Google’s online search activity can help predict spikes in Covid-19 cases up to 17 days in advance, a new study reveals.
Researchers at University College London have created computer models based on online search frequencies to gain insight into the prevalence of the disease in multiple countries, including the UK.
Models based on online searches that were expected to be successful confirmed Covid-19 cases and deaths by 16.7 and 22.1 days, respectively.
The team’s analysis was one of the first to find a link between the incidence of Covid-19 and searches for the symptoms of loss of smell and rash – two symptoms of the disease listed by Public Health England.
Online search data should be used with ‘more established approaches’ to develop public health surveillance methods for Covid and other emerging infectious diseases, the experts say.
Google’s online search data can help inform the public health response to Covid-19, according to a report by the academics at University College London. Previous research has shown that different properties of infectious diseases can be deduced from online search behavior
“This study provides a new set of tools that can be used to track Covid-19,” said the study’s lead author, Dr. Vasileios Lampos of University College London.
‘We have shown that our approach works in different countries, regardless of cultural, socio-economic and climate differences.’
UCL researchers used the Covid-19 symptom profile to develop models of its prevalence by looking at symptom-related searches through Google.
They then recalibrated these models to reduce the bias in these “signals” caused by the public eye, in other words, the effect media reporting has on online searches.
They developed the uncalibrated model by choosing search terms related to Covid-19 symptoms identified by the NHS and Public Health England (PHE).
The three most common symptoms of Covid-19 are a high temperature, a new and persistent cough, and a loss or change in the sense of smell or taste.
PHE also lists fewer different less common symptoms, including aches and pains, headaches, and rashes.
Terms were weighted by their frequency of occurrence in confirmed Covid-19 cases.
This model provided “actionable insights,” including early warnings, and showed the effects of physical distance measures, UCL said.
The calibrated version, taking into account news coverage, enabled academics to provide PHE with a model to more accurately predict UK spikes.
The model has been used in several countries, including the UK, US, Italy, Australia and South Africa.
They found that the same pattern appeared, namely that peaks in cases were predicted by their model.
The graph shows the online search scores for Covid-19 for different countries at the end of 2019 and the beginning of 2020. The frequency of searches is weighted based on the likelihood of symptoms occurring (blue line) and the effects of news media are minimized (black line) . Dates for physical distance or lockout measures are indicated with vertical dashed lines
“Our best chance of addressing health emergencies like the Covid-19 pandemic is to detect them early in order to act early,” said study co-author Professor Michael Edelstein of Bar-Ilan University, Israel.
“Using innovative approaches to disease detection, such as analyzing Internet search activity, to complement existing approaches, is the best way to identify outbreaks early.”
Academics working on the models share their findings weekly with PHE to support disease response, which can be viewed online.
“We are pleased that public health organizations such as PHE have also recognized the usefulness of these new and nontraditional approaches to epidemiology,” said Dr. Lampos.
Internet search analysis is a well-established method of detecting and understanding infectious diseases and is currently used to track seasonal flu. Flu Detector estimates the number of flu-like illnesses in England based on internet searches and is included in Public Health England’s flu surveillance metrics
Analyzing Internet search activities is an established method of detecting and understanding infectious diseases.
The technique is already being used track seasonal flu in the form of UCLs Flu Detector.
The constantly updated online tool estimates the number of flu-like illnesses in England based on internet searches and is included in Public Health England’s flu surveillance metrics.
“Previous research has shown the usefulness of online search activities in modeling infectious diseases such as influenza,” said Dr. Lampos.
The study, “Tracking COVID-19 with Online Search,” is published today in Nature Digital Medicine.