Statistician Herman Steigstra published an article on the website of Dutch pollster Maurice De Hond entitled ‘Excess mortality in 34 countries, failing vaccines?’. Statistical analyzes would show that the corona vaccines lead to excess mortality. Is this correct?
Comparing countries with each other does not always give reliable results
In the article, the author compares the vaccination rates and mortality rates of various European and non-European countries. The data comes from the website Our World In Data, a project related to the University of Oxford.
In medical-scientific research this is called “ecological research”, in which entire regions or countries are compared with each other. This method of research is sensitive to a number of confounding factors and hangs somewhere at the bottom of the “hierarchy of evidence”. This hierarchy roughly indicates how reliable a particular method of research is. At the top are systematic reviews of randomized controlled trials: we can usually trust the conclusions we can draw from them, if the studies are well conducted. The more stairs you go down, the less robust the results of the study are.
Ecological studies are in the bottom half of this hierarchy, as they are sensitive to what are called “confounding variables”. These are factors that confuse statistical analysis by making it appear that there is a relationship between two parameters that in reality have nothing to do with each other.
For example, the article refers to the excess mortality in Bulgaria in recent months. This is low and goes hand in hand with a low vaccination rate. But that doesn’t mean one has to do with the other. Another explanation is the high excess mortality during the first waves of the Covid19 pandemic, resulting in a natural reduction in mortality later on. After all, the number of vulnerable people who would die of something else, if they had not died of Covid19, has been greatly reduced.
Comparing periods also does not take a number of things into account
Steigstra compares the mortality in the period April 2020 to 2021 with the vaccine doses that would be administered later, which were not yet available at the time. He then compares the mortality in the period June 2021 to April 2022 with the administered vaccine doses. The graph looks almost the same: an argument for Steigstra to say that the vaccines are not doing their job.
Such a comparison is difficult, because there was a large wave of covid infections in that second period, so the comparison is not entirely valid. In order to be able to compare, the number of infections would have to be corrected. Which was not done here. But even with that correction, such a comparison is not obvious.
In addition, the countries being compared sometimes have completely different care and populations. In Belgium, for example, there is a marked aging population, while South Africa has a relatively young population. Yet both countries are compared with each other. The analysis does not take into account the differences between the analyzed countries.
Drawing conclusions from a comparison of different countries over different periods should be taken with a grain of salt. Researchers argue that ecological research must always be supplemented with other data in order to arrive at reliable results. That didn’t happen here.
Research shows that vaccines protect against severe Covid19
A lot of research is being done into the effect of the vaccines: research that is higher up the aforementioned hierarchy and therefore has a stronger evidentiary value. A few systematic reviews have been published examining the effectiveness and safety of vaccination. These show that the chance of becoming seriously ill or dying as a result of Covid19 is greatly reduced after vaccination.
Mortality generally not higher after vaccination
One of the most important consequences of vaccination to investigate is the overall probability of death after vaccination. Fortunately, there is also research into this. A systematic review of randomized trials showed that no more people died from any cause, whether they were vaccinated or not.
Some studies that are slightly lower in the weight of evidence hierarchy, two retrospective studies, even show that the chance of death from any cause is lower after vaccination. The fact that the systematic review does not come to this conclusion may be due to the fact that the retrospective studies examined data from many more people.
A report by the RIVM and CBS, the health authority and statistics office of the Netherlands, also points to a reduced risk of all deaths. An analysis of mortality and vaccinations in 2020 and 2021 shows that after vaccination the chance of dying from non-Covid19-related causes is lower.
A side note to the retrospective study and the RIVM report is the chance of the aforementioned “confounding variables”. For example, it is possible that people with a healthy lifestyle get vaccinated more often, which can partly explain the lower mortality rate.
There is effectively excess mortality in the Netherlands, but the cause is unclear
Data from the Dutch government clearly shows that more people are dying than expected and that this cannot be fully explained by Covid19. There is no clear cause. The delayed provision of care may partly explain this. Another possibility is the role of the long-term consequences of Covid19: a large study showed that experiencing a corona infection increases the risk of dying for months, compared to those who were not infected. These – and other potential explanations – are being extensively investigated in the Netherlands.
Conclusion
Merely by comparing different countries over several periods, you cannot demonstrate that the vaccines lead to excess mortality. Better research methods, systematic reviews for example, show that the chance of dying from any cause is the same after vaccination. Mortality from a Covid19 infection is falling sharply. Some other studies show that the chance of dying in general is even lower after vaccination than without vaccination.