tl;dr Swedish vaccination data showed that covid vaccination added almost 3x extra risk of death in younger people.
Jessica Rose’s post is about “Demystifying the Swedish Data…”
Rose accurately points out that the Swedish public health authorities have been engaged in confabulations about the vaccination mortality data. People who were injected with covid vaccines were counted in the unvaccinated statistics. This made the vaccinated statistics look better. This makes one wonder whether the Swedish authorities are corrupt or merely incompetent.
Then Rose goes on to look at vaccination and injection mortality statistics. She sorts data into four categories: (1) never injected, (2) received one injection but less than 21 days had passed since injected, (3) received one injection more than 21 days previously or received two injections but less than 21 days had passed since injected, and (4) received two injections more than 21 days previously–the “fully vaccinated.”
The way that perceptions are skewed depends on how groups (2) and (3) are viewed. The highest mortality risk for each injected person looks to be in groups (2) and (3). If groups (2) and (3) are lumped with group (1), then it looks like the unvaccinated (Rose calls them “uninjected”) are at greater risk. However, if groups (2) and (3) are lumped with group (4), our perceptions of risk change. All of a sudden injections are more dangerous. But wait! Doesn’t the data show that there were more deaths in the uninjected in Feb. than in the injected, even lumping groups (2) through (4) together? Yes, there were more deaths, but the risk was higher for people receiving injections because there were far more people who were uninjected than injected in Feb. You have to take into account the numbers of the injected and uninjected when you consider mortality risk.
Consider an example. Let’s suppose that there were 100 people injected and 1000 people uninjected in a population and 10 of the injected people died and 50 of the uninjected died. The mortality risk for the injected was 10/100, or 10%. The mortality risk for the uninjected was 50/1000, or 5%. So even if more uninjected people died, the risk was still higher in the injected people.
So, let’s talk about normalizing data. This means that we take data for two different-sized groups of people–the injected and the uninjected–and put them on an equal footing as if they were the same size. This is what Rose did in her two charts of normalized data. One graph showed the risk of groups (1) — (3) v group (4), while the other showed the risk of group (1) v groups (2) — (4). The difference is striking. Looking at February data, in the first chart, the mortality risk is 78 for the unvaccinated v. 12 for the “fully vaccinated.” In the second chart, the mortality risk for the uninjected is 54 while the mortality risk for the injected is a whopping 789!
But there’s more to the story. Those who are the injected in Feb. are older and older people have higher risk. Rose failed to do any age adjustment, so the Feb. data should be ignored. However, by April younger people were being injected and we see that in April the mortality risk for the uninjected was 41 per million, while the risk for the injected was 156 per million. So injections added almost 3x extra risk of death in younger people. And the short term mortality risk is carried mostly by groups (2) and (3).