Extreme storms kill and injure Americans every year. But new research from the University of Chicago shows that who suffers most during storms is not simply a matter of being in the wrong place at the wrong time. It is shaped by the social and economic conditions of the communities where people live.
The study, published in Nature Communications, analyzed nearly 112,000 storm events across U.S. counties between 2005 and 2022. It finds that counties with greater concentrations of low-income Black households relative to high-income white households experience dramatically higher rates of storm-related injuries and deaths, even when those counties are not hit by storms more frequently than others.
Counties in the most structurally disadvantaged group experienced direct injury rates nearly seven times higher than those in the most advantaged group (0.82 versus 0.12 per 100,000 people). Death rates were similarly unequal.
“Storm-related injuries and deaths in the United States are not random consequences of extreme weather,” says co-author Kate Burrows, assistant professor of public health sciences at the University of Chicago. “They are the outcome of deeply embedded structural inequalities, shaped by decades of disinvestment and exclusionary policy decisions.”
Why Some Communities Suffer More
Not all storms are equally deadly. But this study finds that the key variable is not the storm itself. It is the community it strikes.
The researchers used a measure called the Index of Concentration at the Extremes (ICE), which captures the degree to which a county’s households are concentrated at opposite ends of the economic and racial spectrum, with high-income non-Hispanic white households on one end and low-income Black households on the other. ICE was developed to reflect the cumulative, historically rooted nature of structural inequality in the United States.
Using 18 years of storm records covering hurricanes, tornadoes, flash floods, coastal floods, and high winds, the team linked each event to the structural conditions of the affected county. The results were striking: each quintile increase toward greater structural disadvantage was associated with 52 percent higher rates of direct injuries and 36 percent higher rates of direct deaths.
The risk was not evenly distributed across the disadvantage spectrum, either. Spline models revealed that the elevated risk was concentrated in the most disadvantaged counties and flattened out considerably as structural conditions improved.
Exposure Is Not the Problem
Critically, the researchers found that more disadvantaged counties were not simply being hit more often. Comparing maps of structural disadvantage and storm frequency side by side, the team found that the two did not systematically overlap. Communities that bore the heaviest burden of storm harm were not necessarily in higher-exposure regions.
This distinction matters. It means that the gap in storm outcomes is not driven by geography or hazard severity alone. It is driven by social and infrastructural context. The same storm, striking two different communities, can produce vastly different consequences depending on the structural conditions of each.
A History Etched in Geography
The researchers situate their findings within a longer history of structural inequality. ICE captures the present-day concentration of economic and racial disadvantage within counties, but those patterns were shaped in part by decades of discriminatory housing and lending policy. The study points to redlining, the federal government’s practice beginning in the 1930s of systematically withholding credit and investment from neighborhoods with large Black, immigrant, and working-class populations, as one example of how structural disadvantage became spatially embedded in American communities. Though redlining was eventually outlawed, its effects on housing quality, infrastructure, and access to resources persist in many places today.
Looking Ahead
The researchers suggest that measures of racialized economic concentration, such as ICE, could complement existing tools used in hazard planning and emergency management to better identify communities where storm exposure is most likely to translate into harm. They call for future research at finer geographic scales and across specific storm types to further clarify how structural inequality shapes disaster outcomes.
“We cannot control where the next storm will strike, but we can change how prepared communities are when it does,” says co-author Jason Cheung, a PhD student at the University of Chicago. “Disaster preparedness should not begin with the forecast — it should begin long before, by investing in communities facing the greatest structural disadvantages. Our study takes seriously the idea that climate vulnerability is, in part, socially produced and therefore socially reversible.”