Shanghai, China’s largest city and a financial hub, is at the epicenter of the country’s worst COVID-19 outbreak since the virus emerged in Wuhan more than two years ago.
The city of 26 million people has entered a fourth week under strict lockdown, with residents confined to their homes and increasingly angry, frustrated and rebellious.
Beijing has steadfastly defended its “zero-COVID” approach, with its mass lockdowns and sweeping testing. China repeatedly cites its low death counts as proof of superiority of the “China model.” But experts have questioned China’s official COVID data since pandemic began.
On April 21, Shanghai reported 11 official COVID-19 deaths, bringing the total from this wave of Omicron-variant infections to 36 – strikingly low compared to the more than 400,000 infections reported so far.
For comparison, Hong Kong has a population less than a third the size of Shanghai. When Omicron swept in back in January, nearly 9,000 reportedly died among 1.18 million cases.
According to The New York Times, growth of Shanghai’s infections closely tracked that of Hong Kong’s. Both have large elderly populations, including many not fully vaccinated.
What’s going on?
Claims of unreported COVID-19 deaths are circulating on social media. Those reports are often censored. China also narrowly defines what’s a COVID-caused death. A broader measure known as “excess deaths” – those above normal levels – suggests a much higher COVID toll.
On March 31, The Wall Street Journal reported on a significant reporting gap: “Shanghai’s government hasn’t reported any COVID-related deaths or outbreaks in its hundreds of elderly-care centers since cases began climbing in the city in March,” the newspaper reported.
Orderlies at one center said “they had witnessed or heard of the recent removal of several bodies from the facility, where they said at least 100 patients had tested positive for COVID-19.”
According to the Associated Press, Chinese media such as Caixin and The Paper also reported on those deaths, but their reports were quickly censored. AP also reported on the case of Muying Lu, a 99-year-old who died on April 1 in a government quarantine facility in Shanghai after testing positive for the virus.
Lu’s family told the news agency that doctors said she died because the virus had worsened her underlying heart disease and high blood pressure. That wasn’t enough to be attributed to COVID-19 in the city’s official tally, however.
“Interviews with family members of patients who have tested positive, a publicly released phone call with a government health official and an internet archive compiled by families of the dead all raise issues with how the city is counting its cases and deaths, almost certainly resulting in a marked undercount,” AP wrote of Shanghai’s statistics.
In fact, China doesn’t count cases in which where COVID-19 merely contributes to or exacerbates conditions that cause death.
Zhengming Chen, a professor of epidemiology at the University of Oxford, told The New York Times that mainland China counts only those who die directly from COVID-related pneumonia. Zuofeng Zhang, a University of California epidemiologist, confirmed that to AP.
“If the deaths could be ascribed to underlying disease, they will always report it as such and will not count it as a COVID-related death,” Dongyan Jin, a virologist at the University of Hong Kong, also told AP. “That’s their pattern for many years.”
That counting method diverges from the rest of the world.
In most countries, including the United States, health authorities follow rules in accordance with the World Health Organization’s (WHO) International Guidelines for Certification and Classification (Coding) of COVID-19 as Cause of Death.
Unlike China’s method, the WHO specifies that deaths in which COVID-19 is a contributing cause must be counted.
“COVID-19 should be recorded on the medical certificate of cause of death for ALL decedents where the disease caused, or is assumed to have caused, or contributed to death,” the guidelines state.
In the United States, the Centers for Disease Control and Prevention (CDC) stipulate that fatality counts include “deaths with COVID-19 as the underlying or contributing causes” and “deaths with confirmed or presumed COVID-19.”
According to the CDC, at least 90% of the 990,592 COVID-19 deaths recorded in the United States as of April 16, 2022, designated COVID-19 as the underlying cause, which means that COVID-19 was the disease that “initiated the train of morbid events leading directly to death.”
“If you apply international criteria,” Oxford epidemiologist Chen told the Times, “the number of deaths would be somewhat high.” According to a study Chen co-authored, COVID-19 deaths in Wuhan during the first three months of 2020 were likely 16 percent higher than China’s count.
In response to AP’s questions about Shanghai’s COVID-19 counts, China’s top health authority, the National Health Commission, said there is “no basis to suspect the accuracy of China’s epidemic data and statistics.”
To be sure, studies have found that official COVID-19 death counts appear to be greatly underreported and underestimated almost everywhere in the world, the United States included.
That is the conclusion reached by researchers who’ve computed “excess deaths,” which are based on a comparison of overall deaths during the pandemic and historical trends.
The British news weekly The Economist built one such statistical model to measure “excess deaths” and estimate how far off the official counts may be.
Disparities varied from country to country. For example, The Economist’s model estimated that the U.S. undercounted COVID-19 deaths by 20%, Russia by 200%, India by 1,000% and Pakistan by 2,800%.
China does not track excess deaths, and The Economist seems to have removed its excess deaths estimate for China. However, on January 2, data scientist George Calhoun cited The Economist’s model and estimated that Chinese official COVID-19 death count was off by 17,000%.
“The Economist estimates that the true number of COVID deaths in China is not 4,636 – but something like 1.7 million,” Calhoun wrote.
Others say The Economist’s model has shortcomings, including a lack of good Chinese data.
“It would be nice if we had similar data from China, where the virus originated,” Calhoun wrote in a subsequent Forbes article published on January 11. “Beijing refuses to provide it.”