How patchy COVID data hampered the pandemic response

A medical worker held a COVID-19 swab test at East Boston Neighborhood Health Center on Jan. 26.Barry Chin/Globe Staff/file

WASHINGTON — On Tuesday, the Centers for Disease Control reported more than 650,000 new positive cases of COVID, a daily figure that serves as a crucial barometer for a nation muddling its way through the Omicron wave.

But the actual number of new infections, which may well be higher, is a mystery — just like the overall number of people who have been infected so far.

Two years after COVID began spreading in this country, the United States is operating with patchy and incomplete data about the virus, a problem experts say has hampered the response to one variant after another and leaves the country just as vulnerable to the next one.

“We miss opportunities to possibly get early indicators and warnings,” said Jessica Malaty Rivera, an epidemiologist with the Pandemic Tracking Collective. “We’ve just been in a constant game of damage control.”

Experts say US officials are missing critical pieces of information, including comprehensive data on breakthrough infections and adequate demographic information about who is catching COVID. Much of the existing data is gathered in a clunky, siloed system dogged by gaps and delays — and there appears to be little prospect of a full overhaul. And many epidemiologists say the United States lags well behind other nations in techniques to quickly and accurately track the virus’s spread, including sequencing virus samples’ genomes to detect emerging variants.

“When you don’t have the data, you cannot make a sound public health decision … and then you’re catching up and the virus is in charge and in the driver’s seat,” said Ali Mokdad, a professor of health metrics sciences at the University of Washington.

The country’s creaky data infrastructure, its critics contend, has contributed to bad public health guidance on masks, delayed the government’s recommendations around booster shots, and slowed the country’s preparations for Omicron. The Biden administration did not step up its response to the new variant by mailing free tests to people’s homes, for example, until mid-way through January — weeks after the variant began tearing through the nation and racking up record case totals.

“We were not prepared because we were not looking at the data and our country was lagging behind collecting the right data and acting upon it,” Mokdad said.

The problem has many scientists looking wistfully at countries with centralized health care systems and faster, more comprehensive systems for data collection and analysis.

“Every country that’s managing the pandemic reasonably well has a weekly report, with everything in it — genomic surveillance, the booster, vaccine effectiveness, I mean, everything. We have nothing. And that’s because we don’t have the data to generate such reports,” said Eric Topol, a professor of molecular medicine at Scripps Research.

While the collection of COVID data has improved since the beginning of the pandemic, the US’s decentralized, understaffed, and outdated public health system puts it at a severe disadvantage compared to many other countries, and CDC officials bleakly admit they are unlikely, because of that, to ever be able to get as broad a look at the state of COVID in the nation as they might like.

“It’s not that we’re flying blind, because we’re using data from different sources,” said Daniel Jernigan, deputy director for public health science and surveillance at the CDC. But, he acknowledged, “there is not a big, comprehensive, all-encompassing view. That may be something we never will have.”

Instead, he said, the CDC pulls together its view of the pandemic through a combination of state and local data, targeted studies and investigations, and other networks that gather patient information. The agency is working on modernizing data collection, he said, and launching a new disease forecasting center.

“Having a coordinated system across the US, across different health care systems, across different jurisdictions that would be a wonderful thing,” Jernigan said. “We have parts of that, but because of the way our health care systems [are] set up… that single, coordinated view across the US for that data, is just not as timely as it needs to be, and as representative as it could be.”

CDC Director Dr. Rochelle Walensky has cited the difficulty of obtaining good, fast data on a rapidly evolving virus as she has sought to defend her agency — and her leadership of it — against charges that it has muddled its messages and issued guidance that had to be revised.

“Sometimes that means that you’re making decisions when the science is evolving and emerging, sometimes grayer than you would like it to be, sometimes more imperfect than you would like it to be,” she said in an interview with the Globe earlier this month.

The reasons for the country’s spotty public health data are many. PCR tests, a key source of data, have been hard to come by. America does not have nationalized health care, which allows countries like the United Kingdom to gather data more easily, and its system is thick with private institutions and regulations that hamper the sharing of data. The CDC does not have the authority to mandate that states or private health systems collect and submit data in any particular way, and some scientists believe the White House and the Department of Health and Human Services have failed to prioritize improving the nation’s data infrastructure.

But in some cases, the CDC has declined to collect crucial information even when it had the power to do so, such as when it decided last year not to track breakthrough COVID cases that did not result in hospitalization or death.

“We have to know the totality of breakthrough incidents, and the CDC has not prioritized getting those raw numbers,” said Rivera, whose organization is working on tracking breakthrough cases. “It’s a huge disappointment.”

Jernigan said the agency limited its tracking of breakthrough cases to ensure it studies only cases in which the outcome is known. But outside experts say that has not yielded enough information.

“This has hampered our ability to come up with a public health recommendation, and tell the public, go and get a booster much earlier,” said Mokdad, the University of Washington professor.

The data holes have drawn bipartisan attention on Capitol Hill. Representative Ayanna Pressley of Boston has been calling for better demographic data since the onset of the pandemic so lawmakers can come up with policies that address the disparate racial and economic effects of the virus. She is also pushing for demographic data on long COVID.

“That is the through line of the power of collective data, knowing where the gaps are, what is the disparate impact, and then fighting for resources to address that accordingly so that we can save lives,” she said, adding that good data will inform decision-making around any future federal relief package.

Beth Blauer, the executive director of the Centers for Civic Impact at Johns Hopkins University, where she is in charge of data for the Coronavirus Resource Center, said that the collection of data around race and ethnicity has improved since the start of the pandemic — but it is still available for only two thirds of cases, and not standardized across the country.

“At what point do we just say this is kind of a continuation of why we have health disparities to begin with in this country?” she asked.

The gaps in data don’t stop there. Many localities and hospital systems are struggling to parse who is in the hospital because of COVID and who is there with the virus as a secondary condition — information that could help deluged hospitals with planning and staffing and give a better sense of the virulence of a given outbreak.

And, although it has ramped up, experts are deeply frustrated that the nation does not have more sophisticated surveillance capabilities to allow officials to use national wastewater data or random sampling to catch outbreaks and new variants much earlier.

Researchers now have evidence that the Omicron variant was already in the country before Thanksgiving. John Brownstein, a researcher and the chief innovation officer at Boston Children’s Hospital said that if real-time surveillance data had shown that to be the case, “that would have impacted how we would have targeted public health messaging.”

“The sooner that we can take that and jump on a variant and tune our response,” he said, “the better the outcomes are.”

Much of the data collected nationally comes from a patchwork of state and local public health departments, which have to navigate a maze of disparate sources to make their reports, compiling figures from hospitals, community health centers, and more.

“The data that goes to CDC that comes from the states is only as good as the data the locals send to the states,” said Dr. Allison Arwady, the commissioner of Chicago’s Department of Public Health, who said gathering it is no easy task. “I still get stuff in the mail, and faxes.”

Some experts warned that, in some ways, the data picture could get worse instead of better going forward. Many states have ramped down their reporting frequency since last year. In late May of last year, 31 states reported COVID data every day, according to the Coronavirus Resource Center at Johns Hopkins University. As of Monday, only 15 states and Puerto Rico were still doing so.

“That, for me, is a signal that everyone involved saw this as temporary infrastructure,” said Blauer.

And the rise of at-home rapid tests — including the billion the administration is mailing to US households — could take another bite out of the data public health officials are able to see, since the results of the tests are often not reported or there is no way to report them.

“They’re releasing these hundreds of millions of rapid tests with no intention of actually collecting any data, even anonymously, about these tests,” Brownstein said. “Why are we not using them for public health surveillance?”

Arwady, the Chicago public health commissioner, said big-city departments like hers are relatively well resourced. But the lack of a more systematized approach to gathering data nationally, as well as better genomic surveillance, has left the country with worrying gaps that could allow a new variant to emerge unnoticed.

“If something emerges … in a more rural part of the state, it would take us quite a while, realistically, as a system to fully identify it,” she said. “We will miss things.”

Jernigan, of the CDC, agreed there is much more to do.

“Are we ready for the next pandemic? No, there’s a ways to go,” he said.