A new study reveals that online search activity extracted from Google can help predict peaks in Covid-19 cases up to 17 days in advance.
Researchers at University College London created computer models based on online search query frequencies to obtain information on the prevalence of the disease in several countries, including the United Kingdom.
Online search-based models successfully predicted confirmed Covid-19 cases and deaths, confirmed by 16.7 and 22.1 days, respectively.
The team’s analysis was one of the first to find an association between the incidence of Covid-19 and research on the symptoms of olfactory loss and rash, two symptoms of the disease listed in Public Health England.
Online search data should be used with “more established approaches” to develop public health surveillance methods for Covid and other new infectious diseases, experts say.
According to a report by University College London academics, online search data extracted from Google may help inform the public health response to Covid-19, according to previous research.
“This study provides a new set of tools that can be used to monitor Covid-19,” said the study’s lead author, Dr. Vasileios Lampos at University College London.
“We have shown that our approach works in different countries, regardless of cultural, socioeconomic and climate differences.”
UCL researchers used Covid-19’s symptom profile to develop models of its prevalence by searching for symptoms through Google.
They then recalibrated these models to reduce the bias in these “signals” caused by the public interest, that is, the effect that media coverage 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 high temperature, a new, continuous cough, and a loss or change in smell or taste.
PHE also lists fewer less common symptoms such as aches and pains, headaches, and rashes.
The terms were weighted according to their occurrence ratio in confirmed cases of Covid-19.
This model provided “useful information,” including early warnings, and showed the effects of physical distancing measures, according to UCL.
The calibrated version, which took into account news coverage, allowed academics to provide PHE with a model to more accurately predict overvoltages in the UK.
The model was applied to several countries, including the United Kingdom, USA, Italy, Australia and South Africa, among others.
They found that the same pattern appeared, as their model predicted case rises.
The graph shows Covid-19’s online search scores for different countries in late 2019 and early 2020. Query frequencies are weighted by the likelihood of symptoms (blue line) and the effects of the symptoms are minimized. news media (black line). Dates for physical distances or blocking measures are indicated by dotted vertical lines
“Our best way to address health emergencies like the Covid-19 pandemic is to detect them early so we can act early,” said study co-author Michael Edelstein of the University of Barcelona. Ilan, Israel.
“Using innovative approaches to disease detection, such as analyzing Internet search activity to complement established approaches, is the best way to identify outbreaks early.”
Scholars working on the models have shared their findings with PHE on a weekly basis to support the disease response, which can be viewed online.
“We are delighted that public health organizations like PHE have also recognized the usefulness of these new and non-traditional approaches to epidemiology,” Dr. Lampos said.
Internet search activity analysis is an established method of monitoring and understanding infectious diseases and is currently used to control seasonal influenza. The flu detector estimates flu-like illness rates in England based on web searches and is included in Health Public England’s flu surveillance metrics
Internet search activity analysis is an established method of monitoring and understanding infectious diseases.
The technique is already getting used to monitor seasonal flu in the form of UCL Influenza detector.
The constantly updated online tool calculates flu-like illness rates in England based on web searches and is included in Health Public England’s flu surveillance metrics.
“Previous research has shown the usefulness of online search activity to model infectious diseases like the flu,” Dr. Lampos said.
The study, “Tracking COVID-19 Using Online Search,” was published today in Nature Digital Medicine.