The Centers for Disease Control and Prevention has published new data on the risks of hospitalization and death from COVID-19 among people who are unvaccinated and vaccinated, with or without booster doses.

The agency recommends booster shots for Americans 12 and older. This is the first comprehensive data on the effectiveness of boosters by age in the United States.

The figures, published Thursday night, confirm that booster doses are most beneficial to older adults, as the CDC has previously reported. But the new numbers for younger Americans were less compelling.

In those age groups, vaccination itself — two doses of the Moderna or Pfizer vaccines, or one dose of the Johnson & Johnson vaccine — decreased the risk of hospitalization and death so sharply that a booster shot did not seem to add much benefit.

The data runs only through the end of December, when the omicron surge had just begun. Because the variant is so highly contagious, booster shots may have helped limit the variant’s spread through the population, an argument for boosters that would not be fully captured in the new research.

Still, several recent studies have found that vaccination alone, without boosters, remained strongly protective against severe illness and death in most people, even after omicron’s appearance.

“I do not think these data support a universal booster rollout for everyone,” said Dr. Celine Gounder, an infectious disease expert and public health researcher at Kaiser Health News.

Instead, boosters seem most essential for older adults, she said, and those who have certain immune conditions or live in long-term care facilities. In younger Americans, it may have made sense to make booster shots available only to those with certain medical risks, she said.

The advantages of booster shots in various age groups were hotly debated in the fall, when the delta variant was the primary form of the virus in the United States. But many scientists came to favor additional doses after the arrival of the highly contagious omicron variant.

“The effect of the booster can be seen in the datasets, but it’s far smaller than the effect of vaccination compared to not,” said John Moore, a virus expert at Weill Cornell Medicine in New York. “The real problem is the carnage among the unvaccinated.”

Unvaccinated people in every age group are at higher risk of infection, hospitalization and death than those who have been immunized, according to the CDC’s data — a persistent trend since vaccines were introduced.

As of Dec. 25, the rate of hospitalization among unvaccinated adults older than age 65 was 246 per 100,000 people. That rate dropped to 27.4 per 100,000 among people who were vaccinated without a booster dose, and to 4.9 among those who were vaccinated and received a booster.

There were roughly 44 deaths per 100,000 unvaccinated adults 65 and older. Vaccinations dropped that number to about 3.6 deaths per 100,000, one-twelfth as much. Booster shots reduced the rate further, to about 0.5 deaths per 100,000, a figure 90 times as small.

But such risk comparisons were less useful in younger people, for whom the rate of severe outcomes was already low.

Among adults 50 to 64, 73 unvaccinated adults per 100,000 were hospitalized, compared with 9 per 100,000 among those who were vaccinated and 2 per 100,000 among those who had also received a booster shot.

Boosters made less of a difference in the number of COVID-19 deaths in this age group. Vaccinations decreased the rate to 0.4 deaths per 100,000 from 8.26 per 100,000. With boosters, that number fell to 0.1 deaths per 100,000 people.

“This is the difference between a relative risk reduction and an absolute risk reduction,” Gounder said. “If you’re starting off with a relatively low risk, and you further reduce that risk, in the big picture that may not be such a big impact.”

The agency did not provide hospitalization numbers for adults 18 through 49, perhaps because the numbers were too small. Gounder added that such data “would be really helpful in guiding decisions about boosters.”

The CDC also did not release data for children 12 and older, possibly because boosters have not been recommended for long enough in that age group to have generated meaningful numbers.

The risk of COVID-19 death among Americans ages 18 to 49 was low. The rate was about 0.9 per 100,000 people among the unvaccinated and plummeted to 0.03 among people who were vaccinated. With the addition of a booster, deaths were too low to measure.

“I’m in favor of boosters, but I don’t want to overstate their importance,” said Dr. Dan Barouch, a virus expert at Beth Israel Deaconess Medical Center in Boston.

“The benefit of a booster dose is clearly greater in the elderly,” he added. It “is progressively less in the lower risk groups.”

One argument for a universal booster recommendation was to limit the number of infections among all groups and reduce the spread of the virus. The omicron variant was able to sidestep immune defenses and infect many more people than previous variants, including many who were vaccinated.

Booster doses cut the risk of infection in vaccinated people by about half across all age groups. The gap between unvaccinated and vaccinated groups was much greater. The numbers most likely reflect only symptomatic cases and may be different for asymptomatic infections, Barouch noted.

The figures may also vary with time: Boosters were administered fairly recently, and early data from Britain suggests that their effect may wane in a few months, he said.

Some people have worried about persistent symptoms even after a mild bout of COVID-19. But a recent study suggested that the risk of so-called long COVID-19 is highest among people with one of four predisposing factors, including Type 2 diabetes and the presence of autoantibodies.

If the CDC had taken a more targeted approach to booster recommendations, younger adults who have one of those risk factors and were concerned about long COVID-19 could still have chosen to receive a booster, Gounder said.

Still, the new data may not make any difference — to the CDC’s recommendations on boosters, or to the general public.

“The booster policy is the booster policy. It’s not going to change,” Moore said.

———

This article originally appeared in The New York Times.