Cell phone location data could be useful when it comes to predicting future trends in Covid-19 cases, according to new research published Monday in the journal JAMA Internal Medicine.
“Perhaps the most important observation of this study was that a decrease in activity at the workplace, transit stations, and retail locations and an increase in activity at the place of residence was associated with a significant decline in Covid-19 cases at 5, 10 and 15 days,” said the researchers, led by Dr. Shiv Sehra at Harvard Medical School.
For example, counties with the greatest use of cell phones in residential locations had a 19% lower growth rate of new cases at 15 days compared with those counties that had the lowest level of home usage.
The researchers also found that activity at grocery stores and areas that were classified as parks was not strongly associated with rates of growth in cases. However, assessing the direct effect of individual activities is difficult, they said.
The researchers used publicly available cell phone location data and new daily reported cases per capita in each US county to evaluate the association between cell phone activity on a given day, in a number of different locations, and the rate of growth in new Covid-19 cases five, 10 and 15 days later.
They found that there was a marked change in activities shortly before stay-at-home orders were issued in individual states, which included less activity in locations outside the home.
Urban counties with higher population levels and higher numbers of cases per capita had a greater increase in cell phone usage inside the home after stay-at-home orders, the researchers found.
However, as the time from the stay-at-home order increased, the use of cell phones at non-residential locations did as well. For example, the researchers said that, on average, there was a 0.5% increase per day at retail locations from the time of the initial stay-at-home order, which suggests “waning adherence to the orders over time.”
Keep in mind: The study did have some limitations, including a potential for selection bias. There could also be other differences at the county level, such as mask mandates during the study period.