Study uses satellite imagery, internet searches to track COVID-19, how reliable is it?


RALEIGH, N.C. (WNCN)- A study out of Harvard Medical School claims COVID-19 could have started spreading in China since early August after studying satellite imagery and internet searches.

Authors of the study said they studied six Wuhan hospital parking lots and found increased traffic in their parking lots in late summer and early Fall 2019.

According to the authors, those traffic increases coincided with internet searches for “diarrhea” and “cough”. Both are known symptoms of COVID-19.

The study claimed in searches for the term “diarrhea” increased in ways not seen during previous flu seasons. They also surpassed searches for “cough”.

The authors hypothesized hospital traffic increased following these searches as a result of community spread. They said those patterns also correlated with new case clusters of COVID-19 in Wuhan in May.

Officials in China reported the virus first made an appearance in December.

Professor Jennifer Kuzma, a professor in the School of Public and International Affairs and Co-director of Genetic Engineering and Society Center at NCSU said we should take this study with a grain of salt.

She says this is a correlation study and correlation does not equal causation.

“I’m not entirely convinced by the author’s argument however, I know that epidemiology and trying to track down the origins and the early spread of the disease is a very difficult scientific process and sometimes any piece of scientific evidence is important,” Kuzma said.

The professor said analyzing tissue or blood samples versus traffic patterns could be stronger evidence.

Kuzma said while any piece of information could be valuable not any one piece of information is enough. While not dismissing the study, she said it is not confirmatory.

“I understand that it is difficult to track down the origin of a disease. This is a new area, this intersection of data science and epidemiology and it’s an exciting area. I think it holds a lot of promise but we’re still going to have to realize there are going to be correlation studies and we’re going to have to show more evidence for by digging deeper into the issues. It’s kind of detective work, epidemiology,” said Kuzma

Among concerns over the study is the small sample size. Kuzma said the changes in traffic are not large enough to be conclusive. Less than 10 hospitals were studied, limiting the scope of the study. Kuzma added while there was a slight increase in an internet search for certain symptoms, they are not large enough to determine a conclusive result either.

The study is not peer-reviewed and self-published not allowing others to weigh in beforehand.

Kuzma said the authors do make a convincing argument that this new piece of evidence can be pieced together with other studies to get a better idea of the situation as a whole.

“If we continue to find more pieces of the puzzle, it could be possible that this virus was around for quite a few months before December 2019 but I don’t think we have conclusive evidence yet,” Kuzma said.

She said scientists could never be absolutely certain about the origin of a virus. She said there will always be one-two percentage of uncertainty but researchers are still far from reaching that mark.

Read the study for yourself by clicking here.

Copyright 2021 Nexstar Inc. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.

Trending Stories

Click here for full list of trending stories