By Daniel P. Oran, Eric J. Topol
On Monday, Maria Van Kerkhove, the World Health Organization technical lead on covid-19, stunned public health officials worldwide when she said during a briefing in Geneva that “it still seems to be rare that an asymptomatic person actually transmits onward to a secondary individual.” The next day, Van Kerkhove and WHO rushed to clarify that much remains unknown about the role of asymptomatic transmission of the virus.
That is true enough, but the answers are slowly emerging. Evidence is accumulating, though it remains inconclusive, that the novel coronavirus can indeed spread silently, transmitted through infected people who will never experience symptoms, just as it is transmitted by the presymptomatic before they become ill. Covid-19’s dual nature — often benign, yet sometimes deadly — puts it in a rare category among the common pathogens that afflict humanity.
Since mid-April, we have been tracking an ever-increasing number of medical studies and news reports that have clarified the prevalence of asymptomatic infection by SARS-CoV-2, the virus that causes covid-19. In our recent review published in the Annals of Internal Medicine, we collected data from 16 sources, spanning the globe and in settings as diverse as cruise ships, nursing homes, homeless shelters and prisons.
In assessing the overall rate of asymptomatic infection, we found some dramatic outliers. For example, among 3,277 prison inmates in four states who tested positive for the coronavirus, an astounding 96 percent had no symptoms at the time of testing. At the other end of the spectrum, a mere 6 percent of 48 infected nursing home residents remained asymptomatic. It appears that a variety of factors, perhaps including age and previous exposure to other, more benign coronaviruses, may account for these extremes.
But these outliers aside, we have been impressed by the uniformity of data from the three representative samples in our collection. By randomly selecting members of the population for coronavirus testing, as studies from Iceland and Indiana have done, or by testing nearly everyone in the population, as happened in one northern Italian town, scientists can accurately gauge the percentage of people who are infected but have no symptoms.
These three studies have yielded asymptomatic infection percentages of 42 percent, 43 percent and 45 percent. Knowing that some people in these cohorts went on to have symptoms, we concluded that the overall rate of asymptomatic infection is likely at least 30 percent and could be as high as 40 percent to 45 percent.
The study in northern Italy had one relatively unique feature: The researchers tested the subjects twice, at the beginning and end of a two-week period. Epidemiologists call this longitudinal data, meaning that it shows how a disease progresses. In this case, none of the asymptomatic people had developed any symptoms by the end of the study. In similar studies from Greece and Japan, only a small fraction of asymptomatic people developed symptoms. More data will be required to clarify the true proportions of presymptomatic and asymptomatic individuals.
By carefully tracing the chain of viral transmission, the Italian researchers discovered that asymptomatic people can infect others. That was also the conclusion of the researchers who studied nursing home residents. And U.S. Navy officials involved in the care of infected crew members aboard the aircraft carrier USS Theodore Roosevelt reported that asymptomatic individuals continue to test positive for the coronavirus — and might possibly be able to infect others — for an extended period of 14 days or more. But how often asymptomatic people actually do transmit the virus to others remains uncertain.
The potential to make others sick is not the only risk of being infected without symptoms. In studies from Japan and South Korea, the lungs of asymptomatic people were found to have abnormalities, revealed as hazy regions on their CT scans — the distinctive “ground-glass opacities” that have become a well-known sign of covid-19.
It is not clear what these lung abnormalities might mean for the long-term health of the asymptomatic — a stark reminder that science is only now beginning to understand the multiple ways that this novel pathogen can harm people.
What should be done with the knowledge that asymptomatic coronavirus infection is common? On an individual level, take confidence that wearing a mask outside the home remains essential because of the significant possibility that you might not have symptoms if infected. If you are unlucky enough to catch the coronavirus, the probability that you will have no symptoms is nearly a coin-flip. To protect others, behave as though you are infected.
On a societal level, knowing that asymptomatic infection is widespread means coronavirus testing should be significantly ramped up — perhaps by a factor of 10 or more. Concentrating testing efforts exclusively on those with symptoms, as has largely been done so far, is a mistake. Public health officials must cast the net far more widely.
Science is often about making the invisible visible — about seeing familiar things in unfamiliar ways. The new knowledge that many people who catch the coronavirus will never have symptoms, yet can still make others sick, means rethinking and revising tactics for controlling — and someday ending — the covid-19 pandemic.
Daniel P. Oran is in the digital medicine group at the Scripps Research Translational Institute. Eric J. Topol, a professor of molecular medicine at Scripps Research, is founder and director of SRTI
