On December 14, 1999, a previously healthy infant was admitted to a Michigan hospital with diarrhea and jaundice. Within hours, doctors diagnosed acute liver failure caused by hepatitis B.

She died three days later.

Her mother had tested positive for hepatitis B at her first prenatal visit. She attended 10 appointments. But somewhere between that positive test and the delivery room, the information vanished. The record provided to the birth hospital stated the mother was “hepatitis-negative.” By the time the infant received her first vaccine at 2.5 months, the virus had already taken hold.

This was not an isolated failure. The Immunization Action Coalition documented more than 500 similar transmissions from 1999 to 2002: positive tests not reported, results lost, records transcribed incorrectly.

Universal hepatitis B vaccination at birth exists precisely because screening systems fail.

On December 5, 2025, the Advisory Committee on Immunization Practices (ACIP) for the Centers for Disease Control and Prevention (CDC) voted 8 to 3 to end the universal birth-dose recommendation for infants born to mothers who test negative for hepatitis B surface antigen (HBsAg). 

In its place, the committee adopted “shared clinical decision-making”—a framework that removes the default and allows delay of the first dose until at least two months of age. Infants born to mothers who are HBsAg-positive, or whose status is unknown, are still recommended to receive the birth dose. For the first time in decades, the lives of more infants will now again depend on screening systems.

The stakes not subtle, the biology unforgiving

Unlike adults, who clear hepatitis B more than 95% of the time, infants are immunologically defenseless. Approximately 90% of those infected in the first months of life will develop chronic infection. One in four will die prematurely from cirrhosis or liver cancer—diseases that emerge decades later, long after anyone remembers the missed vaccine, the lost lab result, the newborn who never had a chance.

When a newborn is exposed at birth, hepatitis B immune globulin (HBIG), which provides immediate but temporary protection, plus the birth dose is approximately 94% effective when given within the first 12 to 24 hours. The new ACIP guidance allows delay of the first dose until at least two months for infants of mothers who test negative. For an exposed newborn, that delay is the ballgame. There is no second chance.

The committee’s decision assumes prenatal screening reliably identifies at-risk infants. It does not.

About 14% of pregnant women in the United States are never tested for hepatitis B. About 2% of pregnant women receive no prenatal care at all. Among those who test positive during pregnancy, only a third receive recommended follow-up, and the National Perinatal Hepatitis B Prevention Program identifies fewer than half of infants born to infected mothers each year. Those are only the gaps we can measure.

The committee’s decision assumes prenatal screening reliably identifies at-risk infants. It does not.

Screening cannot catch what it cannot see. Half of the 2.4 million Americans with chronic hepatitis B do not know they are infected. People can seroconvert (become newly infected) after a negative test. Infants can be infected by household contacts—the virus survives on surfaces for at least seven days, and transmission rates among susceptible household contacts of infected persons range from 14% to 60%. Before universal birth dose vaccination, US-born children of immigrant parents who themselves were uninfected still showed hepatitis B rates of 7% to 11% from community exposure alone.

The birth dose addressed every one of these vulnerabilities. It did not require perfect screening, perfect records, or perfect follow-up. It simply vaccinated every newborn.

There is no new safety signal. Forty years of randomized trials, national surveillance, and long-term follow-up have consistently found the birth dose to be safe. A recent systematic review found no benefit to delaying the first dose and no increase in serious adverse events among infants vaccinated at birth.

The evidence did not change; the committee did

For clinicians, the ACIP vote creates an immediate practical problem: Federal guidance now conflicts with the professional standard of care.

Within hours of the vote, the American Academy of Pediatrics (AAP) issued a statement calling the decision “irresponsible and purposely misleading.” The AAP, the American College of Obstetricians and Gynecologists, the American Academy of Family Physicians, and the Infectious Diseases Society of America have reaffirmed that universal birth-dose vaccination remains the standard of care for their members. 

The Red Book—the AAP’s authoritative reference on pediatric infectious diseases—has not changed. Hospital standing orders, in most systems, have not changed.

Physicians now face a choice between following a federal advisory committee that abandoned its own evidence-to-recommendation framework and instead following the professional societies whose guidelines define their standard of care. This creates immediate medicolegal uncertainty: In a case involving a preventable perinatal hepatitis B infection, a plaintiff’s attorney will cite the AAP guideline, not the ACIP vote. A reconstituted federal panel’s deviation from established science does not constitute a defensible standard of care.

This creates immediate medicolegal uncertainty.

For now, the answer is clear: follow the Red Book. But the question of where to look for guidance is no longer simple. ACIP was designed to provide a unified, evidence-based national standard. That function has been compromised. 

Clinicians must now triangulate between professional society guidelines, state health department directives—California, New York, Illinois, and others have already reaffirmed the birth dose—and their own institutional policies. This is not how vaccine guidance is supposed to work in a functioning public health system.

Insurance implications deserve close attention

Under the Affordable Care Act, private insurers must cover vaccines recommended by ACIP without cost-sharing. The shift to “shared clinical decision-making” can preserve that coverage—vaccines under this designation, like meningococcal B, remain covered—but only if the CDC director formally adopts the recommendation and publishes it on the immunization schedule.

The current Health and Human Services (HHS) secretary has made no secret of his skepticism toward the childhood vaccine schedule. If implementation or language shifts—if HHS declines to ratify the recommendation, or publishes it in a way that suggests the vaccine is “not recommended for routine use”—insurers gain room to add friction. Cost-sharing requirements or coverage limitations would transform the birth dose from a default protection into a financial barrier for families without robust insurance.

The Vaccines for Children (VFC) program will continue to provide the vaccine free for Medicaid-eligible and uninsured children. But VFC administration fees are already inadequate, and the added counseling burden of shared decision-making is not compensated. Safety-net clinics—the providers that serve the populations most vulnerable to screening failures—face a financial disincentive to engage in the very conversations the new policy requires.

This is how vaccine coverage erodes: not through outright prohibition, but through administrative friction, funding ambiguity, and the quiet accumulation of barriers that fall hardest on those with the thinnest margin for error.

Modeling estimates project 1,400 additional chronic pediatric hepatitis B infections per year, approximately 300 additional cases of liver cancer, and 480 preventable deaths over the lifetimes of these children—along with more than $222 million in excess annual health care costs.

Harms will not be distributed equally

The infants most likely to fall through screening gaps are those born to families without consistent prenatal care, who are uninsured or underinsured, or live in communities where health care access is already fragmented. And infants who receive the birth dose are more likely to complete the full vaccine series; removing it raises the risk of children falling behind on all their immunizations.

The birth dose was a safety net. Safety nets exist for patients the system fails to catch. That is who will pay for this decision.

It is extraordinarily rare for countries to roll back a universal birth-dose policy once established—because the whole point is protecting infants from predictable screening and follow-up failures. The United States is now doing so, not because the science demanded it, but because a reconstituted committee, presented with ideology dressed as inquiry, voted to let it happen.

Safety nets exist for patients the system fails to catch. That is who will pay for this decision.

The infant who died in Michigan in 1999 fell through gaps that universal vaccination was designed to close.

On December 5, 2025, ACIP voted to reopen them.

Dr. Scott is a clinical associate professor of infectious diseases at Stanford University School of Medicine and first author of “Updated Evidence for Covid-19, RSV, and Influenza Vaccines for 2025–2026” in the New England Journal of Medicine.