COVID-19's silent spread: Evidence mounting that many infections are going undetected

In Iceland, nearly half of people with COVID-19 had no symptoms when they were swabbed for a scientific study. In the United States, preliminary results from a new antibody survey suggests as many as one in five New Yorkers may have been exposed to the novel coronavirus. And in the waters off the coast of Guam, more than 850 sailors on a navy warship have been infected in a massive outbreak - with roughly half reporting no symptoms when they were tested.

In the early days of COVID, experts were skeptical of initial reports of asymptomatic cases, which weren't thought to play a significant role in spreading the disease. But four months on, there is mounting evidence that "silent spreaders" are a major driver of the pandemic, with top scientists now speculating they could comprise between 25 and 50 per cent of all cases.

Many crucial questions remain but experts say silent spreaders are likely one of the embers keeping this pandemic aflame. And the implications are potentially huge for everything from testing strategies to how vulnerable nursing homes are protected - especially as countries like Canada move toward relaxing restrictions.

"I've done an about-face on this," admits Dr. Isaac Bogoch, an infectious disease specialist with the University of Toronto and Toronto General Hospital. "Early on, I was quite skeptical about the role of asymptomatic people in transmitting it and was quite vocal about that. But I've changed my views to be reflective of the data that's emerged.

"It's important. And when we fully understand who's truly asymptomatic, it will likely be one of several important components that drives the epidemic."

Bogoch said silent spreaders should be thought of in three separate categories: "pre-symptomatic" cases (people who seem healthy when they test positive but later develop symptoms); true asymptomatics, who never so much as suffer a sniffle; and "sub-clinical" infections, where people have mild symptoms but fail to recognize them as being COVID.

The existence of asymptomatic cases was recognized early on by the World Health Organization, which sent experts to China in February to investigate the novel coronavirus. But at the time, officials said there wasn't enough data yet to conclude these cases were playing a significant role.

"There is no evidence that we're seeing only the tip of a grand iceberg, with nine-tenths of it made up of hidden zombies shedding virus," Dr. Bruce Aylward, the WHO expert that led the China mission, told the New York Times in early March. "What we're seeing is a pyramid: most of it is above ground."

But a growing body of science has now emerged to suggest that while symptomatic cases are still the most important source of contagion, a surprising proportion of COVID infections are likely happening below the surface of detection, raising questions over current testing strategies that narrowly target people with symptoms.

Among this emerging literature are 11 "cohort" studies from countries like Iceland, Italy and Japan, where proactive testing has revealed a hidden pool of COVID. In these studies, between 31 and 88 per cent of people showed no symptoms when they tested positive for the virus, said Trish Greenhalgh, a professor with the University of Oxford, in a written statement shared with reporters.

Recent cases in the United States have also attracted dramatic headlines, including the USS Theodore Roosevelt aircraft carrier, where an explosive COVID outbreak infected more than 850 crew members - roughly half of whom said they had no symptoms when they were tested. At a homeless shelter in Boston, blanket testing recently revealed 147 positive cases.

"The number of positives was shocking, but the fact that 100 per cent of the positives had no symptoms was equally shocking," Dr. Jim O'Connell, president of Boston Health Care for the Homeless Program, told a local news station.

There are now fears that silent spreaders could reignite COVID outbreaks in countries that have already beaten back their first epidemic wave.

In China, officials recently ramped up efforts to find and quarantine infected people who don't have symptoms and on April 1, the government started publishing statistics on asymptomatic cases. In the first 24 hours, they reported 166 new infections, of which 130 were deemed asymptomatic, according to a news report in the British Medical Journal. (It's unclear how many of these people may have later developed symptoms.)

In Canada, the federal public health agency says it doesn't know how much of the country's outbreak is attributable to silent spread "but we know that it is occurring among those with close contact or in close physical settings," said Anna Maddison, spokesperson for the Public Health Agency of Canada (PHAC), in an email.

But while Canada isn't actively tracking asymptomatic cases, provinces and territories that have submitted case reports to PHAC have identified at least 220 cases classified as asymptomatic, representing 2.8 per cent of the 7,879 reports where a symptom status was known. (More than 22,200 case reports have been submitted so far.) This is certainly a lowball number, given that testing has been narrowly focused on people with symptoms and surveillance data is incomplete.

Despite accumulating evidence of asymptomatic infections, however, there are still important knowledge gaps to fill before we can fully understand the importance of silent spreaders, said Shelly Bolotin, an infectious disease epidemiologist with Public Health Ontario and the University of Toronto's Dalla Lana School of Public Health.

"We're still in the throes of this first wave of disease," she said. "We don't have all of the pieces of the puzzle yet to really be able to say."

One missing puzzle piece is how effective asymptomatic people are at transmitting the virus, she said. A handful of studies have shown that people can infect others before they develop COVID symptoms. Others have demonstrated comparable viral loads (the amount of measurable virus inside of someone) between patients with and without symptoms - a surprising and counterintuitive finding, Bolotin said.

But proving asymptomatic transmission is particularly tricky. Studies from China showing family clusters, where an asymptomatic person appears to have infected someone else in the household, are compelling - but given the scale of the outbreak at the time, there's always a chance both those people were infected by a third person, she notes.

"I think we can theorize that it's happening, but the devil is in the details," she said. "How infectious are they really? Big picture, I think we're still learning."

Bogoch takes issue with much of the breathless discourse around "asymptomatic COVID" that fails to distinguish between pre-symptomatic, asymptomatic and sub-clinical cases.

He points to two high-profile publications in the New England Journal of Medicine, which some have framed as smoking-gun evidence of "asymptomatic" COVID - one from Iceland, where 43 per cent of positive cases reported having no symptoms, and a published letter from New York doctors, who tested all pregnant women delivering at two hospitals and discovered that 29 out of 33 women who tested positive had no symptoms at admission.

But neither study was able to follow up with every study participant to verify that they didn't later develop symptoms, Bogoch said. It's also possible many people who consider themselves asymptomatic are just failing to recognize their symptoms as being COVID, he adds.

"In the COVID follow-ups I'm seeing, many of these people were a close contact of a symptomatic person (and) classified as asymptomatic," Bogoch said. "But all you've got to do is take a history. 'So, did you have symptoms, did you feel unwell?' And they'll say, 'Well, I actually had some night sweats for a couple of nights.' Well, that's not asymptomatic."

One early study that tried to tease out pre-symptomatic cases from true asymptomatics was an analysis of the Diamond Princess cruise ship, where 634 people tested positive for COVID after being quarantined for two weeks on the ship. After testing people throughout the quarantine period and tracking their symptoms, the researchers used a statistical model to estimate an asymptomatic rate of 18 per cent.

This finding has since been backed up by yet-unpublished research on the American passengers on the ship, who were repatriated and further followed by researchers with the U.S. Centers for Disease Control and Prevention, according to one of the study authors, Gerardo Chowell, an epidemiologist with Georgia State University in Atlanta.

Chowell believes 18 per cent is an underestimate of the fraction of asymptomatic cases in the broader population, however - most of the cruise ship's passengers were older, and younger people are less likely to develop symptoms. Based on what other studies have reported, he estimates the true asymptomatic rate is closer to 30 or 40 per cent of total infections.

But when it comes to truly understanding the hidden "iceberg" of COVID infections, some of the best evidence will come from antibody research, or seroprevalence studies. Preparations are now underway to do this work in Canada, including at Public Health Ontario, but the lab is still waiting on these tests to be validated and approved by Health Canada, Bolotin said.

Antibody studies have already started trickling out from around the world, however, with studies of varying size and quality being released everywhere from Austria to California. But many of these early efforts have generated controversy, with critics pointing to methodology flaws or concerns with tests that may be prone to false positives.

But taken together, this emerging body of evidence suggests the majority of COVID cases are currently going undetected, said David Fisman, an epidemiologist and professor at U of T's Dalla Lana School, who has been analyzing the research published thus far.

"This has huge implications for the effectiveness of case isolation, contact tracing and quarantine," he said. "If we miss most cases, we can't expect those strategies to be effective."

This means that tools like physical distancing become all the more important, he added - a difficult message to hear for many people suffering emotionally or financially under current restrictions.

The role of silent spreaders will have to be urgently addressed in vulnerable settings like long-term-care homes, where they have likely played an outsized role, said Dr. Nathan Stall, a geriatrician and researcher at Sinai Health System. On Friday, a new New England Journal of Medicine study of a Washington state nursing home found that 56 per cent of COVID-positive residents had no symptoms when they tested positive; half went on to develop symptoms, though many were infectious several days before.

Stall suspects asymptomatic workers have played a major role in bringing COVID into nursing homes. And once inside, the virus tends to spread unnoticed, in part because frail and elderly people with COVID tend to have atypical symptoms like confusion, decreased appetite or falls - many of which are easily misattributed to other conditions.

When jurisdictions like Ontario reopen their economies, policymakers will need to ensure vulnerable people are protected from silent spread, he said. In at-risk settings like nursing homes, he added, this could mean ongoing COVID testing for all workers and residents.

"As we loosen these restrictions, and we have people who are asymptomatically spreading, this is going to be a huge issue," he said.

"If you're going to release restrictions, it needs to be done carefully, and most importantly with close monitoring to make sure we're not having a rebound. (Otherwise) you're just going to get hit again with an overwhelming second wave."