FLAWED Booster data was used by FDA for Mass Vax Campaign & We PROVE it in our new PAPER
What this video covers
Dr. Vinay Prasad presents a paper from his own research group arguing that the observational studies the FDA relied upon to justify mass COVID-19 booster campaigns were methodologically compromised. His target is a specific, well-known bias in vaccine observational research: the healthy-vaccinee effect (a form of indication bias), in which the people who get vaccinated differ systematically from those who don’t — they tend to be healthier, more health-seeking, and less frail — making vaccines look more effective against death and hospitalization than they are.
The paper’s case study is a December 2021 New England Journal of Medicine study out of Israel that reported boosters lowered the risk of COVID-19 death by 90%. Prasad says the study’s authors withheld a crucial piece of data — how many people in each group died of causes other than COVID-19 — and revealed it only in a March 2022 letter to the editor: 441 non-COVID deaths in the booster group versus 963 in the non-booster group. His group’s calculation: the “effect” of boosting on non-COVID death was 94.8%, nearly identical to the claimed 90% effect on COVID death — which he argues means the entire result could be residual confounding, since a booster should not lower the risk of car accidents or cancer. He illustrates with an analogy: the study is like finding that babies driven home from the hospital in a Mercedes live longer than those driven home in a Toyota — true, perhaps, but it says nothing about the car. He notes the original authors conceded the key points in their reply, acknowledging “numerous potential confounders” and that boosters were not administered to hospitalized patients already at high risk of death.
Prasad walks through the logic: people on their deathbeds are rarely vaccinated, so deaths concentrate in the unvaccinated group for reasons that have nothing to do with the shot; observational comparisons then credit the vaccine with mortality reductions that actually reflect who chose to get vaccinated. He argues his group’s analysis demonstrates this bias quantitatively in the booster data the FDA used, and that once the bias is accounted for, the apparent effectiveness shrinks dramatically.
The policy implication he draws is sweeping: if the observational evidence base for mass boosting was inflated by healthy-vaccinee bias, then the FDA authorized a population-wide campaign on unreliable data, without the randomized trials that could have settled the question. He frames this as a regulatory failure — the agency accepting weak evidence when stronger evidence was obtainable.
Key points
- Prasad presents his lab’s paper arguing that observational studies underpinning the FDA’s mass booster authorization were distorted by the healthy-vaccinee effect.
- He explains the bias in plain terms: vaccinated and unvaccinated groups differ in baseline health and health-seeking behavior, so crude comparisons over-credit the vaccine for lower death and hospitalization rates.
- According to Prasad, the analysis shows the bias is large enough to meaningfully inflate reported vaccine effectiveness — he claims the paper “proves” the flaw rather than merely hypothesizing it.
- The case study: a December 2021 NEJM Israeli study reported a 90% reduction in COVID-19 death from boosters; Prasad says the authors withheld non-COVID deaths (441 in the booster group vs. 963 in the non-booster group, revealed only in a March 2022 letter) and that his group calculated a 94.8% “effect” on non-COVID death — suggesting the result was residual confounding.
- He illustrates the fallacy with an analogy: a study finding babies driven home in a Mercedes live longer than those in a Toyota says nothing about the car. He notes the original authors conceded the key points in their published reply.
- He argues the FDA should have recognized these biases and demanded randomized trial evidence with clinical endpoints before recommending boosters to the whole population.
- Prasad contends that regulators leaned on observational data precisely because randomized trials might have shown smaller or null effects — an institutional, not scientific, reason.
- He positions the paper as part of a pattern: his group also documented statistical errors in CDC publications and censorship of contrarian work by preprint servers.
- Context: epidemiologists broadly acknowledge healthy-vaccinee effects exist but disagree sharply with Prasad about their magnitude and whether they invalidate the booster evidence base; agencies cited multiple converging lines of evidence, not a single study.