A mathematical model to predict the spread of …

A team of researchers created a mathematical model to predict the spread of the coronavirus in cities by analyzing three variables: where people go throughout the day, how long they stay and how many people are visiting the same place at the same time. According to scientists, this model may help prevent the increase in covid-19 cases.

“We create a computer model to analyze how people from different demographic backgrounds and from different neighborhoods visit different types of more or less crowded places. Based on all of this, we were able predict the likelihood of new infections at any time, place or time, ”says Jure Leskovec, a computer scientist at Stanford University in the United States, who led the study, which also included researchers from Northwestern University.

Where are the contagions

The study, published this week in the prestigious scientific journal “Nature”, combines demographic data of the cities analyzed, epidemiological reports and the mobility of people, collected via the location of cell phones. The results showed that Most virus transmissions occur in ‘super spread’ places: restaurants, bars and gyms, where people stay indoors for longer periods of time..

The researchers assure that the specificity of their model can serve as a tool for public officials to implement infection reduction measures when businesses reopen.

Study co-author David Grusky, professor of sociology at the Stanford School of Humanities and Sciences, notes that this predictive capacity of the model is very valuable as it provides new information on ethnic minorities and low-income people, where most coronavirus cases have been recorded.

“In the past, these disparities were believed to be due to pre-existing conditions and unequal access to health care, while our model suggests that mobility models also contribute to driving these disproportionate risks“, Add.

According to Grusky, the model shows that reopening businesses with lower occupancy limits tends to benefit the most disadvantaged groups. “As places that employ minorities and low-income people are often smaller and more populated, occupancy limits for reopened stores can reduce the risks they face,” warns the researcher.

Grusky thinks that “We have a responsibility to develop reopening plans that eliminate, or at least reduce, the disparities that current practices create”. The investigation also proves that the containment measures made it possible to stop the transmission of the virus.

How was the study carried out

The study tracked the movements of 98 million Americans in 10 of the nation’s largest metropolitan areas – New York, Los Angeles, Chicago, Dallas, Washington, DC, Houston, Atlanta, Miami, Philadelphia and San Francisco – across million commercial establishments

The SafeGraph company then provided location data from anonymous mobile apps that allowed researchers to see which of the 553,000 places visited, for how long, and what the size of each of those sites was. The latter made it possible to calculate the occupancy density of the space per hour

The researchers analyzed the data from March 8 to May 9 in two separate phases. In the first phase, they fed their mobility data model and designed their system to calculate a crucial epidemiological variable: the rate of virus transmission under various different circumstances in the 10 metropolitan areas.

In addition, the researchers took into account how many cases of coronavirus were there in each city according to epidemiological reports. This allowed them calculate the probability of transmission of the virus at different places and times.

The mathematical model

In the second phase, the researchers tested their mathematical model to predict new cases of covid-19. The predictions turned out to be quite close to reports that health officials later provided, or that gave researchers a guideline for the reliability of their model.

Through research, Scientists have been able to demonstrate how low-income people are the ones who have had to leave their homes most often during the pandemic because their jobs demand it and they buy from smaller and more crowded establishments than people high income. who can work from home, use the home delivery service to avoid leaving their home, and move to larger premises

The study found that non-white Americans are twice as likely to buy food. “By merging the mobility, demographic and epidemiological datasets, we were able to use our model to analyze the effectiveness and fairness of different reopening policies,” says Serina Chang, doctoral student at Stanford.

“In principle, anyone can use this model to understand the consequences of different policy decisions for closing businesses and staying at home,” Leskovec concludes.