No one in eastern Kentucky could remember rain as intense as what fell in July 2022. In just five days, more than 14 inches of rain inundated the region's rural counties. On the final day of the deluge, there was too much water for the ground to soak up any more. Flash floods tore through towns, washing away roads, homes and entire neighborhoods.
The floods were deadly. But how deadly, exactly? There are still multiple government death counts for an event that upended the lives of thousands of people. The state of Kentucky determined that 45 people died from the flooding. But the National Weather Service counted 40 deaths. And preliminary data from the Centers for Disease Control and Prevention (CDC) shows only 36 people died.
The disparate death counts in Kentucky are part of a long-standing problem: Despite the growing danger from climate-driven disasters, there is no single, reliable count of who is dying as a result of extreme weather in the United States. For any given weather disaster, multiple government agencies publish independent — and often widely differing — death counts.
The definitive federal accounting of climate change's impacts in the United States, the National Climate Assessment, estimates that upward of 1,300 people die in the U.S. each year due to heat alone and that extreme floods, hurricanes and wildfires routinely kill hundreds more. But those numbers are rough estimates.
That's a problem, the federal government has long acknowledged, because who dies as a result of extreme weather, as well as how they die, is important. That public health information can help protect people from increasingly frequent disasters and can even spur policies that address the reliance on fossil fuels at the root of global warming. And inconsistency over which disaster-related deaths get counted can lead to frustration and even financial losses for the families of those who died.
"The data collection needs to be better," says Samantha Montano, a disaster researcher at the Massachusetts Maritime Academy. "There needs to be a national, publicly available database that everybody has access to that is tracking every single death."
One disaster, multiple death counts
It's unclear how many people in the U.S. officially died in some of the most high-profile and deadly climate-related weather disasters in recent years.
The issue burst into public view after Hurricane Maria hit Puerto Rico in 2017. The official death toll was in the dozens, but survivors and local officials on the ground questioned whether that was an accurate count, in part because reliable electricity wasn't restored on the island for months.
Epidemiologists stepped in and used statistics to compare the number of deaths in the months after the hurricane with the number of deaths during similar periods in previous years when there was no storm. They estimated that the actual death toll was likely much higher. "We went from the federal government saying 89 [people died] to another academic institution saying 2,000 and yet another saying 5,000," says Maureen Lichtveld, an epidemiologist at the University of Pittsburgh. Ultimately, the government of Puerto Rico reported an official death toll of just under 3,000.
Lichtveld is one of the authors of a congressionally mandated 2020 report from the National Academies of Sciences, Engineering, and Medicine about disaster death tolls, written partly in response to public frustration after Hurricane Maria. That report called on the federal government to standardize how deaths are counted after weather disasters. "Significant confusion and disagreement persist," the report notes, "regarding what counts as a disaster-related death."
The CDC is the federal agency officially responsible for mortality statistics for the U.S., including weather-related fatalities. The agency declined to make any of its experts on disaster-related mortality available for an interview, but acknowledged that deaths from weather disasters are potentially being undercounted because of inconsistent information on death certificates, according to a CDC spokesperson who answered NPR questions on background via email.
The CDC says it does not have any estimates of how large the potential undercount might be for different types of disasters, and doesn't have plans to update its guidance for how local officials document disaster-related deaths.
The lack of reliable data is a problem, says Rep. Nydia Velázquez, D-N.Y., who sponsored the bill that led to the National Academies report, along with then-Sen. Kamala Harris. "Death tolls are important," Velázquez says. "They influence public perception about the scope of a disaster and often determine what federal resources are allocated in response."
Despite the congressional attention and the subsequent recommendations from the National Academies, little has changed. Disparities and confusion persist over official death tolls from climate-driven extreme weather.
Hurricane Florence killed 45 people in North Carolina in 2018, according to the state's Department of Public Safety, but the National Hurricane Center reported 40 fatalities from the storm.
The death count disparities from heat waves are particularly large. Heat waves are by far the deadliest type of climate-related extreme weather and are also the most closely linked to climate change. The most extreme heat waves today would be impossible without human-caused warming.
But it's unclear how many people are dying in heat waves in the United States. In 2022, the Texas Department of State Health Services reported 279 heat-related deaths in the state, while the National Weather Service counted just 53 deaths in Texas. That discrepancy is likely because the National Weather Service counts only heat-related fatalities that occur on days hot enough to warrant an official heat advisory.
"You could still have heat-related deaths when the temperature is 95 degrees in Texas," says Gordon Strassberg, the storm data program manager for the National Weather Service. But such weather might not be hot enough to trigger an official heat advisory in a state where very hot weather is common. In that case, the National Weather Service wouldn't count the fatality, but the state government would.
A year earlier, in 2021, the Washington State Department of Health counted 100 deaths during the height of a record-shattering heat dome in the Pacific Northwest, while the National Weather Service reported just seven fatalities. The National Weather Service didn't have access to complete fatality data from the state of Washington, Strassberg says.
Amid concerns that many heat-related deaths might have been miscategorized on death certificates, Washington state epidemiologists used statistical methods to estimate how many extra deaths occurred compared with the same time period in previous years in the area. Those officials found that a more complete death toll for the heat dome is likely closer to 1,000 people.
Some wildfire-related deaths also go uncounted. Research shows that exposure to wildfire smoke contributes to thousands of deaths in the U.S. each year, but there is no national system for counting such deaths. And while the official number of people killed in the 2018 wildfire that destroyed much of Paradise, Calif., is between 84 and 86, that doesn't include dozens of suicide deaths that have been linked to despair and displacement in the aftermath of the fire, none of which are counted by any state or federal agency.
"We know like after a lot of [extreme weather] events that there's an increase in suicide, but there's not a good accounting of that and numbers that we can trust," Montano explains.
Agencies don't agree on which deaths to count
Official death counts vary mainly because it isn’t always obvious which deaths should be attributed to a weather disaster.
Direct deaths are the easiest to count. For example, if a tree falls on someone and kills the person during a hurricane, that individual's death is directly attributable to the storm, Lichtveld explains.
But deaths that are indirectly tied to extreme weather aren't counted as reliably. For example, many people die because of power outages during or after intense hurricanes, wildfires and heat waves. "If there is a disaster and there is no electricity and people who need dialysis can no longer get dialysis, so they go into kidney failure, that's an indirect [death]," Lichtveld says.
There are no standardized criteria to confirm a link between a death that lags an extreme weather event and the disaster itself, and it isn’t always clear where to draw the line. Many people have underlying health conditions, such as heart disease, that are exacerbated by loss of electricity, lack of access to medication or stress from being displaced by a disaster, leading them to die weeks or even months later.
In such cases, the connection between the weather and that person's death is often missing from their official death certificate. Because death certificates are the underlying data for the CDC and most state agencies that keep track of death records, such deaths often go uncounted after disasters.
"There's no uniformity with the death certificate[s]," says Kathryn Pinneri, the former president of the National Association of Medical Examiners. Every state gets to decide which data it gathers about weather-related fatalities. "It really is going to vary among jurisdictions."
This can lead to disparate official death counts. For instance, an agency like a state health department might track deaths that lag the weather event for a longer period or might have more expansive criteria for which deaths to count. In the end, that agency could report a higher death toll than agencies that stop tracking deaths sooner or that choose not to count deaths that are more tenuously connected to weather conditions.
"I think the lack of consistency in all the data reporting does cause a bit of confusion," says Strassberg, of the National Weather Service. He points out that some extreme weather events affect multiple states and that in some areas a single National Weather Service office will be responsible for parts of multiple states. Differences in how deaths are counted or reported can make it difficult to compare disaster impacts across states or over time.
Unlike the CDC, which is responsible for publicly reporting official U.S. death data for all types of fatalities, the National Weather Service gathers weather fatality data primarily for internal use, Strassberg says. For example, information about how many people died in a flood can help inform future flood warnings issued by the local National Weather Service office in that area.
"The fatality data we have is the best information available to our knowledge," he explains, and the numbers are widely used by academic researchers, local emergency managers and even insurance companies. But, Strassberg stresses, "our numbers are not official."
Better disaster death data can save lives
Reliable data about how many people died in a flood, wildfire, hurricane or heat wave, as well as why those deaths occurred, can help save lives during future extreme weather.
"Many of the deaths are avoidable," says Wayne Blanchard, who worked at the Federal Emergency Management Agency (FEMA) for three decades and is now retired. "Particularly for the emergency management community, the more they know, the more enabled they'll be to try to develop mitigation measures" such as evacuation plans, shelters or weather warnings that are tailored to those who are most at risk.
For example, in the 1990s some cities began opening cooling centers during heat waves in response to large death tolls from prolonged heat exposure, Blanchard says. Such cooling centers are now a basic part of managing heat waves across the United States. Today, many city governments are hungry for more information about who is dying from heat exposure, because it can help inform where to place those cooling centers and how to help vulnerable people access them.
And data about drowning deaths in vehicles during flash floods helped spur a federal safety campaign warning drivers "Turn Around Don't Drown" if there is water in the road.
Accurate death counts after disasters can also be painfully personal and even financially important for the families of those who died. The Federal Emergency Management Agency helps pay for funerals for those whose deaths are officially linked to major disasters. But if a death isn't counted as disaster-related, families generally aren't eligible for those relief funds.
The vast majority of applications for FEMA funeral assistance after recent hurricanes were denied, according to a 2019 report by the Government Accountability Office, in part because of missing or incomplete death certificates.
The lack of concrete mortality numbers related to climate change is increasingly problematic at a national and global scale as well. Policymakers around the world rely on data about the human cost of climate change to justify policies that would help curb warming. The higher the death toll, the greater the economic and moral impetus to reduce greenhouse gas emissions that cause climate change.
In a speech before the United Nations General Assembly in 2023, Secretary-General António Guterres invoked the approximately 1,700 people killed in climate-driven floods in Pakistan in 2022 as he argued that world leaders must spend more money to combat climate change. In 2021, a group of European teens joined climate protests in memory of a flash flood victim, drawing a direct line between that teen girl’s death and global climate policies. And senators from Vermont and Hawaii, states that experienced deadly floods and wildfires, respectively, in 2023, noted approximate death tolls from those disasters as they argued in May for long-term federal funding to help families and towns rebuild in more resilient ways.
The personal and the political stakes of accurate death counts were on display in forceful testimony by Rep. Jill Tokuda, D-Hawaii, in September 2023, nearly two months after wildfires killed at least 101 people in the city of Lahaina. "Far too many lives were lost in our tight-knit community," she told fellow lawmakers at a hearing about preventing future catastrophic burns by upgrading the electrical grid. "Some are still waiting anxiously for news of their loved ones. And while they want some kind of resolution, they fear that knock on the door."
"I want answers," Tokuda said. "How do we keep our families safe?"
Transcript
MICHEL MARTIN, HOST:
Every year, public health officials estimate that extreme weather events like wildfires and heat waves kill thousands of people in the U.S. But those numbers are just estimates. The exact death toll from climate-driven disasters is unknown, and it's likely higher than official counts suggest. NPR's climate team has been looking into this undercount more deeply, and reporter Rebecca Hersher is here with us to tell us more about it. Hello.
REBECCA HERSHER, BYLINE: Hi.
MARTIN: You know, this seems strange because we count everything, you know, in this country. So how is it that we don't know exactly how many people are dying in disasters? Doesn't the government keep track of who dies?
HERSHER: Yeah, it is a little strange. You know, the government does keep track of deaths in the U.S., you know, writ large. But it's not that simple for extreme weather. You know, there are multiple government agencies all counting the fatalities from storms and wildfires and heat waves. You have local health departments and emergency managers. You have the National Weather Service, and you have the U.S. Centers for Disease Control and Prevention - the CDC. And for a single weather disaster, like a flood or a hurricane, different agencies might actually come up with different numbers. Sometimes it's just a difference of a few people, but in many cases, the disparities between counts can be hundreds or even thousands.
MARTIN: So the death toll is different depending on who is counting or which agency is counting. Why is that?
HERSHER: So at the most basic level, it's not actually always obvious which deaths should be counted. And here I have to say, we are talking about ways that people actually die every year. I don't want this to be too theoretical. You know, these aren't just hypothetical examples. They represent real people who have lost their lives. So that said, imagine that there is a flash flood, and someone drowns in that flood. Was that death caused by the disaster?
MARTIN: Right. Why wouldn't it be?
HERSHER: Why wouldn't it be indeed. It's pretty clear-cut. That death was caused by the disaster, it's gonna be counted that way pretty much no matter who's doing the counting. But now let's imagine that that person has a neighbor, and that neighbor has a lung condition that requires an oxygen machine.
The flood knocks out the power for the machine for a couple of days. Neighbor initially survives, but the lack of oxygen makes their lung disease worse, and three weeks later they die. Some agencies would count that death. Others would not. And so you can see how you might end up with multiple death toll numbers. And for some types of climate-driven disasters, like heat waves, epidemiologists actually estimate that most deaths are not getting counted because they're like that.
MARTIN: How does digging into this more deeply, how does taking another look at death toll numbers help us going forward?
HERSHER: So disaster managers say that to save lives in the future - and again, climate change is causing more of these extreme events - you have to look at who died in past disasters and then do things to prevent those kinds of deaths from happening again. So for example, knowing that people die in heat waves can help you put cooling centers in the right places. So reliable numbers do save lives. And there's actually widespread acknowledgment among the people who do this work that the lack of reliable numbers is a problem.
MARTIN: So solutions, is there something that can be done to fix this?
HERSHER: You know, honestly, it's been a really tough nut to crack. In 2020, the National Academies put out a big report that called for national standards for exactly what we're talking about, you know, which deaths should be counted as disaster related. When we asked the CDC about this, they acknowledged that deaths from weather disasters are being undercounted and said that the agency is working to do more training for, you know, coroners and emergency managers. But there are no immediate plans to update the official CDC guidance about how to count these deaths. That means that we're headed into another deadly, hot, stormy, potentially wildfire-y summer here without that ability to reliably count how many people will die from that extreme weather.
MARTIN: That is NPR's Rebecca Hersher. Rebecca, thank you.
HERSHER: Thanks.
(SOUNDBITE OF EMAHOY TSEGUE-MARYAM GUEBROU'S "SONG OF ABAYI") Transcript provided by NPR, Copyright NPR.
300x250 Ad
300x250 Ad