Research by Hugi Hernandez and Egreenews Staff 

Extreme heat is the deadliest weather phenomenon in the United States. It kills more people annually than floods, hurricanes, and tornadoes combined. Yet the institutional machinery for responding to it remains fractured, underfunded, and in many jurisdictions, nonexistent. A 2025 GAO report found that between 2018 and 2024, 97 percent of counties across the contiguous U.S. were projected to reach dangerous heat levels, and more than 319 million people lived under such forecasts for at least one day. At the same time, less than 1 percent of FEMA’s primary hazard mitigation grant projects primarily addressed extreme heat.

The public health consequences of this gap are measurable. The CDC estimates 702 heat-associated deaths and 67,525 emergency department visits in the U.S. each year. Those numbers, drawn from death certificates, capture only a fraction of the true toll. Kristie Ebi, an epidemiologist at the University of Washington, has stated that the CDC’s annual estimate of roughly 1,200 heat deaths “is probably at least a tenfold undercount,” with the actual number likely in the range of 10,000 to 12,000 deaths. The gap exists because heat exacerbates cardiovascular, respiratory, and renal conditions that are recorded as the primary cause of death on death certificates.

This report examines the four deepest fissures in the institutional response to heat-related deaths: weak early-warning systems, inconsistent emergency coordination, limited outreach to vulnerable populations, and delayed policy action. These gaps are not abstractions. They are operational failures with body counts. The analysis draws on evidence from 35 cities, 35 counties, and four U.S. territories where heat is a chronic threat and where government action—or its absence—has shaped outcomes in ways that are increasingly well-documented.

Early-Warning Systems—Forecasts Without Follow-Through

The United States has invested heavily in forecasting heat. NOAA issues outlooks for excessive heat eight to 14 days in advance, advisories and warnings when dangerous conditions become imminent, and in 2024 launched the HeatRisk tool—a collaborative effort with the CDC that combines weather forecasts with health data on a color-coded, five-level scale. By May 2025, the CDC had published evidence that HeatRisk forecasts were associated with higher rates of heat-related illness emergency department visits, supporting the tool’s use for risk communication.

The problem is not the quality of the forecasts. It is what happens after they are issued.

A 2021 study published by researchers at the University of Arizona examined how decision-makers determine when and how to issue a heat warning in Phoenix, Detroit, New York, and Philadelphia. The study documented significant variation in the scientific evidence, expert judgments, and procedural triggers used across jurisdictions. The decision to activate a heat health warning system and deploy public health interventions was not standardized; it depended on local capacity, institutional history, and political will.

The gap between a warning and a response can be lethal. NOAA’s own data show that heat-related impacts occur at temperature levels below traditional warning thresholds, illustrating the importance of having a multi-tiered alert system rather than a binary on/off product. Yet for most of the locations examined in this report, the institutional pathway from a HeatRisk magenta alert to the opening of cooling centers, the deployment of outreach teams, and the activation of emergency medical protocols is informal, ad hoc, or nonexistent.

In Yuma County, Arizona—one of the hottest inhabited places in the United States, which recorded 148 days above 100°F in 2020—a CDC study found that 44 percent of surveyed older adults reported recent heat-related illness symptoms, and 18 percent said that electricity cost always or sometimes constrained their use of air conditioning. Barriers to cooling center access among older adults included awareness of location and transportation. The early-warning system existed. The connective tissue between the alert and the cooled space did not.

In New Orleans, a comparative case study by Johns Hopkins University documented that the city had prioritized the installation of backup energy generation following mass blackouts caused by Hurricane Ida in 2021, which left at least 10 residents dead from excessive heat exposure. The study found that the city’s heat policy framework remained underdeveloped compared with its hurricane preparedness infrastructure, even as extreme heat days multiplied.

The Randazza et al. content analysis of 21 local U.S. heat action plans, published in the *American Journal of Public Health* in 2023, found that 100 percent of plans incorporated activation triggers and risk communication, and 95 percent included outreach to at-risk populations. Gaps existed in the specific applications of these broad strategies. Plans often named vulnerable groups without detailing how outreach would be conducted, who would conduct it, or how effectiveness would be measured.

Emergency Coordination—The Nobody’s-in-Charge Problem

When a hurricane makes landfall, the Stafford Act triggers a federal disaster declaration, FEMA coordinates with state emergency management agencies, and resources flow through established channels. Heat has no equivalent architecture.

The GAO reported in 2025 that there has never been a presidentially declared major disaster for an extreme heat event, which would trigger federal assistance for infrastructure repair, emergency protective measures, and mitigation. According to FEMA, past extreme heat events have caused little infrastructure damage—a key criterion for approving federal assistance. FEMA officials told the GAO that, absent extraordinary circumstances, it was unlikely a president would ever declare a major disaster for extreme heat.

A 2026 study from the University of Kansas examined this coordination vacuum and concluded that the lack of a nationwide framework for heat-related emergencies, combined with a shortage of reliable data, complicated efforts to address heat conditions potentially more lethal than hurricanes and tornadoes. Co-author Nathaniel Brunsell stated: “Responsibility varies widely depending on location, with no standard assignment to federal agencies, state governments or local municipalities.” Cities are generally responsible for developing heat-action plans, but many rely on existing infrastructure and lack the resources to fully implement comprehensive plans.

> “Effective heat response does not necessarily require federal mandates, but it does depend on consistency and adequate funding. Local governments are often best positioned to implement solutions, but they frequently lack the resources to do so.” — Nathaniel Brunsell, Director of the Environmental Studies Program, University of Kansas

The consequences of this coordination gap play out jurisdiction by jurisdiction. A qualitative study published in 2025 in *Environmental Research: Health* conducted focus groups with 17 representatives from U.S. local health jurisdictions. Participants identified four factors that influence extreme heat preparedness and response: local conditions (environmental, political, planning); engaging communities and tailoring strategies; partnerships and relational connections; and available resources. The study emphasized the need for activities to be targeted and scaled to the unique climate, population, and needs of each jurisdiction—which is precisely what a fragmented system cannot reliably deliver.

In Harris County, Texas, the coordination gap is compounded by policy contradiction. In June 2023, as temperatures reached triple digits, Governor Greg Abbott signed House Bill 2127, which overrode local ordinances requiring construction workers to receive 10-minute water breaks for every four hours of work. The county commissioners court responded with a resolution supporting workers’ rights to organize, and the City of Houston partnered with Reliant Energy for the “Beat the Heat” program, distributing portable air conditioning units to seniors. These were reactive, piecemeal measures. No countywide heat surveillance system comparable to Maricopa County’s was in place.

The CDC published a summary of evidence on heat response plans in 2020, noting that there is evidence heat response plans can protect health, but mixed evidence on the effectiveness of individual components and the degree of overall health protection. The document was intended to help health departments develop coordinated plans. Four years later, the CDC established a collaborative workgroup in March 2024 for federal, state, and local analysts to discuss heat-related illness data—a recognition that the coordination problem remained unsolved.

 4. Theme Three: Vulnerable Populations—Outreach That Misses the Mark

The populations most likely to die from extreme heat are well-documented: older adults, people experiencing homelessness, outdoor workers, those with chronic conditions, and people without access to air conditioning. The gap lies not in identifying who is vulnerable but in reaching them with effective interventions before they become statistics.

A 2023 study published in the *American Journal of Public Health* examined the placement of 1,402 cooling centers across 81 U.S. cities and found that access to cooling centers differed dramatically among cities, ranging from 0.01 percent of the population in Atlanta, Georgia to 63.2 percent in Washington, DC. On average, cooling centers were located in areas with higher social vulnerability, but access areas were less inclusive of adults aged 65 and older than of younger populations. The study concluded that the current distribution of centers “may be insufficient to protect individuals from the adverse health effects of extreme heat” in the absence of additional risk-reduction measures.

The CDC’s own analysis of Maricopa and Yuma counties from 2010 to 2020 confirmed this pattern. Adults aged 65 and older had higher rates of heat-related hospitalization than those under 65. In a survey of 39 older adults in Yuma County, the barriers to cooling center access were straightforward: they did not know where the centers were, and they had no way to get there.

In Orange County, Florida, a quality improvement initiative called Cool & Connected OC documented that approximately 4 percent of heat-related emergency department visits among older adults originated from residents of small board-and-care homes in heat island zones such as Santa Ana, Anaheim, and Garden Grove. These facilities lacked structured emergency protocols, adequate cooling infrastructure, and culturally appropriate education.

In Puerto Rico, a 2025 study published in the *International Journal of Environmental Research and Public Health* surveyed 500 adults and found that while most participants expressed concern about climate change and high temperatures, fewer than half perceived heat as a high level of personal health risk. Older adults aged 65 and above, despite their higher level of physiological vulnerability, reported lower levels of risk perception and fewer symptoms. Nighttime heat exposure was widespread and strongly associated with heat-related symptoms. Common coping strategies included fans and air conditioning, but economic constraints and infrastructure instability limited access. The study concluded that public health strategies should focus on risk communication tailored to vulnerable groups and address barriers to heat adaptation.

In El Paso, Texas, a UTEP thesis documented that the city experienced its hottest summer in recorded history in 2023: 67 days at or above 100°F, with 44 consecutive such days. The research found that housing inequality, historical redlining, and thermal inequity intersect to concentrate heat risk in majority-Hispanic neighborhoods. The city and the National Weather Service were aware of vulnerable populations and considered them when preparing messaging, but the research identified gaps in the translation of awareness into sustained, resourced outreach programs.

In Los Angeles, the gap between infrastructure and equity is stark. UCLA researchers documented that complex street tree permitting processes have limited the city’s ability to plant trees at the scale needed to meet shade goals, particularly in underserved communities. The findings showed that delays, duplicative inspections, inconsistent procedures, and resource constraints leave trees unplanted and funding unused. A USC study found that South Los Angeles communities average roughly 13 percent tree canopy coverage, with some areas falling as low as 5 percent, compared with the city average of 21 percent. These are not accidental disparities; they are the product of decades of land-use decisions that are now being confronted as heat emergencies multiply.

Policy Action Delayed—The Cost of Waiting

The most consequential gap in the institutional response to heat-related deaths is the one that enables all the others: the systematic delay in adopting policies that evidence suggests would reduce mortality.

At the federal level, OSHA has no standard regulating heat stress hazards in the workplace. Five states have issued their own heat injury and illness prevention regulations. OSHA was first petitioned for a heat stress standard by Public Citizen in 2011. The agency denied the petition in 2012, was petitioned again in 2018 and 2019, and began rulemaking under the Biden administration in 2021. The proposed standard would require employers to provide workers with water and shaded rest areas once temperatures reach 80°F and mandatory 15-minute rest breaks at least every two hours. As of mid-2025, no final rule had been issued.

The Duke University Heat Policy Innovation Hub published a multistate review of heat-related legislation in 11 Southeastern states in 2025. The findings were unambiguous: no Southeastern state has enacted comprehensive safeguards for heat-exposed workers. Some states, such as Florida, have moved in the opposite direction, preempting local governments from passing workplace heat standards. Only a few Southeastern states have moved to regulate utility disconnection during periods of extreme heat. Between 2014 and 2024, Louisiana demonstrated the highest volume and success rate of extreme heat-related legislation, with 14 of 19 introduced bills enacted, but the focus was concentrated in health and education sectors rather than worker protection.

In Florida, the legislative gap is compounded by the state’s climate exposure. The University of South Florida has partnered with the City of Tampa to develop a Heat Resilience Playbook after research found that since 2016, Tampa’s heat index has consistently exceeded 100°F for more than 45 days per year. In 2023, Tampa experienced the hottest July on record with an average high of 93.3°F and an average low of 79.7°F, minimizing the city’s ability to cool off overnight. The Heat Vulnerability Index developed by USF researchers identified a strong correlation between lower-income neighborhoods and higher temperatures. Yet at the state level, comprehensive heat legislation has not materialized.

In Texas, the policy environment is actively contradictory. While the City of Houston distributes portable air conditioning units to seniors through the Beat the Heat program, state law has removed local authority to mandate water breaks for construction workers. The Hobby School of Public Affairs at the University of Houston documented that Texas has the highest number of workers dying from high temperatures, while noting that heat-related deaths are frequently recorded under a different primary cause of injury, making the problem difficult to quantify for policy purposes.

The FEMA gap is particularly revealing. As the GAO documented, FEMA has not evaluated its role in helping tribal, state, and local governments plan for and implement activities that reduce or mitigate future disaster losses from extreme heat events. FEMA has also not assessed how its potential decision to end the BRIC grant program may affect its ability to assist these entities. The GAO made four recommendations, including that FEMA evaluate its role, capabilities, and address any program gaps to assist tribal, state, and local governments in addressing extreme heat events.

At the territorial level, the data gaps are even more fundamental. A 2024 House subcommittee hearing on insular areas documented major gaps in federal data collection for American Samoa, Guam, the Northern Mariana Islands, Puerto Rico, and the U.S. Virgin Islands. Federal statistical products either exclude the territories entirely, or data may be outdated or inconsistently reported. The GAO discovered significant gaps in existing data, primarily because federal agencies do not include the U.S. territories in many statistical products. For Guam and the U.S. Virgin Islands, no location-specific heat mortality data were identified within the date range.

Institutional Capacity vs. On-the-Ground Reality

The contrast between formal institutional capacity and actual outcomes is sharpest where data are richest. Maricopa County, which operates the most advanced heat mortality surveillance system in the country, has also experienced the highest documented death counts. The county’s surveillance catches approximately half of excess heat-related mortality—significantly better than the national rate of 1 to 10 percent, as estimated by David Hondula, director of the Phoenix Office of Heat Response and Mitigation. Phoenix reduced heat-related deaths for two consecutive years through data-driven targeting, with approximately half of all heat deaths occurring among unsheltered residents who make up less than 1 percent of the population. Yet even with this targeted approach, Hondula acknowledged that the city is working with three dissonant data streams: heat-related deaths decreased by approximately 30 percent in one period, while heat-related 911 calls increased approximately 30 percent, and calls coded as heat-related by emergency dispatchers decreased 25 to 30 percent.

The CDC’s Heat & Health Tracker, launched to provide local heat and health information for community preparedness, represents an attempt to bridge the gap between federal data infrastructure and local decision-making. The NSSP’s emergency department data associated with heat-related illness are integrated into the tracker’s dashboards, designed to be accessible to “everyone from grassroots organizers to high-level decision-makers.” Whether this accessibility translates into changed local practice is an open question.

For most locations in this report’s scope—including Corpus Christi, Henderson, Lakeland, Albany, Suffolk, and the territories—verifiable data on heat deaths and government response are either absent or available only at the state level of aggregation. The absence of data is itself an institutional failure. Without mortality surveillance, there is no feedback loop between policy choices and measurable outcomes.

The Periphery and the Center: Yuma and Phoenix

The distance between Yuma and Phoenix is roughly 180 miles. The distance between their institutional capacities for responding to extreme heat is far greater.

Yuma County recorded 148 days above 100°F in 2020—more than Phoenix that year. The county has a significant agricultural workforce that labors outdoors in conditions that can exceed 115°F. The CDC’s analysis of Yuma County found that older adults had higher rates of heat-related hospitalization, that many reported electricity costs constrained their air conditioning use, and that cooling center access was limited by awareness and transportation barriers.

Phoenix, by contrast, has a dedicated Office of Heat Response and Mitigation, an expanding network of cooling and respite centers that served over 35,000 visitors in 2024, and a data infrastructure that allows the city to track heat deaths, 911 calls, and emergency department visits in near real-time. The city has reduced heat mortality for two consecutive years.

The disparity is not explained by temperature differences. It is a function of institutional investment. Yuma County lacks the population base, tax revenue, and public health staffing to replicate Maricopa County’s surveillance system. The federal programs that could close this gap—FEMA mitigation grants, CDC surveillance assistance, NOAA technical support—have not been systematically directed toward smaller, hotter, poorer jurisdictions.

This pattern recurs across the locations studied. Ventura County has developed targeted farmworker outreach with indigenous language communication, supported by EPA recognition. Fresno County, with a larger agricultural workforce and similar heat exposure, has no verifiable equivalent. El Paso has academic research documenting thermal inequity along historical redlining patterns but has not translated that research into a formal heat action plan with dedicated funding. Charlotte-Mecklenburg has conducted urban heat island mapping through a community-academic coalition, but a formal, adopted heat action plan was not identified.

The data suggest that heat preparedness in the United States is a function of local capacity, not of need. The jurisdictions with the highest heat exposure are not necessarily the ones with the strongest institutional response. The jurisdictions with the strongest institutional response tend to be large, wealthy, and politically liberal. This is not a formula for equitable protection.

What Is Known, What Is Not, and What Comes Next

The evidence base supports several conclusions with reasonable confidence. Heat-related deaths are substantially undercounted in every jurisdiction examined, with the true burden likely ten times higher than death certificate data suggest. Early-warning systems have improved technically, but the institutional pathway from forecast to protective action is fractured and under-resourced. Emergency coordination for heat events lacks the legal triggers, dedicated funding streams, and interagency protocols that exist for other disasters. Outreach to vulnerable populations is named in nearly every heat action plan reviewed but is rarely specified with the operational detail necessary for implementation. Policy action at the state and federal levels has been characterized by delay, preemption, and underinvestment.

What remains uncertain is substantial. The data on whether cooling centers reduce mortality, whether heat action plans improve outcomes independent of other investments, or whether specific outreach modalities change behavior are limited. The CDC’s 2020 review noted mixed evidence on the effectiveness of individual heat response plan components. The GAO has identified FEMA’s unevaluated role as a critical gap. The territories remain largely invisible in federal heat mortality data.

For many locations in this report’s scope—Corpus Christi, Jacksonville, Lakeland, Albany, Suffolk, Ventura, Fresno, Bexar County, Shelby County, Mecklenburg County, Guam, the U.S. Virgin Islands, and others—the most important finding is the absence of verifiable data. No surveillance means no measurement. No measurement means no accountability. No accountability means no sustained improvement.

Questions for Further Research

1. What is the minimum viable surveillance infrastructure—in terms of staffing, training, and technology—required for a county of 100,000 to 500,000 residents to produce reliable heat mortality estimates within a single warm season?

2. To what extent do state-level preemption laws that prohibit local heat ordinances affect heat-related mortality among outdoor workers, and can the effect be isolated from confounding variables such as baseline temperature trends?

3. What mechanisms would enable FEMA to declare extreme heat a major disaster under the Stafford Act without legislative amendment, and what would the operational consequences of such a declaration be for local emergency management agencies?

Key Takeaways

1. The CDC’s estimate of roughly 1,200 heat deaths annually is a tenfold undercount of the true mortality burden, according to epidemiological excess-death analyses, which means most jurisdictions are making policy decisions with systematically incomplete data.

2. FEMA has never declared a major disaster for extreme heat, and less than 1 percent of its primary hazard mitigation grants have addressed heat, leaving local governments without the federal funding architecture that exists for floods, hurricanes, and tornadoes.

3. Cooling center access in 81 U.S. cities ranged from 0.01 percent of the population in Atlanta to 63.2 percent in Washington, DC, with older adults consistently underrepresented in access areas despite being the highest-risk demographic group.

No Southeastern state has enacted comprehensive safeguards for heat-exposed workers, and some states, including Florida, have preempted local governments from passing workplace heat standards.

Evidence suggests that state legislatures in jurisdictions where average July high temperatures exceed 95°F may consider establishing a statutory requirement for county-level heat mortality surveillance using a standardized classification protocol, with data reported weekly during the warm season to a publicly accessible dashboard. Technical assistance and partial funding could be provided through an expansion of the CDC’s Climate and Health Program, modeled on Maricopa County’s enhanced surveillance system. The absence of surveillance data in most of the jurisdictions examined is not a data gap—it is a governance gap with lethal consequences.

#HeatMortality #EarlyWarningSystems #EmergencyCoordination #ClimatePolicy #PublicHealthSurveillance