Scoop: The leaked protocol of the CDC-funded Hepatitis B vaccine trial in Guinea-Bissau. “This is another Tuskegee.”
The protocol reveals that the trial will withhold the Hepatitis B birth dose from thousands—without placebos, without universal maternal screening, and with endpoints critics call indefensible.
We have an exclusive that I believe we are first to report, thanks to the HHS employees, past and present, who continue to bravely speak out. Help me amplify them by reading, sharing, and supporting this work. Thanks!
On December 18, the CDC announced a $1.6 million award, without competition, to Danish researchers that would fund an experiment on babies in Guinea-Bissau. The grant, the agency said in a Federal Register notice, would fund a “randomized controlled trial to assess the effects of neonatal Hepatitis B vaccination on early-life mortality, morbidity, and long-term developmental outcomes.” While the details were not public, what was known raised major ethical concerns among experts, especially since the WHO has recommended universal birth doses of the Hepatitis B vaccine since 2009, and GAVI (the vaccine alliance), has committed to funding that policy in Guinea-Bissau starting in 2027.
“This announcement has set alarm bells ringing in the global health community,” Professor Martin McKee, of the London School of Hygiene and Tropical Medicine, told The Guardian.
“Everything is wrong with awarding this unsolicited single-source grant, including the approval process, concerns about conflict of interest, and the dubious ethics of the research,” Dr. Robert Steinbrook, Health Research Group Director at Public Citizen said.
“I’m deeply skeptical that this is a good use of taxpayer money,” UC Berkeley epidemiologist Dr. Arthur Reingold told Science.
And that was before anyone had seen the full protocol.
Here, for the first time in public, obtained by Inside Medicine from a current CDC official granted anonymity to protect his job, is that protocol in its entirety.
What was known, what we’re learning. No placebo? No problem.
Very few people have seen this protocol before. Politico’s Sophie Gardner obtained it in December, but the document was not published at that time. Dr. Arthur Caplan told Gardner that the study was exploitative, later calling it “immoral.” HHS defended the study in multiple outlets, saying that the “highest scientific and ethical standards” would be ensured.
But the trial’s details—closely reviewed by Inside Medicine and Dr. David Boulware, Professor of Medicine at the University of Minnesota, and a veteran designer of high-quality randomized clinical trials—reveal a study that is likely even more flawed than experts had feared.
The “idea” behind this trial is that because Guinea-Bissau is slated to roll out the birth dose of the Hepatitis B vaccine starting in 2027, now is an opportune time to randomize infants to either receiving that dose or not. The problem is that we already know the vaccine works. So what could be learned?
Based on the CDC’s Federal Register notice and Politico’s reporting, it was known that the study would randomize newborns to receive a birth dose of the Hepatitis B vaccine or not. Without explicitly saying so, Politico revealed that there would be no placebo. This seemed to fly under the radar (or behind the paywall). The protocol confirms this. Another CDC document (justifying the lack of competition for this grant) also obtained by Inside Medicine fleshes out precisely what the study will measure:
“This randomized controlled trial (RCT) will evaluate the broader health effects of administering a single dose of monovalent Hepatitis B vaccine (HBV0) within the first week of life in newborns in Guinea-Bissau. The study compares infants who receive HBV0 alongside standard neonatal vaccines (BCG [tuberculosis] and OPV [oral polio vaccine]) with those who receive only BCG and OPV.”
The full protocol states:
“Primary outcome: Mortality and morbidity. The composite outcome of all-cause mortality and non-fatal admission with infection* from randomization to 42 days of age (before the next routine vaccination) in the group given HBV0 compared with the control group.
Secondary outcomes: Neurodevelopmental disorders and atopic dermatitis (AD).
AD by 2 years of age. In a subgroup followed to 2 years of age (see sample size calculation): cumulative incidence of AD symptoms (self-reported at home visits or telephone follow-up) and cumulative incidence of AD diagnosed by an MD with special training.
Neurodevelopment by 5 years of age. In a subgroup of 500 infants followed to 5 years of age: cognitive outcomes assessed by the Mullen Scales of Early Learning and the Kaufman Assessment Battery for Children. These tests
have been tested for validity in Sub-Saharan African children.”
*This is all-cause infections, not just Hepatitis B.
In short, this trial is an effort to find so-called “non-specific effects” of vaccines, an approach favored by Secretary Kennedy’s allies that seeks to find evidence that life-saving vaccines actually have undetected harms. Naturally, non-specific effects have been championed by the very researchers the CDC is funding to do the Hepatitis B birth dose study. (A takedown of their work on “non-specific effects” can be found here, and is also summarized here.)
There are several scientific problems with this and other parts of the protocol.
First, Hepatitis B acquired by newborns is not expected to cause mortality for years. So, trying to detect a decrease in all-cause mortality at 42 days (or 6 months, or 6 years), is doomed to fail. It’s like doing a trial of chemotherapy and asking if the patient is alive a week later. Meanwhile, the justification document states that the “findings from this trial will provide robust evidence on the overall health impact of HBV0, beyond its role in preventing hepatitis B infection.” Indeed.
A second problem is that the second component of the primary outcome includes all infections. That means that any signal of benefit would be diluted, and unlikely to reach statistical significance—notwithstanding the fact that Hepatitis B is expected to cause serious problems later in life, making this outcome a non-sequitur anyway.)
Third, the study does not even track Hepatitis B infection rates (other than as a small sub-study, which is more of an afterthought to look at disease prevalence).
Fourth, the trial fails to state what will and will not constitute “neurodevelopmental” outcomes. The full protocol mentions two tools designed to measure some of this, but each has multiple components. Therefore, it’s statistically probable that something within these surveys will come up positive by chance.
Fifth, while atopic dermatitis is unpleasant it is typically self-limited. So, it's bizarre to see this condition as an outcome in a trial in which the disease being prevented causes chronic liver disease, and in many cases, eventually cancer and death.
Sixth, no FDA-approved vaccines are being used in this study. While all three options in the trial are WHO-approved, it is remarkable that the US CDC is funding a study of vaccines that are not on the US market.
Seventh, there is no placebo. Newborns will either receive two vaccines (polio and tuberculosis) or three (those plus Hepatitis B). Because of this, the parents will not be blinded. (The statisticians will be blinded, thankfully). While single blinding is better than nothing, double blinding prevents many problems, including investigator bias, Dr. Offit pointed out.
More expert analysis.
The study’s design is riddled with problems, says Dr. David Boulware, who read the protocol last night. He listed four “major deficiencies.” Here they are (in bold), followed by my own additional commentary, as well as input from Dr. Offit.
“Failure to test all mothers to exclude Hepatitis B.”
This is an extraordinary ethical violation. The trial’s investigators admit that around 19% of people in Guinea-Bissau are Hepatitis B carriers. Since half of the 14,500 newborns the researchers plan to enroll will not receive the Hepatitis B vaccine at birth, around 1,425 of them (14,500 x 0.19) will be born to mothers who are carriers. The risk of transmission to these babies is very high. It might be ethical to randomize mothers who are known to be negative for Hepatitis B, but even that would be dicey. To consign 1,425 infants to almost certain (70%-90%) Hepatitis B infection (and the chronic liver disease that eventually follows) is unthinkable. Meanwhile, the researchers will perform rapid antigen tests on 250 mothers early in the study to verify the prevalence of the disease. If any of them test positive, their babies will be vaccinated at birth, rather than being randomized (whew). But the other 14,250 dyads (mother-infant pairs) enrolled, among whom around 19% will have Hepatitis B in the mother? No testing. This is absolutely unconscionable, especially since these tests are inexpensive.
Dr. Paul Offit was similarly alarmed. “These investigators will conduct a trial where children who are born to mothers who are HBsAG [Hepatitis B carriers] positive—but don’t know it because they weren’t screened—would not be given a birth dose? That condemns those children to a shorter life.”
Moreover, the researchers say that mothers who are known to be Hepatitis B carriers will be excluded. But they glibly declare that this is “unlikely.” That’s rich. The only reason it is unlikely is that testing is not routine in Guinea-Bissau. So while it is unlikely that a mother knows she has Hepatitis B, it is fairly likely (~19%, per the researchers) that she does. That begs the question: why is this being done?
Of course, if the researchers did test all the mothers, that would mean the trial would have to recruit 20% more people, to account for those who tested out.
Finally, it bears mentioning that the CDC’s Advisory Committee on Immunization Practices recently voted (in the absence of any science) to remove its universal recommendation for the birth dose of the Hepatitis B vaccine. But even RFK’s Jr.’s anti-vaxxer-stacked ACIP did not dare suggest that mothers who hadn’t been tested for Hepatitis B should have the vaccine withheld from their newborns.
“The trial’s endpoints do not measure Hepatitis B incidence in all participants. This is the most direct trade-off of withholding the birth dose of the vaccine.”
Again, withholding the birth dose, especially in a high-prevalence area, is unethical. As Dr. Boulware will explain later, it violates the Declaration of Helsinki, which codifies ethical principles of human subjects research.
“The data safety monitoring board (DSMB) [an external body that keeps tabs on trials like this] has no explicit early stopping boundaries. There is no guidance on what endpoints are to have interim analysis.”
In serious, well-designed clinical trials, the DSMB doesn’t just meet and look things over. They have pre-set “rules” that must be followed. For example, a rule might say that if after 1,000 people have been enrolled, 99% of the deaths are in the control group, the trial must be stopped because it would be unethical to withhold the vaccine from the control group any longer. But from there it gets more complicated. While it’s easy to figure out what to do if the breakdown is 90-10 or 80-20, it gets trickier when the breakdown is 65-35, 60-40, or 55-45. That’s because with smaller numbers, narrower differences are more likely due to chance. Statisticians therefore run the numbers ahead of time so that the DSMB knows exactly what breakdown would be sufficient to stop the trial on these grounds. Without such rules, the DSMB will be winging it. That’s simply not okay.
Dr. Paul Offit agrees. “It’s amazing that there is not a stop protocol for a study that includes death as an outcome.”
“The claim of ‘no known disadvantages’ (Page 13 of the protocol) is clearly false if children develop chronic hepatitis B because vaccination was withheld.”
This is another massive violation. The researchers claim that their design is not likely to harm participants by virtue of participation. But having a vaccine that is known to be effective on hand and not providing it because a study is going on, is the very definition of ethical malpractice.
He also added concerns about the lack of true informed consent in this trial. “It is hard to believe that these parents will be fully informed about the greater risk in the delayed birth dose group,” he told Inside Medicine.
(Previously, Dr. Offit, a co-inventor of the rotavirus vaccine, listed a number of criticisms of this trial based on what was known at the time, some of which overlap with points made here.)
HHS and researchers defend the study.
Asked about certain aspects of the protocol, HHS spokesman Andrew G. Nixon defended the study to Inside Medicine. “The Guinean health authorities plan to introduce Hepatitis B vaccine at birth (HBV0) in 2027,” he said. “This study will randomize more than 14,000 newborns to receive HBV0 (the upcoming policy) or no HBV0 (the current policy) before 2027. The control arm would receive the current standard of care. The planned study represents the world’s first and perhaps only opportunity to test the overall health effects of HBV0.”
Looking for clarification on some aspects of the protocol, I reached out to Professor Frederik Schaltz-Buchholzer, the principal investigator of the trial. He declined to answer my questions or schedule a phone call.
Previously, some of the researchers awarded this grant have similarly claimed that the trial does not withhold vaccines because the birth dose is not yet the standard of care.
Experts I spoke to did not agree.
More on the ethics. (Oh, and is this legal?)
Dr. Boulware believes the protocol to be “ethically deficient,” (by virtue of the Declaration of Helsinki) because “it exposes a foreseeable subset of newborns to preventable, lifelong harm by withholding a proven intervention without adequate risk stratification or mitigation, fails to ensure that risks are minimized and proportionate to potential social value, and does not provide sufficiently robust informed consent or post-trial protections for participants placed at elevated risk.”
Dr. Craig Spencer, who teaches about the history of global public health interventions at the Brown University School of Public Health, said that “the history of global health is littered with interventions, just like this study, that would be deemed unethical at home but are somehow justified when imposed on communities abroad.” Worse, he finds it “particularly jarring” that Secretary Kennedy “gleefully celebrated” the slashing of US support for GAVI, the organization that will be paying for the birth dose of the Hepatitis B vaccine starting in 2027.
So, what should we do instead? Easy, says Dr. Paul Offit. Use the money to start vaccinating newborns sooner. “How many doses of the vaccine does $1.6 million buy?” asked Dr. Offit.
It’s a good question. Given what GAVI expects to pay per dose (encompassing the complete operational chain) the $1.6 million that the CDC is spending on this study could instead be used pay for over a decade’s worth of Hepatitis B vaccine birth doses in Guinea-Bissau. Indeed, if the infrastructure and the money to do this trial exist, the only moral thing to do is to leverage those resources to hasten the sunrise of that nation’s plan to make the birth dose of the Hepatitis B vaccine available to all newborns there in 2026, rather than in 2027 as planned.
“The only acceptable thing is to give these doses out,” the CDC source said. But he went further, adding, “Yes, it is unethical, but there are actual ethical regulations written into federal law. You can’t deny people the gold standard in a clinical trial. No US IRB would ever sign off on this. So, this is illegal, which is being ignored.”
How did this happen? A CDC official speaks, after almost blowing the whistle.
In late 2025, rumors about this study began to circulate at the CDC. The CDC official who spoke to Inside Medicine (and who provided the protocol) was not involved with the Hepatitis B study, but became aware of it shortly before it was announced in the Federal Register.
While not typical, the CDC does occasionally fund randomized controlled trials, particularly to address gaps in the NIH portfolio, such as studies looking at bed nets to prevent malaria.
That wasn’t what bothered this official. Instead, it was that the usual pathway for approving and funding a clinical trial with human subjects had not been followed. Career staff were not asked to complete a “funding determination,” (a concise description of what the study would do) nor a “project determination” (a more detailed readout that would normally include the study protocol, and which would identify whether the research was on human subjects, which would indicate the need for much more oversight, including “Institutional Review Board” (IRB) approval, either the CDC’s or another US-based one.
None of that happened. “Staff in the Office of Science told Trump political appointees that they should not do this,” the official told Inside Medicine. “But they did it anyway.” Career CDC employees were disappointed. “The study should never have been approved, let alone funded,” the official said.
Moreover, he echoed the scientific concerns raised by others. “We would never do a study on a vaccine that has been in use for 40 years. That is a waste of taxpayer resources,” he said, adding that it’s “set up so that they’re likely to find a spurious association.”
Beyond that, though, the official seemed exasperated. “This is another Tuskegee,” he said. “We are allowing children, infants, to be exposed to Hepatitis B when we could prevent it, and then follow them for five years to see what happens. That’s not long enough to see the long-term benefits, but might be long enough to find some non-specific effects.” Nor was he the first observer to invoke that unsettling comparison. Dr. Paul Offit made a similar comment in his Substack.
While the process was not transparent, the official believes the intent was. “This is a study designed to try to find adverse effects of the birth dose. It’s badly designed.” He also believes the reason the usual process was skipped was a combination of ignorance and ideology. “Because it is a study that no reputable IRB would sign off on, HHS gave money to CDC political appointees with specific instructions to fund the University of Southern Denmark.”
So troubled by what he had heard and seen, the CDC official considered filing a whistleblower complaint to Congress in December. He thought it might cost him his job. “I know there are legal protections but they don’t care. I would have been escorted out.”
Then the Federal Register notice came out. He felt that enough of the details had been made public that he didn’t need to risk his job by becoming a whistleblower. The truth, he believed, would eventually come out. Once that happened, he hoped it would lead to more scrutiny which, eventually, might pressure the CDC to cancel the study.
An attempt to justify a non-competitive grant.
That leads us back to the Federal Register Notice from December 18. That announcement revealed not only the grant’s existence, but also that it had been determined that the usual competitive process could be skipped. Below is the CDC’s justification (which comes from another document obtained by Inside Medicine):
“A sole source award is justified due to the unique capabilities of the implementing partner in Guinea-Bissau. This organization operates a well-established surveillance and clinical trial platform with the infrastructure and expertise necessary to conduct large-scale neonatal RCTs in a low-resource setting. Their experience includes neonatal enrollment, longitudinal follow-up, and high- quality data collection on morbidity and mortality outcomes. The study must be initiated promptly to take advantage of a narrow window before HBV0 becomes national policy in Guinea-Bissau. Delays caused by competitive procurement could compromise the feasibility and scientific value of the trial. Furthermore, the partner’s longstanding presence and community trust are essential for successful implementation and retention in this sensitive population. The results of this trial will be critical for informing national and international vaccine policy. By providing high-quality evidence on the broader health effects of HBV0, the study will support evidence-based decisions by ministries of health, WHO, and global immunization partners regarding the optimal timing and delivery of hepatitis B vaccination in newborns.”
Dr. Angela Rasmussen previously called the award “blatant cronyism.” Nothing here appears to change that assessment. Moreover, the request to bypass a competitive grant process was made by two Trump appointees with no scientific background, Stuart Burns and Sam Beyda. Usually, such a request would have been made by a center director or the agency’s associate director for science, another former CDC official told me. “The fact that a permanent or acting Center, Institute, or Office director is not on the request is weird,” the former official said. “They’re doing whatever the fuck they want and don’t really care,” he said, adding that “it’s unethical to be running a study with a vaccine that works.”
The trial has begun. We think.
On January 7, it was reported that this trial had begun. But last weekend, rumors reached my group chats that it had been halted. Reporters raced to confirm the news, but were not able to. Asked whether the rumors that the trial had been halted were accurate, HHS Spokesperson Andrew G. Nixon said, “those rumors are false.”
Assuming that patients have been enrolled, that also means that the CDC and/or the Danish researchers now have 30 days to post the study’s protocol on the ClinicalTrials.gov website. I asked Nixon when the protocol would appear on that website. He didn’t answer that question. Instead he repeated a statement he had previously made to Politico and others, saying that HHS would ensure the “highest scientific and ethical standards are met,” and that the award “supports an independent study designed to answer important questions about the broader health effects of the hepatitis B birth dose.”
When I informed the CDC official who provided the protocol to Inside Medicine of this, he was disappointed. “This study needs to be stopped.”
If you have information about any of the unfolding stories we are following, please email me or find me on Signal at InsideMedicine.88.
As a 15 year veteran of several IRBs, I am gobsmacked and sickened by this. No undergrad designing such a study would pass their course: it's unethical, unable to demonstrate its stated goals, exploitive, unsafe - likely lethal on a large scale, incompetent. That our money will go to enable such corrupt and dangerous "research" is unacceptable. Senator Cassidy take note - you bear much responsibility for RFK and his sequellae. I am ashamed of our government.
What makes this so disturbing is not a single bad study or a disputed endpoint. It is the systematic bypassing of ethical and scientific safeguards that exist precisely to prevent this kind of harm.
A proven intervention is being placed at risk of withholding from newborns in a high prevalence setting, without universal maternal screening, without meaningful endpoints, and without the normal review processes that govern human subjects research. That is not innovation. It is ideological experimentation dressed up as science.
When public health leadership tolerates this, the failure is no longer technical. It is moral. This is why comparisons to Tuskegee are not rhetorical excess but structural accuracy.