Only 1 in 19 people who might have the coronavirus are being diagnosed in Wuhan, new research suggests

• If the Chinese government had the supplies and capacity to test everyone in Wuhan for novel coronavirus, experts say, the number of cases would be far higher than currently reported.
• But medical shortages and the possibility that patients can carry the virus while asymptomatic mean many are likely going undiagnosed.
• A new report estimates that there are 19 times the official number of confirmed cases in Wuhan.
• Visit Business Insider's homepage for more stories.

New estimates of novel coronavirus diagnoses in the Chinese city of Wuhan place the number of infected people far higher than 19,558, the current case count from the Chinese government.

Researchers from Imperial College London calculated in a new report that for every one person who has been tested and diagnosed with the coronavirus in Wuhan, 18 others likely have the infection but are going untested and undiagnosed.

The analysis, released Monday, is based on estimates of the rate of infection in cases outside of mainland China. The researchers looked at roughly 750 passengers who traveled from Wuhan back to their home countries on government-arranged flights. Upon their return, the travelers were all kept in isolation and tested for the coronavirus immediately; many were diagnosed, some of whom didn't present any symptoms.

The researchers applied the incidence of infection calculated in those situations to the population of Wuhan.

The math revealed a severe under-counting of patients in Wuhan.

That speaks to two major challenges in responding to the coronavirus outbreak: the scarcity of resources to test people as the outbreak has grown, and the fact that many infected people only have mild symptoms.

43,000 cases in 6 weeks

More than 43,000 cases of the new coronavirus have been confirmed around the world since the new virus was discovered in Wuhan at the end of December. The death toll passed 1,000 on Monday.

The healthcare system in Wuhan was quickly overwhelmed by the outbreak, with hospitals running out of space and medical supplies - including test kits and medicine. People in China said at the end of January that getting a spit-test kit was like winning the lottery.

The tests identify the coronavirus by its genome. Doctors take samples from saliva, mucus, or phlegm swabs to analyze in labs. The process takes at least 24 hours.

According to Reuters, all virus samples were initially sent to just one lab in Beijing for testing, but they're now also being analyzed in the Hubei province, where Wuhan is located.

That suggests a bottleneck in the testing and diagnosing of patients.

A far higher case total in Wuhan

As foreign nationals in China have evacuated and returned to their home countries, many governments have imposed quarantines. The Imperial College researchers based their analysis on passengers who had left China and returned to Japan and Germany on four flights from January 29 to February 1. All were tested for the coronavirus, and 10 infections were found among the 750 passengers.

According to the paper, that suggests a "detectable infection prevalence of 1.3%."

The researchers estimated that given the exponential rate of the coronavirus' spread, 220 out of every 100,000 Wuhan residents would get the virus on January 31 alone. Since Wuhan has 11 million residents, that means 24,000 new cases on that day. The researchers compared that number to the 1,242 reported and confirmed new cases in Wuhan on February 3 and found a 19-fold under-reporting rate.

That calculation assumes there is a 14-day window in which a patient's coronavirus infection will yield a positive test. The team also ran the numbers based on the assumption that a patient has only a seven-day period of detectable infection. In that scenario, Wuhan would have 300 new cases per every 100,000 residents on January 31. In total, that's 33,000 new cases in the city that day - 26 times the official number on February 3.

The researchers stressed, however, that testing all 11 million residents of Wuhan would be virtually impossible.

"Surveillance is typically biased towards detecting clinically severe cases, particularly at the start of an epidemic when diagnostic capacity is limited," they wrote.

Some people who are infected also may not know they have the virus if they are asymptomatic or very mildly symptomatic. Additionally, scientists are still unsure whether the infection can be transmitted while an infected person is not showing symptoms.

Still, the Imperial College researchers said comparing the death toll with this higher estimated number of cases suggests the coronavirus' fatality rate could be very low.