First Study to Quantify Potential COVID-19 Spread from Hurricane Evacuation: Where Evacuees Go Matters

With the peak of the hurricane season upon us and COVID-19 widespread and abundant in many hurricane-prone areas, the United States is poised to experience the collision of two major disasters. According to a study by scientists at Columbia University and the Union of Concerned Scientists (UCS), a large-scale hurricane evacuation in the U.S. would increase COVID-19 cases in evacuees' origin and destination counties. Directing evacuees to counties with low COVID-19 transmission rates rather than allowing evacuations to follow historical patterns would minimize the increase in cases, according to the study, though keeping virus transmission rates low in destination counties is also essential.

The research, which is awaiting publication in a peer-reviewed journal and is now posted on the medRxiv preprint server for health sciences, is the first to quantify how hurricane evacuation may affect the number and spatial distribution of COVID-19 cases in the United States.

"Many of the country's most hurricane-prone states have recently experienced some of the highest COVID-19 growth rates in the nation," said Kristy Dahl, report co-author and senior climate scientist at UCS. "In every scenario we analyzed, hurricane evacuations cause an increase in the number of COVID-19 cases. Minimizing that increase depends on getting people to destinations with low virus transmission rates and ensuring that those transmission rates stay low even when there's an influx of evacuees."

The study entailed building a hypothetical hurricane evacuation scenario in which residents of Palm Beach, Broward, Miami-Dade, and Monroe counties fled a Category 3 hurricane. Based on studies of evacuation compliance and behavior, the study assumed 2.3 million people would leave the four counties. Post-Hurricane Irma surveys were used to simulate where they would go. That information was then used in a national county-scale model of COVID-19 transmission to determine how many cases would result from the evacuation and where they would occur.

The study assumed that COVID-19 transmission rates in destination counties increased during the evacuation period not at all or by 10 percent or 20 percent, representing the levels of public health directives that were put in place in destination counties and how well they were followed, as well as whether evacuees stayed with friends or family members or in hotels or shelters.

Under the worst-case scenario considered by the authors, if people followed historic evacuation patterns and virus transmission rates increased by 20 percent in their destination counties, there would be roughly 61,000 additional COVID-19 cases in the origin and destination counties combined.

Under the best-case scenario considered by the authors, if people instead evacuated to communities with low COVID-19 transmission rates and transmission rates did not increase in the destination counties, there could be as few as 9,100 additional cases resulting from the evacuation.

"Directing evacuees to destinations with low virus activity and providing housing opportunities and resources that help maintain social distancing, encourage mask usage, and limit opportunities for virus transmission will be essential," said Jeffrey Shaman, senior study author, professor of environmental health sciences, and director of the Climate and Health Program at Columbia University Mailman School of Public Health.

The scientists hope their study will help inform the work of emergency managers and other local decisionmakers as well as federal and state agency staff as the country enters peak hurricane season.