When Killing a Junk Paper Becomes “Censorship”
Bhattacharya pulled one junk paper from a fake journal — and the same press that cheered five years of actual censorship suddenly discovered the First Amendment.
Video Overview
Two weeks ago, interim CDC head Jay Bhattacharya — Great Barrington signatory, lockdown dissident, now also running the NIH — refused to publish a paper in the agency’s flagship journal, the Morbidity and Mortality Weekly Report (MMWR). The paper claimed the 2025-26 COVID vaccines were 50-55% effective against hospitalization. Within days, the Washington Post, CNN (previous link), Science, the Washington Monthly, and numerous others had received their marching orders — start screaming about “censorship” and “political interference.”
Jeremy Faust leaked the manuscript on Substack under the strapline “RFK Jr. and the CDC’s top official, Dr Jay Bhattacharya, don’t want you to read this.” The alt-health reflex is to cheer: Bhattacharya stuck it to the CDC. Fine, but that’s the easy take, and it leaves the real story on the table — three things almost nobody is covering correctly…
The Methodology IS the Story
Bhattacharya didn’t kill the paper because the conclusions were politically inconvenient. He killed it because the method was garbage — the same trick that let researchers “prove” parachutes don't work. The paper used test-negative design (TND) — the same statistical sleight-of-hand Pharma has leaned on for a decade to manufacture “vaccine effectiveness” numbers that don’t survive contact with real-world data.
Here’s how TND works in plain English. You take people who showed up sick at a clinic and got tested. The ones who tested positive for COVID are “cases.” The ones who tested negative are “controls.” Then you compare the vaccination rates between the two groups. If vaccinated people are over-represented in the test-negative group, you declare the vaccine “effective.”
The problem is obvious the moment you say it out loud. The “controls” aren’t healthy people. They’re sick people with something else — flu, RSV, a cold, anything that sent them to a clinic. You haven’t measured whether the vaccine prevents COVID. You’ve measured the ratio of vaccinated-sick-with-COVID to vaccinated-sick-with-something-else, in a population that already self-selected for being sick enough to get tested.
And this doesn’t even begin to touch on behavioral confounders (the vaccinated tend to test more often, seek care earlier, work in healthcare settings), which were baked into the study and almost impossible to scrub out once the cake’s out of the oven. The study’s conclusions are whatever the underlying assumptions are tuned to produce. A fact that Dr John Ioannidis proved in what has been said to be history’s most downloaded study - Why Most Published Research Findings Are False.
This is how you get from negative efficacy in real-world mortality data (or the Cleveland Clinic study) to a dressed-up press release announcing “55% effective.” TND has been the workhorse of CDC flu-vaccine pronouncements for years — the same flu-vaccine pronouncements that have never moved the needle on actual flu mortality.
And it’s a small slice of a much bigger pattern…
Roughly half of all biomedical studies fall into the Invisible & Abandoned category — designed, paid for, and then buried when the results don’t go Pharma’s way. Maryanne Demasi nailed it: “a system that repeatedly turns weak studies into strong claims.” Roger Watson at the Daily Sceptic walked through it in more detail. Everyone else is fighting on the censorship-versus-transparency axis, which is exactly the frame the captured side wants you fighting on.
Argue about whether the paper should have been published, and you’ve already conceded the paper says something true…
MMWR was Never Peer-reviewed
Bhattacharya’s actual kill shot, in his own words: “MMWR is not currently a peer-reviewed journal. But we are working on changing that.” If you’ve been following me since I moved to Substack, my initial series, published one short month ago, on the infamous HHS / Harvard / Lazarus study, contained a graph from a 1999 edition of MMWR that proved exactly the opposite of what the government was trying to prove concerning the overall benefits of national forced vaccination policies.
For 65 years, MMWR has been the CDC’s in-house bully-pulpit dressed up as a scientific journal. Every public-health pronouncement, every vaccine-efficacy figure, every “the science is settled” citation in the legacy press traces back through MMWR. And it has never been peer-reviewed in any sense that the term means at a real journal. The CDC reviews the CDC. Then the CDC publishes the CDC. The New York Times cite the CDC as gospel. Then Google’s algorithm crowns whatever the CDC said as “authoritative” and buries anyone and everyone who publicly questions it.
Now that someone is pointing at the man behind the curtain, Science magazine is rushing to argue that MMWR’s internal review “exceeds what is done at most scientific journals.” Read that again. The trade publication of American science is, with a straight face, defending an in-house newsletter against the charge that it isn’t peer-reviewed — by arguing that internal CDC review is better than peer review. This is what captured ‘research’ (term used loosely) looks like in real time, and the alt-health side should be feasting on it instead of high-fiving over Bhattacharya.
The Role Reversal Nobody is Calling Out
The same MSM press that spent five years cheering deplatforming, Trusted News Initiative pile-ons, Murthy v. Missouri jawboning, the Stanford Virality Project, NewsGuard, and the wholesale erasure of independent doctors from search results — that press class is now, with a straight face, calling it “censorship” when one interim CDC head refuses to publish one methodologically weak paper in one in-house publication.
The paper they’re calling “censored” is freely available on Substack. The Substack post calling Bhattacharya a censor has had more reach than the MMWR paper would have ever gotten. Faust’s tagline — “they don’t want you to read this” — is attached to a piece anyone can read, for free, in five seconds (see earlier link). Meanwhile, the doctors who were actually de-indexed, demonetized, deplatformed, and de-licensed from 2020 onward are still de-indexed, demonetized, deplatformed, and de-licensed. None of those reporters wrote about them.
That’s the story. A methodologically junk paper got pulled from a non-peer-reviewed in-house bulletin by an official who said publicly why, while five years of actual censorship still hasn’t been undone. If that’s the worst “censorship” the legacy press can find now that the boot is on the other foot, maybe the world will get to see what it actually looks like.