A data scientist working in tech, Michael Donnelly became an amateur COVID-19 watcher early in the pandemic. When his vaccinated friends started getting sick following July Fourth festivities in Provincetown, Mass., he documented more than 50 breakthrough cases that ultimately led the CDC to changing its guidance on masking.
Caption

A data scientist working in tech, Michael Donnelly became an amateur COVID-19 watcher early in the pandemic. When his vaccinated friends started getting sick following July Fourth festivities in Provincetown, Mass., he documented more than 50 breakthrough cases that ultimately led the CDC to changing its guidance on masking. / Jason LeCras for NPR

The Centers for Disease Control and Prevention has a lot of ways to pick up on COVID-19 outbreaks, but those methods often take awhile to bear fruit.

Not so with the Provincetown, Mass., cluster that started around July Fourth weekend. "We triggered the investigation as people were getting symptomatic," says Demetre Daskalakis, a deputy incident manager for the CDC's COVID-19 Response. "Pretty amazing — it is warp speed."

How did they do that? It was thanks to a tip from a citizen scientist named Michael Donnelly. A data scientist in New York City's tech sector, he started publishing his own coronavirus data reports early in the pandemic and launched a website, COVIDoutlook.info, with Drexel University epidemiologist Michael LeVasseur.

Following leads from his personal network, Donnelly documented over 50 breakthrough cases coming out of Provincetown, practically in real time, and shared it with the CDC as the outbreak was still unfolding.

Without Donnelly's effort, the agency would have probably detected the outbreak at some point, Daskalakis says, but "it wouldn't have been as rapturous an initiation of an investigation and response as we had."

The speed of the investigation — and the exceptional participation from the mostly gay men involved in the outbreak — helped the CDC learn new information about the delta variant. And it was that new information, in part, that prompted the agency to change its guidance for how vaccinated people should keep themselves safe at this stage of the pandemic — including a return to masking indoors.

It's a testament to the power of citizens engaging with the scientific process, Daskalakis says. "I get goose bumps thinking about it," he says. "Community plus public health is magic."

''Our entire house can't stop coughing''

Donnelly didn't go to Provincetown with his husband for July Fourth, but his friends who were there told him all about it. As they do every year, thousands of gay men arrived for the holiday in this little artist community on the tip of Cape Cod to rent cottages, go to the beach and drag shows and restaurants, and crowd into nightclubs to dance.

Fourth of July festivities in Provincetown resulted in an outbreak among vaccinated and unvaccinated attendees. The delta variant was later determined to be the most prevalent strain of the virus involved in the outbreak.
Caption

Fourth of July festivities in Provincetown resulted in an outbreak among vaccinated and unvaccinated attendees. The delta variant was later determined to be the most prevalent strain of the virus involved in the outbreak. / The Boston Globe via Getty Images

As the festivities from the July Fourth week wrapped up, the updates and gossip streaming into Donnelly's phone quickly took a different cast from years' past.

On July 9, he texted some close friends, "[D]id you boys survive the Fourth in P-town?" The response: "Our entire house can't stop coughing."

More texts started coming in, including from fully vaccinated friends testing positive for the coronavirus. He started thinking to himself, he says, that given that vaccination rates were really high among the gay community and in Provincetown, "the odds don't really add up."

With the pandemic keeping friends apart for so long, there was extra excitement this year, says Zorik Pesochinsky, who traveled to Provincetown from New York City. "The lines were even longer — the bars were even busier and more full," he says.

He and everyone else he knew going were fully vaccinated, so he felt safe. "I was definitely going into it with a mindset of, this is all behind us, we're just going into a super-fun, amazing weekend."

It was a rainy week, which meant everyone was indoors more. "It would get so incredibly hot in these clubs that you would just be wet with sweat, so you'd have to step outside for a moment just to get a breath of fresh air," says Sean Holihan, who came from Washington, D.C.

Halfway through the week, when a few of Cameron Thomas' vacation housemates started coughing, he didn't think much of it. "You're saying hi to so many people, you're in situations where you don't sleep a lot, you're running around — you're going to catch something," he says.

But it wasn't a summer cold that was going around. By the end of that week, news of breakthrough COVID-19 cases started to roll in. And they kept coming.

A community willing to share COVID-19 statuses

As the texts from his friends rolled in, Donnelly's statistician side took over — he wanted to understand what was going on.

"That's what's always drawing me to numbers and math and forecasting and data science — it gives me some better idea of the things that I understand and the things that I can control and the things that I can't," he says.

He knew that breakthrough infections were expected, that no vaccine is 100% effective. But the numbers he was hearing about exceeded what he would have expected the breakthrough rate to be.

"What was really concerning about this was it sounded like entire houses were coming down with breakthrough infections," he says. "That full allotment of people that I know that I would expect over the course of this year to get a breakthrough infection were getting a breakthrough infection in a single week."

He spent the next few days reaching friends and acquaintances from all over the country and taking notes on their COVID-19 statuses and symptoms and vaccination histories and more.

"Michael's known as the COVID data guy," Pesochinsky says. "When he started asking about it, I was like, 'Oh, this makes a lot of sense that he's going to be pulling something together.' "

As soon as Holihan tested positive, he shared it publicly. "I posted on Instagram and Twitter that I had tested positive for COVID and the people who had cold symptoms, who had been in P-town, they should also get tested," he says. "And then I emailed or messaged every person that I could remember coming in contact with and just said, 'Hey, heads up, I have COVID.' " Donnelly saw his tweet and called him up, too.

The men Donnelly was calling were eager to help. After a few days, he had collected information on 51 cases, including COVID-19 and vaccination statuses, symptoms, home city, phone number, whether they were in a house with other breakthrough cases, and more.

It's no accident, Donnelly says, that his friends were so open. "The norms of the gay community say: Share your medical history, share your risks with other people so that they can be responsible and take care of themselves as well," he says. "That came with years of practice within the community, particularly around HIV and AIDS."

That willingness to share information about health and contacts is not a given. Contact tracing COVID-19 in the U.S. has had a lot of problems throughout the pandemic. With a fragmented public health system, tracing outbreaks across jurisdictions has been one challenge. Getting people to report their close contacts and where they were has been another. One CDC report found that two-thirds of people interviewed did not provide any contacts to a contact tracer.

After three days of gathering this information together, Donnelly decided it was time to call in the public health professionals.

CDC starts connecting the dots

As it happened, Donnelly knew the CDC's Daskalakis from the previous year when Daskalakis worked for New York City's health department and Donnelly was putting out his coronavirus forecasts.

Donnelly texted Daskalakis to let him know what he was finding on July 12. Daskalakis asked him to email the spreadsheet and responded the next day, connecting him to the Massachusetts Department of Public Health and other CDC officials.

"It's quite certain that if we didn't have the heads-up from Michael — because of what he was seeing among his friends with his statistician hat on — we wouldn't have heard about it as rapidly," Daskalakis says.

Donnelly soon got a call from the Massachusetts state epidemiologist, Dr. Catherine Brown, "who wanted to know as much as I could tell her," he says. "She had already been aware of some breakthrough cases, but among Massachusetts residents." The spreadsheet gave her a head start in identifying cases connected to Provincetown in cities all over the country.

Donnelly says it's no accident that his friends were so eager to help, sharing information about their vaccination statuses, symptoms, and whether they were in a house with other breakthrough cases.
Caption

Donnelly says it's no accident that his friends were so eager to help, sharing information about their vaccination statuses, symptoms, and whether they were in a house with other breakthrough cases. "The norms of the gay community say: Share your medical history, share your risks with other people so that they can be responsible and take care of themselves as well," he says. / Jason LeCras for NPR

On calls with public health officials, Donnelly says he and LeVasseur tried to explain, not just their spreadsheet and analysis, but the context of what Provincetown is like during these annual gay pilgrimage weeks.

"Even for us — even for gay men who have been in the gay nightlife scene for years and years, it's not the easiest thing in the world to go to a meeting with 10 CDC epidemiology experts and explain the intricacies of Circuit Week versus Bear Week and just how many hundreds of people they can squeeze into those spaces with terrible music," Donnelly laughs. "I've been telling my friends, you haven't lived until you've talked about twinks with the CDC."

In a way, this was an auspicious community to be at the center of an outbreak investigation. The people affected didn't just cooperate when a friend called asking questions — they helped public health officials who called as well.

"I heard in one of those early meetings that the contact tracers had reported back and they said, 'These are the most cooperative people we've ever worked with — we have people giving us their entire itineraries and the names and phone numbers of everybody that were in their house,' " Donnelly says.

Gay men's relationship to public health has been "tempered by fire because of HIV," the CDC's Daskalakis says. "This is an awesome public health moment," he says. "It's a community that believes in science and public health stepping up to the plate."

Findings that changed CDC guidance

The result of the CDC's investigation with the Barnstable County and Massachusetts health departments was a report published on July 30. By the time it was released, news of this unusual outbreak was already out. CDC Director Rochelle Walensky had alluded to it earlier in the week as evidence to support changes in the agency's recommendations on what was safe for fully vaccinated people to do.

The findings of the investigation were striking. Out of 469 positive cases identified, nearly three-quarters of cases were in fully vaccinated people. Delta was determined to be the culprit in these cases, not older strains or some new variant. And — most surprisingly — the amount of virus measured in a subset of people who tested positive was nearly identical among vaccinated and unvaccinated people.

These findings "raised the specter of — is there transmission happening from vaccinated person to vaccinated person? Is the Delta variant a bigger threat from the perspective of transmissibility and vaccine effectiveness?" Daskalakis says. He describes the findings as an "exclamation point" that prompted the agency to change its guidance for what was safe for vaccinated people to do in the context of delta.

Daskalakis says the investigation into this cluster is ongoing. "The next phase will be going into a deeper dive into the various aspects of this outbreak, including what happened outside of Massachusetts," he says.

Donnelly says he's proud of how his community responded to the investigators, and thankful that public health officials listened and wanted to understand.

"I was just the concentrator of this information," he says. It was the way the community responded, he says, that "allowed the CDC to learn something important about the new coronavirus, and — I really hope — protect some folks."

Copyright 2021 NPR. To see more, visit https://www.npr.org.