Disability and Climate Vulnerability: What Different Sides Are Protecting

When Hurricane Katrina made landfall in August 2005, the failures of evacuation were documented in almost every dimension — the failure of levees, the failure of federal response, the failure of the Superdome as a refuge. Less examined was the specific geography of who died. A study published in The American Journal of Public Health in 2006 found that people with disabilities were two to four times more likely to die in Katrina than non-disabled people. Many could not drive, could not walk to higher ground, could not hear emergency warnings, could not navigate the debris fields that blocked wheelchair routes. Nursing home residents in St. Bernard Parish drowned in their beds when the facility did not evacuate; their bodies were found floating in the rooms where they had been left. The facility's administrators were later charged with negligent homicide.

Katrina was not an anomaly. After Hurricane Maria devastated Puerto Rico in 2017, people dependent on powered medical equipment — ventilators, oxygen concentrators, dialysis machines — died during the months-long power outage. The official death toll from Maria was revised upward from 64 to an estimated 2,975 after a study by Harvard researchers, a revision that largely captured mortality from infrastructure collapse rather than direct storm impact. The people most likely to die in the weeks after landfall were those whose survival required the grid to function. During the February 2021 Texas winter storm, power-dependent Texans died at home while their neighbors without medical dependencies lost heat but survived. In the heat domes that have become recurrent features of Pacific Northwest and Great Plains summers, people with multiple sclerosis — whose symptoms worsen rapidly with heat — and people taking antipsychotic medications, which interfere with the body's ability to regulate temperature, have died in apartments without air conditioning that the building code had never required because the climate had never before made it necessary.

Roughly one in four adults in the United States has some form of disability. A significant share of that population faces elevated risk from the specific hazards that climate change is intensifying: heat, flooding, power outage, air quality degradation, and the evacuation failures that disaster events produce. Disability is no longer wholly absent from the federal adaptation architecture: FEMA now has a permanent Office of Disability Integration and Coordination, a 2024 planning toolkit for people with disabilities and others with access and functional needs, and a 2024 third edition of the National Disaster Recovery Framework that defines accessibility and access and functional needs more explicitly than older doctrine did. But these additions still do not guarantee disabled people or disability organizations a governing role in the local and state planning processes that decide evacuation design, shelter conditions, backup power strategy, housing recovery, and continuity of care. The gap between who bears the risk and who shapes the response is the terrain this map is covering.

What universal design advocates are protecting

The principle that disability-inclusive design is not a supplement to good emergency planning but its precondition. Universal design advocates argue that the word "accessibility" has been allowed to function as a synonym for "add-on" — something you do after the real design is finished, to accommodate a subset of users whose needs were not considered at the outset. The result, in emergency management, is a pattern that looks reasonable from the outside and fails catastrophically in practice: evacuation routes designed for ambulatory people, emergency shelters that claim ADA compliance while lacking accessible bathrooms or power strips for medical equipment, alert systems that deliver information through a single channel that hearing-impaired people cannot access. Universal design advocates are protecting the recognition that these are not edge cases being slightly underserved — they are design failures that kill people. Disability inclusion built into planning from the beginning produces better systems for everyone; retrofitted after the fact, it produces systems that pass audits and fail in the field.

The documented failure of the "Special Needs Registry" approach — and the need to replace it with community-embedded systems. Emergency management's primary tool for including disabled people in disaster response is the special needs registry: a database of people with access and functional needs who self-identify in advance and can then be targeted for assistance during an evacuation. The evidence on these registries is not encouraging. After Hurricane Rita in 2005, Texas attempted a mass evacuation in which the special needs registry was used to dispatch transit to listed individuals — an effort documented in subsequent analyses as having failed in coordination, communication, and execution, leaving many registry members unevacuated while the effort had consumed substantial resources. Researchers who have examined special needs registries across multiple jurisdictions find consistent problems: low enrollment rates (disabled people often don't know the registry exists, don't trust government databases, or have changed their contact information), coordination failures between registry managers and transit operators, and the fundamental problem that a registry entry cannot produce assistance when the roads are flooded, the phones are down, and the transit system is overwhelmed. Universal design advocates are protecting the recognition that a tool this comprehensively documented as failing is not a solution — it is a substitute for one, deployed to demonstrate responsiveness while relieving planners of the harder work of designing systems that don't require prior self-identification to function.

The argument that disabled people and disability organizations must be at the table when plans are made — not consulted afterward. Emergency management planning processes are dominated by emergency managers, public health officials, transportation agencies, and elected officials. Disability organizations are occasionally consulted after draft plans are written. The gap between "included in planning" and "consulted on the plan" is not procedural — it is the difference between designing for actual needs and designing for a population imagined by people who don't share those needs. The Disability Integration Act, introduced in multiple congressional sessions but never passed, would have required that people with disabilities be included in the planning, design, and implementation of disaster-related housing and services. Its failure to advance illustrates the extent to which the participation gap is not an oversight but a structural feature of how emergency management works: the people who run these systems are not accountable to the populations most at risk of their failure.

What emergency management incrementalists are protecting

The operational reality of emergency management under resource constraints — and the argument that incremental improvement of existing systems is more achievable than wholesale redesign. Emergency managers operate in environments of permanent resource scarcity and unpredictable demand spikes. A county emergency management director with a staff of three and a budget that has not increased in a decade is not indifferent to disability inclusion; she is managing the gap between mandated functions and available resources with whatever tools she has. Incrementalists are protecting the recognition that the alternative to an imperfect registry is often no systematic attempt at all — that the critiques of special needs registries, while accurate, do not automatically produce better-resourced alternatives, and that telling practitioners their tools are inadequate without providing replacement capacity is a form of policy advice that generates guilt without producing change. The relevant question is not only whether current systems are adequate (they are not) but what specific, fundable, implementable improvements are available given actual operational constraints.

The argument that the ADA already provides the legal framework, and that enforcement rather than new mandates is the gap to close. The Americans with Disabilities Act, the Rehabilitation Act, and FEMA's own guidance already require that emergency programs and activities be accessible and nondiscriminatory. National Council on Disability reports have documented, across multiple disaster cycles, that these requirements are routinely violated — shelters that are not wheelchair accessible, communications that are not provided in accessible formats, recovery programs whose application processes effectively exclude people with cognitive or communication disabilities. Incrementalists are protecting the recognition that the primary gap is not legal authority but enforcement: the Department of Justice and FEMA have rarely used available enforcement mechanisms against jurisdictions that violate disability access requirements in disaster contexts. The case for focusing on enforcement rather than new requirements is that it avoids the political resistance that new mandates generate, builds on legal frameworks that are already established, and addresses the implementation failure where it actually occurs.

The tension between comprehensive planning and operational speed — and the argument that some triage is inevitable when disaster strikes. Emergency response operates on timescales that planning processes cannot fully anticipate. The improvisation required when infrastructure is destroyed, communication systems are overloaded, and resources are simultaneously demanded in multiple locations means that even excellent pre-disaster planning will encounter situations where triage is inescapable. Incrementalists are protecting the recognition that triage itself is not evidence of discrimination — that emergency managers making real-time decisions with incomplete information and inadequate resources will sometimes fail to reach people who needed them, and that designing accountability systems for these failures is different from designing systems that prevent them. The distinction matters because some critiques of disability exclusion in disaster response conflate structural exclusion (the failure to plan for disabled people) with operational failure under conditions of overwhelming demand — different problems requiring different responses.

What disability justice advocates are protecting

The recognition that disability exclusion in disaster planning is not a technical oversight but a reflection of which lives are treated as worth planning for. Disability justice advocates, drawing on a tradition of analysis developed by scholars and organizers including Mia Mingus, Leah Lakshmi Piepzna-Samarasinha, and Sins Invalid, argue that the pattern of disability exclusion in emergency management is not a resource problem — it is a question of whose survival is treated as a default assumption by the systems designed to protect it. The implicit design subject of most disaster planning is a non-disabled, ambulatory adult with a car and access to real-time information through a smartphone or television. This is not a neutral choice. It is a choice that embeds an image of a full person — the person whose needs define the system — and then provides workarounds for everyone else. Disability justice advocates are protecting the recognition that workarounds are a form of condescension: they treat disabled people as exceptions to be accommodated rather than as members of the community whose lives the community is obligated to protect.

The intersection of disability with race, poverty, and housing instability — and the argument that climate disaster makes these compounding vulnerabilities visible rather than creating them. Disabled people are more likely to live in poverty, more likely to live in substandard housing, more likely to be renters without resources for home modification, and more likely to be in congregate settings — nursing homes, group homes, psychiatric facilities — whose institutional character makes disaster response both more essential and more difficult. Congregate settings have been sites of disproportionate disaster mortality across multiple events: nursing homes in Katrina, institutions in Maria, and, most visibly, long-term care facilities during COVID, where people with disabilities who had been institutionalized died at extraordinary rates when the virus entered settings designed for efficiency rather than infection control. Disability justice advocates are protecting the recognition that these intersecting vulnerabilities are not accidents — they are the product of policies that have systematically concentrated disabled people in settings and circumstances that compound their exposure to harm. Climate disaster reveals the structure of that vulnerability; it does not create it.

The critique of "crisis standards of care" — and the argument that protocols designed for resource scarcity have used disability status as a triage criterion in ways that violate civil rights law. During the COVID pandemic, most states developed crisis standards of care protocols — frameworks for how to allocate ventilators and ICU beds when demand exceeded supply. Multiple early protocols included "short-term survival" or "long-term survival" criteria that effectively deprioritized people with preexisting disabilities on the grounds that their life expectancy or likelihood of recovery was lower. Disability rights organizations filed complaints with the Office for Civil Rights at the Department of Health and Human Services, arguing that these protocols violated the Rehabilitation Act and the ACA's prohibition on discrimination on the basis of disability. OCR issued guidance in 2020 affirming that crisis standards of care could not use disability or chronic illness as a criterion for deprioritization. But the underlying question — whether scarcity conditions permit disability status to be used as a triage factor — is unresolved in law and will recur in future mass casualty events. Disability justice advocates are protecting the recognition that the "who do we save first" question in disaster medicine is not a purely medical question. It is a civil rights question, and the answers embedded in protocols drafted before the crisis tend to reflect the values of the people who drafted them, who are rarely disabled themselves.

What fiscal realists are protecting

The resource constraint argument — and the specific claim that universal design mandates impose costs on jurisdictions that are already unable to fund basic emergency management functions. The United States has over three thousand counties and county-equivalents. Emergency management capacity varies enormously — from major metropolitan areas with professional staffs and dedicated preparedness budgets to rural counties where emergency management is a part-time responsibility held by someone who also has other duties. Unfunded mandates for disability inclusion — requirements to modify shelter facilities, deploy accessible communications systems, maintain functional needs support services — impose real costs on jurisdictions that may lack the capacity to meet them. Fiscal realists are protecting the recognition that mandating without funding produces the worst outcome: formal compliance (paperwork, registered systems, posted notices) without operational capability. The argument is not that disability inclusion is unimportant. It is that the choice between a funded, achievable standard and an unfunded mandate that produces nominal compliance without real protection is not obvious — and that advocates who focus on mandate expansion without addressing the funding question are optimizing for legal coverage rather than actual outcomes.

The triage argument — and the claim that the question of who gets priority in resource-constrained emergency response cannot be avoided by refusing to answer it. Every emergency response involves triage, explicit or implicit. The question is not whether to triage but who designs the criteria and on what basis. Fiscal realists are protecting the recognition that insisting on equal priority for all individuals regardless of the resources required to protect them — in contexts where resources are genuinely limited — is not a policy. It is a refusal to engage with the resource allocation problem that every emergency manager actually faces. This does not settle the question of which triage criteria are legitimate (using disability status as a proxy for social worth is not the same as prioritizing the most immediately life-threatened regardless of disability status). But it argues that the debate about disability and disaster cannot be resolved by insisting that resource constraints should not affect how responses are designed.

What the argument is actually about

At the center of this debate is a question that emergency management has been allowed to treat as technical: whose survival is the system designed for? The disability and climate vulnerability debate is not primarily a debate about evacuation logistics or shelter design, though it includes both. It is a debate about who counts as a default member of the community — whose needs are the baseline from which the system is designed, and whose needs are the exceptions that require workarounds. Emergency systems designed primarily for non-disabled ambulatory people with cars and smartphones are not neutral. They encode an answer to the question of who counts, and that answer has consequences that become visible when the system is tested by a disaster.

The disability rights tradition since the ADA has insisted on a specific principle: accommodation is not charity. The obligation to include disabled people in the life of the community — in schools, workplaces, public spaces, and civic processes — flows from the recognition that exclusion is a choice, not an inevitability. The disability and climate vulnerability debate asks whether that principle extends to the most consequential civic processes of all: the ones that determine who survives when the infrastructure fails.

The structural challenge is that climate change is intensifying the hazards while the political conditions for addressing disability inclusion in emergency management remain thin. FEMA is underfunded relative to its mandate. Emergency management is a low-salience political issue between disasters. Disability advocates are a mobilized constituency but not a dominant one in the arenas where emergency management budgets are set. The result is a pattern of documented failure, incremental improvement, documented failure again — with each disaster producing reports, guidance, and promises that the next cycle only partially implements. The question this map cannot answer — but must name — is whether that cycle constitutes a political system working approximately as well as political systems do, or a structural failure to treat a significant portion of the population as full members of the political community whose survival is the system's responsibility.

The people who died in Katrina's nursing homes, in Maria's extended power outage, in Texas's grid failure, were not statistical anomalies. They were the predictable consequence of systems designed around an implicit image of who a full person is — mobile, non-medically-dependent, connected. Climate change is making extreme events more frequent and more severe, which means the systems designed around that implicit image will be tested more often. Each test produces documentation of who was left behind. The question this debate is ultimately asking is whether that documentation will eventually change the design of the system, or whether the design will remain intact and the documentation will continue to accumulate.