Hepatitis B can be a silent killer. It can pass from mother to child during birth and can spread through contact with infected blood or semen. Among babies infected before they turn 1, up to 90% develop a chronic, potentially lifelong infection, for which there are treatments but no cure. People may not even know they have the virus until their liver fails or they develop cancer, years later. Early-in-life hepatitis B infections, left untreated, lead to a 1 in 4 lifetime chance of developing liver cancer.
In the U.S., hepatitis B vaccination became routinely recommended for all infants in 1991; later guidance specified that the first dose should happen within 24 hours of birth. However, in Dec. 2025, the U.S. Centers for Disease Control and Prevention (CDC) under the Trump Administration moved to cease recommending this practice for babies born to mothers who test negative, instead advising a shared decision-making model where parents and clinicians decide together what is medically appropriate: a dose shortly after birth, or a first dose "no earlier than 2 months of age." (For babies born to mothers who test positive or whose status is unknown, the CDC continues to recommend a dose shortly after birth.)
A new study, published today in Annals of Internal Medicine, uses computer modeling to simulate the hepatitis B infection status of an entire year’s worth of children born in the U.S. under the new recommendations over their first 20 years of life. The researchers find that if only about 26% of children are vaccinated—an estimate of what would occur under shared decision-making—chronic hepatitis B cases would rise by 225%.
When should kids get the first dose of hepatitis B vaccine?
The American Association of Pediatrics continues to recommend vaccination at birth against hepatitis B for all infants, and 15 states have sued the Trump Administration over vaccination schedule changes.
“It's been challenged in many states and by professional societies, but we wanted to add some evidence to the conversation about what the potential impact of those changes could be,” says Dr. Amir Mohareb, an infectious-disease researcher at Harvard Medical School and Massachusetts General Hospital who is an author of the new study.
The simulation modeled how many of 3.6 million children, born the same year in the U.S., would develop hepatitis B under a number of scenarios, including if the former CDC recommendations continued to be followed, if the initial dose (also called a “birth dose”) was delayed, and if a shared decision-making scenario was employed.
The model suggests that if the first dose is delayed until 2 months old, and only 75% of infants get it, then there would be a 19% increase in chronic cases compared to if every child received their first dose at birth. With a 26% vaccination rate at 2 months, the number climbed to 225%.
This study modeled only at a single year of births, notes Mahoreb. “Imagine that happening year after year after year after year. That could quickly become thousands and thousands of new infections, and even a small number of new infections could result in lifelong chronic infection that could be detrimental to families, lead to more cases of cancer, and [become] costly to the health care system,” Mohareb says.
Two other recent modeling studies have looked at the impacts of changing hepatitis B vaccine recommendations. One found that if vaccination no longer occurred at birth, a substantial increase in screening pregnant women for the virus would be necessary to keep infection numbers in infants down. Another, looking at the economic ramifications, found that even if every child eventually received the vaccine, delaying the first dose by 2 months yielded an additional $16.4 million in health care spending on new infections.
“All three of these studies complement each other, in a good way, and provide some modicum of evidence so that we don't need to risk doing a clinical trial or testing this on people,” Mahoreb says. “We have done those experiments.” Decades ago, he continues, clinical trials already proved the efficacy of the existing approach, “so that we would have the benefit now of advancing medical progress and not needing to reopen that question.”