Bioethics Forum Essay
When It Comes to Rationing, Disability Rights Law Prohibits More than Prejudice
This week, the Department of Health and Human Services Office of Civil Rights resolved one of many civil rights complaints filed in recent weeks alleging discrimination on the basis of disability–the first instance of federal intervention to enforce civil rights laws in rationing protocols since the beginning of the Covid-19 crisis.
The Office of Civil Rights required the State of Alabama to fully rescind guidance excluding people with intellectual disabilities, dementia, and others from ventilator access and commit to not “include similar provisions singling out certain disabilities for unfavorable treatment or use categorical age cutoffs” in future guidelines.
This represents a significant first step, alongside a March 28th bulletin indicating that care should not be denied on the basis of stereotypes, quality of life assessments, or judgments about a patient’s relative worth based on disability or age. The bulletin also required providers to make decisions about treatment based on “an individualized assessment of the patient based on the best available objective medical evidence.”
Though many state plans still have categorical exclusions based on particular disability diagnoses, the tide is clearly turning against their use. Recently issued state plans in Pennsylvania, Colorado, and Massachusetts have deliberately avoided categorical exclusions from care. But more work remains to be done. Soon, the fight to prevent disability discrimination in rationing will shift to less obvious forms of discrimination.
For many people with disabilities, an impairment is medically relevant, but access to scarce medical resources–like ventilators–still offers a reasonable prospect for recovery.
Should people with disabilities be penalized for impairments that may mean they require more resources to achieve the same result as people without disabilities? Should hospitals be able to consider long-term survival, potentially disadvantaging people with diagnoses that impact future life-expectancy but do not necessarily impact the likelihood of short-term survival? And if we accept that short-term survival is an appropriate consideration, do existing instruments need modifications for people with underlying impairments?
These questions are pressing and reflect the next step in the disability rights movement’s engagement with debates on the allocation of scarce medical resources.
To assist advocates, policymakers, and ethicists, I recently worked with several national disability rights organizations to produce an evaluation framework for state crisis standards of care plans, itself an extension of a principles document issued by a broad array of disability groups. In this article, I will explain how some of the disability organizations on the front lines of these issues are interpreting providers’ obligations under federal disability rights law.
Many state plans, including those in Tennessee, Colorado, and Minnesota, permit prioritization on the basis of anticipated or documented duration of need, either in the initial decision to allocate a scarce medical resource or in a subsequent decision to re-allocate the resource in the event that a patient makes use of it for a greater than typical time period.
Many would argue that this is appropriate because it does not imply any prejudice against the disabled.
But disability rights activists have always put forward a more comprehensive vision of nondiscrimination than simply eliminating irrational prejudice. The Americans with Disabilities Act and Section 504 of the Rehabilitation Act have long required providers to make reasonable accommodations and modifications to policies, practices, and procedures that include no explicit exclusions on the basis of disability. A public building is not just prohibited from posting a “No Wheelchair Users Allowed” sign–it is also required to ensure that a ramp is present. The fact that no particular animosity was intended by stairs is not sufficient.
Similarly, a hospital is not just required to avoid turning away Deaf and hard-of-hearing patients–it must also make available sign language interpretation or CART transcription to ensure that they have an equal opportunity to benefit from care.
These are not acts of charity; rather, they are requirements to avoid discrimination under a civil rights law. Disability rights law has a construct of discrimination that may require entities to expend more resources on certain people in order to afford them equal access.
By the same principle, disability rights law may require a hospital to afford individuals with an underlying disability in need of ventilator support extra time on a ventilator, without subjecting them to a lower priority to access scarce medical resources.
To be clear, this is not a blank check. There are many people with disabilities who require modifications to existing policies that are so intense that they constitute an undue burden or fundamental alteration of the nature of the program in question.
But given the difficulty in predicting how long patients will require use of a ventilator, anticipated duration of need should not be an acceptable criterion for providers to use to determine if a qualified patient first gets access to a ventilator.
In the context of re-allocation decisions concerning patients already receiving ventilation for an acute illness, providers are required to make “reasonable modifications” to policies that allow re-allocation from patients who do not recover fast enough to keep a scarce resource–even if that means that someone with an underlying disability receives more time to show sufficient improvement than someone without an underlying disability. This is necessary to ensure that people with disabilities receive an equal opportunity to benefit from treatment.
Determining how much of an extension of the duration of ventilator treatment a person with a disability is required to receive without having it counted against them in a re-allocation decision depends on the definition of “reasonable.” Determining what is a reasonable modification will no doubt depend on a variety of factors, including the extent of scarcity and the likelihood of recovery for someone given additional time. What is reasonable on one day may be unreasonable on another, as these factors change, requiring regular re-evaluation of re-allocation criteria. And of course, providers can always decline to provide treatment that is quantitatively futile.
But the idea that some additional time must be provided is a natural extension of the basic principles of disability rights law. To claim that this requirement can be tossed aside in a time of crisis would be to shift disability rights from the domain of mandatory civil rights to that of optional charity.
Many of the state plans that have eliminated categorical exclusions have instead turned to the use of long-term survival prospects as a means of prioritizing scarce medical resources, often looking as much as a decade out with the goal of maximizing the number of life-years saved. This should also be a source of concern to disability activists.
Long-term survival projections are significantly less certain than the assessment of short-term survival. Medical innovations such as new pharmaceuticals, surgical techniques and other interventions can shift the long-term prognosis for many conditions. Incorporating comorbidities that do not reduce a patient’s short-term survival prospects into an assessment of whether they will receive care risks incorporating concerning value judgments that will systemically disadvantage people with disabilities and chronic health conditions and reduce the likelihood that they will receive medically indicated care.
The consideration of long-term survival may also threaten the right of patients to receive a truly individualized assessment, since for many patients with rare disorders, clinicians with the specialized expertise to assess long-term survival prospects will not be immediately available.
For example, Massachusetts recently issued Crisis Standards of Care guidance, which indicates that children will not be penalized for predicted remaining lifespan unless death is considered likely within one year. This is justified in part because many children have rare or chronic diseases requiring multispecialist expertise for evaluation. But this same rationale applies to many adults too.
In recent decades, an increasing number of children with rare chronic conditions have survived into adulthood. Adults with conditions such as cystic fibrosis and spinal muscular atrophy would need specialist expertise for a clinically informed, individualized long-term prognosis. People with disabilities must receive equal protection from discrimination based on clinically unjustified assumptions about their lifespan, regardless of age. The consideration of long-term survival places people with disabilities at risk of impermissible denial of care based on the same stereotypes and assumptions that led to the rejection of categorical exclusions.
Avoiding considering such factors would not be unprecedented. For example, New York’s ventilator allocation guidelines explicitly reject the use of long-term survival, arguing that the definition of survival should be “based on the short-term likelihood of survival of the acute medical episode and is not focused on whether a patient may survive a given illness or disease in the long-term (e.g., years after the pandemic). By adopting this approach, every patient is held to a consistent standard. Triage decision-makers should not be influenced by subjective determinations of long-term survival, which may include biased personal values or quality of life opinions.”
New York’s ventilator allocation plan, which also mostly rejects the use of resource-intensity, has received praise from federal authorities, suggesting that its approach of focusing solely on short-term survival may be more likely to pass muster under federal law (though the state still needs to clarify some issues impacting chronic ventilator users). Other states should take notice and remove resource-intensity and long-term survival from their allocation criteria as well.
Short-Term Survival and Reasonable Modifications
If we accept that short-term survival probability is an appropriate consideration, what protections might need to be incorporated into available instruments used to assess short-term mortality to ensure that they do not penalize people with underlying impairments that don’t impact short-term survival?
Instruments designed to evaluate short-term survival probability for typical patients may require reasonable modifications for people with underlying disabilities. For example, many state plans rely on the use of the Sequential Organ Failure Assessment (SOFA), a measure designed to predict short-term mortality, to assess relative survival probabilities. While the use of the SOFA is not unacceptable, state plans must include provisions to ensure that reasonable modifications are made for those whose underlying impairments result in the SOFA penalizing them for their baseline level of impairment prior to the acute care episode.
The Glasgow Coma Scale, a tool for measuring acute brain injury severity, is an element of the SOFA. If administered in its standard form, a patient with a pre-existing speech disability who cannot articulate appropriately intelligible words could achieve a maximum score of 12, adding at least two points on the SOFA. The higher the score, the lower the estimated likelihood of survival. Patients with pre-existing motor impairments are also disadvantaged by this measure — full scoring on the motor response section requires moving in response to verbal commands.
A pre-existing interfering factor should ideally result in a “non-testable” score in this situation, meaning that only the trend (rather than the total score) should be reported. But state and provider level plans have not provided clarity on how to approach such scores within the SOFA. Plans must clearly indicate how instruments like the SOFA should be modified to hold patients harmless for underlying disabilities not documented to be relevant to short-term survival in the context of Covid-19. Given recent reports that individuals residing in congregate care settings are much more likely to develop and die from Covid-19, this clarification is particularly urgent.
Some have argued that disability rights activists should consent to a purely efficiency-oriented approach to the rationing of scarce medical resources, arguing that most disabled people will not be harmed since most disabilities don’t impact resource-intensity needs or long-term survival. But this is an argument for those disabled people who are not at risk to consent to discrimination against those who are. This is unacceptable and is inconsistent with the principle of solidarity at the core of the disability rights movement. As providers consider how to allocate scarce medical resources, they must recall that their obligations under federal civil rights law remain in effect–and that disability rights law requires them to go beyond simply avoiding irrational prejudice.
Ari Ne’eman is a Visiting Scholar at the Lurie Institute for Disability Policy at Brandeis University and a Senior Research Associate at the Harvard Law School Project on Disability. He previously served as President of the Autistic Self Advocacy Network and as a Member of the National Council on Disability. Twitter: @aneeman.
45% of The Hastings Center’s work is supported by individual donors like you. Support our work.