Coronavirus: Cases in China two weeks earlier: study

The SARS-CoV-2 virus was on the loose for at least two weeks before the first officially accepted case of the Covid-19 disease in China in early December 2019.

A medical worker administers a dose COVID-19 vaccine in Wuhan. Picture: Getty Images.

The SARS-CoV-2 virus was on the loose for at least two weeks before the first officially accepted case of the Covid-19 disease in China in early December 2019.

Research published in the PLOS Pathogens journal on Friday revises the date to somewhere between early October and mid-November, with the best estimate at November 17.

“Dating the First Cases of Covid-19” is an analysis of data from 203 countries and territories and argues the virus had spread globally by January 2020, which suggests it travelled earlier and faster than previously thought.

It identifies the likely first occurrences outside China as being in Japan on January 3, 2020; in Thailand on January 7; in Spain on January 12; and in South Korea on January 14. Other likely January cases were the US on January 16 and Australia on January 23.

The first likely case in Africa was probably in Nigeria on February 9, and in South America in Brazil on February 19.

Scientists in Britain and the Czech Republic adapted a method used in conservation science to estimate the timing of extinctions, but reversed it, plotting the first likely cases of the disease, not the last sightings of a species.

The mathematical method uses an algorithm known as optimal linear estimation, taking into account intervals between occurrences of a phenomenon and their distribution, and is not specific to animals and plants. The paper argues that the ­genetically-based method most frequently used to work out the best estimates for first cases of Covid-19 is phylogenetic analysis, which requires much larger numbers and diversity of cases to be ­effective.

By contrast, OLE creates a time series of occurrences, which in the case of Covid-19 was the days on which cases occurred and so is not reliant on the number of cases that occurred on any one day or series of days. From that data the scientists drew inferences about the likelihood that the occurrences represented the disease taking hold.

University of Kent conser­vation scientist David Roberts, the lead researcher, said: “This novel application within the field of epidemiology offers a new ­opportunity to understand the emergence and spread of diseases as it requires only a small amount of data.”

Abrar Chughtai, a public health researcher at the University of NSW not involved in the new paper, said the finding was interesting “but not surprising”.

“Some other indirect evidence also points towards an early origin of SARS-CoV-2,” Dr Chughtai said. “For example, the first case of Covid-19 in the US was reported on 19th January 2020, while the analyses of blood donation samples from American Red Cross showed that the virus was present in the US as early as ­December 2019. We need to conduct more ­… studies to confirm these findings,” he said. “This may also help in finding the origin of SARS-CoV-2.”

The Wuhan Municipal Health Commission first reported a cluster of pneumonia cases on December 31, 2019 before those patients were reclassified as having a novel coronavirus.