From his home in Battle Ground, Clark County, middle school teacher William Baur has paid close attention to any COVID-19 news about schools. He has dug up available data on infections in his district. And he seldom went a day without checking his county’s infection rates. 

So in mid-December, when Washington officials dramatically loosened guidance for reopening school buildings, Baur was perplexed. He hadn’t seen new local data to support such a move, so to find out if there was a good reason, he filed 26 requests with county and state health officials for the number of cases recorded in schools. 

The justification Baur was looking for — a robust and well-studied log of cases in Washington schools that pointed toward safety — wasn’t what led health officials to change the guidance. That research doesn’t exist. 

Rather, it was a decision based on principle. State officials believe that, for academic and mental health reasons, more students should be learning in-person – a view that fits with recommendations from the nation’s top disease experts, including Dr. Anthony Fauci.

Whether to teach in-person or online is a decision left up to individual school districts; Gov. Jay Inslee can’t order schools back in session. But changing the metrics, state officials believed, would encourage more schools to open up.

And it seems to be working. Since the day Inslee announced plans to change school opening metrics, dozens of school districts have started bringing kids back into buildings, including Bellevue, which brought many second graders back on Thursday, and Tacoma, which brought kindergartners back Tuesday. Seattle, Lake Washington, Auburn, Highline and Federal Way have all announced plans to bring the youngest learners back to school in February or March, although some of those plans may depend on negotiations with teachers unions.


State officials say about 85% of students in the state are learning remotely. But with the reopenings, that number will soon change.

The new metrics rely on a flurry of research papers released by Washington disease modelers over the summer, suggesting that hybrid learning and safety measures could help reduce transmission if it’s low in the community, and that districts should start by reopening elementary schools, since young kids are less likely to get sick from, or spread, the virus. And when some schools reopened in the fall, data hinted that schools aren’t superspreaders. 

But there have been few research studies confirming this finding, which is based on crowdsourced, not representative, data. The state acknowledges collecting little data of its own. And there’s limited research to explain a host of other phenomena, such as why some schools where COVID-19 is surging seemed to open safely, while many did not. 

“It’s not about faith in the data, it’s about what we value from a policy perspective,” said Dr. Thomas Tsai, assistant professor of health policy and management at Harvard University’s T.H. Chan School of Public Health. Tsai said we’ve come a long way in devising school safety measures, but acknowledged that the quality of available data on COVID-19 and schools is “poor.” Researchers still have few hard facts on the basics, including infection rates in teachers and students, and transmissibility in schools. 

But, “the question, we’ve realized, is the consequences of keeping schools closed,” Tsai said.

Earlier guidance, which advised keeping most students home until case rates dropped to 25-75 per 100,000 over two weeks (now it’s 350 per 100,000), was a “reasonable approach given what we knew at the time,” said Lacy Fehrenbach, one of the state’s top health officials. 


A confluence of information supported revising the guidance, she said, including evidence from other countries and states, Washington’s outbreak data and a disease modeling group’s findings.

“Those sources of information are aligning to say that if you go back into buildings, and you can do all these health and safety measures, that your chances of having an infection are barely more than if you all stayed at home,” she said.

Washington has logged at least 117 coronavirus outbreaks linked to schools, just a few less than those tied to groceries, retail stores and agricultural work. But DOH has acknowledged that its contact tracing efforts in schools are flawed, so there may be many more.

“Weighing that against the cost of being fully remote — especially for young learners — really motivated us to expand the metrics,” Fehrenbach said. Remote learning has education experts concerned about children’s mental health and their ability to socialize with peers their own age, as well as access to services that support those with disabilities.

To better understand why officials changed course, The Seattle Times examined the research studies they cite, requested the emails of public health officials and spoke with education and infectious disease experts, and reviewed a trove of data Baur received in response to his requests. 

Solo quest

Baur, who teaches science at River HomeLink in Battle Ground, has never had a reason to file a public records request — an inquiry available to any taxpayer who wants information held by the government. But Inslee’s announcement, he said, compelled him to do it.


Over winter break, often with his sleeping infant on his chest, he and a fellow teacher wrote to at least 26 of the state’s health districts and two state agencies asking for data on all COVID-19 exposures in schools since August 2020. They also wanted the number of close contacts who may have been exposed and needed to quarantine.

DOH and the state’s education department responded by saying they had no responsive records. 

But several county health departments did reply with data, including Clark, Pierce, Whatcom and Yakima. In his analysis so far, Baur says, it seems that many school cases are going unreported. Because of this, it’s likely the data only tells part of a bigger story. For instance, Clark County data showed relatively high rates of transmission in schools — but that could be because the county has more robust contact tracing efforts.

“I also found that the data they are reporting has little correlation with the stages of reopening of the school districts or the amount of virus in the community,” Baur said.

Baur’s solo quest to track down data is unique. The state doesn’t keep a school-specific COVID-19 database, and Baur’s venture may help fill in a so-far incomplete picture of the virus’s spread in schools. 

“I want it to influence public policy,” he said. “I want it to give parents an accurate picture of what COVID is like in schools.”


But his attempts to find something concrete come at the same time that policy makers are quickly forging ahead.

Washington now recommends that schools phase in in-person learning for elementary schoolers when cases are “high,” or above 350 per 100,000 residents over two weeks, and when other trends, like cases and hospitalizations, are flat or decreasing. Schools where cases are logged as “moderate” or 50 to 350 per 100,000 should also resume classes for middle schoolers, and if cases are below 50, for all students. These new thresholds are based in part on CDC recommendations, Fehrenbach said, and a review of available scientific literature. 

So far, most school outbreaks have happened in communities with “moderate” case rates — not “high” case rates — a statistic that illustrates how difficult it is to decide when returning to buildings is “safe.” This finding might be explained simply and by the fact that most districts fell into the moderate category in the fall, Fehrenbach said. But even so, she added, “We’re pretty cautious about interpreting that.”

The decision to change the metrics was a monthslong process. Health department officials had been mulling over revisions since at least late October, records obtained by The Seattle Times show. They tried to keep the process quiet at its earliest stages. 

On Nov. 3 of last year, State Health Officer Kathy Lofy emailed county health officials to ask for their input on a draft of new guidance to schools, which they were instructed not to share with any of their colleagues and school officials. With the same warning attached, Jeff Duchin, health officer for Public Health – Seattle and King County, shared his comments on the draft with three colleagues, including Patty Hayes, who directs the county department. 

The proposal “will create some chaos probably,” Hayes replied to him. Another department official recommended a strong public health statement to support school reopenings that would assuage concerns from teachers unions that there “may be a political agenda around these changes.” 


Later that week, DOH presented its proposal to Inslee’s staff. Maddy Thompson, a K-12 adviser to Inslee, said that even with the limitations of data, she felt confident in what DOH presented. 

“I’ve questioned the Department of Health and looked at the list of covid cases,” Thompson told The Seattle Times regarding the challenges of contact tracing in schools. “When you’re at a school, you’re going back every day. It’s a bit different (from) going to a restaurant,” where a different set of strangers gather every day. Local health departments have also prioritized school investigations, she added. 

Inslee met with superintendents in reopened school districts last fall to gather intelligence. The statewide teachers union unsuccessfully tried to sway him against changing the metrics. 

Throwing a wrench

There are new and urgent reasons to question reopening school buildings on a wide scale, some researchers warn.

A new study, for instance, suggests that reopening buildings for 76% or more students in a given location, for hybrid or full-time in-person learning, can drive community spread of the virus. The recent unpublished findings are from a team of University of Washington researchers who used county and district-level data from Washington. 

“There are reasons to be cautious if you start to get into higher case rates, particularly if you’ve got lots of kids in the same buildings,” said Dan Goldhaber, who led the research and is director of UW’s Center for Education Data & Research. 


The presence of a new, fast-spreading coronavirus variant, first recorded in the United Kingdom, is also raising alarm among physicians like Dr. Vin Gupta, a critical care pulmonologist and an affiliate assistant professor at the University of Washington’s Institute for Health Metrics and Evaluation. In mid-January, he said, the variant “should force us to reconsider what we’re doing with schools.”

New variants that don’t follow the same transmission pattern as original strains could amp up risk in school settings, and may throw a wrench in what we’ve come to see as established research findings in schools. 

Dan Klein, a researcher who has led some of the most prominent modeling studies on COVID-19 and schools — including modeling based on King County data that state officials say has guided their policymaking — said he’s “absolutely keeping track” of how new variants are spreading in the United States. Klein, senior research manager at the Institute for Disease Modeling, said he’s also trying to gauge how they might affect the analytical model he uses in his research. 

More highly infectious variants, he said, “would have some implications for the spread in schools and the community overall.”