Skip to Main Content

Health officials in China, racing to try to contain a fast-growing coronavirus outbreak, are principally recording severe cases of disease, using a case definition that cannot capture patients with mild illness, according to experts familiar with the surveillance efforts.

The approach, the experts told STAT, is likely resulting in both an underestimate in the total number of cases and flawed assumptions about fatality rates calculated by those who ignore the repeated caution that it’s too soon to do that math.


The experts were quick to note that the Chinese are not willfully underreporting cases. Rather, the approach is a testament to how challenging data collection can be during the early days of an epidemic. When thousands of sick people show up at hospitals looking for care, there is no time to go searching for people who have mild symptoms and who have stayed home.

“I think right now things are so chaotic in China it may be hard to collect data on the whole spectrum of illness,” said Michael Osterholm, director of the Center for Infectious Diseases Research and Policy at the University of Minnesota.

Officially, an estimated 20% of cases in China are severely ill, according to the World Health Organization. But that calculation is derived based on known cases, and would not reflect mild, undetected ones.


Support STAT: STAT is offering coverage of the coronavirus for free. Please consider a subscription to support our journalism. Start free trial today.

Without knowing for sure what percentage of cases is severe — and how easily the pathogen that causes the disease can be transmitted — it’s impossible to forecast what might happen if the virus continues to spread globally, the WHO’s emergencies chief, Mike Ryan, told reporters Wednesday.

“We don’t understand either of those parameters well enough to make accurate predictions,” he said. Still, he added a warning for people who are concluding that the virus may be less fatal than some other known pathogens: “A relatively mild virus can cause a lot of damage if a lot of people get it.”

The outbreak has infected upward of 7,700 people on the Chinese mainland, and killed 170 since the new virus, known provisionally as 2019-nCoV, was reported to the WHO on Dec. 31. Nearly 20 other countries have reported diagnosing infections in travelers from the Chinese city of Wuhan, the epicenter of the outbreak, but to date there has been little local spread of the virus in other countries.

WHO Director-General Tedros Adhanom Ghebreyesus and Ryan, who were in China earlier this week to consult with the Chinese government and assess the situation, praised the response by officials there. “The challenge is great but the response has been massive,” Ryan told reporters.

The effort, which is of an unprecedented scale, involves quarantining whole cities that are home to tens of millions of people to try to stop spread of the virus. The WHO is still hopeful, Ryan said, that China will be able to stop the outbreak.

If it cannot — or if exported infections seed outbreaks in other parts of the globe — the world could be looking at a more disseminated epidemic, admitted Tedros, as he is called.

Among other factors, epidemiologists are interested in the severity of the outbreak and what is known as the virus’ “attack rate.” The attack rate means the percentage of people who will develop the disease if it spreads.

With seasonal influenza, the Centers for Disease Control and Prevention estimates between 3% and 11% of people become sick with flu every year. But immune systems have experience with influenza; the attack rate might be higher with a virus that is wholly new to humans.

Likewise it’s unclear how many mild infections are being missed. The Chinese are currently only testing people who are sick enough to seek medical care because they have pneumonia — a criterion that automatically excludes anyone on the mild end of the disease spectrum.

With some diseases, there isn’t much mild illness. In the case of the 2002-2003 SARS outbreak, for example, most people who became infected ended up in hospital, noted Dr. Gabriel Leung, dean of medicine at Hong Kong University and a veteran of that city’s battle against the disease.

It’s clear there are at least some mild cases with this new infection. Many of the 100 or so exported cases are people with mild illness; to date there have been no deaths among these cases. Some of the five cases in the United States are only in hospital to ensure they don’t infect anyone else. Such cases would not be tested in China using the current case definition.

“The look and feel of the exported cases, I think, really support the argument that there’s a lot of mild disease that is not being detected in China at the moment for the very good reason that they just can’t do it,” said Dr. Allison McGeer, who fought SARS in Toronto and helped contain hospital outbreaks of MERS in Saudi Arabia. Both SARS and MERS are coronaviruses, related to the new virus.

McGeer, an infectious diseases researcher at Toronto’s Mount Sinai Hospital, noted the types of studies needed to estimate how much mild disease this virus causes are very difficult to do.

Infectious diseases experts use the analogy of an iceberg when they talk about the spectrum of a disease. The most severe cases represent the tip of the iceberg; they are visible, because they are sick enough that they seek health care.

But the portion of the iceberg that is under water is harder to calculate, especially with a new disease. And without having a clear picture of total cases, it’s difficult to come up with an accurate fatality rate.

Wang Linfa, director of the program in emerging infectious diseases at Duke-NUS Medical School in Singapore, is convinced there are a lot of mild infections. “Mild cases don’t go to hospital and there’s many, many cases that remain to be confirmed,” he said, noting laboratories are struggling to keep up with testing amid a reported shortage of test kits.

Leung agreed: “It looks like the submerged bit of the iceberg is fairly large for this thing.”

Some of the answers China cannot currently generate will come from watching what happens with the exported cases, said Dr. David Heymann, who oversaw the WHO’s SARS response and now teaches at the London School of Hygiene and Tropical Medicine. “Now we are really going to begin to see things,” he said.

Countries outside of China are using a broader approach to testing, looking for anyone who has recently traveled to or from Wuhan who has respiratory symptoms. These people, who the CDC calls “persons under investigation” or PUIs, do not need to have pneumonia in order to be tested.

  • Do you consider the lag of 11 days on average for a person to die since the day he is confirmed?
    It may be 2-3% without it. but what about with it, given no isolation is imposed before symptoms show? ( 5.2 days on average)
    Also the average time of infection to appearing symptoms, seems to agree with the doubling time of infections for now.
    This will drop in the future, as we need to start considering, more than one epicenters pretty soon, to understand the rates better.

Comments are closed.