Deconstructing 'Review Calls for Market Removal of COVID-19 Vaccines': A Critical Analysis of Methodological Failure

How Poor Methodology and Misused Data Created a Manufactured Crisis

This article was co-authored by Dr. Jess Steier, a public health scientist (unbiasedscipod); Dr. Nini Munoz (niniandthebrain), a data scientist and risk communicator; and Dr. Liz Marnik (sciencewhizliz), who holds a PhD in immunology and genetics.

---

In the aftermath of the most significant public health crisis of our generation, we find ourselves at a critical juncture. The COVID-19 vaccines represent one of medicine's most remarkable achievements – developed through unprecedented global collaboration and rigorously tested through clinical trials involving hundreds of thousands of participants. These vaccines have prevented millions of deaths and severe illnesses worldwide. Yet today, we face a growing tide of misinformation that threatens to undermine this scientific triumph and, more broadly, public trust in medical science.

Questions about vaccine safety and efficacy are not just valid – they're essential. The scientific process thrives on inquiry, skepticism, and rigorous examination of evidence. However, there is a profound difference between legitimate scientific discourse and the deliberate distortion of data to manufacture doubt. This paper examines one such example: a recent publication that exemplifies how misappropriated data, flawed methodology, and fundamental misunderstandings of epidemiology can be weaponized to create fear and confusion.

---

Last month, a paper published in Science, Public Health Policy, and the Law—(a WordPress blog, not a scientific journal)—called for the immediate withdrawal of COVID-19 vaccines from the market. This publication has been criticized for bypassing rigorous peer review standards and promoting anti-vaccine narratives. In fact, just weeks before this paper was published, another deeply flawed paper that attempted to draw a causal connection between vaccines and autism went live— don't worry, we debunked that one, too.

The paper's credibility is immediately suspect given its publication venue's association with IPAK (Institute for Pure and Applied Knowledge), an organization known for promoting vaccine misinformation. The lead author, Nicolas Hulscher, is affiliated with the McCullough Foundation, headed by Dr. Peter McCullough, who has a documented history of promoting vaccine misinformation and notably serves as Chief Scientific Officer of The Wellness Company, which profits from selling unsupported "cures" for supposed COVID-19 vaccination side effects.

Despite fundamental methodological flaws that undermine its conclusions, this paper has gained significant traction on social media and is being cited by vaccine skeptics worldwide. Its story illustrates how poor methodology combined with misused data can manufacture a crisis where none exists.

The VAERS Analysis: A Lesson in How Not to Use Surveillance Data

The foundation of this paper's safety claims rests on perhaps the most egregious misuse of VAERS data in recent memory. The authors take 19,028 reported deaths and multiply them by an unvalidated factor of 31 - derived from a single non-peer-reviewed analysis - to claim nearly 590,000 vaccine-related deaths. This methodology reveals not just a misunderstanding of VAERS, but a fundamental failure to grasp how vaccine safety surveillance works.

To understand why this approach fails so spectacularly, we need to look at how VAERS actually works when used properly. Consider the case of the Johnson & Johnson COVID-19 vaccine and Thrombosis with Thrombocytopenia Syndrome (TTS) in 2021. VAERS detected an unusual pattern of rare blood clotting events, leading to a rapid FDA and CDC response. Through proper analysis - not crude multipliers but careful statistical signal detection - researchers identified a specific risk profile: approximately 7 cases per million doses in women aged 18-49. This led to revised recommendations favoring mRNA vaccines for this population.

The authors' methodology fails in several critical ways:

1. They fundamentally misunderstand what VAERS reports represent. During the pandemic, heightened awareness and reporting requirements led to unprecedented levels of reporting. Any death following vaccination was required to be reported, regardless of cause. Furthermore, anyone can report data to VAERS, and the claims need to be verified after to ensure they are real. The authors make no attempt to verify these reports or establish causality.

2. They ignore critical context about background rates. In a population where approximately 8,000-10,000 people die each day naturally, some temporal association between vaccination and death is expected by chance alone. It is when the reports occur above the background level that a signal is flagged. The authors make no attempt to compare reported events to these background rates, and VAERS doesn’t provide background rates about the number of doses or individuals who received a vaccine.

3. They make inappropriate comparisons with historical recalls. For example, they cite the Cutter Polio Incident (a manufacturing error) and the 1976 Swine Flu Vaccine withdrawal (due to a small increase in Guillain-Barré Syndrome), but these situations are not comparable to current COVID-19 vaccine safety data.

4. They misuse FAERS data, which tracks adverse drug events, not vaccines - demonstrating a fundamental misunderstanding of different safety monitoring systems.

5. They ignore more relevant comparisons, such as:

   • HPV vaccine (Gardasil) which had thousands of VAERS reports but was thoroughly investigated and deemed safe

   • Annual flu vaccine, which generates similar patterns of post-vaccine mortality reports without triggering recalls because the rates are also not higher than the expected background rate.

   • Furthermore, they neglect that many of the vaccines they are comparing to are administered to a much smaller segment of the population at a given time, compared to the Covid-19 vaccine- so comparing one to the other is inappropriate.

6. They ignore established vaccine safety monitoring precedents. When RotaShield vaccine was withdrawn in 1999 due to intussusception risk, or when H1N1 vaccine was linked to narcolepsy in Europe in 2009, these signals were detected through careful statistical analysis - not by applying arbitrary multipliers to raw numbers.

Dissecting the Supporting Studies: A Pattern of Flawed Research

The paper relies heavily on several studies that have been thoroughly debunked or retracted:

1. Hulscher's previous study claiming 74% of 325 autopsies were deaths due to COVID-19 vaccines was withdrawn due to methodological flaws.

2. The Rancourt et al. study claiming 17 million vaccine-related deaths used unvalidated statistical models, completely disregarded COVID-19 deaths, and made unfounded causal assumptions. The study failed to verify vaccination status of deceased individuals and ignored that mortality spikes aligned more closely with COVID-19 waves than vaccination campaigns.

3. The Mostert et al. BMJ study was retracted, partly due to plagiarism, and notably omitted crucial data showing that 10 countries with lower vaccine uptake had more deaths, while 10 with higher uptake had fewer deaths.

4. The Skidmore study relied on unreliable self-reported survey data of ~2800 individuals and suffered from significant overestimation of rare events when extrapolated to population level.

5. The Pantazatos and Seligmann preprint, has been languishing without peer review since 2021, represents a masterclass in methodological errors. Beyond failing to control for COVID-19 rates and seasonal variability, they made unsupported claims about spike protein cytotoxicity and attempted to validate their findings by "herding" their results toward another flawed paper. They misused European data to extrapolate US findings and proposed Antibody Dependent Enhancement (ADE) as an explanation for observed mortality rather than considering simpler explanations like waning immunity. Most tellingly, they could not explain why their claimed increased mortality only appeared in European data during weeks 0-5 post-vaccination and not thereafter.

6. Another Hulscher and McCullough study cited hasn’t made it past peer-review (no surprises here). The study looks at increase in cardiopulmonary events and cardiopulmonary arrest mortality and tries to connect it to vaccination in King County, WA between 2020-2023, all while completely disregarding the fact that Covid-19, which was spreading uncontrollably, was likely the major contributor to these excess events.

7. The Aarstad and Kvitastein study was all over the place and is not peer-reviewed. Initially led by two authors, it submitted an update as a pre-print (it has not gone through peer-review either) in August 2023 with one of the authors (Aarstad) refuting its prior conclusion. Initially Aarstad and Kvitastein had concluded that all-cause mortality in 31 European countries in 2022 increased over time the higher the 2021 COVID-19 full vaccination uptake. The update presents a contradiction: initially, full COVID-19 vaccination (in 2021) was associated with a decrease in all-cause mortality from January to March 2022, but over the following 14 months, booster vaccinations were linked to an increase in mortality. The studies suffer from the same shortcoming as the rest. There is an ecological fallacy by looking at broad population-level data and doesn’t actually verify the vaccination status of those that die. It associates excess mortality in 2022 to higher vaccination uptake in 2021, without considering the total number of reported Covid-19 cases, hospitalizations and deaths, as well as not doing proper subgroup analysis to look at age, health status and exposure levels.

8. The methodology grows even more questionable with the Alessandria study, which fundamentally misclassified its study groups and made a series of problematic assumptions. The authors began with the demonstrably false premise that excess deaths in 2021 and 2022 couldn't be explained by the ongoing pandemic - a statement that ignores the reality of continued COVID-19 waves. They explicitly chose to ignore COVID-19 deaths because they deemed them "overestimated," a convenient assumption that allowed them to attribute deaths to vaccination instead. The "unvaccinated" category incorrectly included people who later received vaccination, creating a serious methodological flaw. The study failed to analyze whether deaths occurred from COVID-19 itself and made no attempt to account for the timing of deaths relative to vaccination. Most critically, it ignored underlying health differences between vaccinated and unvaccinated populations, a basic requirement for any meaningful comparison.

9. We could not find the cited Lancaster study, so we cannot comment.

10. The Allen study was a statistical analysis of the correlation between excess deaths and the rollout of booster vaccinations in Australia and did not prove causation between the two. The author used regression analysis to examine excess deaths and total vaccinations in the first quarter of 2023. This method just fits a line to data points, measures correlation, and determines whether the observed association is statistically significant or not, but does not establish any causation. Because vaccination campaigns followed surges in disease that led to excess deaths a correlation is expected.

11. Kuhbander and Reitzner claim COVID-19 cannot explain the increase in excess mortality after vaccinations began, which is false. With newer variants that were more transmissible, and often deadly, such as the Delta and Omicron, the number of cases and, thus deaths rose significantly at the end of 2020 and into early 2021, and then later that year. It is also estimated that due to the availability of at-home tests the number of cases during the Omicron wave were vastly underreported. Furthermore, the first vaccines were very effective at preventing infection for the ancestral strain, but the combination of variants and waning immunity meant large segments of the population remained unprotected during these waves.

12. The Rodrigues and Andrade study had significant methodological concerns. In fact, a statement was even issued by the Technical-Scientific Monitoring Committee for Initiatives Related to COVID-19 Vaccines in Brazil. While the authors claimed thatCOVID-19 vaccination nearly doubles the risk of death from all causes after one-year post-COVID infection, The study has significant limitations due to its uncontrolled observational design, inadequate use of Brazil's national health database, and lack of statistical significance. It fails to reference other relevant studies and does not account for key confounding factors such as socioeconomic status, symptom onset, and vaccination timing. Additionally, the database was not designed to test long-term mortality differences, and the study lacks robust statistical methods needed to establish causality.

The "Negative Efficacy" Narrative: A Statistical Stumble

The paper's claims about "negative efficacy" represent another fundamental misunderstanding of vaccine epidemiology and surveillance data. The authors confuse waning immunity with negative efficacy, failing to grasp basic principles of how vaccine protection changes over time. Their analysis ignores crucial testing patterns: vaccinated people, particularly healthcare workers and high-risk individuals, underwent more frequent testing, creating artificial signals of higher case rates in vaccinated populations.

This misinterpretation becomes even more apparent when examining their treatment of confounding factors. The authors fail to account for the evolution of new variants that partially escape immunity, changes in behavior among vaccinated individuals, and variations in healthcare access and testing availability. Perhaps most tellingly, they ignore seasonal variations in transmission patterns that impact case rates independently of vaccination status.

Their most fundamental error lies in their misunderstanding of immunity itself. When they observe higher infection rates in vaccinated individuals months after vaccination, they incorrectly conclude the vaccines are harmful rather than recognizing the natural waning of immunity - a phenomenon well-documented with many vaccines and easily addressed through boosting. They also did not factor in timing of COVID-19 vaccination in the vaccine group, or infection in the unvaccinated. Failure to do this leads to clear confusion between waning protection and negative efficacy and it betrays a deeper misunderstanding of how vaccines work over time.

It is also critical to remind you that vaccines work and are critically important even if they do not prevent infection. We have significant amounts of data (examples can be found here, here, here and here) that outlines the reduced risk of complications such as long COVID and autoimmune disease following infection in those vaccinated.

The DNA Contamination Claims: When Technical Analysis Loses the Plot

The paper's DNA contamination analysis fundamentally misunderstands both testing methodology and regulatory standards. The authors cite McKernan's group in two studies. Both of which are highly problematic. First, one of McKernan’s papers reported widely fluctuating DNA concentrations depending on testing method:

• 2250-3390 ng via electrophoresis

• 312-843 ng via fluorometer

• 12 ng via qPCR

Instead of investigating these massive discrepancies, the authors cherry-picked the highest numbers. They ignored that when tested with proper qPCR techniques - the regulatory standard - contamination levels remained well below the FDA's 10ng threshold for biologics. They also tried to state that higher levels of DNA contamination was associated with more reported adverse events - but the graph they showed actually demonstrated the opposite effect.

The second McKernan paper cited has some very concerning issues that should have resulted in the samples not being used. In this paper the study authors themselves noted

“A limitation of this study is the unknown provenance of the vaccine vials under study. These vials were sent to us anonymously in the mail without cold packs. RNA is known to degrade faster than DNA and it is possible poor storage could result in faster degradation of RNA than DNA.”

This alone is a full stop. If the origin of the vials is unknown, there is no way to verify their handling, storage conditions, or potential contamination. Without a clear chain of custody, any findings from these samples lack scientific validity and cannot be used to draw meaningful conclusions. Furthermore, if the vials were not stored properly then RNA degradation almost certainly happened because RNA can quickly degrade when conditions are not carefully controlled. This is why cold storage is so important for the mRNA vaccines. This information makes any comparison between RNA and DNA levels unreliable.

Many of the other references included in this section also relied on non-regulatory and unvalidated techniques for quantifying DNA. Furthermore, the authors make the claim that the DNA contamination is automatically harmful. However, they ignore the fact that it’s actually very hard for DNA from outside sources to get in and do anything. This is why there is a whole field of research looking into ways of gene-editing to help treat and cure genetic diseases. It is not easy to get foreign DNA into the cells and then actually into the person’s DNA. This is because our body has many safety mechanisms to kill cells that have lots of DNA in their cytoplasm. Normally, DNA lives in the nucleus of the cell. When cells detect DNA outside of that nucleus it triggers alarms that result in destruction of the cell. This happens all the time without us knowing it, for example when we are infected with DNA viruses.

Their analysis ignores that regulatory agencies including the FDA, EMA, and WHO set strict limits on DNA residues, which mRNA vaccines must comply with to receive authorization. The presence of trace DNA fragments is expected in biologics and poses no known safety risks when within these established limits.

The Mortality Analysis: Confounding Factors and Causation Errors

The paper's mortality analysis systematically ignores critical factors affecting death rates during the pandemic. The authors disregard direct COVID-19 mortality, healthcare system disruptions, delayed medical care, demographic changes, and shifts in cause-of-death reporting practices. They fail to account for economic impacts and mental health crises that affected mortality rates during this period.

Perhaps most tellingly, they misinterpret excess mortality data while ignoring a crucial pattern: every high-income country showed consistent declines in excess mortality following vaccine rollout. The authors disregard extensive evidence from the CDC, UK ONS, WHO, and Our World in Data demonstrating that excess deaths were driven by COVID-19, not vaccination. Their analysis completely misses what epidemiologists call the "healthy vaccinee" paradox: because older and medically vulnerable populations were prioritized for vaccination, raw death numbers might appear higher in vaccinated groups even while risk-adjusted mortality is lower. This represents a fundamental misunderstanding of how vaccination programs are implemented, with highest-risk populations receiving vaccines first.

Most critically, they mistake correlation for causation, attributing all post-vaccination mortality to vaccines while ignoring the ongoing pandemic effects and the complex interplay of factors affecting death rates during this unprecedented period.

Overwhelming Evidence Supporting Vaccine Safety and Efficacy

A robust body of scientific evidence contradicts the paper's claims:

1. A systematic review and meta-analysis by Huang & Kuan (2022) analyzing 1.3 million participants found vaccinated individuals had a dramatically lower risk of severe illness (risk ratio: 0.12, 95% CI: 0.04–0.36).

2. A meta-analysis of 54 studies (Rahmani et al., 2022) showed:

   • 71% reduced infection risk after one dose

   • 87% reduced infection risk after two doses

   • 73% reduced hospitalization after one dose

   • 89% reduced hospitalization after two doses

3. A CDC cohort study of 11 million individuals (Xu et al., 2021) found lower non-COVID-19 mortality rates among vaccinated individuals.

4. A Hungarian nationwide cohort study (Pálinkás & Sándor, 2022) reported reduced all-cause mortality among vaccinated individuals during the Delta wave (49-75% effectiveness depending on vaccine type).

5. A retrospective study of myocardial infarction patients (Gupta et al., 2023) found vaccinated individuals had:

   • 42% lower risk of death at 30 days

   • 46% lower risk at six months

Real-world data consistently shows that vaccine rollout aligned with dramatic reductions in COVID-19-related deaths and hospitalizations, even as new variants emerged.

Conclusion

As we emerge from the shadow of a global pandemic, the stakes could not be higher. We're witnessing the consequences of eroded trust in public health measures, from resurging measles outbreaks to pockets of preventable COVID-19 hospitalizations. The paper we've analyzed represents more than just poor science – it exemplifies a broader pattern of manufactured doubt that threatens decades of progress in public health.

The scientific community must continue to uphold rigorous standards while acknowledging that absolute certainty is rare in medicine. We must recognize that behind every skeptical question about vaccines lies a legitimate desire to protect oneself or loved ones. Our response cannot be dismissal, but rather engagement with evidence, transparency about what we know and don't know, and a commitment to distinguishing between reasonable uncertainty and manufactured misinformation.

The COVID-19 vaccines have given us a powerful tool to prevent severe illness and death. They represent not just a scientific achievement, but a testament to what we can accomplish when global expertise and resources align toward a common goal. As we face future public health challenges, from emerging variants to new pathogens, our success will depend on our ability to maintain public trust through honest, evidence-based communication while forcefully countering deliberate distortions of scientific truth.

Stay Curious,

Unbiased Science, Nini and The Brain, Science Whiz Liz