Peter McCullough & Andrew Wakefield: Vaccines, Autism, and the CDC

 

Table of Contents

McCullough and Wakefield Team Up:


On October 27, 2025, Peter McCullough joined forces with Andrew Wakefield to self-publish a lengthy report (Link) reviving the claim that childhood vaccines are a primary risk factor for autism spectrum disorder (ASD). For those who know their history, this shouldn’t be surprising.  McCullough, a cardiologist, gained notoriety opposing the mRNA vaccines during the COVID-19 pandemic (Link, Link) and then later for his opposition to the childhood vaccine schedule, and Wakefield became infamous for publishing a now-retracted paper in The Lancet linking the MMR vaccine with childhood autism in 1998 (Link). Wakefield’s Lancet article was retracted in 2010 for fraud, undisclosed financial interests, ethical violations, and scientific misrepresentations (Link). Following the initial worldwide scare created by Wakefield, numerous rigorous scientific studies were conducted around the world to see if a link between childhood vaccines and autism could be detected. No such link was ever established, despite very large-scale studies involving millions of children followed over many years.

Yet, here we are again back at square one.  It seems only natural, if not inevitable, that McCullough and Wakefield would join forces to revive the link between vaccines and autism. In their report, their central claim is that vaccines, particularly those containing aluminum adjuvants or ethyl mercury, cause or contribute to autism spectrum disorder (ASD). It asserts that over 100 studies support this idea and that only a handful find no link.

Too many vaccines:

The McCullough report claims that when a child gets too many vaccines, it can overwhelm their immune system. The report claims that this stress on the body, particularly on subcellular organelles called mitochondria, produces “mitochondrial stress”, which in turn increases the risk for autism. However, the overwhelming weight of scientific evidence doesn’t back up this notion. There’s no evidence that the large number of antigens (from bacteria, viruses, fungi, etc) that an infant or child’s body is exposed to every day or the small number of antigens in vaccines hurt the immune system in any way, since the human immune system is designed to handle enormous numbers of antigens on a daily basis.

In fact, the number of antigens in vaccines has gone down over the years, even while autism rates have gone up. In the 1980s, for example, the exposure to vaccine antigens was much higher, at ~3000 (mainly because the old whole-cell pertussis vaccine alone contained ~3,000 antigens), than it is today.

Vaccine Doses Related to ASD?

So, while McCullough is correct in saying that there were fewer vaccine “shots” or “doses” a few decades ago (see illustration), this is really irrelevant to the actual degree of antigen exposure or to the claimed increase in ASD with the increase in vaccinations since the 1970s and 80s. After all, the entire vaccine schedule today, from birth to 18 years of age, exposes children to fewer than 165 antigens total (because modern vaccines tend to use purified proteins, toxoids, polysaccharides, and recombinant technologies, which require far fewer antigens to generate strong immunity). This is far less than a baby gets from crawling on the carpet. The typical child gets exposed to between 2,000 and 6,000 antigens per day (Link). A minor respiratory infection may expose a child to 4,000–10,000 antigens in a single event. Most children following the recommended schedule will receive about 30 shots protecting against about 20 diseases by the time they turn 18. This is all by design. A baby’s immune system is designed to handle even billions of antigens simultaneously.  Even if all vaccines were given at once, the immune system would use <0.1% of its theoretical capacity

This clearly supports the conclusion that antigen overload isn’t the problem. And even though “mitochondrial stress” is a popular phrase right now, experts from Science Feedback explain that there’s no proof that vaccines trigger autism, even in people with mitochondrial disorders. They explain that infections, not vaccines, are the real danger for these children. Vaccines actually protect them by preventing fevers and other problems that could lead to brain injury.

“Children have an enormous capacity to respond safely to challenges to the immune system from vaccines,” says Dr. Offit. “A baby’s body is bombarded with immunologic challenges—from bacteria in food to the dust they breathe. Compared to what they typically encounter and manage during the day, vaccines are literally a drop in the ocean.” (Link)

“Given that infants are colonized with trillions of bacteria, that each bacterium contains between 2,000 and 6,000
immunological components, and that infants are infected with numerous viruses, the challenge from the 150 immunological components in vaccines is minuscule compared to what infants manage every day. Indeed, a scraped knee is probably a greater immunological challenge than all childhood vaccines combined.” – Children’s Hospital of Philadelphia, 2018

Summary Chart: Natural exposure far exceeds vaccine exposure

Source of Antigens Approximate Number
Daily natural exposures 2,000–6,000 per day
One common cold virus 4,000–10,000
Entire Current U.S. childhood vaccine schedule ~150–200 total
1980s vaccine schedule >3,000

Low Autism Rates among the Amish:

The argument is also made that communities that have lower vaccination rates, like the Amish, also have lower rates of Autism Spectrum Disorder. It’s not like the Amish don’t vaccinate their children at all.  A 2011 study in Pediatrics, for example, found that only 14% of Amish people surveyed didn’t vaccinate their kids at all. (Link)  Still, several studies have suggested lower overall rates of ASD within Amish communities.

There are several possible reasons for this observation. For example, the Amish have a greater tendency to avoid mainstream medical care and therefore do not take as much advantage of diagnostic criteria or care for ASD. This probably affects the detection rate. After all, there have been studies in certain Amish communities that showed rather high rates of ASD. A 2011 study on autism in Amish children in Ohio, published in the Journal of Child Neurology, found a high rate of autism due to a high degree of genetic consanguinity within the community. The study identified a rate of approximately 27 cases per 1000 children (1 in 37).

Joycelyn M. Lee, et. al., Autism Spectrum Disorders in Amish and Non-Amish Children. Journal of Child Neurology. 2011;26(6):696–701. (see also: Link)

So far, however, it seems as though in certain Amish communities, the rate of ASD may actually be less compared to the national average (1 in 271 vs. 1 in 31). Note, however, that this range is comparable to other non-Amish US communities. For example, the prevalence of autism among 8-year-olds in 2022 was 1 in 31 overall, but ranged from 1 in 103 in a community in Texas to 1 in 19 in California. Several studies have shown significantly lower rates of ASD among Hispanics in Texas compared to non-Hispanic caucasians (Link). In fact, in the U.S., recorded rates can vary 5-fold (or more) between different states or communities. Such variability likely reflects differences in regional genetic characteristics, dietary norms, environmental hazards, access to care, screening, and awareness. In any case, none of these rate variabilities have been associated with vaccination rates, which is the key point. (Link, Link, Link).

Pre-existing Conflicts of Interest:

In addition, note that all of the authors of this report have pre-existing conflicts of interest and a well-known pre-established bias against vaccines:  Here are their backgrounds:

  • Peter A. McCullough, MD, MPH: The foundation’s leader, a cardiologist who became prominent during COVID-19 for spreading vaccine misinformation. He’s had his board certifications revoked by the American Board of Internal Medicine for promoting COVID-19 vaccine misinformation;
  • John S. Leake, MA: Vice President of the McCullough Foundation with no medical qualifications. In fact, he is a non-fiction author and writes true crime novels;
  • Andrew Wakefield, MBBS: Yes, he’s back. The author of the fraudulent 1998 Lancet study that started the entire vaccine-autism myth. His medical license was revoked for research fraud;
  • Nicolas Hulscher, MPH: The report’s primary epidemiologist, who joined the McCullough Foundation immediately after completing his MPH in 2024;
  • Seven other foundation members and associates with previous publications examining vaccines as a “risk factor” for various diseases and adverse outcomes.
  • The report also received funding from the Bia-Echo Foundation, founded by Nicole Shanahan, who served as RFK Jr.’s running mate in 2024 and shares his skepticism about vaccine safety. The foundation’s mission to turn ‘advocacy into actionable research’ is also noteworthy.
  • This report was uploaded to Zenodo, which is an open repository where anyone can post documents without scientific scrutiny. It did not undergo rigorous peer review and is not a scientific journal with credibility. When you can’t get published in a legitimate journal or don’t want to field critiques from external scientists reviewing your work, you self-publish, self-promote, and call it “landmark research.”
  • Despite these clear biases, the authors do not declare any conflicts of interest.

Response from the Medical Community:

Dr. Zachary Rubin:

That figure alone [that over 100 studies support their conclusion] should raise eyebrows. The overwhelming body of peer-reviewed evidence from around the world—including meta-analyses covering millions of children—shows no causal relationship between any vaccine or vaccine ingredient and autism. The authors of this new report don’t reconcile that contradiction; instead, they dismiss every null finding as “biased” or “methodologically flawed” and elevate small, often non–peer-reviewed papers or case reports as proof of causation. This is cherry-picking, not science. What makes the report especially misleading is how it blends legitimate science with distorted interpretation. It correctly notes that autism is multifactorial, shaped by genetics, perinatal conditions, and environmental factors, but then uses that complexity to sneak in a presupposed conclusion: that vaccines are one of those environmental triggers. Correlation is presented as causation. Graphs showing the number of vaccine doses over time are plotted beside autism prevalence curves, suggesting a relationship without any statistical evidence of causality (Figure 2). This is a textbook example of a spurious correlation, the same flawed logic that could “prove” that organic food sales or smartphone ownership cause autism simply because they rose in parallel over the same decades.

The authors claim to have “comprehensively examined” the literature but fail to describe any reproducible method for how studies were identified, screened, or evaluated for quality. There’s no mention of pre-registration, bias assessment, or peer review. The inclusion of “gray literature,” conference abstracts, and preprints, sources that bypass scientific scrutiny, further undermines credibility. Meanwhile, high-quality epidemiologic studies from Denmark, Sweden, Japan, and the United States are brushed aside because they don’t fit the narrative. The result is a document that reads more like a legal brief for the anti-vaccine movement than a scientific analysis.

Zachary Rubin, The McCullough Foundation’s Autism Report Isn’t Science — It’s Propaganda, Do Not Be Fooled, October 28, 2025 (Dr. Zachary Rubin is a double board-certified pediatrician specializing in allergy and immunology)

Review of the 300 References:

We manually reviewed every single one of the 308 references — checking the authors and the journals where they were published. When we highlighted all the problematic sources (those written by known anti-vaccine advocates, retracted papers, and preprints, blogs, and other non-peer-reviewed sources masquerading as journals), the pattern was striking: starting at citation 193 onward, the list is essentially a sea of red flags.

Here’s what we discovered:

The Same Names, Over and Over

When we catalogued the authors cited, we noticed that the same small group of anti-vaccine advocates appeared repeatedly.

    • Papers by Mark and David Geier (Mark’s medical license was revoked in multiple states for unethical autism “treatments”)
    • Studies by Christopher Shaw and Lucija Tomljenovic (several retracted for manipulation)
    • Brian Hooker and Andrew Wakefield’s analyses (repeatedly debunked and/or retracted)
    • The McCullough Foundation authors citing their own previous work

This circular citation creates an illusion of widespread scientific support. They cite their own papers to validate their claims, then use those claims as evidence in new papers. It’s academic three-card Monte — the same bad data shuffled around to look like multiple independent sources, Instead of fairly examining all relevant research and before deriving conclusions, the authors only focused on studies that agreed with their narrative. This is an extreme case of selection bias, where any study that doesn’t fit their narrative is written off as having “serious methodological flaws”.

The Journal Problem:

We also examined where these 308 references were published. Many appeared in:

    • Predatory journals without a peer review process
    • Pay-to-publish outlets where anyone can be published if they hand over enough $$$
    • Journals with editorial boards stacked with anti-vaccine activists
    • Non-indexed journals with no scientific credibility

They even included papers from the Journal of American Physicians and Surgeons (a publication known for promoting HIV denialism and other medical misinformation) and the International Journal of Vaccine Theory, Practice, and Research–both of which are not indexed in PubMed because they do not meet basic scientific standards. In the conclusion, the report claims that of 136 vaccine studies they reviewed, 107 (79%) show a link to autism. Achieving this shocking percentage required some creative accounting:

What They Counted as “Evidence”:
    • Retracted papers, including Wakefield’s original fraud
    • Animal studies using doses hundreds of times higher than human vaccines
    • Test tube experiments that can’t be directly extrapolated as human evidence
    • Online parent surveys with self-selected participants and unverifiable data
    • Misuse of passive surveillance data (e.g., VAERS) — VAERS and similar systems collect raw reports of events after vaccination, but cannot establish causation; treating VAERS counts as evidence of causality is a well-known error
    • Ecological and correlation errors — e.g., correlating aggregate vaccination rates with aggregate autism rates without accounting for detection/diagnostic changes and confounders (this can produce spurious associations)
    • The same data published multiple times by the same authors in different journals
    • Law reviews

For example, a “study” from TheControlGroup.org (citation 291 in the report) is based on mail-in reports, phone calls, and interviews with self-selected participants. The author, Joy Garner, openly admits she’s “neither a PhD, nor a statistician” but rather “a tech inventor (hardware/video games) and patent-holder with an above-average IQ.” That is certainly not high-quality evidence (or evidence, at all).

 

How They Dismissed Contradictory Evidence:

They do cite major studies — including the recent Danish aluminum study that followed 1.2 million children. But these major studies are dismissed, and here’s how:

 

    • The same old “no true unvaccinated” argument: They claim these studies are invalid because they don’t have “genuinely unvaccinated control groups.” With clear evidence that vaccines save lives, withholding vaccines from children merely to create an “uncontaminated control group” would violate clinical equipoise and is extremely unethical. Still, there are good vaccines studies comparing the vaccinated vs. the unvaccinated.
      • A study published in the February 2014 issue of the journal Autism found “the rates of autism spectrum disorder diagnosis did not differ between immunized and non-immunized younger sib groups.” (Link)
      • A retrospective cohort study that assessed MMR vaccination status and psychiatric diagnoses among 537,303 children born in Denmark between 1991 and 1998. The study provides evidence against a causal relation between MMR vaccination and autism. Their data demonstrated a similar risk of autism in vaccinated versus unvaccinated children, a lack of temporal clustering of autism cases after immunization, and that neither autistic disorder nor ASD were linked to MMR vaccination. (Link)
      • A Population-Based Study of Measles, Mumps, and Rubella Vaccination and Autism (2002, by Kreesten Madsen et al.). This study compared ~ 537,303 Danish children; ~ 82% had received the MMR vaccine, ~ 18% were unvaccinated. There was no increased risk of autism or ASD in the vaccinated vs unvaccinated populations (Relative Risk [RR] = 0.92 for autistic disorder; RR = 0.83 for other ASDs) after adjustment.
      • Measles, Mumps, Rubella Vaccination and Autism: A Nationwide Cohort Study (2019, by Anders Hviid et al.). This study included 657,461 children born from 1999–2010 in Denmark, among whom some had not received MMR. Reported an adjusted hazard ratio (HR) of 0.93 (95% CI, 0.85–1.02) comparing MMR-vaccinated vs MMR-unvaccinated, i.e. no statistically significant increase in autism after vaccination.
      • See also: Link
    • Statistical nitpicking: They claim that studies that did not find an association with autism used “overadjusted statistical models,” “registry misclassification,” and “ecological confounding”. By claiming that models were “overadjusted”, the authors are insinuating that they have better subject-matter knowledge regarding a causal relationship than other researchers who have examined this relationship before them. In this vague claim, they fail to make explicit the variables they think are overadjustments and do not provide any additional “expert evidence” to back their claims. Though some degree of misclassification is inevitable in any registry-based study, errors large enough to completely erase an association are extremely rare. Ultimately, these are manufactured and technical reasons to dismiss studies that don’t support their conclusions.
    • The double standard of verification: They complain that “only a few case-control studies verified vaccination through medical records,” while simultaneously including unverified parent surveys and blog posts to support their conclusions.

As a clear example of their dismissal of reputable evidence, let’s take the Danish aluminum study from earlier this year. This study of 1.2 million children found no link between aluminum in vaccines and 50 different conditions, including autism. Rather than acknowledge this as massive contradictory evidence, they simply dismiss it as having methodological flaws. (Link)

They also attempt to discredit vaccine research from Denmark by highlighting Poul Thorsen, a researcher who was arrested for embezzling CDC grant money. Financial fraud is serious, but it doesn’t invalidate the dozens of other studies conducted using Denmark’s comprehensive health registry system. In fact, this is guilt by association: they are dismissing an entire nation’s research infrastructure because of one person’s financial crimes over a decade ago. The hypocrisy here is staggering: they criticize Thorsen’s financial misconduct while including Andrew Wakefield as a co-author–the same Wakefield who was paid by lawyers suing vaccine manufacturers, who manipulated data, and who planned to profit from the MMR vaccine scare he created.

Jess Steier, Anti-Vaccine Mad Libs: The McCullough Foundation’s “Landmark” Report (Same recycled claims, shiny new packaging), Medium, November 4, 2025 

 

 

 

 

Loss of the CDC:

Of course, now that the CDC has been gutted by RFK Jr. (also notoriously opposed to childhood vaccines and well-known for promoting their supposed link with ASD), the new CDC just updated its website in line with the McCullough/Wakefield report, to include a statement linking childhood vaccines with autism. The CDC’s website now says a link between vaccines and autism cannot be ruled out (Link). That’s a reversal from the CDC’s longstanding stance that there is no link.

Response from the Medical Community:

Susan Kressly:

“The Centers for Disease Control and Prevention website has been changed to promote false information suggesting vaccines cause autism,” said Dr. Susan J. Kressly, president of the American Academy of Pediatrics, in a statement. “Since 1998, independent researchers across seven countries have conducted more than 40 high-quality studies involving over 5.6 million people. The conclusion is clear and unambiguous: There’s no link between vaccines and autism. Anyone repeating this harmful myth is misinformed or intentionally trying to mislead parents. We call on the CDC to stop wasting government resources to amplify false claims that sow doubt in one of the best tools we have to keep children healthy and thriving: routine immunizations.” (Link)

Autism Science Foundation:

“The new statement shows a lack of understanding of the term ‘evidence,” the Autism Science Foundation said in a statement the organization provided to NPR, adding, “No environmental factor has been better studied as a potential cause of autism than vaccines.” (Link)

Allison Singer:

“The facts don’t change because the administration does”, said Alison Singer, president of the Autism Science Foundation. At this point, it’s not about doing more studies; it’s about being willing to accept what the existing study data clearly show. You can’t just ignore data because it doesn’t confirm your beliefs, but that’s what the administration is doing.”

“The CDC has always been a trustworthy source of scientifically-backed information but it appears this is no longer the case”, added Singer. “Spreading this misinformation will needlessly cause fear in parents of young children who may not be aware of the mountains of data exonerating vaccines as a cause of autism and who may withhold vaccines in response to this misinformation, putting their children at risk to contract and potentially die from vaccine-preventable diseases.” (Link)

Sean O’Leary:

Sean T. O’Leary, M.D., M.P.H., FAAP, chair of the AAP Committee on Infectious Diseases, called the latest move to put misinformation on the trusted CDC website “madness” and “a tragic moment for this country” and said he does not blame the career CDC scientists.

“For many decades, we (could) rely on CDC to provide the American public with the best available science,” Dr. O’Leary said. “Now our government is using it as an apparatus to spread falsehoods and lies.” (Link)

:Independent researchers across seven countries have conducted more than 40 studies involving over 5.6 million people to conclude that there is no link between vaccines and autism.” (Link)

Paul Offit:

“It’s a statement that’s confusing by design”, said Dr. Paul Offit, a pediatrician and director of the Vaccine Education Center at the Children’s Hospital of Philadelphia. “These are the usual anti-vaccine tropes, misrepresentation of studies, false equivalence,” he says. “They might as well say chicken nuggets might cause autism because you can’t prove that either.” (Link)

The CDC acknowledges in a footnote on its main webpage on autism and vaccines that it still carries a header reading “Vaccines do not cause autism*” and says it hasn’t “been removed due to an agreement with the chair of the U.S. Senate Health, Education, Labor, and Pensions Committee that it would remain on the CDC website.” (Link)

 

Video Clips:

Dr. Noc:

Dr. Rossome:

Dr. Mona:

See also Dr. Mona’s original post on Instagram: Link

Overall Summary Critique:

1) Title / subtitle / opening claims

Report: “Determinants of Autism Spectrum Disorder” — claims to be “the most exhaustive synthesis” and that “Combination and early-timed routine childhood vaccination emerges as the single most significant driver of autism risk.” Zenodo

Critique (lines of evidence):

  • Grandiose framing (“most exhaustive”) is a rhetorical claim, not scientific evidence. Credibility depends on methods (search strategy, inclusion/exclusion criteria, bias assessment). The report’s public pages list >300 references, but a credible systematic review requires transparent methods (pre-registered protocol, explicit search strings, inclusion criteria, risk-of-bias assessment, PRISMA flowchart). The Zenodo record provides a PDF and supplement, but the report does not appear to be published in a peer-reviewed journal or to include the transparency (prospective protocol / independent quality appraisal) that standard systematic reviews require. That weakens the claim of being definitive. Zenodo

2) Abstract / key result statements — representative lines and critique

A. Report statement (abstract): “Of 136 studies examining childhood vaccines or their excipients, 29 found neutral risks or no association, while 107 inferred a possible link….” Zenodo

Critique:

  • Counting raw numbers of papers that “infer a possible link” is misleading without weighting by study quality, sample size, design (ecologic vs cohort vs registry vs randomized), and risk of bias. High-quality national cohort studies (large, registry-based, with medical-record vaccination verification and multivariable adjustment) carry far more evidentiary weight than small case reports, ecological analyses, or biased surveys. The report’s tally (107 vs 29) gives equal weight to low- and high-quality evidence unless it performs a formal risk-of-bias and evidence-grading step (GRADE or equivalent). I could not find a transparent, systematic GRADE-style evaluation in the PDF’s front matter; without it, the numeric tally is unreliable as an index of truth. Good systematic reviews (e.g., Institute of Medicine/National Academies or major meta-analyses) evaluate study quality rather than simple counts. NCBI+1

B. Report statement: “12 studies comparing fully vaccinated versus completely unvaccinated children found superior overall health outcomes and dramatically lower risks of autism among the unvaccinated.” Zenodo

Critique:

  • This is a red-flag claim because the largest, best-powered cohort and registry studies (Denmark, other national datasets) do not support an increased autism risk with vaccination; they find no association (and in many cases slightly lower or null HRs for vaccinated vs unvaccinated). For instance, Hviid et al. (Denmark, n≈657,461) found an adjusted HR of 0.93 (95% CI 0.85–1.02) for MMR-vaccinated vs MMR-unvaccinated — i.e., no increased risk; Madsen et al. (NEJM 2002, n≈537,303) likewise found no association. Large registry studies often have far greater representativeness and power than small practice-based series or convenience samples. PubMed+1

  • Many “practice-based” or convenience studies that compare vaccinated and unvaccinated children suffer from strong selection bias: unvaccinated children are a nonrandom subgroup (different health-seeking behavior, socioeconomic characteristics, reporting patterns) — those differences can explain apparent outcome differences (confounding), not vaccine effects. Robust analyses therefore rely on registry data and validated outcome ascertainment, which the big Danish studies use. The report’s claim appears to cherry-pick smaller, biased studies and treat them as equivalent evidence. New England Journal of Medicine

C. Report statement: “The neutral association papers were undermined by absence of a genuinely unvaccinated control group… registry misclassification… ecological confounding… averaged estimates that obscure effects within vulnerable subgroups.” Zenodo

Critique:

  • It is true that some studies classified “unvaccinated” imprecisely, and that ecological studies have limits. But the largest registry cohorts (Denmark, several other nationwide databases) do have verified vaccination records and individual-level linkage to medical diagnoses — reducing many of those concerns. Critiquing all null studies by invoking generic measurement error is logically possible but must be supported by concrete evidence of systematic misclassification bias in those particular datasets; the report does not demonstrate that such bias reverses the null results in the large cohorts. In other words: you can point to limitations, but you must show that those limitations are sufficient to overturn the results — the report asserts that without substantiating why large, high-quality studies are invalid. Compare how the Institute of Medicine (IOM) and subsequent meta-analyses handled the evidence and limitations — they weighed quality and still rejected causality. NCBI+1

3) Mechanistic claims — representative lines and critique

Report: “Antigen, preservative, and adjuvant (ethyl mercury and aluminum) induced mitochondrial and neuroimmune dysfunction … culminating in neurodevelopmental injury and regression.” Zenodo

Critique:

  • Mechanistic plausibility is a standard part of causality assessment, but the existence of a hypothesized mechanism alone does not establish causation. Good causality evaluation requires consistent epidemiologic association + dose-response + temporality + biological mechanism that is demonstrated reproducibly. The large epidemiologic evidence base does not show consistent population-level associations that would be expected if the mechanism operated at scale. The National Academies / IOM review (and subsequent studies) concluded that the epidemiologic evidence did not support MMR or thimerosal as causes of autism; mechanistic hypotheses remained theoretical unless connected to consistent population signals. The McCullough report collects mechanistic studies and case reports, but does not provide direct, reproducible human experimental evidence that vaccines at standard doses produce the described chain of mitochondrial → neuroimmune injury → ASD in real-world populations. NCBI+1

  • Regarding thimerosal (ethylmercury): thimerosal was largely removed from childhood vaccines in many countries decades ago, yet autism prevalence rose (if anything) after its removal — that temporal pattern argues against a main-effect role for thimerosal in population trends. Similarly, aluminum adjuvants are used widely, and large epidemiologic assessments (and toxicologic profiling) have not shown patterns consistent with a causal role in ASD at the exposure levels present in vaccines. The report cites mechanistic literature, but mechanistic claims must be reconciled with the epidemiology; the report emphasizes the former and downplays the latter. NCBI

4) Claim: “Clustered vaccine dosing and earlier timing … increase the risk of ASD.” Zenodo

Critique:

  • This is a specific, testable hypothesis (schedule intensity/timing). The authoritative summary (as of the earlier literature and large cohort studies) is that no high-quality study has demonstrated that the recommended U.S. childhood schedule increases autism risk. Some smaller, practice-based studies have suggested associations with alternative schedules, but they are subject to selection bias and confounding. Large population studies that have looked at schedule/timing and cumulative exposure find no association. The report asserts the schedule effect but does not present large, well-controlled evidence that isolates timing as causal — and it does not show why the many studies that failed to find such an effect are invalid. New England Journal of Medicine+1

5) “To date, no study has evaluated the safety of the entire pediatric vaccine schedule for long-term neurodevelopmental outcomes through age 9 or 18 years.” (Report) Zenodo

Critique:

  • This statement is technically true in the sense that there is no randomized trial comparing a full-schedule vs never-vaccinated cohort followed prospectively through age 18 (which would be unethical). However, the scientific standard for policy relies on large observational cohorts, registry linkage, and studies targeted at particular vaccines/components. The absence of a single trial is not, by itself, evidence that the schedule is unsafe — it merely indicates a gap in the exact experimental design. Authorities (IOM, WHO, CDC) evaluate causality using the full body of evidence (mechanistic + observational + toxicology), not only RCTs. The report uses the absence of one trial as if it were a decisive gap when in reality the existing observational data (large cohorts) do provide strong evidence relevant to schedule safety. That is a classic “argument from absence” framed as a major criticism. NCBI+1

6) On the report’s use of studies: cherry-picking and weighting problems

Evidence from my reading of the report:

  • The report’s list of 136 vaccine-related papers includes many small, mechanistic, or case-report studies. The authors count any study that “infers a possible link” as supportive. They pair those with wording that discredits large registry null studies by alleging misclassification without demonstrating that the misclassification is large enough to invert the conclusions. This is classic cherry-picking and weight-by-counting bias: many low-quality small studies + counting = appearance of majority support, but without rigorous bias adjustment or quality weighting. Zenodo

  • The strongest epidemiologic evidence comes from very large, population-based registries with validated vaccination data (Denmark studies) — these should be given far greater evidentiary weight than small uncontrolled case reports. The report does not convincingly justify downgrading the Danish/registry studies. See Hviid et al. (2019) and Madsen et al. (2002) as benchmarks. PubMed+1

7) On including Andrew Wakefield as an author and its implications

Fact:

  • Dr. Andrew Wakefield is listed among the creators. Wakefield’s 1998 case-series paper linking MMR to autism was retracted and his medical license was revoked in the UK for ethical and scientific misconduct; his work is widely discredited. The presence of Wakefield as an author reduces the credibility of a report in the eyes of many scientists and journals because of his history of fraudulent practices. That doesn’t automatically invalidate every claim, but it raises the prior probability that unusual claims will be poorly supported and that extraordinary evidence is required. The report does not adequately meet that burden. (Background widely documented — see many summaries and the Wakefield retraction story.) Voices For Vaccines

8) On policy statements and impact claims (e.g., “urgent moral, scientific, and public-health imperative” to reassess the schedule)

Critique:

  • Policy proposals must rest on high-quality, reproducible evidence that a recommended change will reduce harm without causing more harm (e.g., increased vaccine-preventable disease). The report’s call to “reassess” hinges on reinterpretation of existing evidence and on studies it favors (smaller, biased designs). Major public-health bodies (CDC, WHO, national immunization technical advisory groups) will require robust replicated evidence before changing schedules — and the report does not provide that level of replicated, high-quality epidemiology. Meanwhile, altering the vaccine schedule without good evidence risks outbreaks of measles, pertussis, etc., with predictable morbidity and mortality. The report does not balance those risks adequately. NCBI

9) Specific methodological red flags (examples you can look for in the PDF)

When reading the report, watch for:

  • Counting studies without formal risk-of-bias weighting (counts ≠ evidence strength). Zenodo

  • Heavy reliance on ecological correlations or temporal associations (e.g., “vaccine use rose and autism prevalence rose”) without controlling for diagnostic substitution, increased awareness, or service-access changes — temporal correlation alone is weak evidence. (IOM and other reviews emphasize this.) NCBI

  • Use of small practice-based cohorts as representative of the population. These can be useful but are highly vulnerable to confounding. Zenodo

  • Failure to engage with the strongest counterevidence (large registry cohorts, meta-analyses) in a way that explains why those studies are invalid: mere assertion of “misclassification” is insufficient without empirical demonstration. PubMed+1

10) Bottom-line evaluation (short)

  • The McCullough/Wakefield report is not a peer-reviewed, methodologically transparent systematic review according to standard epidemiologic norms. The report compiles many studies but treats raw counts and mechanistic hints as if they outweigh large, high-quality registry studies. It emphasizes mechanistic plausibility and small studies while downplaying large, well-controlled epidemiologic evidence showing no association between standard vaccines (including MMR) and autism. The presence of Wakefield as a co-author and the document’s distribution outside established peer-review outlets further reduce its persuasiveness to the scientific community. NCBI+3Zenodo+3PubMed+3

11) Key evidence you can use to check the report’s central claims

  • Large Danish cohort (Hviid et al., 2019 — 657,461 children): MMR vaccination not linked to increased autism (adjusted HR 0.93, 95% CI 0.85–1.02). PubMed

  • Madsen et al., NEJM 2002 (n≈537,303): no link between MMR and autism. New England Journal of Medicine

  • Institute of Medicine (IOM) review (2004) and subsequent systematic reviews conclude that epidemiologic evidence favors rejection of a causal link between MMR/thimerosal and autism; mechanistic hypotheses remain theoretical. NCBI

  • Reputable public-health rebuttals and fact checks (Voices for Vaccines, Children’s Hospital of Philadelphia vaccine safety summaries) that summarize the larger evidence base and explain why the new report’s counting method is misleading. Voices For Vaccines+1

_______________

Resources

From VEC

The Vaccine Education Center (VEC) has compiled the following resources addressing concerns surrounding vaccine safety as it relates to autism. These materials are based on the scientific findings of many scientists studying hundreds of thousands of children on multiple continents. The one clear and consistent finding is that vaccines do not cause autism.

Print materials

Webpages

Videos

From other groups

References

DeStefano, R., T.T. Shimabukuro, The MMR vaccine and autism, Ann Rev Virol (2019) 6: 1.1-1.16.
Autism is a developmental disability that can cause significant social, communication, and behavioral challenges. A report published in 1998, but subsequently retracted by the journal, suggested that measles, mumps, and rubella (MMR) vaccine causes autism. However, autism is a neurodevelopmental condition that has a strong genetic component with genesis before one year of age, when MMR vaccine is typically administered.

Hviid A, Hansen JV, Frisch M, Melbye M. Measles, mumps, rubella vaccination and autism. Ann Int Med 2019; epub ahead of print.
The authors evaluated the relationship between receipt of MMR vaccine and the development of autism in more than 650,000 Danish children born between 1999 and 2010. During the study period, about 6,500 children were diagnosed with autism. The authors found no increased risk of autism in those who received one or two doses of MMR vaccine compared with those who didn’t. The authors also found that MMR vaccine did not increase the risk of autism in children with specific risk factors such as maternal age, paternal age, smoking during pregnancy, method of delivery, gestational age, 5-minute APGAR scores, low birthweight, head circumference, and sibling history of autism. Further, by evaluating specific time periods after vaccination, the authors found no evidence for a regressive phenotype triggered by vaccination. The authors concluded that MMR vaccination did not increase the risk for autism or trigger autism in susceptible children.

Jain A, Marshall J, Buikema A, et al. Autism occurrence by MMR vaccine status among US children with older siblings with and without autism. JAMA 2015;313(15):1534-1540.
The authors evaluated about 100,000 younger siblings who did or did not receive an MMR vaccine when the older sibling had been diagnosed with autism spectrum disorder (ASD). For children with or without older siblings with ASD, there were no differences in the adjusted relative risks of ASD between no doses of MMR, one dose of MMR or two doses of MMR. The authors concluded that receipt of MMR vaccine was not associated with increased risk of ASD even among children whose older siblings had ASD, and, therefore, were presumed to be at higher risk for developing this disorder.

Taylor LE, Swerdfeger AL, Eslick GD. Vaccines are not associated with autism: an evidence-based meta-analysis of case-control and cohort studies. Vaccine 2014;32:3623-3629.

The authors conducted a meta-analysis of case-control and cohort studies that examined the relationship between the receipt of vaccines and development of autism. Five cohort studies involving more than 1.2 million children and five case-control studies involving more than 9,000 children were included in the analysis. The authors concluded that vaccinations, components of vaccines (thimerosal), and combination vaccines (MMR) were not associated with the development of autism or autism spectrum disorder.

Hornig M, Briese T, Buie T, et al. Lack of association between measles virus vaccine and autism with enteropathy: a case-control study. PLoS ONE 2008;3(9):e3140.
The authors evaluated children with GI disturbances with and without autism to determine if those with autism were more likely to have measles virus RNA or inflammation in bowel tissues and to determine if autism or GI symptoms related temporally to receipt of MMR.  The authors found no differences between patients with and without autism relative to measles virus presence in the ileum and cecum or GI inflammation. GI symptoms and autism onset were unrelated to the receipt of MMR vaccine.

Uchiyama T, Kurosawa M, Inaba Y. MMR-vaccine and regression in autism spectrum disorders: negative results presented from Japan. J Autism Dev Disord 2007;37:210-217.
MMR vaccination was only utilized in Japan between 1989 and 1993, given as a single dose between 12 and 72 months of age.  The authors examined the rate of autism spectrum disorders (ASD) involving regressive symptoms in children who did or didn’t receive MMR during that period. No significant differences were found in the incidence of ASD regression between those who did or didn’t receive an MMR vaccine.

Afzal MA, Ozoemena LC, O’Hare A, et al. Absence of detectable measles virus genome sequence in blood of autistic children who have had their MMR vaccination during the routine childhood immunization schedule of UK. J Med Virol 2006;78:623-630.
Investigators obtained blood from 15 children diagnosed with autism with developmental regression and a documented previous receipt of MMR vaccine. Measles virus genome was not present in any of the samples tested. The authors concluded that measles vaccine virus was not present in autistic children with developmental regression.

Honda H, Shimizu Y, Rutter M. No effect of MMR withdrawal on the incidence of autism: a total population study. J Child Psychol Psychiatry 2005;46(6):572-579.
MMR vaccination was only utilized in Japan between 1989 and 1993, given as a single dose between 12 and 72 months of age. The authors found that while MMR vaccination rates declined significantly in the birth cohort of years 1988 through 1992 (~70% in 1988, < 30% in 1991 and < 10% in 1992), the cumulative incidence of ASD up to age 7 years increased significantly. The authors concluded that withdrawal of MMR in countries where it is still being used will not lead to a reduction in the incidence of ASD.

Smeeth L, Cook C, Fombonne E, et al. MMR vaccination and pervasive developmental disorders: a case-control study. The Lancet 2004;364:963-969.
The authors reviewed a major United Kingdom database for patients diagnosed with autism or other pervasive developmental disorders (PDD) over a 28-year period and similarly aged patients without those diagnoses to determine if the receipt of MMR vaccination was associated with an increased risk of autism or other PDD. They found no association between MMR vaccine and risk of autism or other PDD.

Madsen KM, Hviid A, Vestergaard M, et al. A population-based study of measles, mumps, and rubella vaccination and autism. N Engl J Med 2002;347(19):1477-1482.
The authors conducted a retrospective review of all children (> 500,000) born in Denmark between 1991 and 1998 to determine if a link existed between receipt of MMR vaccine and diagnosis of autism or autism spectrum disorders. No association was found between ages at the time of vaccination, the time since vaccination, or the date of vaccination and the development of autistic disorder.

Taylor B, Miller E, Farrington CP, et al. Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association. The Lancet 1999;353:2026-2029.
The authors determined whether the introduction of MMR vaccine in the United Kingdom in 1988 affected the incidence of autism by examining children born between 1979 and 1998. They found no sudden change in the incidence of autism after introduction of MMR vaccine and no association between receipt of the vaccine and development of autism.

Reviewed by Paul Offit, MD, on Sept. 27, 2025

____________

Dr. Sean Pitman is a pathologist, with subspecialties in anatomic, clinical, and hematopathology, currently working in N. California.

Leave a Reply