• Sebastian Leck

FACT CHECK: Do four-fifths of people infected with COVID-19 show no symptoms?

MISLEADING. The figure represents only one day of numbers from China in April. Further studies suggest that somewhere between 20 and 45 percent of people infected with COVID-19 never show symptoms, but the true percentage is difficult to measure.

On April 2, the British Medical Journal published a news article titled “Covid-19: four fifths of cases are asymptomatic, China figures indicate.” It has since been used on Twitter and Facebook to suggest that the virus is not as serious as public health authorities claim. (Data from Crowdtangle shows that the article has been shared more than 3,800 times on Facebook accounts with a combined total of 4.5 million followers.) It was also included in a list of “COVID-19 Facts” on Swiss Policy Research, a website known for spreading conspiracy theories, which claimed that “up to 80% of all test-positive persons remain symptom-free.”

The original BMJ article is a summary of new cases in China on the day of April 1. During that 24-hour period, Chinese authorities found that 130 out of the 166 new cases of COVID-19, or 78 percent, showed no symptoms at all. The story then goes on to quote Tom Jefferson, an epidemiologist at Oxford, who says that if the numbers are representative in China, the virus had been circulating “for longer than generally believed” and that a lockdown was “unlikely at this point to slow or halt viral circulation.”

However, the results described in the BMJ article can’t be taken at face value. There are two different possibilities for asymptomatic patients: they could be presymptomatic, meaning they are not showing symptoms but will in the future, or asymptomatic—meaning they don’t show symptoms and never will. (There’s also a third option: that they are experiencing very mild symptoms and mistakenly labeled it as asymptomatic.) The numbers from China were only for a 24-hour period, and the BMJ doesn’t seem to have followed up. So it’s possible that many of the patients developed symptoms later, since it can take up to fourteen days for symptoms to appear.

Angela Rasmussen, a virologist at Columbia University, took issue with the conclusions of the piece. She noted that the cases could be presymptomatic rather than asymptomatic, and that it was a study of only 166 patients among more than 82,000 confirmed cases in China—which is far from representative.

There were also potential problems with the source of the numbers; it was based on a press release from the Chinese National Health Commission (a version in Mandarin can be found here), and one response to the BMJ article argued that its author had misinterpreted the numbers.

Further studies of asymptomatic cases have put the figure at lower than 78 percent. People with no symptoms rarely go to a hospital for testing, so the best case studies are in places like cruise ships where everyone gets tested regardless of their symptoms. For example, a study of passengers on the Diamond Princess, a cruise ship in Japan that had a large outbreak of COVID-19, calculated an asymptomatic rate of 17.9 percent after adjusting the numbers to account for passengers developing symptoms later.

The American Centres for Disease Control and Prevention now pegs their “current best estimate” figure for asymptomatic cases at 40 percent of all patients. That number lines up with a review of existing research in the Annals of Internal Medicine, which found that patients without symptoms seem to account for around 40 to 45 percent of all cases.

Overall, the claim that four-fifths of cases of COVID-19 are asymptomatic misrepresents the article it is based on. The numbers from the British Medical Journal are highly suspect, given that they are based on only 24 hours of data, and should not be used to draw larger conclusions about the COVID-19 virus.