COVID-19: Mobile phone data exposes venues for ‘superspreader’

A SARS-CoV-2 transmission model indicates that the majority of infections are accounted for by a limited number of venue categories, such as restaurants, hotels, and religious venues. The model also helps understand why people live in poorer areas are affected overwhelmingly by infections.

Friends
Image credit: vgajic/Getty Images

There is still significant confusion about the best ways to restore economic and social life to something close to normality almost a year after the start of the COVID-19 pandemic in Wuhan, China.

The model offers a method for defining high-risk venues and exploring possible ways out of lockdown, according to the developers of the new model at Stanford University, CA, and Northwestern University in Chicago, IL.

Computer scientists and medical researchers worked on the model, which reported the movements of 98 million citizens in the United States between March 1 and May 2, 2020, using anonymized location data from mobile phone networks.

At about 553,000 venues, the team modelled the number of infections occurring hour by hour, which they divided according to usage into 20 groups. They also took into account each venue’s floor space.

In 10 of the largest metropolitan regions, including Chicago, New York City, and San Francisco, their simulations correctly predicted daily reported infections.

The model indicates that reopening gyms, full-service restaurants, cafes, hotels, and religious venues, due to the high densities of individuals and their long periods of stay, contributes to the greatest rise in infections.

According to the model, a relatively small number of these “superspreader” venues account for the majority of new infections.

For instance, the model found that 85 percent of all infections accounted for 10 percent of all venues in the Chicago metropolitan area.

Capping numbers of visitors

On the upside, the model shows that it is more efficient and less disruptive to limit the number of people allowed into venues at any time than to uniformly reduce the freedom of movement of all.

The model, for instance, predicts that restricting a venue’s occupancy to 20% of its full capacity eliminates new infections by more than 80%.

Since people are likely to respond by spreading their visits more thinly during the day, however the measure reduces by a relatively modest 42 percent the total number of visits.

“The researchers write in their paper explaining the model, which appears in the journal Nature, “[O]one can achieve a disproportionately large reduction in infections with a small reduction in visits. “Precise approaches like these could be more successful than less focused steps, with slightly lower economic costs being incurred.”

“Our study emphasizes that it doesn’t have to be all or nothing,” Stanford University senior author Jure Leskovec said at a press conference to reveal the results.

“We can select different levels for different types of locations, he said and our model offers a tool for policymakers to basically manage and make the right choices for these trade-offs.”

Disadvantaged groups

In racially and socioeconomically deprived communities, the model correctly predicts higher infection rates.

During lockdowns, individuals within these classes have been shown to decrease their mobility levels to a lesser degree, possibly as a result of operating in critical services.

The model also showed that a few types of location, such as full-service restaurants, were driven by high infection rates in deprived areas.

The venues tended to be smaller compared with more affluent places, contributing to higher customer densities. Furthermore, individuals going to these venues appeared to stay longer.

In low-income areas for example, grocery stores had 59 percent more people per square foot than in higher-income areas, and their customers remained, on average 17 percent longer.

As a result, Leskovec said the model showed that the risk of infection was approximately twice as high for a low-income person visiting a grocery store compared to a high-income individual.

Cell phone data

The researchers drew on data from SafeGraph, a company that aggregates anonymized cell phone details.

The hourly movements of 98 million people from 57,000 neighbourhoods, which they call “census block groups,” or CBGs, to 553,000 separate locations, which they call “points of interest,” or POIs, were mapped using this data.

Their basic model of infection known as the SEIR model, predicts how individuals in each neighborhood and location transfer between four categories: vulnerable to infection, exposed to infection, contagious, and removed (recovered, self-isolated, or died).

Importantly, not only the locations of cell phone users are revealed by SafeGraph data, but also the floor space of venues they visit and their purpose, such as a place of worship or grocery store.

By using the local infection and death rates that The New York Times publishes online, the researchers “fine-tuned” their model.

Finally, they reported that the model could probably be used to forecast future rates of infection and death.

Missing groups of people

Dr. Julian Tang, a clinical virologist who was not involved in the research at the University of Leicester in the United Kingdom, welcomed the research but cautioned that its results might not be uniformly applicable.

“The use of cell phone data and modeling are powerful tools to help us understand how the virus spreads, but we need to be cautious about the interpretation as what may apply may not necessarily apply elsewhere in one population,” he said.

Kevin C. Ma and Marc Lipsitch of the Harvard T. H. Chan School of Public Health in Boston, MA, point out in a comment article accompanying the paper that the model does not account for infections among infants, older adults, and those in jail.

They write:

“Further model testing is needed, but given the challenges in gathering and interpreting other relevant data types, these findings could have a valuable role in guiding policy decisions on how to reopen society safely and minimize the harm caused by movement restrictions.”

Related Articles

Back to top button