Covid Market Origins Hypothesis Challenged by Statisticians

A seminal study on the origins of the COVID-19 pandemic has been brought into question after drawing criticism from statisticians. Study authors and experts told Newseek what they think of the matter.

In 2022, evolutionary biologist Michael Worobey, together with a global team of immunologists, virologists, biologists and statisticians, published a paper in the journal Science on the origins of the COVID-19 pandemic. By plotting the locations of the earliest known COVID-19 cases, along with the geographical locations of the earliest viral lineages, environmental samples, and circumstantial evidence, the team concluded that the Huanan Seafood Wholesale Market in Wuhan was the most likely epicenter of the coronavirus pandemic.

However, in a recent paper, published in the Journal of the Royal Statistical Society Series A, Dietrich Stoyan, a professor in mathematics and statistics at the Technical University of Bergakademie Freiberg in Germany, and Sung Nok Chiu, a professor of mathematics at Hong Kong Baptist University, described this original analysis as "fundamentally flawed."

"The paper authored by Worobey et al. (2022) faced criticism from several individuals, primarily centered around the poor quality of the data utilized," Stoyan and Chiu told Newsweek. "The main points of critique revolved around the data being incomplete in terms of numbers, imprecise in terms of addresses of infected individuals containing errors, and notably, lacking information on the times of infection.

"Our approach thoroughly examines the statistical methodology employed by Worobey et al. and demonstrates that even with excellent data, their methods fail to produce reasonable results."

One of their primary concerns was that the clustering of data around the Wuhan market was used as confirmation that the market was indeed the epicenter of the pandemic. "It is important to note that centrality does not imply causality," Stoyan and Chiu said.

"Worobey et al.'s use of the center-point to identify the 'center' of a point cloud is analogous to the use of the median to measure the central tendency of a set of numerical data, obviously under the (unestablished) assumption that the 'center' of a cloud of locations of cases starting from an origin of the infection process is close to this origin," they write.

Using the same data, Stoyan and Chiu mapped out the cluster of data points used by Worobey et al. around the area of the Huanan market. However, they used a different system of coordinates to map out the case locations, and used a different established method to measure the shortest distances between these data points.

From their analysis, Stoyan and Chiu identified several other landmarks which they consider to be potential alternative candidates for the coronavirus's origin.

"Worobey et al. did not consider alternative points of origin. In close proximity to the market, there are at least three other potential locations that could be epicentres," Stoyan and Chiu told Newsweek. "Hankou railway station, Wanda Plaza, and the Wuhan CDC are examples of plausible centres of the point cloud of the cases. However, as we mentioned above, centrality does not imply causality."

From their analysis, Stoyan and Chiu conclude that the origins of the COVID pandemic are still a mystery. "Unfortunately, all is open," they said. "Neither the lab-leak hypothesis nor the zoonosis hypothesis is rejected or proved, and hence further investigations are needed. With the available data, we find it difficult to draw any definitive conclusions."

But how have the original authors responded to these criticisms?

"This paper is an exercise in motivated reasoning and tunnel vision," Michael Worobey, first author on the cited paper and a professor and department head a the University of Arizona, told Newsweek. "Aside from a long list of factual errors the authors make, they just miss the big scientific picture."

Florence Débarre, an evolutionary biologist at the national Center for Scientific Research in France who was not involved in the original study, echoed this sentiment: "[Stoyan and Chiu] completely miss the forest for the trees," she told Newsweek. "They need to zoom out.

"Early cases are around the market, not around either meaningful location of the Wuhan Institute of Virology. Yes, the center of all locations does not fall exactly on the market. No one expects it to in real life. But the center and the market are close enough to confirm that the market did play a role, as is known from epidemiological investigations."

Indeed, co-author Marion Koopmans, a virology professor and head of the viroscience department at the Erasmus MC WHO collaborating center for modelling, evolution and control of emerging infectious disease, emphasized that their paper "does not claim [the market] was the only possible place."

"What the Worobey study did is a careful analysis of the available data to try to address an important question," she told Newsweek. "It claims that the analyses point at the central market as epicenter. Could the virus have lingered somewhere else before exploding? For sure."

Worobey pointed towards an "outsized proportion" of the earliest COVID cases occurring in those who worked at the market. "And those who didn't work at the market tended to live unexpectedly close to and centered on the market," he said. "If the virus had already been widespread in Wuhan in December 2019, then case locations would be expected to reflect the population density of susceptible individuals throughout the city. In a city more than 8000 square kilometers in size, the area where early cases were most likely to be drawn, in reality, turns out to be a tiny circle containing the market. All of this is utterly unexpected had the market not been the early epicenter of the pandemic."

Ignoring this circumstantial evidence was a major oversight by Stoyan and Chiu, Débarre said. "[Their] article is uninformed and narrow-minded. The authors [...] ignore the fact that many cases were epidemiologically linked to the market. This does not just mean that the people lived close: this means that the people worked or shopped there. Some of these people even belong to documented clusters from inside the market. The market is therefore not a random landmark; it is a place that has been identified as critical as early as December 2019."

Worobey added: "[Stoyan and Chiu's] argument is akin to concluding that, because the messy area of destruction around one of four explosives factories in a huge city encompasses a train station, a hotel, and a shopping center—and is centered a few feet from the exact middle of the factory floor—that there is not overwhelming evidence that the explosives factory was the source of the explosion."

Definitive evidence for the origins of the COVID-19 pandemic is still lacking but, based on a plethora of circumstantial, genetic and geographical data, the Huanan Seafood Market still appears to be a very likely candidate.

Update 01/18/2024 17:09 ET: This article was updated per request.

Update 01/19/2024 03:37 ET: This article was updated with a locator map of the Huanan Seafood Wholesale Market.

Is there a health problem that's worrying you? Do you have a question about coronavirus? Let us know via health@newsweek.com. We can ask experts for advice, and your story could be featured in Newsweek.