At a June 8 press conference, a World Health Organization scientist confusingly suggested that asymptomatic transmission of the coronavirus is “very rare” — a statement that many scientists found problematic, and which some politicians and those on social media seized upon as evidence that certain public health measures were not necessary.
It is well documented that people without symptoms can transmit SARS-CoV-2, the virus that causes COVID-19. But it is not yet known how often that happens, or how frequently that involves someone who is completely asymptomatic.
Maria Van Kerkhove, an infectious disease epidemiologist and the WHO’s COVID-19 technical lead who was responsible for the remarks, clarified the next day that the amount of transmission from people who are asymptomatic is a “a big open question.” She said her original comment applied to only a subset of studies and data from member states, and pertained only to those who are truly asymptomatic.
“We do know that some people who are asymptomatic, or some people who don’t have symptoms, can transmit the virus on,” she said in a live online Q&A event. “What I was referring to yesterday in the press conference were a very few studies, some two or three studies, that have been published that actually try to follow asymptomatic cases.”
While we are not privy to the member state data, Van Kerkhove is mostly right that the few epidemiology studies that exist have found little onward transmission from truly asymptomatic people, although she may have exaggerated. One study that the WHO cited in its mask guidance identified no secondary transmission from asymptomatic cases, but another identified spread from nine people, or 14% of asymptomatic cases.
Dr. Oyuka Byambasuren, a researcher at Bond University’s Institute for Evidence-Based Healthcare in Australia who co-authored a systematic review and meta-analysis addressing asymptomatic transmission, told us that based on the available data, WHO was correct to state that asymptomatic transmission is low, but that the term “very rare” was “probably inappropriate.”
Van Kerkhove’s initial summary also failed to include modeling studies that have found significantly higher contributions to transmission from people without symptoms, which she acknowledged.
Much of the core confusion, however, stems from semantics and the use of the term asymptomatic, which in scientific circles typically means a person who never has any symptoms throughout the disease course. That’s in contrast to someone who is pre-symptomatic, and goes on to develop symptoms later, and to those who have only very mild symptoms.
Right now, there’s substantial evidence that people without symptoms can spread the disease, which is why public health officials are encouraging the use of face masks and social distancing, along with other pandemic prevention efforts.
But exactly how much spread is from pre-symptomatic people versus those who never develop any symptoms whatsoever is less clear.
As Dr. Isaac Bogoch, an infectious disease specialist at the University of Toronto, told us, there’s “pretty good data” showing that pre-symptomatic people are responsible for some transmission of COVID-19. But there’s less information about people who stay symptom-free.
“We don’t actually know, to date, the degree to which people who have no symptoms ever — throughout the time they have this virus — to what extent they’re responsible for driving this pandemic,” he said.
This wrinkle may have been what Van Kerkhove was trying to discuss, but the message many people heard was that the virus doesn’t spread when people don’t have symptoms.
One Facebook meme falsely claimed that the WHO was “now saying they were wrong” and that “it seems like the ‘conspiracy theorist’ were right again.” The post went on to say, “This was never about a deadly virus,” and that the WHO meant that an “asymptomatic person cannot transfer the virus.”
That’s not true. Not only did Van Kerkhove not say that, but as WHO Director General Dr. Tedros Adhanom Ghebreyesus noted in a June 10 press conference, the agency has said since early February that “asymptomatic people can transmit COVID-19, but that we need more research to establish the extent of asymptomatic transmission.”
Another post, which was a snapshot of a tweet from the conservative commentator Liz Wheeler, pointed to lockdowns and business closures for healthy people, saying, “The WHO now says asymptomatic transmission is rare. We’ve been duped. Again.”
Some Republican politicians also jumped on the news. Rep. Jim Jordan of Ohio declared, “Reopen America!” on Twitter while retweeting a CNBC story about the WHO’s comments.
Sen. Rand Paul of Kentucky, sharing the same story, said in a tweet, “Good News! People who catch coronovirus but have no symptoms rarely spread the disease. Translation: sending kids back to school does not require millions of test kits.”
The CNBC story was subsequently updated, and another story explaining the WHO’s clarification was published the morning of June 9. The article appeared after Paul’s tweet, but minutes before Jordan’s sharing of the first article. Neither politician shared news of the WHO’s clarification about the matter.
We’ll review what the WHO said, the evidence on asymptomatic transmission and why the issue has become so muddled.
What WHO Said
Van Kerkhove’s first comments came at the end of the WHO’s June 8 press conference, when a Reuters reporter asked about asymptomatic spread, given reports in Singapore that at least half of new cases were without symptoms.
Van Kerkhove started by noting that in many countries, when COVID-19 cases have been contact traced, many of the seemingly asymptomatic cases were not completely without symptoms, but rather, had very mild symptoms.
“They may not have developed fever yet, they may not have had a significant cough, or they may not have shortness of breath,” she said, acknowledging that there were some people who were truly asymptomatic as well.
She then turned to the question of how many asymptomatic people transmit the virus, saying that that assessment requires careful monitoring of individuals to look for secondary transfer.
“We have a number of reports from countries who are doing very detailed contact tracing. They’re following asymptomatic cases, they’re following contacts, and they’re not finding secondary transmission onward,” Van Kerkhove said. “It’s very rare.”
She added that much of that evidence is not published, and pointed to a few other published studies.
“We are constantly looking at this data and we’re trying to get more information from countries to truly answer this question,” she said. “It still appears to be rare that an asymptomatic individual actually transmits onward.”
Van Kerkhove then said during a social media Q&A on June 9 that she wanted to make “some clarifications” to address “perhaps some of the misunderstandings” from what she said the day before.
Her first point was that it’s not clear how many people are truly asymptomatic, with some estimates ranging from 6% to 41% of the infected population.
“We know that people who are infected with COVID-19, many develop symptoms. But there are some people who do not,” she said. “And to truly understand how many people don’t have symptoms, we don’t actually have that answer yet.”
On transmission from asymptomatic people, and her statement the day prior, Van Kerkhove said she was referring to “some two or three studies” along with information from member states.
“I was responding to a question at the press conference, I wasn’t stating a policy of WHO or anything like that, I was just trying to articulate what we know,” she said. “And in that, I used the phrase ‘very rare,’ and I think that that’s [a] misunderstanding to state that asymptomatic spread globally is very rare.”
Calling the amount of asymptomatic transmission a “major unknown,” she also said that she had neglected to mention modeling studies which have arrived at much higher estimates of transmission from people without symptoms.
“There’s a big range from the different models, depending on how the models are done,” she said. “But some estimates of around 40% of transmission may be due to asymptomatic [spread]. But those are from models. And so I didn’t include that in my answer yesterday, but wanted to make sure I covered that here.”
Unclear Definitions
Some scientists considered Van Kerkhove’s first remarks a misunderstanding, while others were more critical.
“I think that that comment was taken out of context and misunderstood,” Peter Hotez, the dean for the National School of Tropical Medicine at Baylor University, told us. “If you’re communicating a lot, eventually you’ll have missteps.”
The Harvard Global Health Institute issued a statement over email saying that the WHO “created confusion,” and that “[a]ll of the best evidence suggests that people without symptoms can and do readily spread SARS-CoV2, the virus that causes COVID-19.”
“In fact, some evidence suggest[s] that people may be most infectious in the days before they become symptomatic — that is, in the pre-symptomatic phase when they feel well, have no symptoms, but may be shedding substantial amounts of virus,” the statement continued.
Bogoch called her comments a “mistake,” noting that the WHO is under a lot of pressure, but that it has to get its communication right. “It was a bit of an oversimplification, and required clarification and some correction,” he told us.
But Bogoch also said that the scientific community had not been diligent in differentiating between the mildly symptomatic, the pre-symptomatic and the completely asymptomatic.
“From the very beginning, the studies that have been published have not really been entirely clear on distinguishing between these groups,” he said. “And it’s done us a tremendous disservice.”
He pointed to several high-profile studies that on the surface seem to suggest that a relatively high number COVID-19 infections are asymptomatic, but which are not able to fully make those claims because of incomplete reporting.
For example, one study in Iceland, which was published in the New England Journal of Medicine in April, observed that 43% of participants who tested positive for COVID-19 did not have symptoms at the time, but it did not follow-up with participants to see if they developed symptoms later.
Another widely cited but unpublished paper from Vo, Italy, which found that around 40% of cases were asymptomatic, Bogoch said, was problematic because the authors limited the symptoms to fever and cough.
“We know that’s not fair,” he said. “There are a lot of other symptoms for COVID-19, and fever and cough are not sufficient.”
“Most studies, or most data that are reported, are from people tested at a single time point. So they are a mix of people who will develop symptoms (pre-symptomatic) if followed for long enough and those who will remain asymptomatic,” explained Dr. Nicola Low, an epidemiologist at the University of Bern in Switzerland.
Low is a co-author of a systematic review and meta-analysis posted to the preprint server medRxiv synthesizing the available information about asymptomatic COVID-19 cases. Based on 28 studies with adequate follow-up, she said, she estimates that about 15% of people remain asymptomatic throughout infection.
A separate systematic review and meta-analysis, posted to medRxiv and cited by the WHO in its mask guidance, came up with the same percentage for asymptomatic cases.
Bond University’s Byambasuren, who led the review, told us that the results were based on nine “good studies,” and that the team only included “studies with a valid sampling process and followed people for at least 7 days after their initial positive test” to tease out the asymptomatic cases from the pre-symptomatic ones.
The Centers for Disease Control and Prevention, meanwhile, has pegged the approximate number of asymptomatic cases, based on data as of the end of April, at 35%.
Evaluating how frequently people who are truly asymptomatic pass on the virus is even trickier than pinning down the percentage of asymptomatic cases.
The minimal data that does exist, Bogoch said, does point to relatively uncommon transmission.
A small study of an outbreak in a South Korean call center, for example, found four asymptomatic cases out of 97 infections, and none of those people’s household contacts acquired secondary infections.
A contact tracing investigation in Taiwan also found no transmission from nine asymptomatic patients to 91 close contacts.
But, Bogoch said, those glimmers of evidence aren’t enough to really come to a conclusion about asymptomatic spread. “It’s not entirely clear if it’s very rare just yet,” he said.
In her review, Byambasuren included four studies — the two previously mentioned ones plus two others — with estimates of asymptomatic transmission that could be compared with symptomatic transmission. For those few studies, the asymptomatic transmission rate ranged from 0% to 2.2%, while the symptomatic transmission rate ranged from 0.8% to 15.4%.
“From our research we estimate that asymptomatic people transmit about 40% less than symptomatic people,” she said, noting that could be because those people are infectious for a shorter period of time and because they produce fewer infectious respiratory droplets because they’re not coughing or sneezing. “Unfortunately, based on the available data we cannot rule out that both cases could transmit equally.”
Low said that rigorously determining how often asymptomatic people transmit the virus in a study is incredibly hard to do, so the alternative is to look at mathematical models of transmission.
“They generally find that 40-60% of all transmission comes from people who didn’t have symptoms at the time,” she said.
Separately from her other analysis of the proportion of asymptomatic cases, Low assessed four modeling studies, three of which evaluated pre-symptomatic transmission only. According to her review, the fourth study, published in the journal Science by researchers in the U.K., found that asymptomatic transmission was responsible for 6% of transmission while pre-symptomatic transmission contributed 47%.
Some experts nevertheless cautioned against making too much of modeling studies.
“You’ve got to model data,” Bogoch said. “And every model’s going to have assumptions.”
“We need actual data rather than modeling studies,” Byambasuren said.
Policy Implications
Multiple experts told us that while the debate over asymptomatic transmission may appear to be an academic exercise, it’s one that ultimately matters.
“Contact tracing and testing relies on most people becoming symptomatic so they can be isolated and their contacts – symptomatic or not – be quarantined,” said Byambasuren. “Pre-symptomatic cases make this slower, but can still be detected; whereas fully asymptomatic cases make it much harder – and rely on being captured by the contact tracing of symptomatic cases.”
Accordingly, if more asymptomatic spread is happening than previously thought, more testing might be necessary to contain the virus.
Bogoch said that it could be a major boon to public health if it turns out that many of the people who are thought to be asymptomatic are actually mildly symptomatic or exhibit more unusual COVID-19 symptoms. If the public could be better educated about that, more people could be caught earlier in infection and their contacts could be traced, limiting community spread.
Still, even if true asymptomatic spread is rare — which is still up for debate — scientists said the takeaway should not be that people can abandon precautionary measures.
“There is evidence that people can transmit to others before they develop symptoms (pre-symptomatic period),” explained University of Florida biostatistician Natalie Dean in a Twitter thread. “This means that people without current symptoms can be infectious.”
And despite the confusion, Van Kerkhove noted at the end of her social media Q&A that the uncertainty about coronavirus transmission — including the amount stemming from any type of person who lacks symptoms — is why physical distancing, hand washing, wearing face masks and following stay-at-home orders is so important. “That’s why all of that is really, really critical,” she said.
Editor’s note: FactCheck.org does not accept advertising. We rely on grants and individual donations from people like you. Please consider a donation. Credit card donations may be made through our “Donate” page. If you prefer to give by check, send to: FactCheck.org, Annenberg Public Policy Center, 202 S. 36th St., Philadelphia, PA 19104.