Fact-checking claims coronavirus might have started in August 2019

The outbreak of the coronavirus was first observed in Wuhan - but did it circulate earlier than expected?
There was criticism of a study from the USA, according to which the corona virus could already have been present in the Chinese city of Wuhan in August last year.
The Harvard University study, published earlier this month, was rejected by China and questioned by independent scientists.
What did research say?
The research, which has not been reviewed by experts, is based on satellite images of traffic movements in hospitals in Wuhan and the tracking of online searches for certain medical symptoms.
The number of vehicles parked in front of six hospitals in the city before the end of August until December 1, 2019 is said to have increased significantly.
According to the Harvard report, this coincided with an increase in the search for possible coronavirus symptoms such as "cough" and "diarrhea".
The researchers monitored the traffic patterns in Wuhan
This would be an important finding, as the earliest reported case in Wuhan was only reported in early December.
The scientists write, "Although we cannot confirm whether the increased volume is directly related to the new virus, our evidence supports other recent work that shows that it occurred before it was identified in the Huanan Seafood market."
The Harvard study has been well received in the media. President Trump, who was extremely critical of China's pandemic response, tweeted a Fox message highlighting the results of the researchers. The tweet has been viewed more than three million times.
So is their evidence up?
The study claims there has been an increase in online requests for coronavirus symptoms, particularly "diarrhea", on the popular Chinese search engine Baidu.
However, officials from the Baidu company have denied their results, claiming that the search for "diarrhea" has actually decreased during this period.
So what's up?
A diagram of Internet search traffic with coronavirus symptoms in Wuhan
The term used in Harvard University's work is translated from Chinese as a "symptom of diarrhea".
We checked this with Baidu's tool, which allows users to analyze the popularity of searches like Google Trends.
The search term "symptom of diarrhea" actually shows an increase in requests from August 2019.
However, we also used the term "diarrhea", a more common search term in Wuhan, and it actually showed a decrease from August 2019 to the onset of the outbreak.
A lead author of Harvard newspaper Benjamin Rader told the BBC that "the search term we chose for" diarrhea "was chosen because it best matches confirmed cases of Covid-19 and was suggested as a related search term for coronavirus" .
We also looked at the popularity of finding "fever" and "difficulty breathing", two other common symptoms of the coronavirus.
The search for "fever" increased slightly after August, similar to "cough", and the requests for "breathing difficulties" decreased over the same period.
Questions were also asked about the study, which used diarrhea as an indicator of the disease.
A large-scale UK study of nearly 17,000 coronavirus patients found that diarrhea was the seventh most common symptom, well among the top three: cough, fever, and difficulty breathing.
What about the number of cars?
From August to December 2019, an increase in the number of cars parked in hospitals was reported in the Harvard study in all six hospitals.
However, we have identified some serious shortcomings in their analysis.
The report states that images with trees and shadows were excluded to avoid over or under counting vehicles.
However, satellite images released to the media show large areas of hospital parking lots blocked by tall buildings, which means that it is not possible to accurately estimate the number of cars available.
In the tweet below, we marked the areas covered by the tall buildings with white boxes.
Benjamin Strick
@ BenDoBrown
June 11, 2020
Reply to @BenDoBrown
Oh man. I haven't looked at the report's SEO analysis page, but it seems that this paper could contain some banger errors:
https: //
twitter.com/WillMaStat/sta
tus / 1270575010093182976? s = 20
...
Will Ma
@ WillMaStat
A team led by @johnbrownstein from @harvardmed used satellite imagery and search engine data to indicate that #coronavirus may have hit Wuhan in August. However, they seem to have used incorrect Chinese words for the search. Your results in hospital traffic have no statistical significance.
Benjamin Strick
@ BenDoBrown
The angle of the images shown in the @ ABC report on this block blocks parts of the parking lot that would cloud the data on which a trend in cars in hospital parking lots is based. H / T to @Fang__z to highlight this first.
5
1:51 p.m. - June 12, 2020
Twitter Ads info and privacy
See Benjamin Stricks' other tweets
There is also an underground garage in the Tianyou Hospital, which is visible in Baidu's Street View function, but only the entrance is visible on satellite images - not the cars under the ground.
One of the study's authors, Benjamin Rader, said: "We cannot definitely consider the underground parking garage in any period of the study, and this is one of the limitations of this type of research."
Benjamin Strick
@ BenDoBrown
June 12, 2020
Reply to @BenDoBrown
The angle of the images shown in the @ ABC report on this block blocks parts of the parking lot that would cloud the data on which a trend in cars in hospital parking lots is based. H / T to @Fang__z to highlight this first.
Benjamin Strick
@ BenDoBrown
A fundamental mistake in @ Harvard's report examining cars in hospitals, indicating early signs of COVID, was the failure to inspect the parking garage. Here is the entrance to the underground car park of the Tianyou hospital. It was full in 2017:
https: //
map.baidu.com/search/wuhan/@
12727583.74,3550200.65,21z, 87t, 159.68h # panoid = 09000200121905181536293348Y & panotype = street & Heading = 169.04 & Pitch = -4.62 & l = 21 & tn = B_NORMAL_MAP & sc = 0 & newmap = 1 & shareurl = 1 & p194812
... H / T @crushspread & @Wanyuan_Song
11
1:58 p.m. - June 12, 2020
Twitter Ads info and privacy
See Benjamin Stricks' other tweets
There are also concerns about the choice of hospitals for the study.
The Hubei Women's and Children's Hospital is one of the included locations, but children rarely need hospitalization for coronavirus. In response, the authors say that overall their results would still show increased parking garage use, even if this hospital were excluded from the survey.
The researchers could also have compared their data with hospitals in other Chinese cities to determine whether the increase in traffic and searches was specific to Wuhan, where the outbreak first became known.
Without this comparison, in addition to the questions we raised about searching for medical symptoms online, the evidence for Wuhan residents who have been treated for coronavirus since August last year remains highly controversial.
However, we don't know much about the early spread of the virus in Wuhan.
Banner picture with the words "More about Coronavirus"
SIMPLE GUIDE: How do I protect myself?
IMPACT: What the virus does to the body
RECOVERY: How long does it take?
LOCKDOWN: How can we remove restrictions?
ENDGAME: How do we get out of this mess?
banner
Reality check branding
Read more about Reality Check
Send us your questions
Follow us on Twitter

Click to receive the most important news as a notification!

Last News

Trump Creates Radical Coalition Declaring ‘No International Right’ To Abortion

Post-race brawl between Gordon and Keselowski

Laptop connected to Hunter Biden linked to FBI money laundering probe

Judge dismisses 1 charge against former cop in Floyd’s death

Trump violates agreement and releases "60 Minutes" interview

This Cyclist Just Set a New World Record for Riding the Furthest Distance in a Week