DREDF’s Comment Letter on the NASEM COVID-19 Vaccine Allocation Framework

Sent electronically to COVIDVaccineFramework@nas.edu

October 1, 2020
Committee on Equitable Allocation of Vaccine for the Novel Coronavirus
National Academies of Sciences, Engineering, and Medicine
500 5th St NW
Washington, DC 20001

RE: Comments on Discussion Draft of the Preliminary Framework for Equitable Allocation of COVID-19 Vaccine

Dear Committee Members:

The Disability Rights Education and Defense Fund (“DREDF”) appreciates the opportunity to provide comment to the National Academies Sciences, Engineering, and Medicine (NASEM) on its Discussion Draft of a Preliminary Framework for Equitable Allocation of COVID-19 Vaccine (draft). DREDF is a national cross-disability law and policy center that protects and advances the civil and human rights of people with disabilities through legal advocacy, training, education, and development of legislation and public policy. In the more than 40 years that have passed since our founding, we have persistently fought for the right of people with disabilities to be fully integrated within all aspects of community life. DREDF’s work is based on the knowledge that people with disabilities of varying ages, racial and ethnic backgrounds, genders, and sexual orientations are fully capable of achieving self-sufficiency and contributing to their communities with access to needed services and supports and the reasonable accommodations and modifications enshrined in U.S. law.

Public Comment Outreach and Timeline

As a first matter, DREDF strongly encourages the Committee to provide additional public outreach, expand the scope of public comment, and accept for the record public comments that have been submitted after September 4, 2020 on the vaccination distribution report commissioned by the Centers for Disease Control (CDC) and the National Institutes of Health (NIH). As NASEM notes on the webpage where the draft was first made available online September 1, 2020, “[I]nput from the public, especially communities disproportionately affected by the COVID-19 pandemic, is essential to produce a final report that is objective, balanced, and inclusive.”[1] Yet the public comment period was slated to end on September 4, approximately 86 hours after its release. While NASEM released a media advisory on the formation of the Committee on Equitable Allocation of Vaccine for the Novel Coronavirus,[2] and the Committee held a “public workshop” to discuss its development of a framework,[3] most members of the population groups identified by the Committee as disproportionately affected by the coronavirus specifically Black, Hispanic or Latinx, and American Indian and Alaska Native, low income persons, older persons, people with disabilities and health conditions are unlikely to subscribe to NASEM, CDC, or NIH newsletters and media releases. In July and August, advocacy groups for these populations were still deeply engaged in fighting for equitable COVID-19 care, testing and distribution of personal protective equipment (PPE), fielding on-the-ground complaints of discrimination, and fighting for renewed federal funding to reduce conditions such as the crowded congregate living conditions that exacerbate vulnerability to virus infection for identifiable groups.

While DREDF understands the urgent need for a vaccine distribution framework that will guide federal and state law and policymakers once a vaccine is appropriately developed and tested, the Committee has been working and periodically meeting for over a month. Some of the Committee meetings were virtually open while others were closed. Recorded meetings appear to be captioned, but it is unclear if real-time language interpretation was available, or if meeting materials were available in language translation or alternative formats, or fully accessible online. It is similarly unclear, even now, if the draft report is available in any non-English translation, alternative formats, American Sign Language, or the kind of plain language or pictorial formats that would allow individuals with intellectual or developmental disabilities to understand and respond to the recommended prioritizations made in the draft framework. Taking additional time, and providing additional means (e.g., toll-free number for voice messages, physical mail address, and so forth) to capture public comments beyond those submitted by professional stakeholders with full internet capacity will only strengthen the transparency and public trust that the Committee rightly acknowledges as crucial to the report’s broad acceptance.

DREDF appreciates that the Committee did not and does not undertake its work in any covert way, but if the Committee, CDC and NIH wish to receive input from those most directly impacted by the disparities fully exposed by the coronavirus, each organization must develop a proactive outreach plan that engages community-based health organizations, directs outreach to health and civil rights advocacy organizations, and makes full use of social media. Outreach materials must be written in plain language. And other aspects from the Committee’s Statement of Task such as the “issues related to vaccination program administration, evaluation, and assessment (to ensure effectiveness and equity); vaccine hesitancy, demand, and promotion; and risk communication and strategies for community engagement,”[4] must equally be made available for public comment. Black Americans, disabled men and women, and Indigenous persons and women prisoners have all been subject to officially sanctioned medical experimentation,[5] eugenics programs,[6] and sterilization respectively,[7] without informed consent. The insight of affected groups and individuals who have been subject to disparate coronavirus infection and death rates will be particularly valuable for developing effective strategies to evaluate equitable vaccine allocation, overcome vaccine hesitancy, and communicate and engage with their own communities. The Committee must solicit thorough public input into these “implementation” issues, as well as the framework for allocation, before issuing its final report.

Full Inclusion of People with Disabilities as a Broad Demographic Group in the Preliminary Framework

The Committee’s desire to ensure equality in the preliminary framework is evident throughout the draft, and enunciated in statements such as the one at pg. 33 that asserts the framework’s design “does not perpetuate discrimination and inequities” The Committee’s foundational principles for formulating a preliminary framework also reflect a commitment to equity: maximization of benefits (reducing morbidity and mortality), mitigation of health inequities, equal regard, fairness, transparency, and evidence-based.[8] The Committee’s discussion of these principles recognizes that the pandemic has caused higher rates of infection, serious illness, hospitalization and death among specific populations, such as people of color and people in certain work environments and living arrangements. DREDF supports these principles and the goal of not perpetuating discrimination and inequality, but finds it remarkable that people with disabilities of varying ages, who have been disproportionately subject to the impact of the pandemic on numerous grounds including comprising the great majority of the population in most congregate living situations, have not been recognized as being subject to disproportionate impact, or even acknowledged as a demographic group.

People with a range of disabilities have been deeply impacted by the coronavirus but the draft framework does not actually acknowledge or engage with their vaccination as a population group. Some disabled persons are included in the framework’s categorization of older persons, people with co-morbid conditions, homeless persons, and individuals in prisons, while additional disabled persons may be caught because they are persons of color who live in geographic areas highlighted by the CDC’s Social Vulnerability Index (SVI) tool. But the failure to recognize disability broadly as an identifiable group that is disparately impacted by COVID-19, as well as a heterogenous group that encompasses a range of diverse housing, living situations, races and ethnicities, genders, and needed services and supports means the preliminary framework ill serves disabled persons as a whole. The framework’s criteria are applied in the examples of Draft Table 2 in ways that unnecessarily fragment and overlook people with disabilities who face highly similar risks from COVID-19 and a similar lack of choices. The draft also fails to acknowledge unique discriminatory barriers people with disabilities commonly encounter when seeking COVID-19 or any other healthcare treatment, does not consider disabled persons will be forced to deal with a complex and barrier-filled vaccine administration process, and further perpetuates the ongoing data vacuum that makes it so hard to establish disability as a demographic factor that results in health and healthcare disparities which can be clearly traced to historic and ongoing limited access to Social Determinants of Health. We will further discuss these below in the order mentioned.

Fragmenting and Overlooking People with Disabilities

The degree to which COVID-19 has devastated the population of nursing home residents has been increasingly documented.[9] Approximately 40% of U.S. deaths from the coronavirus have occurred in nursing homes across the 46 states and the District of Columbia meeting reporting requirements.[10] This is a population comprised of mostly older persons, but just as evident, this is a population of people with disabilities: people with varying functional limitations who need assistance with basic activities of daily living (ADLs) such as bathing, eating or toileting or with instrumental activities of daily living (IADLs) such as preparing meals, shopping or tracking one’s finances. The congregate living common in nursing homes, in addition to the prevalence among nursing home residents of comorbid and underlying conditions places this group at high risk of both infection by COVID-19 and death once acquired, prompting this group’s placement in Phase One of the preliminary framework. But this same level of unavoidable risk of infection, because residents have no choice regarding their crowded setting and often require close physical personal assistance, combined with a high prevalence of co-morbid and underlying conditions, occurs across all congregate living settings, including assisted living facilities, intermediate care facilities for people with developmental and/or intellectual disabilities, psychiatric hospitals, and board and care homes.[11] Moreover, residents in these facilities may be older, and also may reflect disproportionate racial and ethnic residence levels,[12] but it is impractical, unfair, and impedes vaccine efficacy to allocate vaccination among residents on the basis of age, differentiating between people who eat together and share close common space.

The Committee’s failure to recognize how widely people with disabilities as a group are doubly subject to COVID-19 infection and death may also contribute to overlooking people with disabilities who receive home and community-based services (HCBS). This includes the over 2.5 million individuals who receive Medicaid HCBS,[13] ranging from mandatory state plan home health services, such as intermittent nursing services and medical supplies, to “optional” state plan services such as assistance with ADLs/IADLs and attendant support that help keep people with disabilities out of institutions. In addition to those who provide paid HCBS, there are an estimated 53 million family caregivers who provide unpaid services and supports that are valued at over $500 billion to older and disabled persons.[14] These are services provided to individuals who cannot choose to simply shelter at home and avoid visitors because care assistance from third parties is necessary to their ongoing survival. Paid support persons often work in the homes of multiple employers with disabilities and these employers may have their own families, while both paid and unpaid support persons will have their own homes and families as well. They all have the same need for PPE and regular virus testing as other front-line healthcare workers but likely have less access to both and less knowledge of how to use PPE properly. Furthermore, the home care workforce also raises racial/ethnic equity concerns as “the care workforce is disproportionately composed of women of color, Black women and immigrant women.”[15] The entire group of disabled persons who need long-term services and supports, including HCBS, as well as the home healthcare workers and personal care assistants who work with them, need to be as clearly included in Draft Table 2 and Draft Figure 2 as nursing home residents and staff.

Some people with disabilities are also captured by the preliminary framework’s recognition of people with multiple and single comorbid conditions. However, just as with the factor of congregate living, the Committee overlooks how people with disabilities as a group are subject to numerous COVID-19 risk factors including high rates of cooccurrence with multiple health conditions (depending on the given functional limitation), and little or no choice in settings since many people with disabilities often have a range of health events that require urgent, acute or preventive visits to health facilities, including hospitals. Pointing to studies that identify people with multiple underlying conditions as having a much higher chance of undergoing severe COVID-19, the Committee finds that “given the high risk of adverse outcomes in individuals with select comorbid conditions and the evolving evidence on this topic, it will be critical to monitor how the nature and number of comorbid conditions affect morbidity and mortality at the individual level.”[16] Further on, the Committee comments as follows on its assumption that time will help further the identification of specific risk factors and result in ongoing refinement of vaccine prioritization for those who have a single comorbid condition:[17]

Other comorbid conditions may be considered for this phase as evidence emerges. In addition to CDC’s list of comorbid conditions that put individuals at increased risk, CDC has also compiled a list of comorbid conditions that might put individuals at increased risk. This list includes asthma (moderate-to-severe); cerebrovascular disease; cystic fibrosis; hypertension; immunocompromised state from blood or bone marrow transplant, immune deficiencies, HIV/AIDS, use of corticosteroids, or use of other immunosuppressive medicines; neurologic conditions; liver disease; pregnancy; pulmonary fibrosis; smoking; thalassemia; and type 1 diabetes mellitus (CDC, 2020c).

The above references are troubling because they seem to indicate that for comorbidity to be considered a risk factor that triggers a higher priority for vaccine allocation, there must be a “match” with a particular medical diagnosis that is scientifically verified as potentially linked to increased rates of infection by, and serious illness or death from, COVID-19. People with disabilities as a community have a long history of being categorized according to their medical diagnosis and conditions, not only for healthcare treatment, but for educational, economic, and social purposes. This “medical model” generally ignores the multiple environmental and social factors that greatly influence what people with disabilities can do, where they can go, and whether and how they participate in their communities.[18] Strict adherence to a medical model submerges the myriad non-physiological systemic healthcare delivery factors that place individuals with disabilities at elevated risks of infections and serious illness and death after infection; these will be the focus of the next section. In addition, elevating a need for diagnostic information will likely undercount those individuals whose comorbidity is a factor in COVID infection and death because there are significant problems with comorbidity data collection from emergency rooms and ICUs.

In a supplemental Morbidity and Mortality data report, the CDC released surveillance morbidity and mortality data summarizing characteristics of death from COVID-19 in both the U.S. and China, reporting that “approximately three fourths of decedents had one or more underlying medical conditions reported (76.4%) or were aged 65 years (74.8%),” and for those “aged <65 years, 7.8% died in an emergency department or at home.”[19] The data was an attempt to analyze additional information about underlying medical conditions, race/ethnicity, and location of death, and add to the prior established understanding that COVID-19 mortality is higher among people with pre-existing conditions and those over 85 years. The supplemental data established important disparities relating to deaths rates among Hispanic/Latinx populations, but also highlighted data shortcomings, particularly with regard to underlying conditions, that have significant implications for the preliminary framework.[20] As the CDC notes:

First, despite >90% completeness for age and race/ethnicity variables in the supplementary data set, the proportion of missing data for some variables, such as underlying medical conditions, clinical course, and race/ethnicity in case-based surveillance, and location of death, was higher than that for other variables; accordingly, the proportions reported for these variables should be considered minimum proportions rather than robust estimates. Second, reporting practices varied by jurisdiction, and several states bundled underlying medical conditions into organ systemspecific categories (e.g., hypertension was included as cardiovascular disease) or did not code specifically for a given condition (e.g., immunosuppression was only specifically coded in 10 of the jurisdictions). These differences in reporting structure precluded evaluations of specific conditions other than diabetes using the entire data set. [The data went on to list additional data shortcomings.]‍

The select comorbid conditions listed in the preliminary framework are not bundled categories of organ system-specific categories. The idea that individuals with specific disabilities must wait for those specific disabilities to be linked to increased COVID-19 risk when the underlying evidence and data needed to establish the link does not even recognize specific disabilities is obviously problematic. The same reasoning applies to individuals who have less known underlying conditions and disabilities that may never reach the case numbers required for recognition as a statistically significant risk factor for COVID-19. Moreover, the mere fact of a vaccine allocation that emphasizes an apparent need for recognized diagnoses and conditions linked to COVID-19 inevitably leads to the presumption that people who have a somewhat or slightly different diagnosis, or who may not have specific diagnoses but are higher weight and therefore subject to many of the same barriers to COVID-19 treatment specified below, will be told to wait, left off a prioritized vaccine list, irrespective of their vulnerability to COVID for all the reasons that follow.

Barriers for People with Disabilities Seeking COVID-19 Healthcare

Mounting pandemic infection rates in the early spring of 2020 prompted some states and hospital systems, to revive or develop “crisis standard of care” (CSC) policies that would guide the anticipated need to ration medical resources in the event that COVID-19 patient needs overwhelmed available ICU beds, medications, equipment such as ventilators, and trained personnel. Many of the unearthed policies explicitly discriminated against people with certain underlying conditions and disabilities and/or gave medical personnel the discretion to make COVID treatment prioritization decision on the basis of the professional’s own assessment of the quality of life of the patient with disabilities. This was the case despite numerous studies establishing how health care professionals commonly hold implicit bias and entrenched negative assumptions about the quality of life of disabled persons that differ widely from the quality of life experienced by people with disabilities.[21]

DREDF worked with other national and state advocacy organizations to analyze and file formal complaints against several states that seemed to advance or endorse CSC policies that discriminated on the basis of disability or age, legitimizing the denial or lower prioritization of COVID-19 treatment to people with disabilities, older adults, and individuals with certain comorbid conditions. There appeared to be a widespread misperception that disability civil rights did not apply under emergency conditions. DREDF and others filed complaints with the US Department of Health and Human Services’ Office for Civil Rights (OCR) on behalf of individuals with disabilities who feared being denied effective healthcare and COVID-19 treatment if they went to the hospital, or even having their own personalized medical equipment such as CPAPs and BiPAPs taken away after admission and given to other patients assessed with a higher treatment priority under “surge” conditions. OCR issued a Bulletin on March 28, 2020 on “Civil Rights, HIPAA, and the Coronavirus Disease 2019 (COVID-19)” advising that:[22]

‍[P]ersons with disabilities should not be denied medical care on the basis of stereotypes, assessments of quality of life, or judgments about a person’s relative “worth” based on the presence or absence of disabilities or age. Decisions by covered entities concerning whether an individual is a candidate for treatment should be based on an individualized assessment of the patient based on the best available objective medical evidence.

Well after the above Bulletin was issued, disability discrimination complaints continued to be filed against states and healthcare entities. Issues raised in Washington, Alabama, Kansas, Tennessee, Pennsylvania, New York, Oklahoma, Connecticut, Oregon, Nebraska, Arizona, and Texas included not only discriminatory CSC policies but the failure of states and hospital systems to establish clear policies on reasonable accommodations, policy accommodations, and effective communication for people with disabilities. At this point, OCR has resolved some but not all of these complaints. These same issues have been raised by the Consortium of Citizens with Disabilities (CCD) and other disability advocates who submitted comments on the preliminary framework earlier in September.

The fact that such a bulletin needed to be issued in the first place, that complaints continue to be filed, that literally thousands of hospital and hospital systems are developing their own CSC standards because states generally only provide non-binding “guidance” on CSC policy, and the general lack of transparency from hospitals on CSC policies, show how heavily the scales are tipped against people with disabilities trying to defend themselves from being triaged out of medical care should they become infected with the virus. They cannot be asked to bear the burden of advocating for themselves while in the throes of the infection or any other urgent health event that requires hospitalization. Restrictive visitation policies mean that many people with disabilities cannot rely on having an advocate there on their behalf to fight for needed treatment accommodations and policy modifications, without which a disabled person may decompensate into even worse health, further playing into a pre-existing expectation that they cannot survive treatment and belong on hospice care. There have also been examples in the press of how disability and race combine in ways that subject individuals to multiple stereotypes and generalizations about quality of life, worth, and ongoing support, and potential lead to decisions to stop COVID-19 treatment even over the objections of a key family support person.[23]

DREDF knows that these are not theoretical examples. We have heard from numerous individual disabled persons and their families, some of whom have become our clients, after experiencing healthcare disability discrimination after the advent of the pandemic. In April, a woman contacted us because her mother was hospitalized with COVID and was denied the use of her of her own BiPAP, which she needed for a diagnosis of Chronic Obstructive Pulmonary Disease (COPD). The caller knew her mother was getting increasingly ill and incoherent due to the carbon dioxide build-up. After constant advocacy, the hospital finally administered a blood gas text which showed a very high carbon dioxide level, and even then only increased the patient’s oxygen which does little or nothing to alleviate the problems of carbon dioxide build-up. While this patient was eventually transferred to a negative pressure room and allowed to use her BiPAP, if CSC policies had been administered while her condition was worsening without her BiPap machine, she could easily have been deprioritized for COVID-19 care on the basis of having failing health, increasing disorientation, and likely being unresponsive to treatment.

Ms. Sarah Johnson, an 86-year-old Black woman hospitalized with COVID in mid-April was not so fortunate.[24] Brought to a New Orleans hospital by her family, Ms. Johnson’s seemed to recover from the respiratory impact of the virus quite quickly, but was dehydrated and her “modest chronic kidney disease” had become acute renal failure so the hospital admitted her for treatment. A succession of doctors came to evaluate her. The last one made the assessment that “[b]ased on physical appearance and presentation, patient likely has advanced [or] end stage dementia.”[25] This diagnosis was made despite the high prevalence of temporary delirium among acutely ill elderly patients and the recommendation that dementia not be diagnosed in an acute care setting, and without speaking to Ms. Johnson’s family who had not noted signs of dementia in their mother. The last doctor also concluded that Ms. Johnson was “severely malnourished,” despite a dietician’s opposite conclusion made a few hours early and entered on the chart.[26] These two conclusions, along with Ms. Johnson’s “advanced age” and her poor functional status let the doctor to conclude that her patient was a candidate for a do-not-resuscitate order and hospice care rather than active COVID-19 treatment. Informed of this, Ms. Johnson’s family agreed to hospice care, though she was kept in the hospital for the week-end. By Monday morning, new labs showed great improvement in Ms. Johnson’s kidney functions and the irregular heartbeat noted at her admission had stabilized. Nonetheless, the decision to discharge her for hospice care at home moved forward, with the hospital “citing her kidney problems as the main reason, along with dementia and a general ‘failure to thrive.'”[27] Meanwhile, her sodium levels had risen very high without remark or action, even the dehydration that likely caused it could be readily treated with IV fluid, and lack of treatment could lead to confusion, erratic behavior, and eventually coma and death. Ms. Johnson’s eventual death at home, thirteen days after discharge, found her unable to swallow despite being thirsty, increasingly weak and in pain, the victim of a cascade of faulty medical judgements fed by discriminatory assumptions as much as the virus itself.

The specific diagnoses and non-congregate living situations of the two women would not necessarily qualify them for first priority vaccine allocation under the draft preliminary framework. But as individuals with disabilities, they were each subject to unique risk factors such as a denial of needed medical equipment and care decisions based on intersecting assumptions about age and disability instead of individualized medical assessments. The resulting is an increased risk of mortality from COVID-19 that arises from how they are treated in the healthcare system, and may arise from physiological susceptibility as well, but that link will not be successfully captured for a very long time when hospitals are not required to uniformly capture granular information about underlying health conditions, functional limitations, and whether or not accommodation needs were met.

DREDF also worked with other advocacy organizations in California to file administrative complaints from patients and their families who sought treatment from hospitals around the state for non-COVID treatment. The examples include:[28]

  • One 26-year old man with several developmental disabilities who laid in pain after developing complications from an emergency appendectomy while his mother and regular supporter was refused entry for four days;
  • A daughter and primary caregiver of a nonverbal 68-year-old patient was not allowed to provide necessary overnight care to prevent aspiration and other medical complications, and questioned by staff about a do-not-resuscitate order that was not on her mother’s record;
  • The mother of a 60-year old woman with an intellectual disability was repeatedly asked to authorize an invasive feeding tube even as the hospital refused to allow a family member or known staffer to provide mealtime support;
  • A 77-year old woman with dementia, aphasia, quadriplegia, and a traumatic brain injury was denied the assistance of her daughter in the hospital;
  • A 57-year old patient with psychiatric disabilities who could not speak for herself was denied her support person for hours on a Saturday, forcing the latter to seek a disability rights lawyer.

These are rare examples only in the sense that the individual and families involved managed to gain the assistance of a trained advocate. The circumstances that raise the need for assistance are all too common. Many hospitals make no distinction between the visitation restrictions that have been reasonably put into place as an infection control measure during COVID, and the support needs of people with disabilities who have long-standing arrangements for personal and communication assistance. Other facilities have explicit disability exceptions written in their policies but frontline security and medical staff are insufficiently trained in how the policy works. The result is the same: people with a range of disabilities or conditions that may never be scientifically substantiated as a comorbidity that physiologically raises the risk of COVID-19 vulnerability, but who nevertheless face a greatly elevated risk of bad health outcomes when they enter a hospital during the pandemic.

Despite nearly five decades of federal disability discrimination laws,[29] people with disabilities who seek healthcare encounter specific barriers ranging from a lack of accessible equipment for something as common and necessary as finding a patient’s current weight, to programmatic barriers[30] that prevent effective communication and full examinations, to provider stereotypes and discrimination about disability. The healthcare profession is primarily trained to cure non-disabled people, not work consistently over time with disabled people and their families to maintain function and health.[31] Nowhere is this gap more apparent than in urgent care hospitals and emergency departments. If the Committee ignores the testimony of people with disabilities and the documented existence of disability health and healthcare disparities, including how disability and race/ethnicity intersect to compound unequal health,[32] it perpetuates the longstanding healthcare inequalities experienced by people with disabilities.

The preliminary framework does recognize disability in a few specific instances. For example, the Committee states that “the ability of frail or disabled individuals to access vaccination location must be taken into account while operationalizing vaccine access and delivery.”[33] These acknowledgements of the real-life impact of living with a disability are few and far between and none of them seem to bear on the actual analysis of risk that was used to determine vaccine allocation. The limits of the Committee’s superficial approach to disability can be seen when considering persons who are blind, as one example. Blindness, in and of itself, is unlikely to present as a comorbidity that exacerbates the likelihood or severity of COVID-19 infection. However, blind individuals preserve their independence and navigate their world primarily through touch. Taking public transit, shopping for groceries, finding the right elevator button, and reading public signage all add to a blind individual’s viral exposure and load, irrespective of the individual’s age or race/ethnicity, though living in a lower-income area may increase the need to take public transit to reach grocery stores, pharmacies and medical facilities.

The Committee’s entire analysis notes neither the real-life impacts of the pandemic on people with specific disabilities beyond diagnostic comorbidities nor the risks that medical rationing, lack of accommodation and congregate living pose to people with disabilities as a broader community because of COVID-19. The initial failure to recognize disability, broadly defined, as a pandemic risk factor with independent and intersectional significance for vaccine allocation may reveal a lack of knowledge, but after receiving numerous public comments including ours, continued adherence to the current preliminary framework is a judgement that disability is irrelevant to vaccine allocation. If the Committee comes to this conclusion, state, local, tribal, and territorial governments are even less likely to explicitly incorporate disability in their vaccine allocation, distribution, or data collection.

Alternatively, if the final allocation framework does not recognize people with disabilities as a demographic population, the Committee seems to be making the judgement that equity is satisfied when at least some people with disabilities are sufficiently caught by equality measures such as the Committee’s recommended use of the CDC’s Social Vulnerability Index (SVI). The SVI’s analysis of social determinants of health is useful for identifying geographic areas whose occupants are particularly vulnerable to health and healthcare disparities, and DREDF agrees that locating vaccination sites in such areas will be helpful to disabled individuals and their families who live in those areas, many of whom are lower-income and people of color.[34] But the conclusion, that “catching some is good enough,” runs entirely counter to the Committee’s commendable recommendation to include all critical risk workers as a priority vaccination group, without excluding those individuals who are not U.S. citizens or green card holders. DREDF strongly agrees with that recommendation because all critical risk workers are subject to similar conditions of high virus exposure, inadequate and inconsistent access to PPE, reduced choice to avoid exposure, and disincentives to use the healthcare system (immigrants have financial and legal reasons as noted in the draft[35]); immigration status is irrelevant to considerations of equity and the goal of reducing overall exposure to the virus. People with a range of disabilities of all ages who live in any type of congregate setting, who receive home and community-based services (including the HCBS providers), and who are subject to medical rationing fears and healthcare barriers, including discrimination, should receive high and equal prioritization under the framework allocation rather than be forced to rely upon having additional personal characteristics.

Complexities of Vaccine Administration for People with Disabilities

DREDF believes that the heterogenous and cross-cutting nature of disability across other risk factors explicitly recognized by the Committee, such as age, congregate living and race/ethnicity (through the incorporation of the CDC’s Social Vulnerability Index), will place a significant and unnecessary burden on people with disabilities and entities responsible for vaccine administration. The failure to recognize disability as a population that is particularly vulnerable to COVID-19, except for those who have an established comorbidity with the virus, means that the entities that will be physically distributing and administering the vaccine county, tribal government, hospital, insurance plan, congregate living facility will presumably be responsible for adhering to the vaccination prioritization that is adopted. This will foreseeably raise practical and equitable barriers both for those who cannot establish their priority through something as relatively easy to prove as age (I.e., comorbid conditions), those who share congregate living situations but are not “older,” and those persons with disabilities living in the community who are at elevated risk of COVID-19 infection or severity but who fall out of the preliminary framework for allocation.

As one example, the allocation framework’s primary goal of “[M]aximizing societal benefit by reducing morbidity and mortality caused by transmission of the novel coronavirus novel coronavirus” places older persons in nursing homes high on the prioritization list. But younger people with disabilities also live in nursing homes, albeit often unwillingly. A published report in 2010 found that working-aged people from 31-64 years increasingly made up the nursing home population, comprising just over 14 percent according to data from the federal Department of Health and Human Services.[36] Back in 2010, one key factor in the growth of younger persons in nursing homes was that across nearly all states in the preceding decade, budgetary gaps led to cuts of the home and community-based services that people with disabilities need to stay in their communities.[37] The coronavirus has placed states across the country in similar situations of deep budget shortfalls and the institutional bias of institutional long-term care as a required Medicaid service while home and community-based services are option are likely to lead to similar results. The latest data available from CMS indicates that 15.5 percent of the nursing home population was under 65 in 2014.[38] If a state agency, county, or nursing facility elects to adopt the Committee’s framework will it be expected to distribute the first vaccine shipments received and administer them only to “older persons” in the nursing facility, leaving younger residents vulnerable despite sharing the same environment, the same staff to provide physical assistance if required, or even the same bedroom? And if younger residents in congregate living facilities are included as they logically should be, equity also requires that individuals with disabilities who have the same nursing facility level of care as their institutionalized peers, but who have fought to remain integrated in their communities, should not be penalized for exercising their legal right to receive home and community-based services and supports.[39]

This then raises the practical allocation issue of how disabled people living in their communities would presumably have to prove that they have an established medical diagnosis for “a select comorbid condition” that the CDC recognizes as linked to higher severity or infection with COVID-19. This may be easier for individuals with a regular source of care, but how will the one in three adults with disabilities aged 18 to 44 years without a usual health care provider[40] establish their prioritization to the vaccine? Most of us do not carry our electronic health record with us, and many of us may have inaccurate or incomplete records even if we had easy access. Physical distribution of the vaccine to homeless shelters, with vaccination stations established close to homeless encampments and located within geographic areas identified through the SVI to capture allocation priority groups, raises the fundamental problem of achieving only patchwork immunization among individuals who are living in very close proximity will also come up if allocation prioritization distinctions are attempted among residents in the context of any congregate setting such as psychiatric facilities, prisons and jails, and group homes.

There will be practical difficulties of vaccine distribution regardless of whether people with disabilities as a group are prioritized for vaccine allocation. But the fragmented recognition of some disabled persons among other recognized priority groups leaves people with disabilities, both those captured in an allocation priority group and those who are not, subject to myriad assumptions and stereotypes about what people with “real disabilities” look like, where they live, and how they can be served. If the Committee’s finalized recommendations keep a fragmented approach to disability, disabled people will have to deal with employees of vaccinating entities on the ground who will gatekeep the vaccine as part of their job. Approximately 1.5 million people living in nursing homes[41] and over 800,000 assisted living residents[42] represent only a tiny fraction of the 61 million people with disabilities in the U.S.[43] How will younger persons with disabilities “prove” their comorbidity, establish that they receive HCBS, or receive needed accommodations for such barriers as physical inaccessibility, diminished public transportation, and severely compromised immune systems? How will paid and unpaid providers of HCBS establish that they are healthcare providers who regularly come into contact with vaccine prioritized individuals. The preliminary framework rightly prioritizes persons in homeless shelters, but where does that leave people with disabilities who are at imminent risk of losing their homes, particular persons with long-term care needs, many of who have particularly precarious and inadequate housing?[44] How much training and discretion will be given to individual employees administering the vaccine?

It could be argued that the preliminary framework should not be concerned with such practical problems of administration as how an individual establishes that they “belong” in a priority group, but questions of administration squarely raise equity issues, evident from the discussion above.

Perpetuating the Ongoing Lack of Disability Data

We discuss the data issues here because they are directly related to the topics above, and to the Committee’s foundational principles of maximizing benefits, equal regard, mitigating health inequities, fairness, evidence-based, and transparency.[45] DREDF applauds the Committee’s attempt to judge the allocation criteria against an overarching commitment to equity. Nonetheless, equity in the context of people with disabilities requires acknowledgement of, and a response to, the degree to which national population surveys fail to provide granular and consistent data on disability and functional limitation, access to health services, health expenditures, health outcomes, and patient experience. The National Health Interview Survey (NHIS), Medical Expenditures Panel Survey (MEPS), Survey of Income and Program Participation (SIPP), and Behavioral Risk Factor Surveillance Survey (BRFSS), have begun to use the same validated set of six functional limitation questions to identify individuals with disabilities that are used in the American Community Survey (ACS),[46] but when survey data is reported out and analyzed, it rarely includes information on health services use or health outcomes and disparities by disability status. Section 4302 of the Affordable Care Act mandated use of a minimum data collection standard for the demographic category of disability in major national surveys, but did not necessarily mandate how often information about disability must be collected, or when or how disability information should be reported out. As a result, disability information is missing across years and across surveys. This makes it particularly difficult to detect trends over time for people with disabilities as small sample sizes require pooling across time to obtain more reliable results. As well, many state and local health surveys, including state Medicaid data collected for CMS administrative data sets, still lack any questions for collecting basic demographic information about disability.

Identifying people with disabilities and underlying conditions is a first step to understanding how widely people with disabilities live, work, have families, and participate in their communities. But the next step of analyzing this data in conjunction with other key demographics such as race/ethnicity, and measuring how disability interacts with social determinants of health, is equally critical. The lag in disability data collection, analysis, and the understanding of disability as a demographic factor that leads to health and healthcare disparities leads directly to the kind of fragmented approach to disability taken by the Committee. Despite examples of the current gaps in disability-related data and calls for better and more consistently collected data,[47] we continue to see information lags and failures to fully incorporate disabled people within health equity and research initiatives continues.

In the specific context of COVID-19, there has been limited study of coronavirus infection, spread and fatality that factors in disability as a demographic factor rather than the biological impact of specific conditions on the course of the virus. Without such studies, it is difficult to establish how principles of equal regard, mitigation of inequity, and fairness should apply to people with disabilities. One of the few studies that attempts such an analysis looks at social inequities in who has contracted COVID-19, and where this has occurred.[48] The author collects data that “suggests that people with disabilities are experiencing a ‘multiple jeopardy’ based on the convergence of their disability, racial/ethnic minority, and poverty status. Moreover, the author highlights a recent study that found higher rates of COVID-19 cases among patients under 18 years who have intellectual and developmental disabilities. Ultimately, the author calls for better disability data collection during the pandemic.

. . . the findings emphasize the need for COVID-19 data collection and surveillance systems to incorporate disability identifiers, in addition to existing indicators of age, race/ethnicity, and gender. To understand and address the needs of people with disabilities affected by this pandemic, more detailed information is urgently needed on rates of infections, hospitalizations, outcomes, and fatalities, disaggregated by types of difficulty experienced by people with disabilities, as well as their demographic and socioeconomic characteristics.

A longstanding failure to adequately study the health and healthcare disparities commonly faced by people with disabilities cannot be allowed to initiate a self-perpetuating cycle of overlooking people with disabilities during a pandemic, in emergencies, and for vaccine distribution. DREDF does not contest that people with disabilities and underlying conditions will be infected and will die during a pandemic. We contest leaping to the further conclusion that every disabled individual who dies during a pandemic is an unavoidable outcome that would have occurred irrespective of health and healthcare disparities, disability discrimination, vaccine allocation, and a data vacuum.

Recommendations and Conclusion

Broad principles such as “maximizing benefit” and “fairness” are challenging to apply in real life. DREDF recognizes the Committee faces a task that is made even more difficult by the ongoing daily toll of lost lives. We make the following recommendations because we are a disability-led organization that has worked in and with disability communities for over four decades. We are deeply familiar with the multiple ways in which disability is overlooked as a demographic factor, a vector for explicit and implicit discrimination, and a reason for health and healthcare disparities. The U.S. does not yet have a safe and fully tested coronavirus vaccine, but when this occurs, the allocation framework must appropriately include people with disabilities in the following ways:

  • Recognize people with disabilities as a demographic group that is more likely to acquire COVID-19 and to suffer serious health consequences due to comorbidities as well as multiple non-physiological systemic and systemic healthcare factors, including stereotypes about disability, quality of life with a disability, the administration of Crises Standards of Care, and the common use of morbidity measures in emergency care that are not appropriately modified for individuals with chronic conditions.[49]
  • Include the full range of congregate living situations, including assisted living facilities, intermediate care facilities for people with developmental and/or intellectual disabilities, psychiatric hospitals, and board and care homes, in the same prioritization category of the allocation framework as individuals (residents and staff) in all of these congregate living situations experience the same lack of choice regarding living/work situation, increased risks of morbidity and mortality, and ongoing lack of data collection on illness & death, whether caused by COVID-19 or otherwise.
  • Include people with disabilities outside of institutions who receive home and community-based personal assistance with Activities of Daily Living and Instrumental Activities of Daily Living, and their personal care assistants and direct care workers, within the allocation framework as a priority population that is subject to the same COVID-19 risks and lack of choice as institutionalized persons.
  • Explicitly reference disability and other non-discrimination and civil rights laws as standards that continue to apply both during a pandemic and to the actions and vaccination policies of all federal, state, and local governments, as well as entities that receive federal funds.
  • Highlight the importance of any vaccine undergoing the full existing federal gamut of vaccine approval and human trial protocols before triggering of the allocation framework and any targeted or large-scale distribution given the historical justified fears of medical experimentation held by people with disabilities, Black Americans, and people of color.
  • Call for all data on COVID-19 vaccination, including distribution, administration and refusal rates, adverse effects, timing and community benefit, to include voluntarily obtained granular information on people with disabilities as this will have vitally important implications for the full inclusion of people with disabilities in ongoing scientific trials on the coronavirus, as well as implications for disparities and intersectional analysis.
  • Develop mechanisms for maintaining transparency and accountability in vaccine allocation that will span the numerous entities and providers, from state and local government agencies to tribal governments, private insurers and health clinics, that will have a role to play in vaccine distribution.
  • Stress the need for community engagement and direct representation of people with disabilities, people of color, low-income and critical workers, and other COVID-19 vulnerable populations in vaccination appeal procedures.

We note that some of the above recommendations address chapters of the final report that were not open for public comment. As noted in the beginning of our comments, we stated our strong recommendation for receiving public comment on the issues that were to be addressed in forthcoming chapters such as vaccine hesitancy, program assessment for effectiveness and equity, and strategies for community engagement. People with disabilities worldwide have encapsulated their fight for equal inclusion with the phrase “Nothing About Us Without Us.” We call on the Committee to fully engage with disability conceptually and procedurally, and with disabled individuals, communities and advocates who are expert in disability health needs as well as how people with disabilities experience inequitable healthcare in the U.S.

Thank you for the opportunity to provide comments on the proposed allocation framework for a COVID-19 vaccine. We would be happy to answer any questions on the above.

Sincerely,

Silvia Yee,
Senior Staff Attorney


[1] National Academy of Medicine President Victor J. Dzau, NASEM news release (Sept. 1, 2020) available at https://www.nationalacademies.org/news/2020/09/national-academies-release-draft-framework-for-equitable-allocation-of-a-covid-19-vaccine-seek-public-comment.

[2] NASEM Media Advisory (July 21, 2020), available at https://www.nationalacademies.org/news/2020/07/national-academies-launch-study-on-equitable-allocation-of-a-covid-19-vaccine-first-meeting-july-24.

[3] National Academy of Medicine Event Notice and registration page, available at https://nam.edu/event/public-workshop-to-inform-equitable-allocation-of-covid-19-vaccine/.

[4] “A Note from the Committee Co-Chairs, Discussion Draft at 3.

[5] Vann R. Newkirk II, A Generation of Bad Blood, The Atlantic, June 17, 2016 at https://www.theatlantic.com/politics/archive/2016/06/tuskegee-study-medical-distrust-research/487439/.

[6] Alexandra Minna Stern, EUGENIC NATION: FAULTS AND FRONTIERS OF BETTER BREEDING IN AMERICA (American Crossroads) (2d ed., 2018).

[7] Brianna Theobald, A 1970 Law Led to the Mass Sterilization of Native American Women. That History Still Matters, November 28, 2019, at https://time.com/5737080/native-american-sterilization-history/.
;Hunter Schwarz, Following Reports of Forced Sterilization of Prison Inmates, California Passes Ban, Sept. 26, 2014, at https://www.washingtonpost.com/blogs/govbeat/wp/2014/09/26/following-reports-of-forced-sterilization-of-female-prison-inmates-california-passes-ban/.

[8] Discussion Draft, pg. 34, lines 690-706.

[9] Kaiser Family Foundation, STATE DATA AND POLICY ACTIONS TO ADDRESS CORONAVIRUS (September 2020) at https://www.kff.org/report-section/state-data-and-policy-actions-to-address-coronavirus-maps-and-data/#long-term-care-cases-deaths; Sarah True et al., UNDER THE RADAR: STATES VARY IN REGULATING AND REPORTING COVID-19 IN ASSISTED LIVING FACILITIES (June 2020) at https://www.kff.org/coronavirus-covid-19/issue-brief/under-the-radar-states-vary-in-regulating-and-reporting-covid-19-in-assisted-living-facilities/.

[10] Id.

[11] DREDF also recognizes the disproportionate presence of people with disabilities in jails, prisons and similar involuntary congregate settings but since the committee has explicitly incorporated all incarcerated persons within the preliminary framework, we are not in this comment letter addressing the issue.

[12] American Psychiatric Association, MENTAL HEALTH DISPARITIES: DIVERSE POPULATIONS (2017) (“Inpatient mental health services were used more frequently by black adults and those reporting two or more races.”), at https://www.psychiatry.org/File%20Library/Psychiatrists/Cultural-Competency/Mental-Health-Disparities/Mental-Health-Facts-for-Diverse-Populations.pdf.

[13] Molly O’Malley Watts, MaryBeth Musumeci & Priya Chidambaram, MEDICAID HOME AND COMMUNITY-BASED SERVICES ENROLLMENT AND SPENDING, (February 2020), at https://www.kff.org/medicaid/issue-brief/medicaid-home-and-community-based-services-enrollment-and-spending/.

[14] AARP, CAREGIVING IN THE U.S., Executive Summary (May 2020), at https://www.caregiving.org/wp-content/uploads/2020/05/Executive-Summary-Caregiving-in-the-United-States-2020.pdf.

[15] Josephine Kalipeni & Julie Kashen, BUILDING OUR CARE INFRASTRUCTURE FOR EQUITY, ECONOMIC RECOVERY, AND BEYOND (September 1, 2020), at 7, at https://caringacross.org/wp-content/uploads/2020/08/Building-Our-Care-infrastructure_Aug1_FINAL.pdf.

[16] Discussion draft, page 63, lines 1408-1410.

[17] Discussion draft, page 69, lines 1578-1585.

[18] Sara Goering, Rethinking Disability: The Social Model of Disability and Chronic Disease, 8 CURRENT TRENDS IN MUSCULOSKELETAL MEDICINE, 134 (2015) at https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4596173/.

[19] J.M. Wortham et al., Characteristics of Persons Who Died with COVID-19 United States, February 12May 18, 2020, MMWR Morb Mortal Wkly Rep 2020;69:923-929. DOI: http://dx.doi.org/10.15585/mmwr.mm6928e1 at https://www.cdc.gov/mmwr/volumes/69/wr/mm6928e1.htm

[20] Id.

[21] Mary Crossley, Ending-Life Decisions: Some Disability Perspectives, 33 Ga. State Univ. L. Rev. 90001 (2017) (citing several studies finding that health care providers’ often perceive people with disabilities to have a low quality of life when, in reality, most report a high quality of life and level of happiness, especially when they have access to the health care services and supports that they need to equally participate in and contribute to their communities); M. C. Wilson & K. Scior, Attitudes Towards Individuals with Disabilities as Measured by the Implicit Association Test: A Literature Review, 35 RES DEV DISABIL., no. 2, 294-321 (2014) (the public holds moderate to strong negative attitudes toward individuals with disabilities with little or no association with the explicit attitudes that are expressed); Kenneth A. Gerhart, et al., Quality of Life Following Spinal Cord Injury: Knowledge and Attitudes of Emergency Care Providers, 23 Annals Emergency Med. no. 4, 80712 (1994): (finding that only 18 percent of emergency care providers expressed that they would be glad to be alive after experiencing a spinal cord injury, in contrast to the 92 percent of actual spinal cord injury survivors who report a high quality of life) (also finding that misperceptions about disability are more pervasive among less-experienced physicians, who are more likely to consider a treatment futile than more-experienced physicians).

[22] At https://www.hhs.gov/sites/default/files/ocr-bulletin-3-28-20.pdf.

[23] Joseph Shapiro, One Man’s COVID-19 Death Raises The Worst Fears Of Many People With Disabilities, July 31, 2020, at https://www.npr.org/transcripts/896882268.

[24] Annie Waldman & Joshua Kaplan, Sent Home to Die, PROPUBLICA, Sept. 2, 2020, at https://www.propublica.org/article/sent-home-to-die

[25] Id.

[26] Id.

[27] Id.

[28] Complaints of Disability Discrimination California Hospital Visitation Policies at https://dredf.org/2020/08/27/complaints-of-disability-discrimination-california-hospital-visitation-policies/.

[29] Section 504 of the Rehabilitation Act of 1973, 29 U.S.C. 794, is generally recognized as the first federal disability civil rights act in the U.S. Section 504’s implementing regulations in the area of healthcare are found in 45 C.F.R. Parts 84 and 85, enacted in 1977.

[30] Nancy Mudrick & Silvia Yee, PROMOTING PROGRAMMATIC ACCESS: WHAT DOES PROGRAMMATIC ACCESS MEAN IN HEALTHCARE SETTINGS, at https://dredf.org/wp-content/uploads/2012/08/Healthcarepgmaccess.pdf.

[31] HEALTHCARE STORIES, A series of videos in which a diverse group of individuals with disabilities speak of their healthcare experiences in their own words is available at https://dredf.org/healthcare-stories/.

[32] Silvia Yee, Mary Lou Breslin, et al., COMPOUNDED DISPARITIES: HEALTH EQUITY AT THE INTERSECTION OF DISABILITY, RACE, AND ETHNICITY, Nat’l Acads. Sci., Eng’g, & Med. (2017), available at https://dredf.docksal.site/wp-content/uploads/2018/01/Compounded-Disparities-Intersection-of-Disabilities-Race-and-Ethnicity.pdf.

[33] Discussion Draft, page 77, lines 1806-1807.

[34] The SVI’s accounting of persons with disabilities seems to rely upon a single demographic variable gathered in the American Community Survey (presence of persons older than age 5 with a disability in the section for household composition). If this question is not included in the ACS during a particular round of testing, it is not included in the index. See CDC, PLANNING FOR AN EMERGENCY: STRATEGIES FOR IDENTIFYING AND ENGAGING AT-RISK GROUPS (1st Ed. 2015) at p. 14, at https://svi.cdc.gov/Documents/Publications/SVI_Community_Materials/atriskguidance.pdf.

[35] Discussion draft, page 67.

[36] Joseph Shapiro, A New Nursing Home Population: The Young, Dec. 9, 2010, at https://www.npr.org/2010/12/09/131912529/a-new-nursing-home-population-the-young.

[37] Id.

[38] Elaine K. Howley, Nursing Home Facts and Statistics, Oct. 11, 2019, at https://health.usnews.com/health-news/best-nursing-homes/articles/nursing-home-facts-and-statistics.

[39] The U.S. Supreme Court has recognized the right of people with disabilities to appropriately receive services and supports in their communities. See Olmstead v. L.C., 527 U.S. 581 (1999).

[40] Centers for Disease Control and Prevention, DISABILITY AND HEALTH INFOGRAPHICS, at https://www.cdc.gov/ncbddd/disabilityandhealth/infographic-disability-impacts-all.html.

[41] Howley, supra.

[42] American Health Care Association and National Center for Assisted Living, FACTS AND FIGURES, at https://www.ahcancal.org/Assisted-Living/Facts-and-Figures/Pages/default.aspx last visited September 27, 2020).

[43] CDC, “Disability and Health Infographics,” at https://www.cdc.gov/ncbddd/disabilityandhealth/infographic-disability-impacts-all.html.

[44] Kartik Trivedi, Tatjana Meschede, & Finn Gardiner, Unaffordable, INADEQUATE, AND DANGEROUS: HOUSING DISPARITIES FOR PEOPLE WITH DISABILITIES IN THE U.S., (April 2020). Community Living Policy Center. Brandeis University, Waltham, MA. Available at: https://heller.brandeis.edu/community-living-policy/images/pdfpublications/2020aprilhousingbrief.pdf.

[45] Discussion Draft, pages 47-48, lines 1044-1083.

[46] The set of six disability questions has faced some criticism for its ability to adequately capture mental health and episodic disabilities, and a lack of nuance concerning the severity of functional limitations. Nevertheless, the core set of questions provides a validated base on which to build. Section 4302 of the Affordable Care Act mandated use of the set of six disability questions in major national surveys as the minimum data collection standard for the demographic category of disability, but did not necessarily mandate how often information about disability must be collected, or when or how disability information should be reported out. This makes it difficult to detect trends over time as disability information is missing across years and across surveys.

[47] Gloria L. Krahn, Deborah Klein Walker & Rosaly Correa-De-Araujo, Persons with Disabilities as an Unrecognized Health Disparity Population, 105 (Suppl 2) AM J PUBLIC HEALTH S108-S206 (2015), doi:10.2105/AJPH.2014.302182; Yee, supra at pages 10-21.

[48] Jayajit Chakraborty, Social Inequities in the Distribution of COVID-19: An Intra-Categorical Analysis of People with Disabilities in the U.S., DISABILTY AND HEALTH J. ONLINE, In press, Journal Pre-Proof, Sept. 18, 2020, at https://doi.org/10.1016/j.dhjo.2020.101007.

[49] Emily Cleveland Manchanda, Cheri Couillard & Karthik Sivashanker, Inequity in Crisis Standards of Care, N Engl J Med (May 13, 2020), at https://www.nejm.org/doi/full/10.1056/NEJMp2011359. (Most CSC guidelines reference use of the Sequential Organ Failure Assessment (SOFA) score as a way to objectively assess acute disease severity, but SOFA scores are not developed for individual mortality assessment, fail to account for the impact of functional limitations, and “one study found that nearly half of ICU patients with a SOFA score in the worst category survived.”)