On the list the CDC turned over, nineteen of the twenty studies were the wrong studies.

Aaron Siri’s client, the Informed Consent Action Network, had spent years asking the agency through FOIA for the research it cited in support of the public claim that vaccines administered in the first six months of life do not cause autism. The CDC eventually produced a list of twenty studies. Nineteen were MMR research, and MMR is not administered before twelve months of age. The twentieth was an Institute of Medicine literature review of DTaP, a survey of existing work rather than a study designed to establish causation in either direction.

That stipulation, signed by Department of Justice attorneys on behalf of the CDC, is the kind of document Siri’s new book is built around. “Vaccines, Amen: The Religion of Vaccines”, reviewed Sunday by Maryanne Demasi at Brownstone, is not a polemic placed in front of the documents. It is a polemic made of them.

Siri has the docket to do it. He is the managing partner of Siri & Glimstad, the New York firm that built a national vaccine-litigation practice on FOIA, depositions, and the unfashionable view that federal agencies should be made to produce what they claim to possess. The credit line on the case most readers will remember reads Public Health and Medical Professionals for Transparency v. FDA. In that suit the FDA asked U.S. District Judge Mark Pittman in the Northern District of Texas for seventy-five years to release the data it had relied upon to license Pfizer’s COVID-19 vaccine. Pittman, in his January 2022 order, gave them eight months. As of the agency’s most recent status filing in January, roughly 4.5 million pages had been produced and another 1.2 million pages remained outstanding.

The same KFF Health News story carries another fact worth setting next to the book. When the Department of Health and Human Services under Secretary Robert F. Kennedy Jr. reorganized FDA and laid off most of the agency’s FOIA staff this year, the workers spared were the ones processing the Pfizer and Moderna vaccine records on Siri’s cases, a workflow run through nine full-time and one part-time contractors at a cost of $3.5 million. Of every FOIA pipeline the cuts touched, the one HHS chose to keep funded is the one producing Siri’s records.

The book opens earlier than COVID. Its historical pivot is the National Childhood Vaccine Injury Act of 1986, which moved vaccine-injury claims out of state tort courts and into a federal compensation program funded by an excise tax on each dose, shielding manufacturers from civil liability for design defects. Demasi’s review describes Siri framing this as the load-bearing structural change: a product class on which the manufacturer has no economic incentive to find or address rare serious harms, because Congress has assumed the loss. That is not a contestable claim about the structure. It is what the statute does. Whether the program has functioned the way Congress described it in 1986 is the separate question, and the record on which Siri draws is the one that says it has not.

The book’s most contested chapter, on Demasi’s account, is the one about the unpublished study. The Henry Ford Health study, led by infectious-disease specialist Dr. Marcus Zervos, reviewed records for 18,468 children born between 2000 and 2016: 16,500 had received at least one vaccine, and 1,957 had not. The researchers concluded that exposure to vaccination was independently associated with an overall 2.5-fold increase in the likelihood of developing a chronic health condition, with the largest signals in asthma, autoimmune disease, and neurodevelopmental disorders.

The paper was completed in 2020 and never submitted. In a secretly recorded conversation Siri introduced to Senator Ron Johnson’s September 9, 2025 subcommittee hearing on corruption in science, Zervos said: “If I publish this, I might as well retire.” Henry Ford Health’s chief of clinical enterprise, Dr. Adnan Munkarah, told Michigan Public the analysis had a significant baseline mismatch between cohorts, that the wrong statistical method had been used, and that the decision not to publish was routine quality control. “We have hundreds of papers like that,” he said, “sitting in drawers at the present time.”

Both sides of that line belong on the record. The vaccinated cohort was larger, more diverse, and followed for longer than the unvaccinated one, and the methodological concerns are real. The Henry Ford draft does not establish that the disparities would have survived peer review. What it does establish is the institutional fact: the lead researcher said on tape he could not publish without ending his career, and the institution he works for did not publish. Whether the analysis would have held up is a different question from why nobody was permitted to find out.

The pattern, repeated across the book, is the actual argument. The CDC could not produce the studies it had publicly cited. The FDA asked the court for seventy-five years to release the data underlying a product administered to most of the country. The 1986 Act stripped the manufacturer’s incentive to look. A draft study at a major hospital system showing differential outcomes in vaccinated children was shelved, and the lead author said the reason was professional survival. Each rests on a filing, a FOIA production, or a recording.

What the book does not establish, and Siri on Demasi’s reading does not claim it establishes, is the clinical question itself. The documents are about the institutions, not the biology. They show what was asked for, what was produced, what was buried, and who declined to answer. They do not adjudicate which vaccines on the childhood schedule cause which harms in which children. That adjudication would require the studies the CDC told the court it does not have, the data the FDA wanted seventy-five years to withhold, and a publication record at hospital systems where the lead researcher believes publication ends careers.

Aaron Siri has the federal docket. The agencies have the answers. When HHS reorganized FDA this year, the FOIA workflow it chose to preserve from the cuts was the one releasing Pfizer’s vaccine records. The 1986 statute remains in place, and the no-liability structure it created remains the operating environment for every dose. “Vaccines, Amen” is a synthesis of what one attorney’s litigation has been able to pull out of the federal government over two decades. The argument it makes, on its strongest ground, is that this much extraction should never have been necessary.

Sources

  1. Brownstone Institute, Maryanne Demasi review of “Vaccines, Amen” (June 2026)
  2. Amazon listing for “Vaccines, Amen: The Religion of Vaccines,” Aaron Siri
  3. KFF Health News, RFK Jr.’s FDA FOIA staff cuts spared the workers on Siri’s COVID vaccine lawsuits
  4. Bloomberg Law, why a judge ordered FDA to release Covid-19 vaccine data on an eight-month timetable (PHMPT v. FDA, N.D. Tex.)
  5. Aletho News, DOJ stipulation on the CDC’s autism-studies list produced under FOIA
  6. Brownstone Institute, inside the Henry Ford vaccine controversy (Zervos draft study)
  7. Michigan Public, Henry Ford Health didn’t publish a vaccine study, was it buried or rejected science
  8. Siri & Glimstad LLP, Aaron Siri firm biography and litigation record