New study contained methodological limitations and failed to show that COVID-19 vaccines increase disease burden, despite claims by Patrick Flynn

In a Facebook post from February 2025, chiropractor Patrick Flynn claimed that people vaccinated against COVID-19 were “sicker” than those who were unvaccinated. He cited a December 2024 study by Harald Walach and Rainer Klement as evidence, alleging that vaccinated individuals had more doctor visits, higher COVID-19 rates, and increased musculoskeletal issues. Flynn specifically blamed mRNA COVID-19 vaccines for these effects.

Science Feedback previously debunked several inaccurate claims by Flynn about vaccination. We also reported that Walach and Klement previously published a COVID-19 vaccine safety study that was later retracted due to serious methodological flaws.

Flynn’s latest claim is also misleading. Not only does it misrepresent the study’s findings, but the study itself was performed in a way that introduced several risks of bias, thereby undermining the reliability of its conclusions about vaccine safety.

What did the study do?

The researchers investigated whether people who received COVID-19 vaccines reported more health issues than those who didn’t. They surveyed about 1,050 volunteers in Germany, asking them about their vaccination status, medical conditions over the past two years, and personal beliefs. Roughly 80% of respondents were vaccinated, while 20% were not.

According to the survey results, vaccinated individuals were more likely to report health problems. About 42% of vaccinated respondents said they had seen a doctor for a medical issue in the past two years, compared to 30% of unvaccinated individuals, a difference that was deemed statistically significant.

They also reported slightly higher rates of COVID-19, with around 30% of vaccinated respondents saying they had been infected, compared to about 23% of unvaccinated respondents.

The study found that 21% of vaccinated participants mentioned musculoskeletal issues, compared to 15% of the unvaccinated group. On average, vaccinated respondents reported experiencing about 1.6 health issues, while for unvaccinated individuals, this figure was around 1.3.

The authors thus concluded that vaccinated people experienced more health issues overall, though they acknowledged that further studies using clinical data would be needed to confirm or refute their findings.

Flynn misrepresented the study’s results

Flynn’s claim that the study demonstrated a higher risk of COVID-19 and musculoskeletal issues in vaccinated individuals isn’t supported by the study’s own data.

The authors of the study set a statistical significance threshold at a p-value of 0.01. In the context of this study, the p-value measures how likely it is that an observed difference between vaccinated and unvaccinated individuals can be explained by chance alone. The smaller the p-value, the less likely that the difference observed can be explained by random chance. Conversely, this indicates that other factors apart from chance are more likely explanations for the observed difference. That said, a p-value alone cannot determine what the causes of the difference are.

A p-value above the threshold of significance suggests that the observed differences could very well be due to chance. In this study, the reported p-values for both COVID-19 (0.0535) and musculoskeletal issues (0.059) were above the 0.01 threshold, meaning the differences weren’t statistically significant. Simply put, this indicates that the difference that the researchers observed in the risk of COVID-19 and musculoskeletal issues could simply be due to chance alone.

While statistical significance can help determine whether an observed difference is meaningful rather than the result of chance, it doesn’t address other potential flaws in how the data was collected or analyzed. Even when a result is statistically significant, biases in the study design or data quality could still affect its validity. This is what we’ll address in the next section.

Study’s methodological limitations prohibit drawing conclusions about vaccine safety

The study has several important limitations that may distort its results, meaning it can’t be used to support claims that COVID-19 vaccines are unsafe.

One major issue is selection bias, which occurs when the way participants are recruited leads to differences between groups that make the results unreliable. Unlike a randomized clinical trial, where people are randomly assigned to vaccinated and unvaccinated groups, this study relied on voluntary participation in an online survey.

This means that vaccinated and unvaccinated people may have had different reasons for taking part in the study. For example, people who believe they experienced side effects from vaccination could have been more motivated to participate, while unvaccinated individuals who distrust medical research could have been less likely to take part. As a result, the two groups aren’t necessarily comparable, and any differences in health outcomes between them could be influenced by these underlying differences rather than vaccination itself.

Another issue is recall bias, which occurs when people don’t accurately remember or report past health events. Since the study relies on self-reported health issues over a two-year period, it is highly susceptible to this bias. For instance, people may simply forget some medical problems they experienced, especially if they were mild or occurred long ago.

However, the tendency to recall and report past health issues may not be the same for vaccinated and unvaccinated individuals. Vaccinated people, for example, may be more health-conscious and more likely to notice and remember medical issues. On the other hand, individuals with anti-vaccine views, who are more likely to be in the unvaccinated group, might downplay or misattribute symptoms that could have been due to an illness that vaccination would have prevented^[1].

Since the study didn’t verify these reports with medical records, there is no way to confirm whether the reported health issues accurately reflect reality. This makes it unclear whether the differences in reported health outcomes are due to actual differences in medical events or simply differences in how participants recall and report their health histories.

A third major issue is confounding bias, which happens when factors other than vaccination status influence health outcomes. In this study, the vaccinated and unvaccinated groups weren’t comparable—they differed in several important ways that could affect health risks independent of vaccination status.

According to Table 1, vaccinated individuals were generally older, had a higher body mass index (BMI), and took more medications than unvaccinated individuals. These differences were statistically significant, meaning they were unlikely to be due to chance. Age, excess weight, and pre-existing medical conditions are all well-known risk factors for a range of health issues, including more frequent doctor visits, hospitalization, and cardiovascular problems. The fact that the vaccinated group had more of these risk factors from the start means that their higher rates of reported health issues may have nothing to do with vaccination itself.

Finally, another factor that could skew the results is healthcare-seeking behavior. Vaccinated individuals may be more proactive about their health, meaning they may visit doctors more often, get tested for COVID-19 more frequently, or undergo routine screenings that lead to earlier diagnoses of medical conditions. If unvaccinated individuals avoid doctors or medical tests, some health issues in that group may go undetected and underreported, creating the illusion that they are healthier.

Because these differences weren’t fully accounted for in the study, the reported health disparities between vaccinated and unvaccinated individuals could simply reflect underlying differences in age, health status, and healthcare engagement—rather than the effect of vaccination.

Conclusion

Flynn misrepresented the study’s findings, particularly in claiming that vaccinated individuals had a higher risk of COVID-19 infection and musculoskeletal issues—outcomes that weren’t statistically significant in the study itself.

The study by Walach and Klement doesn’t provide reliable evidence that COVID-19 vaccines increase the risk of health problems. By contrast, other studies didn’t report a lower overall health following COVID-19 vaccination^[2].

While the authors reported that vaccinated individuals experienced more medical issues, their findings relied on self-reported survey data, which is prone to bias and lacks medical verification. Differences in age, pre-existing health conditions, and healthcare-seeking behavior may have influenced the results, but the study’s design didn’t adequately control for these factors.

REFERENCES

• 1 – Salmon et al. (2005) Factors associated with refusal of childhood vaccines among parents of school-aged children: a case-control study. Archives of pediatrics & adolescent medicine.
• 2 – Elisabeth O’Regan et al. (2024) A register and questionnaire study of long-term general health symptoms following SARS-CoV-2 vaccination in Denmark. Vaccines.