Every flu season, Americans are presented with official statistics telling them how well the latest influenza vaccine performed. Percentages are published, recommendations follow, and the public is expected to trust that the numbers reflect reliable science.
But what happens when the method used to produce those numbers becomes the subject of serious scientific criticism?
That is precisely the controversy now surrounding the Centers for Disease Control and Prevention (CDC), after a veteran epidemiologist challenged the research methods behind the agency’s annual flu vaccine effectiveness estimates.
The questions go beyond whether this year’s flu shot works better or worse than last year’s.
They strike at something far more fundamental: How reliable are the statistics being used to shape public health recommendations?
And if the research methods themselves are vulnerable to bias, how much confidence should the public place in the conclusions?
Epidemiologist Challenges the CDC’s Vaccine Effectiveness Studies
Dr. Eyal Shahar, professor emeritus of epidemiology at the University of Arizona, has raised concerns about the CDC’s continued reliance on a research method known as the test-negative design.
In an October 8 report published by Children’s Health Defense’s The Defender, Shahar argued that the approach can produce misleading estimates because of statistical biases built into the selection of study participants.
Shahar is not a newcomer to epidemiological research. His academic background includes extensive published work, teaching in epidemiological methodology, and editorial experience with the American Journal of Epidemiology.
______________________________________________
🔴 Help keep this independent voice alive and uncensored.
Support independent journalism and help keep Chris Wick News going.
👉 Support Chris Wick on GoFundMe
His central criticism is straightforward.
The CDC continues using a research design that he believes is fundamentally vulnerable to bias, even though alternative approaches exist.
According to Shahar, the problem is not simply that researchers occasionally make mistakes when analyzing data.
It is that the underlying study design may introduce distortions before the final calculations are even performed.
That distinction matters enormously.
A sophisticated statistical calculation cannot automatically rescue a study whose underlying comparisons are biased.
What Is the Test-Negative Design, and Why Does It Matter?
The test-negative design is widely used in influenza vaccine effectiveness research.
Researchers examine patients who seek medical treatment for respiratory symptoms and undergo influenza testing.
Those who test positive for influenza are compared with those who test negative. Researchers then examine vaccination status and adjust for measured factors that could influence the results.
The approach is practical, relatively inexpensive, and allows researchers to evaluate vaccine performance using real-world healthcare data.
But critics argue that convenience comes with methodological risks.
The first concern involves selection bias.
Because participants must have sought medical care and received testing, the study does not directly represent everyone in the broader population.
People who never visit a doctor, experience mild symptoms, or have no symptoms at all may be excluded.
Another concern involves collider bias, a statistical problem that can arise when researchers condition their analysis on a factor influenced by multiple variables.
Under certain circumstances, this can create misleading associations between vaccination and infection outcomes.
Shahar argues that these issues can distort effectiveness estimates.
However, an important distinction must be made: the existence of possible bias does not automatically prove that every CDC estimate is wrong.
The direction and magnitude of any distortion depend on the assumptions, populations, and analytical methods involved.
The CDC maintains that test-negative studies can reduce certain biases associated with healthcare-seeking behaviour.
The scientific dispute therefore concerns whether the method’s advantages sufficiently outweigh its limitations.
CDC’s Own 2025–2026 Numbers Reveal Why Study Methods Matter
The CDC’s preliminary estimates for the 2025–2026 influenza season provide an important example.
In its March 2026 report, the agency estimated that influenza vaccination reduced flu-associated outpatient visits by approximately 22% to 34% among adults, depending on the surveillance network.
For adults, estimated protection against influenza-associated hospitalization was approximately 30%.
Among children and adolescents, reported protection against influenza-related outpatient visits was approximately 38% to 41%, with an estimated 41% effectiveness against hospitalization.
These figures were derived using test-negative case-control methods.
The CDC interpreted the findings as evidence that influenza vaccination continued to provide protection, although effectiveness was lower than in some recent seasons.
But the agency also acknowledged limitations.
Its report identified possible unmeasured confounding, incomplete vaccination records, differences between surveillance networks, and uncertainty surrounding preliminary estimates.
That acknowledgement is significant.
It demonstrates that even official vaccine effectiveness figures are estimates derived from observational research, not absolute measurements of protection.
When a government agency reports a percentage, the public deserves to understand what that percentage actually measures, what assumptions produced it, and how much uncertainty remains.
A Cleveland Clinic Study Adds Another Layer of Controversy
The Defender also highlighted a Cleveland Clinic study examining influenza vaccination during the 2024–2025 season.
According to the report, researchers using a cohort approach found an association between vaccination and a 27% higher risk of influenza among the healthcare workforce studied.
The finding attracted attention because it differed sharply from positive vaccine effectiveness estimates reported through other surveillance systems.
However, the interpretation requires caution.
An observational association does not establish that vaccination caused the increased risk. Differences in exposure, testing, previous immunity, and other factors may influence the results.
Furthermore, findings from a particular healthcare workforce cannot automatically be generalized to the entire population.
Nevertheless, the conflicting findings raise a legitimate research question.
Why can different observational approaches produce substantially different estimates?
And what additional evidence would be necessary to determine which results most accurately represent vaccine performance?
These are questions that deserve rigorous investigation rather than automatic dismissal.
The Debate Reaches the Highest Levels of American Public Health
The controversy extends beyond one epidemiologist.
According to The Defender, National Institutes of Health Director Jay Bhattacharya publicly criticized the use of the test-negative design in connection with a COVID-19 vaccine effectiveness study earlier in 2026.
The report described a dispute over whether research employing that methodology should appear in the CDC’s Morbidity and Mortality Weekly Report.
Bhattacharya argued that the design could produce biased estimates without reliably establishing the direction of the bias.
The episode illustrates a broader disagreement over the standards of evidence that public health institutions should require.
Supporters of the test-negative approach consider it a useful observational tool, particularly when randomized trials are impractical.
Critics contend that researchers should place greater emphasis on carefully designed cohort studies and, where appropriate and ethical, randomized trials.
Neither position eliminates the need for transparent data, independent scrutiny, and replication.
The central issue is whether public health institutions are willing to subject familiar methods to the same critical examination they demand of competing research.
Why Independent Verification Matters
Scientific credibility does not come from institutional authority alone.
It comes from methods that can withstand scrutiny, findings that can be replicated, and researchers who are willing to confront evidence that challenges their assumptions.
That principle should apply equally to government agencies, pharmaceutical manufacturers, academic institutions, and independent researchers.
The public should also understand that vaccine effectiveness is not a single universal number.
Protection can vary according to age, influenza strain, previous immunity, season, and the outcome being measured.
A vaccine’s estimated effectiveness against medically attended influenza is not necessarily equivalent to its effectiveness against infection, transmission, hospitalization, or death.
Those distinctions can become lost when complicated research findings are compressed into simple headlines.
The public deserves more than reassuring percentages. It deserves the evidence, the limitations, and a clear explanation of how conclusions were reached.
The Bigger Question: Is Public Health Science Open to Challenge?
The debate surrounding CDC flu vaccine estimates exposes a tension that extends far beyond influenza.
Public health agencies must communicate recommendations clearly while acknowledging the uncertainty inherent in scientific research.
At the same time, researchers must remain free to challenge established methods without having their arguments accepted or rejected merely because of their institutional affiliations.
Shahar’s criticism deserves examination on its methodological merits.
So does the CDC’s defense of its research approach.
A credible evaluation would compare competing study designs, identify the conditions under which bias occurs, and assess whether different methods consistently produce different estimates.
It would also distinguish uncertainty about the precise magnitude of vaccine effectiveness from evidence that vaccination provides some protection.
Those are not the same question.
Final Thoughts: Trust Should Be Earned Through Transparency
The real issue is not whether Americans should automatically accept or reject the flu vaccine.
It is whether the institutions responsible for measuring its effectiveness are providing the strongest evidence available and explaining that evidence accurately.
The CDC’s estimates have documented limitations. Shahar argues those limitations may be more fundamental than the agency acknowledges.
His criticism is serious enough to warrant careful scientific examination, but it does not by itself invalidate all evidence supporting influenza vaccination.
That distinction matters.
Public trust should never depend on discouraging questions. It should depend on answering them.
If the methods are reliable, transparent independent scrutiny should help demonstrate why.
If improvements are needed, the public deserves to know.
And if competing studies produce conflicting conclusions, the answer should be better research, not unquestioning loyalty to whichever result supports a preferred narrative.
Because when public health decisions affect millions of people, the standard should never be simply whether an institution says the science is settled.
The standard should be whether the evidence can survive being challenged.
Sources
1. Children’s Health Defense — The Defender (October 8, 2026)
Can You Trust CDC’s Flu Vaccine Effectiveness Estimates? Epidemiologist Says No
https://childrenshealthdefense.org/defender/can-you-trust-cdcs-flu-vaccine-effectiveness-estimates-epidemiologist-says-no/
2. Centers for Disease Control and Prevention — MMWR (2026)
Interim Estimates of 2025–26 Seasonal Influenza Vaccine Effectiveness
https://www.cdc.gov/mmwr/volumes/75/wr/mm7509a2.htm
3. CDC — Preliminary Flu Vaccine Effectiveness Data for 2025–2026
https://www.cdc.gov/flu-vaccines-work/php/effectiveness-studies/2025-2026.html
4. CDC — Factors That Influence Vaccine Effectiveness
https://www.cdc.gov/flu-vaccines-work/php/effectivenessqa/outcomes.html
5. CDC — Preliminary Flu Vaccine Effectiveness Data for 2024–2025
https://www.cdc.gov/flu-vaccines-work/php/effectiveness-studies/2024-2025.html
