The analysis of web searches has already been used to predict the spread of other infectious diseases, and the analysis of Google web searches for specific keywords can also be used to predict infection hotspots in COVID-19, the researchers report. Their study results were published in the journal “Mayo Clinic Proceedings”.
The number of COVID-19 cases is currently rising rapidly and new hotspots of infections with the new coronavirus SARS-CoV-2 are suddenly appearing again and again. A research team at Mayo Clinic (USA) reports that web search queries could help to predict where such hotspots will appear.
Dr. Mohamad Bydon’s research team analyzed possible connections with the COVID-19 outbreak for ten search terms. The search queries were selected based on the frequency of their use and the emerging patterns on the Internet and in Google News at the time, the researchers explain. The following search queries were considered:
The searches were conducted days before the first COVID-19 cases were reported in certain regions, and each of the keywords had a different correlation with the number of cases, Dr. Bydon said.
Ten keywords evaluated
Overall, strong correlations between the corresponding keyword searches and COVID-19 outbreaks were found in parts of the United States and these correlations were observed in some states up to 16 days before the first reported cases, the researchers report. “Our study shows that there is information in Google Trends that precedes the outbreaks,” concludes Dr. Mohamad Bydon.
“A look at the Google Trends data showed that we were able to use keywords to identify predictors for hotspots that would emerge over a period of six weeks,” he continued. This could also be used to better allocate resources in terms of testing, personal protective equipment, medication and more.
The evaluation of web searches is a good way to predict where future trouble spots will arise. So far, the evaluation has only included ten search terms and “if we had looked at 100 keywords, we might have found even stronger correlations to the cases,” said the study leader.
“Any delay in providing information could mean that opportunities are missed to improve preparedness for an outbreak in a particular location,” Dr. Bydon emphasizes. And “if you wait for the hotspots to appear in the media coverage, it will be too late to react effectively.
COVID-19: Using search queries to predict hotspots
As the pandemic progresses, people will also search for new and different information, but the search terms will have to be adapted again and again, explain the researchers. Overall, however, this offers a promising option for predicting the occurrence of the infection. (fp)
WashingtonNewsday Health and Wellness.