Here we go again – bad published research tries to convince us that COVID-19 rates are unrelated to the uptake of COVID-19 vaccines in several countries and US counties. That would be interesting if not for the poor design and analytical methodology utilized by these researchers.
Of course, the second this kind of paper is published, every COVID-19 denier jumps on board to say “here’s the official proof that the COVID-19 vaccines are useless.”
I’m going to set aside the irony that anti-vaxxers love to argue that the bulk of vaccine research is garbage, usually bought by Big Pharma. But any “research” that supports their pre-conceived conclusion is, of course, Nobel Prize-worthy.
It’s frustrating, but my job is to critique this type of research. And this new paper that states that COVID-19 rates are not related to uptake levels of vaccines is a prime example of research filled with bad methodology, poor analytical procedures, and lots of bias.
Here we go again.
COVID-19 rates are unrelated to vaccines – the paper
In a paper published in the European Journal of Epidemiology, a moderate quality journal, S. V. Subramanian & Akhil Kumar, proposed that increases in COVID-19 rates are unrelated to uptake levels of vaccines across 68 countries and 2947 US counties.
Subramanian and Kumar wrote:
At the country-level, there appears to be no discernable relationship between percentage of population fully vaccinated and new COVID-19 cases in the last 7 days (Fig. 1). In fact, the trend line suggests a marginally positive association such that countries with higher percentage of population fully vaccinated have higher COVID-19 cases per 1 million people. Notably, Israel with over 60% of their population fully vaccinated had the highest COVID-19 cases per 1 million people in the last 7 days. The lack of a meaningful association between percentage population fully vaccinated and new COVID-19 cases is further exemplified, for instance, by comparison of Iceland and Portugal. Both countries have over 75% of their population fully vaccinated and have more COVID-19 cases per 1 million people than countries such as Vietnam and South Africa that have around 10% of their population fully vaccinated.
There is a lot to take down here, so I’m going to do it point by point:
- They chose to look at 68 countries, except they don’t list out any inclusion criteria for choosing which countries to include. And in their list, France, the UK, and Germany were not included – all of which have shown a huge drop in COVID-19 rates as a result of high vaccine uptake. This is a huge sign of bias and cherry-picking.
- They include low GDP countries where vaccination rates are low, but testing levels are similarly low. Using data from these countries may provide with a unreasonably low level of COVID-19.
- Country-level research is not the way to do proper epidemiology, which is why I am surprised that this paper was published in an ostensibly higher quality epidemiology journal. Confounding variables, from access to healthcare to income, can influence these results. In fact, the level of testing itself is a huge confounding factor that is completely ignored by the authors. Why? Accounting for confounding is one of the most important aspects of good epidemiological research.
- The same can be written about county-level data in the USA. There is a wide range of testing and reporting across all of the counties. It’s difficult to make any reasonable comparison between Los Angeles County in California and Brevard County in Florida.
- The worst part of this study is that it focused on cases, which is one criteria. However, there are others that are actually more important in determining effectiveness of COVID-19 vaccines – hospitalization, ICU, and mortality rates tell us more information about these vaccines.
Let’s talk about Israel
The authors, like many others, are using Israel as a case study for the failure of vaccines. Unfortunately, their commentary about Israel’s experience with the Pfizer vaccine, like almost everything else they write in this article, is filled with bias and poor analysis.
Yes, Israel was one of the first countries to vaccinate the majority of their population against COVID-19. Yet, over the past few months, they have had extremely high rates of infection.
But a deeper analysis of what’s happening in Israel can lead us to different conclusions:
- Yes, we see waning immunity. This happens with all vaccines, that’s why a booster is being proposed. Also, the mRNA vaccines allow sceintists to adjust the antigens created by the vaccine to make it better, but that takes time. Unless you’re in love with the Nirvana fallacy, even a vaccine with waning immunity is better than nothing.
- The Delta variant caused more breakthroughs. Mutations happen, that’s why we need a new flu vaccine every year.
- Even if you get a breakthrough infection after being vaccinated, generally the vaccinated have lower rates of hospitalization, ICU stays, and death than unvaccinated. Subramanian and Kumar completely miss this point in their paper.
- Even though Israel had a high rate of vaccination, that was for eligible individuals over the age of 12. Actually, only 58% of the population was vaccinated, which is not high enough. It’s hard to tell what the exact herd immunity level is for COVID-19, but it’s probably close ot 88%. Israel may have been highly vaccinated, but not nearly highly enough.
- Vaccines aren’t everything – facemasks, restrictions on group gatherings, and other mitigation efforts cannot be ignored even if vaccinated.
- Despite the claims of Subramanian and Kumar, that immunity from a natural infection is better than the vaccine, one must remember that getting immunity from the disease means you’ve had to contract COVID-19 which kills.
Israel isn’t an argument against the COVID-19 vaccines, it shows issues with how to mitigate COVID-19 rates with vaccines.
I am quite annoyed by papers like this one. It’s one of the laziest ways to analyze epidemiological trends – you can literally Google the data and put together a paper without doing any hard work. Population-level numbers just don’t tell us anything about whether COVID-19 vaccines have any relationship to COVID-19 rates.
These types of analyses require hard work to provide accurate information. Subramanian and Kumar didn’t do hard work, so they provide an analysis that has little to no meaning in understanding the effectiveness of vaccines. Sadly, it will probably be used to push a false narrative about the usefulness of these vaccines.
- Subramanian SV, Kumar A. Increases in COVID-19 are unrelated to levels of vaccination across 68 countries and 2947 counties in the United States. Eur J Epidemiol. 2021 Sep 30:1–4. doi: 10.1007/s10654-021-00808-7. Epub ahead of print. PMID: 34591202; PMCID: PMC8481107.
- Long vax — rare neurologic syndrome after COVID vaccines - 2023-12-03
- Review of the book “We Want Them Infected” by Jonathan Howard - 2023-11-28
- Flu vaccine reduces heart attacks - 2023-11-27