The association between smartphone-derived population-level mobility and COVID-19 in 2020: a census tract-level analysis of King County

Abstract

Stay-at-home orders were an important tool in mitigating the COVID-19 pandemic. This ecological study used small area estimation models to estimate the associations between mobility and social vulnerability with COVID-19 case counts under different lockdown conditions for all 397 census tracts in King County, Washington in 2020. Data was used from Safegraph’s Social Distancing Metrics dataset, King County’s COVID dashboard, and American Community Survey data. Census tracts with higher mobility and increased social vulnerability were more likely to have higher numbers of COVID cases, especially as the county entered into their reopening plan. In early months of the year the association was stronger in northern tracts, particularly around the Seattle and Eastside areas. In later months, the southwestern tracts had a stronger association. This spatial analysis illustrates the many factors that influence higher transmission in certain areas, as well as the need to address geographic inequalities and fine tune public health response appropriately.