How Deadly Was China’s Covid Wave?

After China eased the world’s strictest Covid-19 restrictions in December, the virus exploded. Clues of the surge are everywhere: Hospitals are turning away patients. Crematories were overwhelmed with corpses. The wave of top scholars is dead.

But China’s official Covid death toll for the entire pandemic remains low: 83,150 people as of February 9. front.

While a precise accounting is impossible, epidemiologists have been working to unravel the mystery of the outbreak that accelerated in December. Four separate academic teams have converged on the same estimate: China’s Covid wave could kill between a million and 1.5 million people.

All researchers consulted by The New York Times warned that without reliable data from China, the estimates should be understood as known forecasts, with uncertainties – even if the estimates correspond to better evidence than official figures.

The question of the death toll has enormous political relevance for the ruling Communist Party. At the start of the pandemic, China’s strict lockdown largely contained the coronavirus. Xi Jinping, the supreme leader, has portrayed his previous successes as evidence of China’s superiority over the West, a claim that will be difficult to maintain given the high death toll.

The difference between China’s figures and researchers’ estimates is dramatic. Official numbers would give China the lowest death rate per capita of any major country during the pandemic. But in its estimated death rate, China has certainly exceeded the official death rate in many Asian countries that have never curbed as long or as aggressively.

At the same time, China will rank below Germany, Italy, the United States and other countries where the outbreak accelerated before a vaccine was available.

Two of these estimates are in papers published in academic journals or submitted for peer review, while two other analyzes were shared by epidemiologists in response to inquiries from The Times.

Researchers are using a variety of approaches to measure how many people are infected and – the key question – how effective China’s homegrown vaccine is in preventing deaths. Some drew on the behavior of the virus during past outbreaks in Hong Kong and Shanghai, where data were more reliable, and some used detailed computer models to simulate the epidemic.

There is still more to official sampling data, based on China’s systematic testing of hundreds of thousands of people, to create a model where the estimated death toll will exceed the government’s tally.

“If the data says what we think, it’s an explosive wave,” said Lauren Ancel Meyers, a professor of biology and statistics at the University of Texas at Austin.

Why official data does not reflect China’s outbreak

China has a narrow definition of what counts as a Covid-19 death.

When the crematorium was flooded in December, Chinese officials only announced deaths caused by respiratory failure, leaving out those infected who died of liver, kidney or heart failure — an omission met with widespread skepticism. In mid-January, the government began releasing data on other deaths, but the numbers are still incomplete.

Most notably, they did not include those who died outside the hospital. Although it is impossible to know exactly how many deaths at home are missed, from 2018 to 2020, only about a fifth of all deaths in China occurred in hospitals.

The official figure “definitely underreports all Covid deaths,” said Yong Cai, a demographer at the University of North Carolina at Chapel Hill who studies deaths in China. “There’s no question about it.”

While government data shows that China has doubled the number of intensive care beds since 2020, hospitals are still overcrowded as they increase. Experts believe that hospital deaths may account for only a small proportion of total deaths.

“With the rapid spread, ICU beds are certainly not enough to deal with the peak,” said Shengjie Lai, an epidemiologist at the University of Southampton.

The number of people infected is unknown, which complicates understanding the scope of the outbreak. After two years of widespread testing and quarantine, the Chinese government in December closed its ubiquitous testing centers and made self-reporting of test results voluntary.

Other data is missing. At least nine cities in various parts of China, including Beijing, have stopped publishing monthly cremation totals.

Previous estimates, based on the Shanghai outbreak

One estimate, published last year by scientists at Fudan University in Shanghai, used previous Omicron outbreaks in Shanghai to estimate how quickly the virus would spread in mainland China.

The virus spread through the city early last year, before lockdowns and other social restrictions had a chance to slow it down. Researchers use data from that time to inform disease models that predict how future outbreaks will occur if strict control measures are removed.

The researchers made several assumptions: how many ICU beds are available, when the lockdown will end and how quickly people will receive additional vaccines.

But if anything, the estimate may be conservative, said Bruce Y. Lee, an infectious disease modeler at the City University of New York who was not involved in the research.

The study assumes outbreaks occur during the spring and summer, when more people are outdoors, meaning transmission rates will slow. But the virus died out in China during the winter.

“The evidence is that this virus shows seasonality,” said Dr. Lee. “If you had to guess, you would expect reproductive rates to increase during the winter.”

The focus of the paper is how treatment, vaccination and other measures can slow the wave and reduce the number of victims. But the work does not lead to a final conclusion: Ending the “zero Covid” policy will likely overwhelm the health care system, causing about 1.6 million deaths.

More recent estimates, based on travel patterns

The toll of China’s outbreak will also be influenced by the age, and movements of those infected.

In a more recent paper, three scientists at the University of Hong Kong estimated deaths by looking at the number of people in each age group who died during previous outbreaks in other countries, and adjusting the data for China’s demographics. Several researchers made similar calculations.

Hong Kong researchers also modeled how travel during the Chinese Lunar New Year, the busiest travel time of the year, would help spread the virus. They estimate that the surge could kill as many as 970,000 people by the end of January.

Bill Hanage, a Harvard epidemiologist who was not involved in the work, said the rate of transmission affected by holiday travel would be difficult to pin down precisely. But he says the approach is a good one.

“I think it’s done very well,” Professor Hanage said.

Retrospective estimates, now the peak has passed

A third team of researchers shared another estimate with The Times, using information available after the worst of the outbreak had passed.

Researcher – Dr. Meyers at the University of Texas and Zhanwei Du, an epidemiologist at the University of Hong Kong – found a unique way to answer another important question: How many people are infected? Even after China scrapped its mass testing program, health officials continued to test hundreds of thousands of people across the country between mid-December and mid-January to track infection rates, according to a report from China’s CDC.

Based on these data, he concluded that 90 percent of the population was infected within a month.

While the numbers are high, epidemiologists not involved in the project say the levels are credible. And in January, a leading government epidemiologist said on Weibo, China’s social media platform, that 80 percent of the population had been infected. Some European companies’ operations in China saw a 90 percent infection rate among employees in December, Joerg Wuttke, president of the European Union Chamber of Commerce in China, told The Times in an interview.

When the researchers combined the duration of the outbreak, the estimated death rate and the effect of vaccination into a statistical model, they found that the outbreak could have killed about 1.5 million people. Because of uncertainties — like how quickly the vaccine takes effect — a plausible range for the estimate is 1.2 to 1.7 million deaths, Dr. Meyers.

Many factors can affect how the sampling program in China measures the true number of infections, Dr. Meyers warned. He called the numbers “uncertain” and pointed out that any inaccuracy would affect the estimate.

A back-of-the-envelope calculation based on mortality rates in the United States

Even the simplest calculations by disease models reveal that the number of deaths may be an order of magnitude higher than the official tally.

Jeffrey Shaman, an epidemiologist and professor of environmental health sciences at Columbia University, started with a simple assumption, that the death rate for infected people in China is about the same as it is today in the United States: 0.15 percent, or about 1 in. 650 people.

Various factors can balance it, says Dr. Shaman. China uses a different vaccine than the U.S. But China’s population was less exposed to the virus during the outbreak, making it more vulnerable.

At a fatality rate similar to America’s, if 40 to 65 percent of China’s population were infected — a conservative estimate — then 900,000 to 1.4 million people could die, he said.

Ben Cowling, an epidemiologist at the University of Hong Kong, reached the same death rate by considering only the 82 million people in China aged 60 and older who had not been vaccinated or had received less than three doses of the vaccine by the end of November. If 80 percent of the group were infected, he would expect more than a million people to die, due to limited immunity and exposure to the virus, he told the China Foreign Correspondents Club last week.

China, after all, is the only country in the world that faced the first major wave of infection without trying to slow it down, causing what Dr. Cowling is the fastest-spreading respiratory pandemic virus in modern history.