Lipstick on a Pig: Why the Amended Hepatitis B Birth-Dose Trial in Guinea Bissau Remains Ethically Indefensible

The last 48 hours have been a veritable roller coaster regarding the study I first wrote about in December 2025. This is a study funded by the CDC through HHS, designed to examine the so-called non-specific effects of childhood vaccines in Guinea-Bissau, using a trial design that plans to randomize 7,000 newborns to not receive the hepatitis B birth dose. For a few hours yesterday, myself and researchers around the world who have been following and speaking out about this planned unethical clinical trial breathed a sigh of relief when The Guardian (US) reported that the trial may not proceed as planned.

That sense of reprieve was short-lived. Very quickly, a revised protocol surfaced online, the study’s backers (HHS) made it clear that they intend to proceed on schedule. These contrasting reports murky the water on what the actual status of the trial is. After reviewing the amended protocol, one conclusion is unavoidable, this revision is little more than lipstick on a pig. The core ethical violations that prompted global concern remain intact, and in some respects the investigators have doubled down.

At the heart of the study, nothing fundamental has changed. The investigators still plan to randomize approximately 7,000 newborns to not receive the hepatitis B birth dose a vaccine that is globally recommended and known to prevent perinatal transmission of a lifelong, potentially fatal infection. That single fact alone should give any clinician, ethicist, or public health professional pause. Hepatitis B acquired at birth is not a benign exposure. In high-prevalence settings like Guinea Bissau, perinatal infection carries a 70–90% risk of chronic infection, with downstream risks of cirrhosis, hepatocellular carcinoma, and premature death. In Guinea-Bissau, where up to 18% of the population is hepatitis B surface antigen positive, these are not speculative harms for babies exposed at birth. Randomizing infants to forego a proven preventive intervention at the moment of highest risk is deliberately exposing children to preventable harm.

The revised protocol makes procedural clarifications, but it does not address the ethical fault line at the center of the trial. The investigators continue to argue that there is “equipoise”( the core ethical idea that a clinical trial can only proceed if there's genuine, balanced uncertainty within the expert medical community about the relative merits of the interventions being compared) because hepatitis B birth-dose vaccination is not yet universally implemented in Guinea-Bissau and because many mothers are not screened for hepatitis B. That argument fails scientifically and ethically.

Absence of implementation is not absence of obligation. The fact that a health system has not yet achieved universal coverage does not create moral permission to randomize children away from a WHO-recommended standard of prevention especially when that standard exists precisely because the consequences of perinatal infection are so severe. The amended protocol itself acknowledges both the high prevalence of hepatitis B among women in Guinea-Bissau and the effectiveness of the birth dose in preventing vertical transmission, while still choosing a design that withholds that protection from half of enrolled infants.

A question that continues to arise is whether there is any way this trial could be conducted ethically. The answer is yes but with important caveats.

A stepped-wedge or phased rollout design is a possible ethical alternative. In theory, such a design can be defensible, but only under very specific conditions. The key distinction is whether delayed access is unavoidable, or whether it is being manufactured by the research design.

A phased rollout can be ethically justifiable if a government and its partners genuinely cannot vaccinate all newborns immediately because of binding constraints e.g. limited vaccine supply, cold-chain capacity, staffing shortages, or transport barriers and the rollout would look essentially the same whether or not a study existed. In that case, the research is observing scarcity, not creating it. But those constraints must be real, documented, and not solvable with resources already in hand. That is not what is happening here.

This trial is well-resourced, externally funded, and operationally capable of delivering vaccines yet it chooses not to deliver the hepatitis B birth dose to all newborns in order to preserve a control group. That is not unavoidable scarcity rather it is engineered scarcity, and that distinction matters ethically. There are additional guardrails that any ethical phased rollout would require and this study meets none of them.

First, the study must not determine who gets vaccinated first. Rollout order should be driven by public health logic facility readiness, geography, disease burden and not by what produces the cleanest statistical comparison. Randomization is only ethically acceptable if it occurs between rollout sequences the Ministry of Health in Guinea Bissau would plausibly adopt anyway, not sequences invented to optimize study power.

Second, the standard-of-care floor must always be respected. Infants in later phases should receive everything they ordinarily would: maternal screening where feasible, counseling, linkage to care, and vaccination as soon as capacity allows. Vaccination should never be delayed simply because “the study isn’t finished.”

Third, the research question itself must justify the risk. Using phased rollout to study implementation challenges or confirm well-established benefits such as reductions in perinatal transmission is one thing. Using scarcity to chase poorly defined “non-specific effects” in newborns is far harder to justify, particularly given how sensitive such outcomes are to confounding, secular trends, and care-seeking behavior. These are precisely the weaknesses stepped-wedge designs are most vulnerable to.

Finally, strong local governance and transparency must anchor the entire process. Any phased rollout must be clearly ministry-led, shaped by local priorities, and communicated honestly to communities. Families must understand that some sites start later because of rollout capacity not because their children were randomized to receive less care.

None of these conditions are met by the current study.

Many scholars have pointed out that this study and the global reaction to it reflects a familiar and deeply troubling pattern in global health research: what Seyi Abimbola has described as the foreign gaze.

The foreign gaze is not about bad intentions. It is about who gets to define the problem, whose uncertainty counts, and whose lives absorb the risk when evidence is deemed incomplete. It is about research agendas shaped primarily by external institutions and priorities, while affected communities are positioned as sites of data collection rather than as equal moral stakeholders.

That pattern is unmistakable here.

The principal investigators are Danish. The sponsor is a Danish institution. The funding includes support from the U.S. government through HHS and the CDC. Yet the only ethical approval explicitly cited is from Guinea-Bissau’s national ethics committee. There is no indication of parallel ethical scrutiny by institutions in the countries where the study was designed, funded, and justified.

At a press conference this week, Dr. Yap Boum of Africa CDC indicated that Africa CDC is ready and willing to support redesigning the study in a manner that aligns with ethical standards. That involvement from regional public health leadership is welcome and important. It makes clear that concerns about this trial are not coming from outsiders alone but are shared by African public health leaders themselves.

An important voice still missing from this discourse, however, is that of public health authorities in Guinea-Bissau. Their perspectives are the most important ones here, and I hope they will weigh in soon. The country is currently navigating a difficult political moment, having recently faced an attempted coup, and that context matters.

Some may argue that criticism of this study from voices outside Guinea-Bissau particularly those from outside the African continent risks reinforcing the very dynamics of the foreign gaze it seeks to critique. To that I would say this: ethical concerns anywhere in the world must be open to scrutiny, especially when powerful institutions are involved. The responsibility is not to remain silent, but to call out harm while actively amplifying local expertise and leadership.

Abimbola’s work helps clarify the deeper issue. In global health, uncertainty is often treated as a reason to delay interventions in African settings, while the same level of uncertainty would rarely be tolerated to justify withholding protection in Europe or North America. The evidentiary bar quietly shifts depending on who bears the risk. This is how the foreign gaze operates: uncertainty becomes a virtue when it slows delivery of care to marginalized populations; scarcity becomes a justification rather than a failure to be corrected. Procedural ethics are satisfied, while structural inequities persist.

Seen through that lens, the problem with this trial is not only about hepatitis B vaccination. It is about whether African newborns are once again being positioned as acceptable sites of uncertainty for questions that would never be answered this way elsewhere.

On the Investigators’ Defense

After my initial substack in December, the lead author of the trial protocol reached out with a detailed defense of the study’s ethical basis. I responded directly and point by point. I will link that full exchange here so readers can review it in its entirety here. If nothing else, the widespread critique clearly did not fall on deaf ears. But it was deeply disappointing that the response ultimately refused to grapple with the study’s ethical blind spots and instead doubled down on defending what remains indefensible.

If the investigators truly care about the health of children in Guinea-Bissau, there is a far more ethical, impactful, and immediate use for the $1.6 million allocated to this study. That funding would save more lives if it were used now to ensure that as many newborns as possible receive the hepatitis B birth dose rather than to fund a trial that withholds it from thousands of infants while the government plans to roll out the intervention nationally in 2027 anyway.

The question is not whether hepatitis B birth-dose vaccination will be implemented in Guinea-Bissau. That decision has already been made. The question is what we choose to do in the meantime. Do we protect children when we know how or do we study them while leaving them exposed?

History will not judge this trial by its statistical power, its composite endpoints, or its theoretical contributions to the literature on vaccine interactions. It will judge it by a far simpler standard: When we had the resources to prevent harm, did we choose to do so?

For now, the fact that this study is paused, under close scrutiny and, that further revisions may yet come feels like a battle won. But the war is far from over. We will continue to watch closely, to speak plainly, and to insist that the protection of study participants comes first.

If you enjoyed reading this piece and value evidence-based analysis, ethical clarity, and accountability in public health, global health, consider subscribing. I write regularly about infectious diseases, vaccines, public health policy, and the real-world consequences of scientific decisions.