Another data rabbit from Top Hat by RIVM

Read the full article: Another Top Hat Data Rabbit by RIVM

reading time: 4 Minutes

In collaboration with Egbert Hidding

88000 hospitalizations saved

On Monday, RIVM published a report that vaccination led to a decrease of 88,000 hospital admissions. This has been covered in almost all media. If you read the letter and carefully analyze the numbers, I wonder how an official body could possibly bring up something like this.

The letter states that the scientific publication will be published later this year. But the graph and interpretation show what was done to arrive at this figure of 88000. A number mentioned in the letter is still a conservative estimate.

First the chart published by RIVM:

The yellow bars are the real hospital admission. The purple bars are the calculated number of hospital admissions prevented.

According to the RIVM system explanation, the calculation of the purple bars is based on the VE (vaccination efficacy) value for each age group per day, together with the vaccination coverage for each age. We will only be able to read exactly how this was done once the scientific publication is published. The components for each age group would be the vaccination percentage, VE value, and the number of patients admitted who were vaccinated and who were not.

But for the sake of understanding: since the effectiveness of vaccination is supposed to be higher, the purple bars will also be higher. Because then more people would have been taken away from the hospital. Over the entire study period, the total number of hospital admissions saved was 2.1 times higher than the number of true hospital admissions, according to RIVM.

If you take these ingredients together and apply a formula to them, the numbers will appear. But the question is whether the results of those formulas are realistic. Let’s say I’m working on something completely different. Based on a complex equation, my result is that there are currently 30 million people living in the Netherlands, then I know I made a mistake and I should check if I made an arithmetic error or if there is something wrong with my calculation. Equation.

put in context

Only if you look at this graph and put the hospital admission numbers from March 2020 in front of it, you’ll see something very strange.

When comparing the chart from August 2021 to the previous period, three things stand out in particular:

• Without vaccination, the number of hospitalizations was 7,800 patients at the end of November 2021, according to RIVM. This is 2.7 times more than the highest point in March 2020, when there was no vaccination and no one had contact with the virus.
• The peak of 7800 at the end of November 2021 was four times higher without vaccination than its peak the exact year before when there was no vaccination and there were many more people who had not yet been infected in November 2021, a year later this was the case
• Holland started relatively late with the first booster. It was at the end of November 2021. That was more than six months after the older age groups had received their primary vaccination. All kinds of global studies show that vaccine protection drops sharply after a few months. That’s why the booster was needed, right? In this graph, the number of hospital admissions saved at the end of November 2021, when older adults had already received their last vaccination about 6 months ago, is still about 2.5 times higher than the true number of hospital admissions.

November 2021 vs March 2020

For example, I will only focus on the comparison between November 2021 and March 2020. Hospitalization shows a difference of 2.7. This equation is indicative for the rest of the period:

In March we had a big outbreak. Nobody has antibodies yet. The highest number of hospitalizations in a week was 2,800. No one was vaccinated at that time.

It was the highest peak in the chart, including prevented hospital admissions, after 20 months, according to RIVM. Meanwhile, not only were 66,000 people hospitalized, but of course many more were infected. Let’s start from the numbers given by RIVM itself at the time about the relationship between infected people and hospitalizations. (1.25%). Then these 66,000 patients indicate that before the end of November 2021, about 5 million people were infected in the Netherlands (30%). Their risk of becoming infected again in November 2021 was much lower than those who did not become infected.

This means that in November 2021, 70% of the Dutch population had not yet been infected. But according to RIVM’s calculation, in November 2021 without vaccination, there were 2.7 times more people hospitalized than in March 2020. But since 30% had already been infected in November 2021, this 2.7 should be offset and you’d come to a factor of about 3.5%.

So this graph from the new RIVM letter shows that under fairly equal conditions (no one vaccinated) at the end of November 2021, Covid patients could have ended up 3.5 times in hospital compared to March 2020!

If RIVM gets to such unlikely numbers with their calculations, they’d be better off scratching their heads again and thinking about what I did wrong than starting to post it.

I will give a hint: in the formula, the effectiveness of the vaccine is – to a large extent – exaggerated.