The Appointment of David Geier: A Predictable Breach of Scientific Integrity in Vaccine-Autism Research

How Placing an Unqualified Science Denialist in Charge of Public Health Research Endangers Us All

This analysis is co-authored by a multidisciplinary team of experts with extensive training in public health, data science, clinical medicine, immunology, virology, and science policy —scientists committed to evidence-based research and safeguarding public health through rigorous scientific standards.

Introduction

The appointment of David Geier by the U.S. Department of Health and Human Services to analyze data for a federal study examining potential links between vaccines and autism—during an active measles outbreak in multiple states—represents not just poor judgment but a dangerous breach of scientific integrity that threatens public health. This decision undermines decades of rigorous scientific research that has consistently found no causal relationship between vaccines and autism spectrum disorders and violates fundamental principles of scientific methodology and medical ethics.

Before we tell you about why this particular choice is so alarming based on David Geier’s history, we should let you know in advance what this study is going to show– because it’s very predictable, as the anti-vaccine movement has been following this formula for decades. Let’s discuss…

The Dangerous Endgame: Playing Out the Logical Conclusion

Scientific research begins with a falsifiable hypothesis — a prediction that can be definitively supported OR rejected based on collected data. The scientist must be willing to look at the data generated and throw out the hypothesis if the data does not support it. The data cannot be manipulated to fit that hypothesis – otherwise, it is not real science; it's just sparkling confirmation bias.

This means that how the data is collected and interpreted really matters. When doing human subject research, you want the two groups you are comparing to be as similar as possible in distribution for factors like gender, age, chronic health conditions, socio-economic factors, etc. When this isn’t possible, the researcher must adjust and account for these differences or other potential confounders. If this is not done correctly, the data are not reliable.

Following well-established scientific processes is crucial for objective, reliable, and repeatable research, ensuring that conclusions are based on evidence and not personal biases. This is particularly important in the context of science that informs government decisions that impact public health and safety. Because science is so integral to informing policy decisions, federal agencies have developed scientific integrity policies ensuring the trustworthiness of research, protecting against bias and misconduct, and fostering a culture of open and transparent science. The Centers for Disease Control and Prevention’s (CDC) scientific integrity policy recognizes that scientific information “be developed under and subjected to well-established scientific processes, free from inappropriate interference that undermines impartiality, nonpartisanship, or professional judgment."

The appointment of a known pseudoscientist with a clear anti-vaccine agenda to lead this study raises profound concerns about upholding scientific integrity and the reliability of the study's outcome. Given David Geier's history of methodologically flawed research and predetermined conclusions, there is a high probability that this study will be designed specifically to "find" a link between vaccines and autism where none exists.

Should such a methodologically compromised study claim to identify a vaccine-autism connection, the consequences would be catastrophic for public health:

1. Undermining Vaccine Confidence: Even a single government-backed study claiming a vaccine-autism link would create devastating confusion among parents, potentially triggering widespread vaccine hesitancy during an active measles outbreak.

2. Regulatory Uncertainty: Such findings could prompt calls to rescind vaccine approvals or modify vaccination schedules, creating regulatory chaos and threatening decades of carefully established immunization protocols. Since insurers, including Medicare/Medicaid and the Vaccines For Children program, are required to cover vaccines recommended on the official schedule, disruptions could jeopardize coverage, access, and availability.

3. Legal and Political Implications: Conclusions based on misinterpreted data could fuel a new wave of vaccine litigation and provide political cover for scaling back vaccine requirements in schools and other settings.

4. Resource Diversion: Precious public health resources would be diverted to addressing a manufactured crisis rather than focusing on the genuine causes of autism and providing support to autistic individuals and their families.

5. Stigmatization: This approach inherently stigmatizes autism by framing it as a vaccine "injury" rather than a natural variation in neurodevelopment, harming autistic individuals and their families.

6. Undermining Public Trust in Federal Agency Science: Public trust in federal agency science erodes when research is not independent, objective, and transparent.

The timing of this appointment coincides with an active measles outbreak in multiple states and elevated measles cases worldwide—a public health emergency that requires clarity and evidence-based messaging. If this study produces misleading results suggesting a vaccine-autism link, the immediate impact could be devastating: parents may delay or refuse vaccinations during an active outbreak, directly contributing to the spread of a highly contagious disease that can cause pneumonia, encephalitis, and death. The potential for preventable disease outbreaks to expand exponentially in undervaccinated communities creates an urgent public health risk that makes this appointment not just scientifically indefensible but immediately dangerous.

A Case Study in Flawed Research: The 2017 Geier Thimerosal Study

Remember how we said a scientist should be willing to throw out a hypothesis if the data does not support it? Well, here is a reason that perfectly exemplifies why we can’t trust Geier to do this…

The Geiers published a 2017 paper in the Journal of Trace Elements in Medicine and Biology. An examination reveals critical methodological issues that would never pass rigorous scientific scrutiny:

1. Biased study design: While claiming to be "hypothesis-testing," they selected an arbitrary follow-up period for controls (8.05 years) based on a formula they created (mean age of autism diagnosis plus twice its standard deviation), with no epidemiological justification for this specific cutoff.

2. Cherry-picked endpoints: They specifically examined "atypical autism" (ICD-9 code 299.80) rather than all autism spectrum disorders, meaning the results are not generalizable to all children with autism.

3. Temporal mismatch: Their cases (n=164) were born 1991-1998, while controls (n=15,216) were born 1991-1992, creating a fundamental temporal bias. This period spanned significant changes in diagnostic practices and awareness of autism, which they failed to account for. This also ignored that there were major changes to the uptake of hepatitis B vaccines, as the universal recommendation was made in 1991. Between 1991 and 1998, the vaccination rate went from 10% to 90%. This artificially causes a major increase in the chance of your children with autism being vaccinated and the control group (children without autism) being unvaccinated. In other words, the choice of time periods creates an association that isn’t real.

4. Statistical manipulation: The authors report three separate analyses for exposure within the first month, first two months, and first six months of life – creating multiple opportunities to find a significant association by chance alone, without appropriate statistical correction for multiple comparisons.

5. Implausible odds ratios and absent confidence intervals: They report odds ratios like "indeterminate" with confidence intervals of "11.4-indeterminate" and "10.0212-indeterminate" – statistical impossibilities that suggest data manipulation. In legitimate research, confidence intervals have defined upper and lower bounds, and they tell us how certain we are of the results. Reporting an "indeterminate" upper bound is statistically impossible (every calculation has a numerical answer) and suggests fabrication or manipulation of data. Not providing proper confidence intervals makes a point estimate meaningless, as it fails to capture how accurate and reliable the findings are—essentially signaling the results cannot be independently verified or reproduced by other scientists; a foundational principle of science is that reliable findings must be reproducible by independent recreation of a study’s analysis.

6. Inconsistency in basic data: In Table 1, for their third statistical test, they report 18 cases and 358 controls with "37.5 μg Hg [mercury] within first 6 months" and 0 cases and 928 controls with "0 μg Hg within first 6 months." Yet, in Table 4, they show a completely different distribution with 18 (10.98%) cases and 358 (2.35%) controls receiving 37.5 μg Hg, inconsistent with their other reported numbers.

7. Circular reasoning: They claim biological plausibility for their findings by citing their own previously published (and widely criticized) papers, creating a closed loop of self-citation. Science works by following the consensus of data- it must be reproducible by other investigators to be meaningful. If the only evidence you can find is what you generate and your results can’t be replicated, it’s far more likely that you are wrong than that everyone else is.

8. Misrepresented funding sources: While claiming funding from "non-profit" organizations (Institute of Chronic Illnesses and CoMeD), they fail to disclose that these are their organizations operated from their private residence. Failure to disclose funding sources and associated conflicts of interest is considered among the most serious ethical breaches a researcher can commit.

This paper demonstrates the Geiers' fundamental approach: beginning with a predetermined conclusion and working backward through selective analysis, arbitrary cutoffs, inconsistent numbers, and inappropriate comparisons to manufacture their desired outcome rather than following evidence where it leads.

The authors' willingness to publish such fundamentally flawed research – which contradicts dozens of properly conducted large-scale epidemiological studies – reveals why allowing them to lead a federal study is scientifically indefensible and a breach of public trust.

Furthermore, a large mountain of evidence looking at thimerosal and ASD shows a stable or even inverse correlation, where removal was associated with either stable rates or even higher case rates:

• Sweden/Denmark (1990-2000): Autism rates rose after thimerosal was removed from vaccines in 1992, despite stable rates during its use.

• Denmark (1961–2000): No link between thimerosal dose (0–200 μg) and autism incidence.

• Canada (Quebec, 1987-1998): Thimerosal exposure did not correlate with autism diagnosis.

• Denmark (1990–1996): No differences in autism risk between children who received thimerosal-containing or thimerosal-free vaccines. It also found that the rates of autism increased after the removal of thimerosal from all vaccines.

• USA (1991-1999): No relationship between thimerosal exposure and autism in over 140,000 children.

• UK (1988–1997): No link between thimerosal exposure and autism in over 100,000 children.

• California (2008): Autism rates rose in California even after thimerosal was removed from vaccines.

How To Lie With Statistics

Given David Geier's history of methodologically flawed research, we have deep concern that similar basic scientific errors may lead to an erroneous conclusion. Even a study demonstrating no link between vaccines and autism would be useless if the methodologic errors above were present.

Cherry-picking data: By selectively analyzing only certain subgroups or time-periods that show the desired outcome while ignoring others, researchers can manufacture associations that don't truly exist.

Improper Classifications or Binning: Misclassification means putting data in the wrong category, and improper binning means grouping data into ranges (bins) in ways that mislead. While these techniques don't change the raw data, they change how we interpret it by either washing away trends or exaggerating them.

Multiple comparisons without correction: When researchers make dozens or hundreds of statistical comparisons without proper adjustment, false positives are guaranteed. For example, if 100 different analyses are run at a standard p-value of 0.05 (accepting a 5% chance of false positives), approximately 5 comparisons will appear significant by chance alone.

Ignoring critical confounding variables: Studies that don't account for factors that influence both vaccination status and autism diagnosis can create artificial associations. A prime example is healthcare access – children with better access to healthcare are both more likely to be vaccinated and more likely to receive an autism diagnosis when appropriate.

Inadequate sample sizes: Studies with small sample sizes have limitations in both directions - they may miss real effects (Type 2 errors) but can also produce misleading associations, especially when using biased sampling methods (as in Wakefield's case). The extensive body of large-scale studies examining millions of children has consistently found no link between vaccines and autism, providing much more statistical power and reliability than smaller studies.

Misinterpreting correlation as causation: The classic error of assuming that because two things happen around the same time (autism diagnosis often occurs around the same age as several routine vaccinations), one must cause the other.

Statistical significance without clinical significance: As a sample size grows larger, the estimates of differences between the subpopulation of the sample become increasingly precise, meaning that it is easier to show that a difference is significant. However, this means even very small effect sizes can be made to look meaningful when they aren’t. A large sample size does not undo the effects of confounding variables and does not automatically show a causal link.

Lying about the data: This is the most extreme form of statistical manipulation (if it can even be called that), but it has happened before. For example, one of the early trials of ivermectin in COVID-19 claimed to show a massive reduction in mortality with ivermectin- but it was riddled with manipulated and, in some cases, invented data, including counting deaths that occurred before the study time-period and plagiarism in the text. While we do not know of instances in which the Geiers have done this particularly, the anti-vaccine lobby as a whole is not concerned with being honest brokers (as evidenced by CHD staging a website impersonating the CDC but with anti-vaccine duplicities throughout), and you should fully expect that this is possible. Methods to detect faked data exist, but they are extremely labor-intensive and require specialized expertise.

These methodological flaws can transform normal statistical noise into the appearance of a meaningful signal. That's why the scientific consensus against a vaccine-autism link is based on multiple large-scale studies using diverse methodologies reproduced across different populations – all consistently reaching the same conclusion. Even if all the methodologic concerns above were not present, a single large-scale study would not negate the mountains of evidence already present in the minds of trained scientists and clinicians.

Problematic Background, Qualifications, & Questionable Ethics

David Geier lacks the appropriate scientific and medical credentials for this role. While he earned a Bachelor of Arts degree from the University of Maryland, Baltimore County in 2002, he possesses no medical training nor advanced scientific degrees, and was disciplined by Maryland regulators in 2011 for practicing medicine without a license, resulting in a civil fine of $10,000. Both David Geier and his father, Mark Geier, have published papers claiming vaccines increase autism risk—assertions repeatedly discredited by rigorous scientific scrutiny.

The Maryland State Board of Physicians suspended Mark Geier's medical license for putting autistic children at risk through an experimental protocol involving Lupron, a powerful hormone-suppressing drug approved for prostate cancer and precocious puberty but not for autism. Their treatment approach was based on an unsupported theory connecting mercury, testosterone, and autism—a link explicitly rejected by the Institute of Medicine. The Geiers have also been criticized for failing to provide adequate informed consent to patients regarding experimental treatments and for misdiagnosing precocious puberty in autistic children to justify administering Lupron therapy.

Equally concerning are questions about the Geiers' institutional review board (IRB), which reportedly consisted of family members and business associates—a clear conflict of interest that violates established ethical standards for human subjects research. This arrangement fails to provide the independent oversight essential for protecting research participants.

Recognized Science Denialists

In 2010, the prestigious scientific journal Nature Medicine identified Mark and David Geier as prominent "science denialists"—individuals who reject scientific consensus in favor of radical and unsupported claims. The journal noted they were "among the first to publish claims that the thimerosal preservative used in certain vaccines causes autism" and described their promotion of unproven and potentially harmful treatments for autism.

Their placement alongside AIDS denialists and other fringe figures highlights how thoroughly the scientific community has rejected their methodologies and conclusions. That such individuals have now been granted authority over federally funded research represents an alarming development for the welfare of the public.

These additions would provide concrete examples of the Geiers' problematic research approaches, address the rhetorical strategies used to justify the study, and emphasize how thoroughly their work has been rejected by the scientific community.

Methodology Matters: Revealing the Hidden Flaws

Multiple scientific authorities have systematically criticized the Geiers' research. The Institute of Medicine specifically concluded that their studies have "serious methodological flaws, and their analytic methods were nontransparent, making their results uninterpretable and therefore noncontributory with respect to causality." Multiple courts have also questioned the Geiers’ expertise, with one declaring him "a professional witness in areas for which he has no training, expertise, and experience."

Medical journals have retracted the Geiers' work due to serious concerns about methodology, data collection, and conflicts of interest. In one case, the journal Autoimmunity Reviews retracted a paper by the Geiers following detailed criticism of their research methods and institutional review board structure. Similarly, in 2015, the journal Science and Engineering Ethics retracted a paper co-authored by the Geiers that claimed conflicts of interest among public health officials, citing errors and the Geiers' failure to disclose their conflicts of interest.

Overwhelming Scientific Consensus Against Vaccine-Autism Link

The scientific consensus on vaccines and autism is clear and has been established through numerous large-scale, methodologically rigorous studies conducted over several decades:

1. A comprehensive 2004 review by the Institute of Medicine examined the hypothesis that vaccines, specifically the measles-mumps-rubella (MMR) vaccine and thimerosal-containing vaccines, were associated with autism. It concluded that there was no evidence supporting a causal relationship.

2. A decade-long study of over half a million children in Denmark published in 2019 showed the MMR vaccine does not increase the risk of autism, providing robust statistical evidence reinforcing what was already medical consensus.

3. A 2014 meta-analysis examining studies involving millions of children found no relationship between vaccination and autism or autism spectrum disorders and no relationship between any vaccine components and autism.

4. Multiple studies examining the effects of thimerosal, a mercury-containing preservative previously used in vaccines, have found no association with autism, despite this being the specific focus of the Geiers' claims.

5. Despite the removal of thimerosal from childhood vaccines by 2001, autism rates have continued to rise, further contradicting the hypothesis that this preservative contributes to autism.

The overwhelming nature of this evidence cannot be overstated. Studies using every major epidemiological design—from systematic reviews and meta-analyses to large cohort studies and case-control studies—have reached the same conclusion: There is no link between vaccines and autism. These findings have been replicated by different scientific teams worldwide using various methods and populations.

While autism diagnosis rates have increased over time, this is largely explained by broadened diagnostic criteria, increased awareness among healthcare providers, earlier diagnosis, diagnostic substitution (children previously diagnosed with other conditions now receiving autism diagnoses), and improved reporting systems—not environmental factors like vaccines.

The False Equivalency of "Just Asking Questions"

Asking questions is fundamental to science. Confidence in vaccines doesn’t come from blind faith—it’s built on rigorous research, exhaustive peer review, strict regulatory oversight, and continuous monitoring. Every step of the scientific process to show the effectiveness and safety of vaccines is scrutinized. But reopening long-settled questions, especially ones that have been repeatedly studied, tested, and confirmed, is not scientific inquiry. It’s science denial masquerading as curiosity.

Proponents of this study may argue that "more research is always a good thing" or that they're "just asking questions." This rhetoric ignores three fundamental issues:

First, this question has been asked and answered through dozens of large-scale, methodologically rigorous studies conducted across multiple countries, spanning decades. The scientific consensus against a vaccine-autism link is based on extensive evidence involving millions of children.

Second, those claiming previous studies are biased because researchers who "support vaccines" apply a blatant double standard. If they were genuinely concerned about bias, they should be equally outraged—if not more so—that someone with no relevant expertise, a documented history of disciplinary action, and a clear agenda against vaccines has been selected to lead this study.

Lastly, we have the fundamental problem of economics: we will always have unlimited wants and limited resources to address those wants. We have a finite amount of money to devote to scientific research, which means we have to be judicious about which studies get funded and which do not. Giving money to re-examine a link between vaccines and autism, which has been rejected repeatedly by multiple high quality studies across diverse populations, by unrelated researchers, and across multiple time periods necessarily means that another study doesn’t get that money; a study that could address far more important and worthy questions- like how to best support autistic children throughout their life.

Scientific inquiry requires rigorous methodology and a willingness to accept findings even when they contradict initial hypotheses—qualities conspicuously absent from the Geiers' work. A genuine commitment to scientific integrity would demand independent researchers with appropriate qualifications, not individuals with a documented history of producing methodologically flawed and heavily biased research that supports their predetermined conclusions.

Conclusion: Dire Prediction

The appointment of David Geier to conduct federal research on vaccines and autism represents a dangerous deviation from evidence-based policy-making and scientific integrity. His lack of appropriate qualifications, history of disciplinary action, and promotion of discredited theories make him unsuitable for this role. This decision not only undermines public health but also threatens to further stigmatize autistic individuals by perpetuating harmful misconceptions about autism's causes.

unbiasedscipod

A post shared by @unbiasedscipod

We call on scientific and medical organizations, autism advocacy groups, and concerned citizens to demand independent oversight of any federally funded vaccine research and to reject findings from studies conducted under such clear conflicts of interest and methodological concerns.

Stay Curious,

Unbiased Science (Dr. Jess Steier), Dr. Elisabeth Marnik, Dr. Aimee Pugh Bernard, Dr. David Higgins, Dr. Jacob Carter, Dr. David Higgins, Dr. Nini Munoz, Ed Nirenberg, and Dr. Jeff Dewey

P.S. Want to support our work? The best way is to subscribe to our Substack and share our content with your friends. While all our articles are always completely free to read (and always will be), paid subscriptions help sustain our in-depth reporting on public health topics. Thank you for considering it!