Find Us

Satellite images suggest COVID-19 started in the fall, not winter 2019, say Harvard researchers

Scientists conduct research with corona viruses at the Helmholtz Centre for Infection Research HZI, in Brunswick, Germany, May 8, 2020.
Scientists conduct research with corona viruses at the Helmholtz Centre for Infection Research HZI, in Brunswick, Germany, May 8, 2020. Copyright Julian Stratenschulte/dpa via AP
Copyright Julian Stratenschulte/dpa via AP
By Alice Tidey
Published on
Share this articleComments
Share this articleClose Button

A sharp increase in hospital attendance in the Chinese city of Wuhan was first observed in August 2019, researchers from Harvard Medical School have said.


Satellite images of Wuhan hospitals' parking lots and search engine queries from the Chinese city suggest the COVID-19 outbreak started much earlier than previously thought, according to research from Harvard Medical School.

Researchers used more than 110 satellite images of six hospitals located in Wuhan from between January 2018 to April 2020 as well as daily data for symptom-related searches such as "cough" and "diarrhoea" on Chinese search engine Baidu from Wuhan from April 2017 to May 2020.

They found that "increased hospital traffic and symptom search data in Wuhan preceded the documented start of the SARS-CoV-2 pandemic in December 2019."

They noted for instance that a "steep increase" in hospital occupancy as measured by the parking lot volume proxy was first observed in August 2019, culminating with a peak in December 2019.

August also saw "a unique increase in searches for diarrhoea which was neither seen in previous flu seasons or mirrored in the cough search data".

Queries of the respiratory symptom "cough" show seasonal fluctuations coinciding with yearly influenza seasons, but "diarrhoea" is a more COVID-19 specific symptom and only shows an association with the current epidemic, researchers stressed.

The study, which has not yet been peer-reviewed, does admit to limitations such as the fact that many of the satellite images received were not usable because cloud cover, buildings' shadow or construction made it impossible to count the numbers of cars on parking lots with accuracy.

Search engine data are also "vulnerable to fluctuations related to events we might not be aware of" and depends on internet penetration which "can be highly variable" in China, researchers said.

But, they argued that "while we cannot confirm if the increased volume was directly related to the new virus, our evidence supports other recent work showing that emergence happened before identification at the Huanan Seafood market".

"These findings also corroborate the hypothesis that the virus emerged naturally in southern China and was potentially already circulating at the time of the Wuhan cluster" and "also hint at the missed early signals of COVID-19 in current surveillance systems for respiratory pathogens," they added.

China has been heavily criticised by US President Donald Trump, who accused the country of not alerting international health authorities about the novel coronavirus early enough. He has also threatened to withdraw US funding to the World Health Organisation (WHO) arguing it is to close to Beijing to be impartial.

WHO member countries voted last month for the organisation to carry out an independent probe of the international response to the outbreak.

More than 406,000 people worldwide have succumbed to the deadly virus while more than 7 million are known to have been infected, according to a tally from Johns Hopkins University.

China has so far communicated 4,638 deaths and 84,195 confirmed cases.

Share this articleComments

You might also like

Coronavirus spread: These animations show how COVID-19 has progressed in key countries

In the 'WHO blockade' of Taiwan, everyone is a loser

Scotland's Nicola Sturgeon says she felt 'overwhelmed' by COVID pandemic during testimony