A retired sheriff’s deputy who drowned when his truck was trapped in a raging river. A pair of firefighters who had been rushing to rescue someone when their engine was crushed by a falling tree. Two Georgians whose mobile home was flipped over by powerful winds. At least 147 fatalities have already been directly attributed to Hurricane Helene as southeastern states begin to dig out from the wreckage of one of this century’s worst storms.
By Sarah Kaplan
But Helene’s lasting damage to homes, health systems, economies and social networks could end up resulting in thousands of additional deaths in the coming years, new research suggests.
An analysis of more than 500 tropical cyclones that have hit the United States since 1930 found that the average hurricane leads to as many as 11,000 excess deaths — a figure hundreds of times higher than official mortality estimates. By combining weather records with Centers for Disease Control and Prevention mortality data from months before and after the storms struck, the study’s authors revealed how death rates in affected states remain elevated for 15 years after a storm makes landfall.
The research published Wednesday in the journal Nature is the first comprehensive assessment of what researchers call “indirect deaths” from hurricanes. Unlike the fatalities from turbulent floodwaters and collapsing homes, these deaths can occur years after a disaster. They are often associated with medical conditions — cancer, heart disease — that can be exacerbated by the stress and disruption of a storm. And they tend to strike those who are already most vulnerable in society: infants, the elderly, racial minorities.
In total, the study suggests, as many as 5.7 million Americans died in the aftermath of hurricanes between 1930 and 2015 — accounting for 3 to 5 percent of all deaths across the contiguous United States each year. The staggering toll points to major shortcomings in the country’s disaster-recovery systems that leave communities poorer, sicker and more isolated years after the floodwaters recede, the authors said.
“It shows we need to do a lot more to not just leave people out to dry,” said co-author Rachel Young, an environmental economist at the University of California at Berkeley. “These folks need support to rebuild their lives.”
The ‘only explanation’ for a striking pattern
The idea that storms cause a long-term spike in death rates comes as no surprise to W. Craig Fugate, a Florida native and veteran of too many hurricanes to count. In more than a decade leading Florida’s emergency management division and the Federal Emergency Management Agency, Fugate has seen how “the trauma is not over when you’re finished rebuilding.”
Fugate said he has observed firsthand how suicide rates increase after a disaster destroys livelihoods. City and state budgets get reorganized to pay for infrastructure repairs, leaving less money for school food programs and prenatal health clinics. Formerly close-knit communities become fractured, leaving vulnerable people with no one to care for them in times of crisis.
“It’s about time somebody put a study together,” Fugate said.
But the authors said they didn’t expect to uncover such a strong and lasting effect when they began this research several years ago. Co-author Solomon Hsiang, director of the Global Policy Laboratory at Stanford University, had published a previous study showing that cyclones significantly suppress economic growth, even in relatively wealthy nations such as the United States. Since research indicates poorer people tend to die younger, it seemed reasonable to predict that hurricanes would trigger a small increase in deaths lasting a few months after the storm.
Instead, Hsiang said, they found “one of the sharpest signals I’ve ever seen.” Death rates spiked immediately after a storm made landfall and continued to increase for an average of almost six years afterward. Cyclones with higher wind speeds led to larger increases in death rates. In states that are routinely hit by storms, including Florida, the Carolinas and Louisiana, hurricane-attributable mortality accounted for roughly 1 in 10 deaths that have occurred in the past century.
Because states are so frequently hit by new storms before they have recovered from the previous one, the researchers used a statistical technique called “deconvolution” to show how the death toll from each event stacks on top of the ones that came before. Hsiang compared the technique to identifying individual notes ringing from a bell that has been struck multiple times.
“Being able to pull out what each storm is doing allows us to explain the pattern that is affecting so many people,” he said.
Tatyana Deryugina, an environmental economist at the University of Illinois who was not involved in the Nature study, called the findings “very unexpected.”
“The common wisdom is that disasters cause a lot of death in poor countries, but in richer countries most of the damage is to capital,” she said. “This suggests that there’s some mechanism where the mortality impacts [in the United States] are growing over time.”
But Deryugina said there is nothing in the study’s methodology to make her question its findings. Because Hsiang and Young were so shocked by their own results, they spent months stress-testing their study to uncover spurious connections and analytical flaws. They compared states of varying sizes and demographics to make sure the changes in mortality rates they were seeing weren’t actually reflecting differences in income or race.
“We systematically were able to cross out other explanatory factors,” Young said. In the end, “hurricanes were the only thing that explained it.”
‘If the hurricane didn’t hit, they would still be alive’
The analytical methods used in the study don’t allow the researchers to pinpoint specific deaths that were caused by a disaster — or exactly why those deaths happened. But that doesn’t mean the link isn’t robust, Hsiang said. Similar techniques were used to initially identify the connection between tobacco use and mortality rates — a connection that was ultimately borne out by a huge body of medical research.
“We can’t say for sure whose death is caused by the hurricane,” Hsiang said. “But there are people who, if the hurricane didn’t hit, they would still be alive.”
The demographics of those who die help point to why this phenomenon occurs, the authors said. Nearly half of excess mortality after a hurricane is among people older than 65, and many of the deaths come from increased rates of cardiovascular disease and cancer.
These diseases are known to be exacerbated by stress, the researchers said — an indication that the anguish of surviving a storm and its aftermath could make people sicker. Severe storms can also change the demographics of communities, pushing working-age people to relocate in search of better opportunities. Elderly residents may die sooner without younger relatives and friends to care for them.
Older people are also more likely to have chronic illnesses that become difficult to manage in a hurricane’s aftermath. Andrew Schroeder, vice president for research and analysis at the medical charity Direct Relief, said someone with diabetes may have trouble maintaining a healthy diet if they have to spend their savings on home repairs. A kidney disease patient whose car gets damaged may be unable to reach their dialysis clinic.
The group most imperiled by hurricanes was infants, the Nature study found. Though children under age 1 make up roughly 1 percent of the U.S. population, they account for 14 percent of excess deaths after a hurricane — including babies who weren’t born yet when the disaster struck.
To Hsiang, this suggests that storms can irrevocably alter environments and economies of the places they hit, making these communities more dangerous for children. Mothers struggling with unemployment in a storm-battered economy might struggle to afford food and medical care for themselves and their babies, he said. A damaged hospital might put off upgrades to its obstetrics unit so it can pay for repairs.
This finding might also help explain why more babies die in southeastern states than the rest of the country, Hsiang added.
Regardless of age, members of marginalized communities face the highest increases in hurricane-related risk. For any given storm, the study shows, a Black victim is three times more likely to die than a White person who experienced the same disaster — a finding that highlights the economic and health disparities that affect Black Americans even before a hurricane strikes.
“One of the principles that we tend to abide by in health for emergency response [is that] the people that are most vulnerable prior to the disaster are most vulnerable during the disaster,” Schroeder said.
In his work providing medical assistance after disasters, Schroeder has seen how Black households are less likely to have insurance and more likely to be near facilities — such as factories and landfills — that release pollution during storms. Studies have also found that people of color are less able to access federal assistance when a catastrophe occurs.
But this unequal toll also hints at ways to insulate people against hurricanes’ long-term consequences, Hsiang said.
“Things like the gap between the Black and White experience of the storms points to the fact that there are some things we can do to protect populations,” Hsiang said. “The key is to figure out what those measures are so everybody can have access to them and maybe scale them up.”
A rising threat
The United States has dramatically improved its disaster-response systems, with far fewer people dying during hurricanes than did 100 years ago. But the researchers said their study points to significant shortfalls in support for communities dealing with the long-term aftermath — and the dire need for improvement as rising global temperatures fuel stronger, wetter storms.
The research shows that the number of indirect deaths from hurricanes has steadily increased since the 1930s — largely due to increases in the population of southeastern states. Since 2001, a spike in the number of storms making landfall has caused the hurricane-related death rate to accelerate further.
Climatological factors can explain only about 12 percent of the increase in excess mortality since 1930, according to the study.
“This is happening now, it was happening in the past,” Young said. “It’s not just a climate change phenomenon.”
Yet climate researchers warn that warmer oceans could lead to higher-intensity hurricanes making landfall in the United States — a trend that the Nature study indicates would lead to higher long-term death rates.
Other research suggests that rising temperatures could cause tropical cyclones to reach peak intensity at higher latitudes, leading to worse storms in states that are unaccustomed to dealing with them.
Yet there are few federal resources available to help communities cope with the aftermath of hurricanes, said Fugate, the former FEMA administrator. The agency’s mandate is to respond to the immediate emergency, and most of its support programs end within a few months of a disaster. And when FEMA faces a cash crunch — as it does this year — it diverts resources from longer-term relief and resilience projects to meet immediate needs.
Meanwhile, funding for state and local public health departments has fallen significantly. Supplemental Nutrition Assistance Program benefits haven’t kept up with the rising cost of living.
“The problem with Congress and federal funding is they’ll throw money in the acute phase, but they have not shown a lot of interest in the long term,” Fugate said. “What are we doing five years out, 10 years out, when the attention has moved away, yet people are still dealing with it?”
To Direct Relief’s Schroeder, the Nature study emphasizes the challenges facing communities in North Carolina and elsewhere as they grapple with the devastation Helene has wrought.
“The water will go down. You’ll repair the roofs. The pictures in the news will go away,” he said. “But the health problems will continue way beyond this.”
The connection between hurricanes and the threats of poverty, loneliness and disease can be hard to see, he acknowledged. But survivors lives’ may depend on governments and nonprofits taking these dangers as seriously as raging floodwaters and roaring winds.
“It’s the invisible stuff that’s most consequential,” Schroeder said. “We can’t let the victims of these kinds of disasters become invisible because we don’t connect them to the shock that affected their community.”
correction
A previous version of this article incorrectly said that a new analysis published in the journal Nature suggests that long-term deaths from hurricanes account for 3 to 5 percent of all deaths each year in the continental United States, which is made up of all states except Hawaii. The study examined data from the contiguous United States, which excludes both Alaska and Hawaii. The article has been corrected.
