Israeli research: Google search can help predict COVID-19 spikes
Analyzing search activity currently stands as an established method for tracking and monitoring the seasonal flu.
By ZACHARY KEYSERA Google search page is seen through a magnifying glass in this photo illustration taken in Brussels(photo credit: FRANCOIS LENOIR / REUTERS)
Online search activity could help predict spikes in COVID-19 cases, according to a recent study performed by the University-College of London (UCL) in cooperation with researchers from Bar-Ilan University.Using online search data, experts could predict peaks in coronavirus cases "an average 17 days in advance."UCL explains that analyzing search activity currently stands as an established method for tracking and monitoring the seasonal flu.The researchers profiled the COVID-19 symptom list and developed an analytical model that studies symptom-related Google searches and compares them against spikes in COVID-19 cases.The team also worked toward eliminating media bias within the results, meaning searches may spike in areas due to media coverage of a certain subject, where it does not indicate a rise in cases.The model used symptoms identified by the NHS and Public Health England (PHE), and weighted the terms according to how frequently they occurred in confirmed COVID-19 cases.Lead author of the research Dr. Vasileios Lampos of UCL's computer science department claimed that this research was one of the first to associate skin rash and the loss of smell to be confirmed COVID-19 symptoms.“Adding to previous research that has showcased the utility of online search activity in modelling infectious diseases such as influenza, this study provides a new set of tools that can be used to track COVID-19," Lampos said.“We have shown that our approach works on different countries irrespective of cultural, socioeconomic and climate differences," he added. "Our analysis was also among the first to find an association between COVID-19 incidence and searches about the symptoms of loss of sense of smell and skin rash."We are delighted that public health organizations such as PHE have also recognized the utility of these novel and non-traditional approaches to epidemiology.”
Throughout the pandemic, UCL has shared its research with PHE on a weekly basis while conducting the study, to support the government efforts of containing the viral spread. The researchers published their findings in Nature Digital Medicine.The analytical model was also applied in several other countries, including the UK, as well as the United States, Italy, Australia and South Africa, among others, where UCL claims the same patterns were discovered and typically linked to a rise in coronavirus cases within the areas of focus.“Our best chance of tackling health emergencies such as the COVID-19 pandemic is to detect them early in order to act early," said Co-author Michael Edelstein, Professor of Population Health at Azrieli Faculty of Medicine of Bar-Ilan University."Using innovative approaches to disease detection, such as analyzing internet search activity to complement established approaches, is the best way to identify outbreaks early,” he added.The researchers purport that these systems will be put in place to advance conventional epidemiological models for the future.“We can at least use the plethora of data sets around COVID-19 for further experimentation and validation of such techniques in an attempt to complement current epidemiological approaches and be better prepared for the next pandemic,” Lampos concluded.
