By Faye Flam / Bloomberg Opinion

This week, the Centers for Disease Control and Prevention announced it will overhaul itself in response to pandemic mistakes. The first thing the CDC should do is to clarify what those mistakes were.

While many experts think those mistakes are obvious, half of the public assumes the mistakes involved too many, overly strict rules that were kept in place too long, and the other half assumes the mistakes all revolved around rules that were too loose and abandoned too soon.

Some are furious that the agency suggested vaccinated people could take their masks off in the spring and summer of 2021. Others are furious that mask mandates returned and proliferated as the country dealt with vaccine-evading variants.

The CDC is also in the business of conducting studies and here, too, some people say the agency erred in promoting its own studies before they were peer reviewed. Others are accusing the CDC of being too slow to make its data public.

It looks like a no-win situation for the organization. But transparency could help placate both sides.

The purpose of the CDC is to serve the public, and part of that is to communicate with us clearly and honestly. That means honesty about uncertainty, which is always an issue in science but more so when dealing with something that’s never happened before. (Yes, there was the 1918 flu, but covid-19 is a very different pathogen spreading in a changed world.)

Early in the pandemic, the CDC had to act on a novel situation which was threatening to cause a collapse of the health care system unless they did something right away; before there was time to do studies. That’s when total transparency about the level of uncertainty would have helped people the most. It would have seeded the ground to help people understand and accept that polices would be changing as scientists’ knowledge advanced. But the CDC was not up front about all they didn’t know.

Moreover, they never sufficiently shared the goals behind some of their recommendations. What were the goals of all the stay-at-home orders that became known as lockdowns? While it worked to “flatten the curve” in New York City and Boston, the message was they’d be lifted when it was “safe,” a goal that was known to be impossible even back then. And the lockdowns were imposed and lifted before the big surges in places like Texas, Oklahoma and South Dakota. Why was that timing so off? And as lockdowns were replaced with universal masking, what were the goals there? Under what circumstances would masking end?

By May 2020, with people still clamoring for more information about relative risks of different scenarios and activities, I found most of the useful data came from other countries, not the CDC. The media disseminated the most important consumer data about local case counts and hospitalizations, with outlets like The Atlantic and The New York Times offering user-friendly maps and graphs. And after vaccines became available, it was Bloomberg News that came out with a vaccine tracker.

This week I chatted with Baruch Fischhoff, an engineering and public policy professor at Carnegie Mellon. He emphasized that the CDC didn’t do enough to give people detailed information about covid risks and recommendations. Instead, we got too many decrees without enough explanation.

Rochelle Walensky, the current CDC director, was slammed for ending and then re-instating mask recommendations after the vaccine rollouts, though at least she did explain that her U-turn came from a study out of Provincetown demonstrating how fast the new variant, delta, could spread among the vaccinated.

There was less explanation for her recent announcement to stop recommending screening of people with no symptoms, end quarantines for people who are exposed, and to shorten the isolation period for people who’ve tested positive from 10 days to five. (The last of these changes, in particular, seems to fly in the face of studies that show people can be infectious well after five days, and that testing can be an effective way evaluate whether they remain infectious.) The CDC should more clearly lay out the science.

The CDC, as well as the National Institues of Health and the Food and Drug Administration, should also adopt the same stance as universities and private companies in allowing staff scientists to talk to the press unimpeded. As it stands now, journalists have to get approval from a public affairs office, which can often take days. There’s no benefit to the public in this. Journalists often consider questions or details that the experts hadn’t or find clearer ways to explain studies, data and policy changes. Generally, the more help journalists get from scientists, the better their stories.

Science communication doesn’t have to be dumbed-down to be clear. In fact, if a decision is logical and honest, then explaining it in full detail is going to be clearer than leaving out steps. But not all policies are honest. As risk-communication expert Peter Sandman has pointed out on my podcast and his website, sometimes public health officials shade the truth because they think it’s better for us; that is, they tell noble lies. But nobility is often only in the mind of the liar. Honesty is the best policy in the long run; and the CDC’s director could start by clarifying which mistakes she things warrant an overhaul and why.

Faye Flam is a Bloomberg Opinion columnist covering science. She is host of the “Follow the Science” podcast.