The Giant Experiment That Could Keep Schools Open

The bright-blue tents appeared shortly after the close of winter break. Each Tuesday, Thalia Ruark and her classmates at the Bromfield School in Massachusetts, line up single file, spaced a neat six feet apart, for their weekly coronavirus test. The 11-year-old sixth grader still spends most of her classroom time on a computer at home, in accordance with Bromfield’s hybrid-learning model. But the school’s new testing measures are meant to keep her and her peers safer while they’re at the school, which is in the rural town of Harvard, some 30 miles west of Boston. They enter the tents one by one after sanitizing their hands and blowing their noses; a gowned, gloved, and masked nurse swivels a soft-tipped swab into each of their nostrils. “It kind of makes your eyes water,” Thalia told me. “But it doesn’t really hurt. And it makes my nose feel really clean after we’re done.”

Each week, more than 300,000 students and school staff in Massachusetts are tested as a part of an ambitious and unprecedented statewide experiment, designed to screen for the coronavirus. The state-run pilot funds testing using a cost-saving tactic called pooling, in which multiple people’s samples are processed at once. Already, more than 1,000 schools have opted in, according to Russell Johnston, a senior associate commissioner at the Department of Elementary and Secondary Education, or DESE.

The program is part of the state’s multipronged response to its fractured education system. Since the fall, new guidance around masking, distancing, and ventilation has allowed a good number of private schools and well-resourced public schools to resume at least some in-person learning; in February, Governor Charlie Baker announced that full-time education on campus might return for elementary-school students as early as April. But many institutions lack the resources to regularly screen students for asymptomatic cases of the coronavirus—infections that could silently seed an outbreak and shut down a district in days.

[Derek Thompson: The truth about kids, school, and COVID-19]

State-sponsored pooling, run by DESE, promises to change that for those who opt in, at least for the time being. The state is footing the bill for participating schools for the two-month pilot, from mid-February through mid-April, at a cost of $40 million to $60 million, Johnston told me. After that, districts will retain access to labs, testing-services companies, and state-brokered prices, but will need to pay for the program themselves.

As schools around the country continue to inch toward reopening, Massachusetts’s pooling program will serve as a sort of beta test—one other states and districts can watch and critique—and potentially as a crucial precedent. Frequent screenings of students and staff could confirm that infection-prevention measures in schools are working; they could also help districts avoid outbreaks, and bring peace of mind to parents, students, and staff. Additionally, the data collected on transmission in schools—which won’t always mirror how the virus spreads in the community at large—can give public-health officials key information on where, and among whom, spread is occurring.

The Massachusetts model is proof that such a system can be implemented, and is likely to make a difference. But a program of this scale will also push some districts to make tough trade-offs as they try to fund their students’ way back to in-person learning. Accurate tests come with a high price tag—and the state’s program is still likely to leave behind communities that can least afford to protect themselves from the virus.


At the heart of the state’s program is the strategy of pooled testing, which batches samples from multiple people into a single tube. Each batch is then processed by a polymerase chain reaction, or PCR, the mainstay technique of coronavirus testing. Pools that test negative clear everyone in them as presumably virus-free; only samples that end up in positive pools must be evaluated again, either in smaller pools or on an individual basis. Testing a school’s worth of individuals every week takes money and lab time; pooling can be a massive boon for the wallet and the watch. The strategy substantially cuts down on manual labor and the costs of chemicals and laboratory supplies, and can stretch a lab’s testing capacity several times over.

Typically, the pooling step happens in a laboratory, where technologists combine a fraction of each person’s sample into a single tube. They save the rest in case a pool turns positive, and each sample needs to be retested. The DESE pooling program isn’t doing that. Instead, the state recommends that schools pool samples on campus and ship tubes containing up to 10 swabs to a lab, which can then test each batch for $30 to $50. (Ginkgo, one of the labs processing test samples, will allow pools of up to 25.)

This version of pooling avoids the hefty prices of storing, shipping, and retesting individual samples, but it punts a large chunk of the process to schools. Because the labs don’t have the original specimens, all members of positive pools have to return to campus and take a rapid antigen test called the BinaxNOW—a cheap, user-friendly diagnostic that the Trump administration purchased en masse last year. The tests have since made their way to states by the millions; Massachusetts schools already use the BinaxNOW to diagnose students and staff with possible COVID-19 symptoms.

[Monica Gandhi: Overcaution carries its own danger to children]

The tests provide an extra layer of assurance that might make it easier for districts to offer in-person classes. “I think it’s our path to full-time learning,” says Kathy Campbell, a parent in Medway, one of 200 or so public-school districts that have signed on with the DESE program. Her son, Will Dowling, described his first testing experience, conducted in the school hallway, as safe and minimally disruptive. “It took, like, a minute,” the high-school junior said. “It’s definitely worth it.”

Armand Pires, Medway’s superintendent, told me that his team had started looking into testing last summer. District officials were able to set up weekly pooled testing for school staff partway through the fall semester, and had planned for high-school and middle-school students, who are currently in hybrid learning, to join soon after. But Pires knew early on that testing students would swiftly squeeze the town’s resources. “The economics were really not feasible for us until we were able to partner up with the state,” he said. The DESE plan has made it possible for the district to include elementary-school students in the program. Pires intends to keep the program up for the remainder of the school year, and is already thinking about similar protocols for the fall.

But the state’s unusual approach to pooling hasn’t sat well with everyone—especially the choice of a rapid antigen test to tease apart positive pools. “That’s not the way these tests are generally intended to be used,” David Pride, a pathologist and microbiologist at UC San Diego who pioneered pooled-testing efforts for the coronavirus, told me. Tests such as the BinaxNOW are cleared for use only in people with symptoms. They’re also less sensitive than PCR tests, making them less likely to detect low-level infections, and seem to be less precise when used in kids. Baha Abdalhamid, a pooled-testing expert at the University of Nebraska Medical Center, calls the reliance on the BinaxNOW “impractical.”

Johnston, the DESE official, told me he stands by the rapid test and its performance. He noted that if everyone in a positive pool gets a negative BinaxNOW result, each person then has to take a third test: a non-pooled PCR.

[Read: Fewer Americans are getting COVID-19 tests]

But that points to a different drawback of the program—it threatens to be a Pandora’s box of logistical decisions. With so much of the testing process thrust upon districts, school nurses are now working more than ever. Every time samples need to be collected, a trained health-care worker has to don a set of pricey personal protective equipment; each follow-up test increases the risk of an infected person exposing others. Colleen Nigzus, a nurse leader in Harvard, told me that, along with testing coordination, contact tracing, and community education, these new duties have increased her workload by at least 50 percent, bleeding into nights and weekends, without overtime pay. The larger the district, the more unwieldy the system gets. (Most Massachusetts nurses I talked to have at least gotten their vaccines.)

Resampling people with a different test on a different day also raises the risk that an infection will be missed. People afraid to test positive might not return for the retest. And there would be no way to tell if the BinaxNOW caught one infection, but missed two others in the same pool.

“On a shoestring budget, you have to make some compromises,” Pride told me. “But I don’t think these are compromises many of us would recommend to make.”

Pires, the Medway superintendent, admitted that some of the data on the BinaxNOW have made him uneasy. Still, without it, “we wouldn’t know if anyone was positive, unless they became symptomatic,” he told me. “It’s still a better solution than the alternative, which is, we just don’t do testing.”


Not all districts have had to settle for the state’s take on testing; many are using their own funds to amend the program. One is Harvard—Thalia’s district—which began a nearly identical pooled-testing program before the state announced its own. Early on, school officials uncovered a positive pool. No BinaxNOW tests caught the positive sample; PCR tests did. “So we said, We need to keep doing the PCR test, to be more accurate,” Nigzus told me.

Harvard still signed on with DESE to take advantage of state funding, and it’s using the BinaxNOW. But it checks each one with a PCR test via a second swab—an additional step that the DESE program doesn’t cover. It’s a fail-safe, Nigzus said, and will provide the schools and scientists with valuable data on how well the rapid test performs.

Nigzus’s district also disregards DESE’s guidance on people in positive pools; according to the state, they don’t need to quarantine or isolate before they receive their follow-up test results. In Harvard, “we still tell them they need to quarantine or isolate until they know more,” Nigzus said.

Other DESE-enrolled districts have dispensed with the BinaxNOW entirely. In wealthy suburban Concord, people in positive pools have to find their own PCR tests for follow-up, and then submit the results to the school nurse. The guidance, which effectively decentralizes the diagnostic process, has caused some confusion and frustration, one Concord parent told me. (When I reached out over email, the district declined to comment.)

A number of districts with homegrown pooling programs, some of which sprouted long before the state’s, have found little reason to hop on the state bandwagon. Wellesley, one of the highest-income towns in Massachusetts, has been regularly testing staff and students since October. Its pooled testing pilot grew out of the efforts of local parents, school leaders, and scientists last summer; the town has spent more than $500,000, largely bankrolled by private funding from residents and local education foundations, on school testing overall, with the greater part going to the pooling pilot.

[Read: A simple rule of thumb for knowing when the pandemic is over]

The town’s DIY testing initiative ostensibly skirts many of the problems that have given other districts pause about the DESE program. Wellesley’s scientific partner is a Brooklyn-based company called Mirimus, which offers weekly pooled saliva testing to the town’s high-school and middle-school students and staff. (Kids in elementary school aren’t enrolled in the program, because younger children are considered lower risk for transmission and severe COVID-19.) Participants spit into testing kits at home on Monday or Tuesday mornings, then shuttle the samples to campus—no assisted collection necessary. Every individual specimen is then shipped to Mirimus. Katey Goehringer, a Wellesley parent, told me her ninth-grade son now considers his weekly spitting routine “second nature.”

Mirimus takes care of pooling and retesting, using the same specimens that were fed into the first round of tests. The entire process takes about 24 to 48 hours from the time the samples reach the lab. Since the program began, some 35,000 tests have been run, identifying about three or four dozen coronavirus cases—including one that was likely picked up at the high school in November, according to Linda Corridan, the director of nursing services for Wellesley Public Schools. The campus promptly shut down for two weeks, potentially averting an outbreak. The program seems to have greatly eased anxiety among parents and teachers in the district, Jesse Boehm, a Wellesley parent and one of the leads of the district’s testing efforts, told me.

By the time the state started to publicize its own program, Wellesley had hit its stride, and decided to stick with the protocols it already had in place. Any money saved on shipping extra samples, Boehm told me, would have been rapidly eaten up by many of the hidden costs of in-school testing, including extra personnel, PPE, even tech support.

There’s also no beating the convenience of the at-home spit sample, which can be done without professional assistance or taking time away from the classroom, Corridan told me. And with Mirimus in charge of retesting, the risk that infected and uninfected students and staff mingle during follow-up testing is far lower. That can’t be said for the BinaxNOW, says Linda Chow, the chair of the Wellesley School Committee. “For us, that was a nonstarter.”


Wellesley represents one extreme end of a large and complex spectrum of privilege in Massachusetts. The town’s financial wealth is buttressed by intellectual capital: Many residents, including Boehm, are research scientists, who started the pandemic armed with an arsenal of knowledge about biomedical research or molecular diagnostics. Here—as in other well-off parts of the country—the coronavirus has largely been brought to heel, and will likely stay that way for the foreseeable future.

The DESE program is meant, in part, to make that level of protection more accessible. But the pilot’s prerequisites have kept it from being a true equalizer: Only schools that have begun some form of in-person learning are eligible to participate.

About a fifth of the state’s districts, including several in a low-income knot of cities just north of Boston, remain closed because of untenably high rates of community spread. “We’re in the red, we’ve been in the red—the deep, deep, deep red,” says Dianne Kelly, the superintendent of Revere, where test positivity rates were in the double digits until a month ago. They now hover around 5 percent.

Kelly and Almudena Abeyta—the superintendent in neighboring Chelsea, the early epicenter of the pandemic in Massachusetts—joined a multidistrict collaborative last summer to bring pooled testing to schools. But with their district COVID-19 rates blocking the possibility of hybrid learning in the fall, the pair constantly felt like the odd ones out. Many of Revere’s and Chelsea’s students live in multigenerational homes with essential workers, raising their risk of exposure and upping the stakes of unknowingly bringing the virus home. For months, Abeyta listened silently as her colleagues discussed testing initiatives that, to her, still sounded like pipe dreams. “I just thought, What am I doing here?” she told me. “We’re still in remote learning. No, we are not doing COVID-19 testing today.”

Following months of rampant spread, Revere’s numbers have finally plunged. After nearly a year of fully remote learning, the district plans to begin a hybrid model on March 8, and will be able to use DESE’s program to get a head start on testing—first with staff, then with students.

The timeline to in-person learning is hazier in Chelsea. Abeyta told me that about 100 of the district’s more than 6,000 students are completing remote-learning sessions in school buildings to support their virtual education; she plans to use DESE funds to test these kids, as well as the essential staff members still coming to campus, even though this technically falls outside the state program’s criteria. The district will, eventually, creep back toward a hybrid model, perhaps in April; early testing, Abeyta said, might help it accelerate that process.

But none of that has clear bearing on what will happen when the DESE program ends. Though the state’s negotiations have driven the price of testing down, government help is explicitly temporary. “After the launch, every district is going to have to have some serious discussions about how to spend limited resources,” says Nira Pollock, an infectious-diseases diagnostic expert at Boston Children’s Hospital, who has advised several school-testing efforts in Massachusetts. Although a forthcoming influx of federal stimulus funds will help, schools have had to pour resources into many aspects of virtual learning, and many districts have only so much to spare.

“The costs of testing are real,” says Chelsea Banks, the chief of opportunity and response at Salem Public Schools, which launched its own pooled-testing program at the end of last year, but has since switched to the state’s. Salem, a working-class city north of Boston, was able to put some of its federal CARES Act funding toward testing, and has allocated money to keep its program going after April 18, Banks told me. Schools with less flexibility in their budget might find their programs petering out, potentially widening disparities between districts that have had easy paths back to in-person learning and those that have not, she said.

Kelly, for instance, says that to continue testing, Revere will also need to dip into its remaining CARES Act funds—money that might otherwise staff nurse’s offices, or send laptops to kids who don’t have reliable access to virtual learning. And she worries that the BinaxNOW won’t cut it as the district’s only follow-up-testing tactic. Many children in Revere live in low-income families without access to a vehicle, and wouldn’t be able to drop by campus for a test on a moment’s notice. Kelly plans to wrangle a mobile-testing unit to meet these students where they live, in the event that an infection is suspected.

In the long term, a successful testing program is also contingent on buy-in, which is no guarantee. The first week of Medway’s pilot, about 50 percent of the students signed up, according to Pires, the district’s superintendent. Pires expects that number to climb. But he and others told me that, across districts, distrust of the program might concentrate in certain communities that have been denied resources by the medical community before.

Even in districts where enthusiasm for testing has been high, skeptics have raised concerns. Nigzus, in Harvard, said some teachers have been wary of being pooled with students, worrying that a student’s infection will needlessly keep them out of school. In Watertown—a suburban middle-income city that is running its own PCR-only pooling program—two school nurses, Michelle Laracy and Bianca Jones, have been battling calls from employees and students who have been asking about the identities of the infected. “They say, ‘I don’t feel safe,’” Jones told me. “Everyone wants to know who the positive is.”

None of these issues is easy to solve, or exclusive to pooled testing, state-funded or not. Other, less well-resourced states—ones that don’t have the scientific firepower housed in Boston laboratories, for instance—might face further hurdles. Many school staff members across the nation have yet to be called into the vaccination queue; the timeline for pediatric shots remains even hazier. But if in-person education has any hope of keeping pace with the pandemic, schools will soon need to have their reckoning: The coronavirus is likely here to stay, which means testing for it will need to endure as well. With one lost year already under their belts, American children can’t afford another pandemic semester.

In Harvard, Thalia Ruark and her sister Siena, a 14-year-old freshman at Bromfield, are now using testing to navigate their school’s new normal. Things have been a bit tense, Siena told me. A year into the pandemic, she still sees classmates with loose-fitting masks, dangling off the tip of their nose. With Bromfield’s multitude of precautions in place, the risks of catching the coronavirus are low. But they aren’t zero.

“It would certainly be safer if I were remote,” Siena said. “But I don’t think I could handle that.”