Full Breakdown
FDA’s Handling of COVID-19 Vaccine Safety Studies Under Scrutiny
5/8/2026, 11:28:37 PM
FDA Suppresses COVID-19 Vaccine Safety Signals
HHS says the Biden-era FDA sidelined senior medical officer Ana Szarfman after her updated safety-data-mining method identified 25 statistically significant adverse-event signals missed by the agency’s default system in March 2021. Internal directives later ordered her to stop the analysis, citing potential anti-vaccine rhetoric.
Historical Surveillance Practices
The FDA traditionally uses a proprietary algorithm for vaccine safety monitoring. Szarfman, working with its inventor, added a cross-checking approach using electronic health records. The Trump-era FDA previously blocked two peer-reviewed studies for overstated conclusions.
Principal Actors
Ana Szarfman (senior medical officer); Sen. Ron Johnson (R-WI, investigation chair); Peter Marks (former CBER director); Patrizia Cavazzoni (former CDER director); Steven Anderson (ex-OBP director); critics Jessica Adams, Paul Thacker, Alex Berenson; Jeffrey Morris (U-Penn biostatistics).
Timeline of Key Developments
Key dates: March 2021 – Szarfman’s method flags ~25 signals; October 2023 – FDA orders withdrawal of two COVID-vaccine safety studies; November 2023 – STAT News reports the withdrawal; January 5 2025 – 65+ Medicare preprint released; March 2025 – Anderson starts consultancy; August 2025 – Under-65 study presented; April 29 2025 – Sen. Johnson previews findings.
Data and Methodology
Szarfman’s analysis uncovered 25 adverse-event signals. The 65+ study examined Medicare beneficiaries >=65; the under-65 study covered enrollees 6 months-64 years. Both used a self-controlled case series design.
Official Statements & Responses
HHS said the FDA “marginalized” Szarfman’s findings and ordered a pause. Former CBER chief Peter Marks warned the method could create “erroneous conflicts” for anti-vaccine narratives. HHS noted the Trump-era block of two peer-reviewed studies.
Criticism & Opposition
Jessica Adams questioned whether the FDA prioritized suppressing methodology over safety. Paul Thacker called the New York Times’ coverage “blatant politics.” Alex Berenson labeled the 65+ study “junk.” Jeffrey Morris called the drafts “generally well done” but urged revisions.
Verbatim Quotes
- “No such energy for Dr. Sztarfman's [sic] studies,” — Jessica Adams, former FDA regulatory review officer
- “Shouldn't the NY Times do a journalism and report the blatant politics, instead of pretending this is 'science'?” — Paul Thacker, former Senate pharmaceutical-payola investigator
- “generally well done.” — Jeffrey Morris, director, University of Pennsylvania biostatistics division
- “Thus the days immediately following vaccination are likely to see a DROP in serious side effects" because jabs must come with "very serious problems to overcome that inherent bias in their favor," he said.” — Alex Berenson, former Times drug-industry reporter
Conflicting Reports & Gaps
Sources differ on Steven Anderson’s status: the 65+ preprint lists him as an FDA employee, yet his LinkedIn shows departure in December 2024. The self-controlled case series design removes confounding but can bias outcomes if vaccination timing reflects health, and media omitted authorship and algorithm-inventor details.
What’s Next
The Permanent Subcommittee on Investigation will hold a full hearing on the findings, and HHS has pledged further review of FDA data-mining practices. Additional disclosures of the underlying datasets and methodological revisions are anticipated in the coming months.
