(CNN) Federal health officials surprised just about everyone this week when they announced a change to guidance for fully vaccinated people, saying most should resume wearing masks inside.
The US Centers for Disease Control and Prevention also said everyone should wear masks in school -- students, teachers, staff and visitors.
CDC Director Dr. Rochelle Walensky said new data had convinced her the Delta variant was "behaving uniquely." She said the evidence indicated that fully vaccinated people who have breakthrough infections involving Delta may be as likely to transmit virus to others as unvaccinated people are.
Plus, she said, people living in areas of high or sustained transmission should start wearing masks in public again -- even if they are vaccinated -- because of the higher risk of becoming infected when more virus is circulating.
But she didn't present the data, and the CDC's website still indicates only 5,914 people have become severely ill with Covid or died despite having been vaccinated.
This as the CDC issued a health alert and held a rare briefing Tuesday to warn that the Delta variant was spreading more rapidly than previous variants, and urging people to mask back up and get vaccinated.
Had the CDC been doing more to report breakthrough infections, the change to the masking guidance might not have taken the country so much by surprise.
Instead, the seemingly sudden change in guidance looked like a flip-flop.
And it calls into question President Joe Biden's repeated promises that his administration would not only follow the science, but would be transparent and open about it.
No vaccine is 100% effective and studies are showing that while the three authorized vaccines protect well against serious disease and illness, they are less effective in preventing any type of infection, including a mild case of Covid-19 or asymptomatic infection.
The CDC has tracked serious infections requiring hospitalization and deaths among fully vaccinated people. But it's not publishing that data in a timely manner, and it has stopped national tracking of mild and asymptomatic infections.
That makes it hard to know if fully vaccinated people are, in fact, less likely to pass along the infection to others. It's hard to say whether people infected with the Delta variant despite having been vaccinated are more likely to infect others.
And it's hard to be confident that the virus is not infecting millions of fully vaccinated people and perhaps evolving in their bodies under pressure from a vaccinated immune system -- a scenario that is unlikely but that could, in theory, give rise to new vaccine-impervious variants.
"I think it is important to track the symptomatic breakthrough infections. It is important to track hospitalizations and deaths -- but if you don't track all symptomatic ones, from the epidemiology point of view you don't know the frequency and severity of those symptomatic breakthrough infections," David Holtgrave, dean of the School of Public Health at the University at Albany, State University of New York, told CNN.
"Plus, you miss the opportunity to follow up with a sample of people with symptomatic breakthrough infections to see if they develop long haul symptoms. I don't think we have a good handle on that. We could use better data."
This lack of data could also have the effect of reducing vaccine confidence, Holtgrave said.
"If you want to be able to assure the nation that breakthrough infections are rare -- and I believe symptomatic breakthrough infections are relatively rare even in the time of the Delta variant -- you have to be able to say 'we went out and looked very hard for those infections and we didn't find many and when we did, they weren't that severe,'" Holtgrave said.
"You are only as reassuring as the extent to which you went out and looked."
The CDC did do this in looking for rare side-effects from the Covid-19 vaccines, including a rare blood clotting condition linked with the Johnson & Johnson vaccine and an unusual inflammatory response affecting the heart called myocarditis linked with the Pfizer and Moderna vaccines.
Walensky promised more data is forthcoming.
"In the coming days, you will actually see the published information on the science that motivated this change," she told CNN's New Day Wednesday.
She made a similar promise Tuesday.
"We are collecting passive reporting data on people who are hospitalized and who have died, but we recognize that, epidemiologically, that is not going to give us the best information with regard to rates of breakthrough infection, because passive data collection is generally underreported," Walensky told reporters in a telephone briefing.
"In order to counter that, we have been collecting data through more than 20 cohorts of people. These include tens of thousands of people who we are following nationwide, and they include health care workers, essential workers, long-term care facilities, and in some of these cohorts, we're collecting PCR (test) data from every person in them weekly," Walensky added.
"So we are absolutely studying and evaluating breakthrough infections in many different sites, many different people, across the country. We are looking at those data on a weekly or bi-weekly basis and we will be reporting on those soon."
Following people and testing them weekly, whether they show symptoms or not, is the best way to track asymptomatic infections.
Having the number of mild and asymptomatic breakthrough infections would also help people better judge the risk of developing a severe infection despite having been vaccinated, said Jennifer Nuzzo, an epidemiologist and senior scholar at the Johns Hopkins Center for Health Security.
"The number gets reported but not having the denominator makes it harder to put into context," Nuzzo told CNN. It might reassure people to know that while there are many breakthrough infections, only a very small percentage of them are severe enough to cause concern.
"There is a public fascination with people who contracted Covid despite being vaccinated and I think we need a way to answer that," Nuzzo added.
It wouldn't even be necessary to count every single case, she said. A systematic sample -- like the cohorts Walensky described -- could be sufficient.
Dr. Tom Frieden, a former CDC director who is now CEO of health advocacy group Resolve to Save Lives, makes a similar point.
"It is neither necessary nor wise for our public health systems to attempt to track every single breakthrough case that causes no or only mild symptoms," he wrote in a recent opinion piece for CNN.
"There are simply going to be too many, and there is little benefit to tracking illnesses that are no worse than a cold. Instead, it makes sense -- as is being done -- to do special studies of all breakthrough infections as part of specific population-based studies, and also to try to track every severe Covid illness after vaccination," he added.
"This should give us much better information about who is most at risk of serious illness from Covid after vaccination,and might lead to different dosing recommendations for some groups."
Having more data on breakthroughs might also help epidemiologists better track changes in how the virus is spreading, Nuzzo said.
"I'd like to see more sequencing to see chains of transmission and clusters and where people are getting sick," she said.
Better data on breakthrough infections would also support the argument for vaccinating the entire population to provide herd immunity.
The lack of data to back up the changed guidance gave an opening to Arizona Gov. Doug Ducey and other Republican governors to say the changing guidance would diminish confidence in the vaccine -- and to double down on their refusal to do more to get their citizens to get vaccinated and wear masks.
It also creates the impression that the new guidance was issued for political reasons.
"Do you really think that CDC has changed mask guidance for vaccinated adults, to use Dr. Walensky's words, because ... a vaccinated person may transmit virus to an unvaccinated person? No," Dr. Jonathan Reiner, a professor of medicine and surgery at George Washington University, told CNN Tuesday night.
"They changed guidance to do the right thing, which was to force basically unvaccinated people to be vaccinated."
Some people who are resisting vaccination say it's a personal choice, affecting only themselves.
Dr. Paul Offit would beg to differ.
"There's an old study of measles transmission between 1999 and 2000. It's going to sound counterintuitive, what they found but if you think about it, it makes sense," Offit, a vaccine adviser to the FDA and director of the Vaccine Education Center at Children's Hospital of Philadelphia, told CNN's Jim Acosta last week.
"Obviously, you were least likely to get measles if you were vaccinated living in a highly vaccinated community. But you were actually more likely to get measles if you were vaccinated living in a highly unvaccinated community than if you were unvaccinated living in a highly vaccinated community," Offit said.
That's in line with what the CDC said Tuesday. The CDC might benefit from showing its sources, also.