What does it mean when you have COVID-19, but you don’t know how you contracted it?

A nurse practitioner conducted a COVID-19 test in the parking lot at Brockton High School.David L. Ryan/Globe Staff/file

Imagine this: You’ve been working from home since March. You always wear a mask in public over your mouth and your nose. You’re fastidious about hand-washing. Your social calendar primarily includes trips to the supermarket and Zoom happy hours. You’ve perfected the art of turning down invitations to parties, weddings, and Scrabble nights, until a coronavirus vaccine becomes available.

But somehow, somewhere you still got infected with COVID-19. What gives?

Social media is rife with tales from dejected virus sufferers, who claim to have followed all the rules, as laid down by public health experts, and yet they caught the virus anyway. Meanwhile, the source of their infection remains a mystery.

Recently, for example, Arne Duncan, the former secretary of education under President Barack Obama, lamented on Twitter that he and his family caught COVID-19 despite wearing masks, social distancing, avoiding crowds, and keeping people out of their home.

“How did we catch it? I don’t know,” he said.

Dr. Abraar Karan, an internal medicine physician at Brigham and Women’s Hospital, has been hearing stories like those from his patients for months. They aren’t surprising, he said, as they’re indicative of high levels of community spread.

“Unless you have some major outbreak — like a super-spreading event where a lot of people are giving you the same story — it’s extremely difficult to figure out where transmission is happening, and that’s actually one of the reasons why the epidemic will probably never be fully stopped,” Karan said. “You’ll always have a trickle of cases ongoing at small levels.”

Reverse contact tracing is even harder when community transmission is so widespread. In Massachusetts, the Department of Public Health has traced thousands of COVID-19 infections to clusters at a variety of settings, including restaurants, schools, nursing homes, and child care facilities.

But the vast majority of new clusters — almost 94 percent — identified between Oct. 18 and Nov. 11 occurred in households, accounting for roughly 16,000 new infections during that four-week period. Under the state’s guidelines, a “household” cluster, defined as two or more confirmed cases at the same residence, is not associated with another cluster, meaning contact tracers were unable to learn whether those infections were linked to workplaces, social gatherings, or other activities.

“COVID is just pretty ubiquitous and it can be hard to tell exactly where it is that you picked it up,” said Samuel Scarpino, an assistant professor and head of the Emergent Epidemics Lab at Northeastern University. And as the state’s positivity rate climbs, the probability of coming into contact with someone with the virus increases.

“In the summer, we were down around 1 percent positivity, and now in some places, we’re up over 10 percent,” Scarpino said. “So just the number of times in which you may find yourselves inside of a building with someone with COVID is so much higher than it was before.”

In Massachusetts, the seven-day average positivity rate, based on each individual tested for COVID-19 and excluding repeat tests, has surged from less than 2 percent in August to 9.7 percent as of Nov. 21. Still, Massachusetts is doing better than most states in COVID-19 incidence. Over the past week, the state has averaged 38.5 new cases per 100,000 residents. By comparison, North Dakota and Wyoming, which lead the nation in infections per capita, have averaged 166.6 and 140.1 new infections, respectively, per 100,000 people.

“All these things we can see in the data . . . are contributing little by little by little, and that all adds up to a huge amount of risk,” Scarpino said. “There’s just so much COVID out there now — especially compared to where we were in the summer — that anytime you’re in an enclosed space with people outside your household, that’s potentially high risk again.”

Linger too long in one place, too close to someone else, and your chances of getting sick tick up.

It’s also possible, Karan added, that people are contracting the virus in “places that you just don’t normally think about.” He pointed to an unexpected outbreak over the summer at Baystate Medical Center that had been linked to employee break rooms, where staff members had been congregating without wearing masks.

“It’s really hard to rely completely on people telling you where they got infected because they just may not remember,” he said.

Dr. Ali Raja, an emergency medicine physician at Massachusetts General Hospital, has also begun hearing more stories from perplexed patients he’s diagnosed with COVID-19, surprised they got sick when they said they did everything right. Raja worries mask-wearing has given some people a “false sense of security,” leading them to ignore other health precautions, like physical distancing.

“Mask-wearing has really become the most visible part of this and most people I see, they’re wearing masks,” Raja said. “It’s easy to forget that mask wearing alone is not perfect. We still have to maintain distance and wash our hands and use alcohol hand sanitizer.”

Dr. Shira Doron, an infectious disease specialist and epidemiologist at Tufts Medical Center, agreed. She invoked the “Swiss cheese” model of infection control: No single approach to protecting oneself from COVID-19 is foolproof and multiple layers of protection are better than one.

“You have all these layers of Swiss cheese; they all have holes in them. None of them is perfect,” Doron said. “But even if the virus slips through the holes in cheese number one, cheese number two, and cheese number three, it’s still caught by cheese number four.”

Those “slices of cheese” are proper mask-wearing, social distancing, avoiding crowds, and keeping time spent with people outside of the household to a minimum.

“When you add all of those layers of cheese,” she said, “one of them is going to be solid.”