Expert Biostatistician Debunks Flawed Study Linking Vaccines to Chronic Illness

At a recent Senate hearing discussing the integrity of scientific research related to vaccines, an unpublished study gained significant attention for its bold claim that vaccines increase the risk of chronic diseases in children. The study, conducted in 2020 by researchers at Henry Ford Health in Detroit, suggested that vaccinated children are more likely to develop long-term health issues. These findings are now part of public discussions and are supposed to be featured in a forthcoming documentary titled "An Inconvenient Study."

The study analyzed medical records of approximately 18,500 children born between 2000 and 2016, of whom about 16,500 had received at least one vaccine, and around 2,000 were completely unvaccinated. Researchers concluded that vaccinated children had 2.5 times higher incidence of "any chronic disease" compared to unvaccinated children, with some specific conditions showing 3 to 6 times higher rates. Despite these claims, a seasoned biostatistics expert from the University of Pennsylvania, Jeffrey Morris, pointed out fundamental flaws in the study’s design that undermine its conclusions.

The primary issue lies in the study’s methodology. The researchers followed vaccinated children longer and into older ages than unvaccinated kids, resulting in a surveillance bias where more diagnoses are simply recorded because these children were monitored for a longer period. The study’s authors attempted to adjust for this bias by analyzing subgroups with extended follow-up, but the bias persisted due to uneven follow-up durations—vaccinated children remained under observation for a longer span, leading to more opportunities for diagnosis.

Furthermore, the study’s data depended solely on Henry Ford’s medical records. Children who sought care elsewhere, especially after infancy, potentially had diagnoses outside the studied system. This reliance introduces detection bias, as children with more frequent doctor visits—more common among vaccinated children—are more likely to be diagnosed with chronic conditions. Indeed, vaccinated children averaged about seven doctor visits annually, compared to only two for unvaccinated children, skewing the data towards higher recorded health issues in the vaccinated group.

The statistical comparison is also complicated by confounding variables. The vaccinated and unvaccinated groups differed in several characteristics from birth—such as gender, race, birth weight, and maternal health—many of which influence health outcomes. Although adjustments were made, many other factors like socioeconomic status, living environment, and environmental exposures were not accounted for, making it difficult to isolate the effect of vaccines.

In addition, the follow-up period was inconsistent and often too short to detect conditions that typically manifest after age 5, like asthma, ADHD, or learning disabilities. Unvaccinated children were not tracked as long as vaccinated children, which further complicates the comparison and introduces surveillance bias.

Henry Ford Health issued a statement clarifying that the study was not published initially because it failed to meet the rigorous scientific standards required by a leading research institution. As Morris explains, strong scientific conclusions require robust methodology that accounts for such biases and confounding factors. As it stands, the study’s findings are not reliable evidence of a causal link between vaccines and chronic diseases.

In summary, this case exemplifies why careful study design, appropriate follow-up, and thorough statistical analysis are critical. While questions about vaccine safety are vital, claims based on flawed research can mislead the public. Experts emphasize the importance of evaluating scientific claims critically and relying on well-designed studies to inform health policies.

This analysis underscores the necessity of skepticism towards sensational claims and the value of rigorous scientific standards in public health research.