How Google search data can predict COVID-19 outbreaks
Although some of the behaviors that lead to SARS-CoV-2 infections are obvious, new waves of COVID-19 cases usually do not always follow predicted patterns.
Now, however, a report from researchers at New York University’s Courant Institute of Mathematical Sciences describes a possible means of spotting contamination surges before they happen through the evaluation of online searches.
The researchers learned a correlation between a surge in searches associated with activities outside the residence - activities that could put persons vulnerable to SARS-CoV-2 infection - and a growth in COVID-19 cases 10-14 times afterward. Attacks fell when there is an increase in searches relating to stay-at-home activities.
Analysis author Anasse Bari, a medical assistant professor at the Courant Institute, notes that professionals have already successfully used data mining “on finance to create data-driven investments, such as studying satellite pictures of cars on parking lots to predict businesses’ earnings.”
“Our research shows the same tactics could be put on combatting a pandemic by spotting, in advance, where outbreaks are likely to occur,” says senior author Megan Espresso of the Division of Infectious Disease & Immunology at the New York University (NYU) Grossman School of Medicine.
Identifying with greater accuracy those behaviors that create infection spikes can help epidemiologists and policymakers better shape public policies relating to closures, lockdowns, etc.
The system that the analysis paper describes avoids privacy issues by involving only large clusters of anonymized info.
The analysis appears in Social Network Analysis and Mining.
Mobility vs. isolation
The researchers’ first rung on the ladder was to build up categories based on search phrases or keywords that they could then track.
Both key categories that they tracked were called the mobility index and the isolation index.
The team assigned particular searches to the mobility index track, including “theaters near me,” “flight tickets,” and different inquiries about activities that involve leaving the home and being in physical proximity with others.
As Bari sets it, “When someone queries the closing period of an area bar or looks up directions to an area gym, they provide some insight into what future risks they could have.”
For the isolation index track, the experts collected search queries - such as for example “at-residence yoga” or “food delivery” - that indicated an purpose to stay home and isolated.
The experts based their categorization of keywords on the Democracy Fund + UCLA Nationscape survey - a report where respondents listed the things that they would be undertaking if “restrictions were lifted on the advice of public health officials regarding activities.”
The survey found that the most notable three activities that persons missed were “likely to a stadium/concert,” “going to the films,” and “attending a sports event.”
According to Bari, “This is a first step toward creating a tool that will help predict COVID-19 case surges by capturing higher risk actions and intended mobility, which searches for gyms and in-person eating can illuminate.”
THE WEB Movement Index
The researchers collected search info for March through June in 2020 from all 50 states in america. They used Google Styles to track tendencies in the data, allowing them to develop the mobility and isolation indexes.
The researchers also created a “Net Movement Index” to point the relationship between the two indexes. An increased Net Activity indicated a change toward flexibility search queries and from isolation searches.
The authors write, “We theoretically expect that a abrupt decline in net activity (i.e., more people staying home) would correspond to a decrease in COVID-19 pass on, with a lag equal to the incubation amount of COVID-19.”
Source: www.medicalnewstoday.com
