Annual Report Finds Harvard Kennedy School Faculty Remains Largely White, Male
Harvard Square Celebrates Oktoberfest
Harvard Corporation Members Donated Big to Democrats in 2020 Elections
City Council Candidates Propose Strategies for Supporting Low-Income Residents at Virtual Forum
FAS Dean Gay Hopes to Update Affiliates on Ethnic Studies Search by Semester’s End
The novel coronavirus may have begun spreading in China as early as August 2019, new research from a team of scientists led by Harvard Medical School professor John S. Brownstein suggests.
The research — published as a preprint on June 8 — was motivated by a desire to overcome the limitations of currently available coronavirus data, according to Benjamin M. Rader, an author on the paper and Boston University graduate student.
“By the time someone's in the hospital, or the CDC gets information about the disease, there's some lag of information going to public health agencies and getting aggregated and collated and cleaned,” Rader said.
In the preliminary study, researchers analyzed satellite imagery and common queries from the Chinese search engine Baidu to investigate the initial emergence of COVID-19 in Wuhan, China.
The team partnered with RS Metrics, a satellite data collection and analysis company, to pore over images taken within the past two years of Wuhan hospital parking lots, and found a significant increase in traffic around major Wuhan hospitals during fall 2019 compared to fall 2018. Certain images displayed up to 90 percent increases in parked cars.
After they observed the increased hospital traffic, Rader said, the team set out to determine whether the spike had any correlation with COVID-19 cases in Wuhan. After analyzing index information from Baidu, they found that “cough” and “diarrhea” queries on the website surged during at the same point in time.
“While queries of the respiratory symptom “cough” show seasonal fluctuations coinciding with yearly influenza seasons, “diarrhea” is a more COVID-19 specific symptom and only shows an association with the current epidemic,” the researchers wrote in their manuscript.
Rader said using data from a search engine can reveal information about virus spread and presence that may be hard for public health officials to detect.
“There's some balance between the number of people searching for something where it would get picked up by a search engine, but still low enough where it might not get picked up by public health authorities,” Rader said.
The researchers believe this was potentially the case in Wuhan during the months leading up to December 2019.
Despite its conclusions, the study — which is now submitted for peer review — faced many limitations, according to Rader. He said the full story of how COVID-19 emerged largely remains a mystery to public health experts.
“It's really just a sliver of information that exists and needs to be interpreted as part of a broader picture,” Rader said.
Brownstein said that such limitations leave the door open for much further work on the topic, underscoring the need to “strengthen our public health surveillance globally.”
“Our work wasn’t meant to point any fingers,” Brownstein said. “Signals around disease get missed all the time, so it’s all about how do we improve the ability to identify things and take the lessons learned to prevent the next pandemic.”
“I think it’s not one paper that is going to clarify the origins of COVID-19; it's a lot of work among many scientists who are going to have to really think deeply,” he added.
—Staff writer Virginia L. Ma can be reached at email@example.com.
—Staff writer Meera S. Nair can be reached at firstname.lastname@example.org.
Want to keep up with breaking news? Subscribe to our email newsletter.