isn’t it suppose to be a reduction in the number of cases?

There are number of plausible explanations for this. Is this evidence for Antibody Dependent Enhancement? Could be; contrary to false so-called fact-checking websites and articles that claim that the mRNA vaccines encode the protein that contributes least to ADE, the mRNA vaccine produce the protein second most likely to cause ADE. Is it Pathogenic Priming? My analyses from April 2020 and follow-up validation of my predictions suggest yes, sure, some people might experience Pathogenic Priming. Original Antigenic Sin? Possibly.

There are other possible causes of worsening public health under widespread vaccination. Perhaps people who vaccinate more are also more likely to receive thimerosal-containing influenza vaccines, which silences the protein ERAP1, essential for antigen folding for APC (antigen presentation) cells? Or perhaps the fiscal incentives in play rewarding medical facilities for diagnosis are correlated with financial incentives in play for vaccination?

Either way, it is accurate to say, at a minimum, that current real-world data suggest that total vaccination uptake is not reducing the number of newly reported cases of COVID-19.

This can be seen in the 1:1 comparison of rates between Israel in the United States, as has been pointed out by Tony Lin:

My analysis was limited in a number of way. First, by virtue of the limited access to vaccination rate data at OWID, it was to countries for which I could access data on vaccine uptake from their graph, and those did not distinguish between once- or twice-jabbed rates. I also excluded the data from China due to an unbelievably low new case rate of 17.

If you know details about OWID I don’t know that could explain these trends, please share them in the comments below. I’m open for rational discourse. Ad-hominem BS will be ignored.

Here is spreadsheet I used for my analyses:

Read some more by Dr. Lyons-Weiler’s Substack Newsletter, Popular Rationalism:

9/9/2021 – Dr. Ionnidis – I’m Sorry to Have to Be So Blunt, But In Public Health vs. Science, Science Always Wins

Click on Science Breaking Chains to find out how to support


  1. I imagine that the slope is even more dramatic and here’s why:

    The bias of “vaccines work” is so prevalent even the CDC came out in April and announced they weren’t going to be tracking breakthrough cases the same. Mixed in with policies to not test vaccinated individuals at the same rates as unvaccinated individuals. This means as populations are vaccinated whole groups of individuals will be shifting from the likely to be tested to unlikely to be tested.

    This will severely bias any baseline on actual population infection samples. It’s the proverbial streetlight effect bias.

    With actual data though, it’s difficult to say if that would change the 3/1000ths to 3/500ths or 3/100ths etc.

    1. Yes, the propaganda bias could be in play.
      It sure would have been nice to have long-term RCTs to track long-term health consequences of these vaccines.
      Unfortunately, all of the control groups in the early RCTs were vaccinated to prevent such knowledge from existing.

Leave a Reply