Sharp fall in coronavirus cases undermined by questionable data
As President Xi Jinping vows to restart China’s economy, a marked slowdown in reported new coronavirus cases is helping his cause.
On Monday, Chinese health authorities reported only 11 new cases outside of Hubei, the central province where the virus originated. Over the past two weeks, the number of new confirmed cases has dropped by more than 80 per cent.
China’s count have been swung by changes to how individuals are officially diagnosed, leaving experts concerned over their ability to draw conclusions from the figures. “We don’t know whether the case numbers tell us about the real trend in incidents or are just a result of testing practices,” said Benjamin Cowling, head of epidemiology at Hong Kong University.
In the space of a week in mid-February, China’s government switched to a broader measure of diagnosis and then rolled it back again. Many foreign epidemiologists, as well as doctors in China, argued that the broader approach was necessary, as the virus overwhelmed local capacity to conduct diagnostic laboratory tests.
So China’s National Health Commission instructed Hubei to start reporting “clinically diagnosed cases” — those diagnosed with viral pneumonia through a CT lung scan. The new category was an addition to “confirmed cases”, whereby evidence of the virus was found with a nucleic acid lab test kit, and “suspected cases”.
All other provinces continued to report only confirmed and suspected cases. The result was that on February 13, Hubei reported a surge of more than 14,000 new cases — more than 10 times the average number reported over previous days.
“It made a lot of sense for Hubei to switch to this ‘probable’ case definition,” said Prof Cowling. “These tests are difficult, the reagents [used in lab tests] are expensive and not every hospital can do it.”
But the change was revoked by the NHC after less than a week, leading to a sudden decline in Hubei’s infection numbers on February 20. At a press conference that day, an NHC expert explained that Hubei’s testing capacity had “hugely improved” and that “nucleic acid tests were no longer a problem”.
Sceptics point out that the shift in numbers has come after Mr Xi took a more prominent role in the “war” against the disease and flagged the importance of restarting the world’s second-largest economy. The fallout from the virus has taken a heavy toll, with factories across the country prevented from reopening and workers kept away from manufacturing sites.
This has led some experts to questioned whether the count has become subject to political influence. “Over the last week the virus has become politicised, with Xi Jinping stepping back centre stage,” said Rodney Jones, a veteran Asia watcher at Wigram Capital Advisors. “The dilemma for local governments is that high infection rates would now be seen as directly undermining Xi.”
Bruce Aylward, head of the World Health Organization’s foreign experts’ mission to China, acknowledged questions had been raised over the statistics. But, speaking on Monday at a press conference organised by the NHC, he added: “The decline that we see is real.”
Neil Ferguson, professor of epidemiology at Imperial College London, said he stood by his initial estimate that only 1 in 10 infections in China had been detected. But he added that cases were likely to have declined as a result of the large-scale shutdown of people and businesses.
The view that China’s lockdown has had an impact was shared by a virus researcher in Beijing, who wished to remain anonymous. “Hubei has been doing a blanket search and quarantine of all suspected infected cases, which possibly reduced the transmission rate,” the person said.
As the overall trend points towards a decline, there remain concerns about potential under-reporting outside Hubei. Medical experts in southern Guangdong, the hardest-hit province after Hubei, said official figures there were likely to be underestimates.
According to one officer at Guangdong Provincial Center for Disease Control and Prevention, who asked not to be identified, some cases had gone undetected because local authorities could not fully track migrant workers who had returned from Hubei. He also suspected that the incubation period was probably longer than the current 14-day standard — meaning people who tested negative initially might become ill later. Two hospital staff in the Guangdong city of Shenzhen also said that they were seeing more virus cases than were being officially reported.
It is not just China under scrutiny. There are concerns that Iran has under-reported the spread of the virus within its borders. Iran reported 61 infections and 12 deaths on Monday, an almost 20 per cent fatality rate compared to about 3 per cent in China, implying that dozens if not hundreds of other cases there have gone undetected.
Jianghu Dong, a biostatistics expert at the University of Nebraska Medical Center, said the virus epidemic has been “a big lesson for our government” in China. He added: “We really hope we can set up a transparent [testing and reporting] system.”
Additional reporting by Robin Yu in Hong Kong and Qianer Liu in Shenzhen