The Role of Courts in Shaping Health Equity

Abstract

United States' courts have played a limited, yet key, role in shaping health equity in three areas of law: racial discrimination, disability discrimination, and constitutional rights. Executive and administrative action has been much more instrumental than judicial decisions in advancing racial equality in health care. Courts have been reluctant to intervene on racial justice because overt discrimination has largely disappeared, and the Supreme Court has interpreted civil rights laws in a fashion that restricts judicial authority to address more subtle or diffused forms of disparate impact. In contrast, courts have been more active in limiting disability discrimination by expanding the conditions that are considered disabling and by articulating and applying the operative concepts “reasonable accommodation” and “other qualified” in the context of both treatment and insurance coverage decisions. Finally, regarding constitutional rights, courts have had limited opportunity to intervene because, outside of specially protected arenas such as reproduction, constitutional law gives government wide discretion to define health and safety goals and methods. Thus, courts have had only a limited role in shaping health equity in the United States. It remains to be seen whether this will change under the Affordable Care Act or whatever health reform measure might replace it.

Over the past fifty years, courts have played a limited, yet key, role in shaping health equity in the United States in three areas of law: racial discrimination, disability discrimination, and constitutional rights. In this article, I examine in various ways the roads courts have taken, the roads not taken, and possible future paths.

Racial Discrimination

Desegregating Hospitals

Despite the troubling racial disparities that exist in many areas of health care delivery, it is important to reflect on the substantial improvements in equitable access that occurred in prior generations. These improvements are largely attributable to the civil rights era, including its signature law, the Civil Rights Act of 1964. Prior to this law, at the end of the 1950s, most northern hospitals were integrated but almost no southern hospitals were (Reynolds 1997). Instead, in the South, hospitals almost uniformly either refused admission to blacks, sending them to separate, but definitively not equal, black-only facilities, or admitting them only to segregated wards. Even in the North, segregation existed in terms of medical staff membership. Most hospitals across the country would not accept African American interns or residents, and only a quarter would grant them staff privileges (Reynolds 1997).

By 1966, however, 85 percent of hospitals had desegregated and were no longer refusing black physicians (Reynolds 1997). Within a few additional years, overt racial discrimination among patients became unheard of. As the leading scholar summarizes (Smith 2015), within a remarkably short time span, US hospitals “went from the nation's most segregated private institutions to its most integrated.” As Rosenbaum and Schmucker (2017) report in this issue, “intentional, legal segregation has disappeared from the scene of publicly funded programs and services.”

These remarkable gains occurred mainly as a result of administrative actions and broader social movements, however, rather than resulting from court decisions. Whereas judicial leadership was needed to desegregate schools and other important social and government institutions, hospital integration occurred primarily as a result of regulatory and quasi-regulatory action. David Barton Smith recounts that, following the 1964 enactment of the Civil Rights Act, early enforcement efforts were targeted to southern hospitals. Building on that, Medicare's enactment in 1965 put pressure on all hospitals to comply with antidiscrimination norms that were imposed as a condition of hospitals' eligibility for Medicare funding. The Johnson administration's eagerness to ensure widespread hospital participation led it to enlist the support of the American Hospital Association (AHA) to convince member hospitals of the need to abandon past discriminatory practices. The AHA's supportive position also led to private enforcement of antidiscrimination standards through the Joint Commission on Hospital Accreditation.

Although administrative and accreditation authorities led the way to hospital desegregation in the 1960s, a key court decision helped to clear the path. Earlier federal law also prohibited racial discrimination by hospitals that had received construction funding from the Hill-Burton Act of 1946, but the enforcing agency maintained that “separate but equal” facilities constituted compliance. That position was successfully challenged in Simkins v. Cone (323 F.2d 959 [1963]), where the Fourth Circuit ruled that the government's separate-but-equal position was unconstitutional. When the Supreme Court declined to hear the case the following year, that action was widely viewed as endorsing the ruling. The prominence of this decision gave support to legislative efforts in Congress to adopt the Civil Rights Act soon thereafter.

Exempting Physicians and Disparate Impact

Following the rapid integration of hospitals in the 1960s, courts have not continued to open new routes for addressing racial discrimination in health care. Limited judicial action under the civil rights laws has been due in large part to the roadblocks established by the governing law and administrative interpretations, as detailed in Rosenbaum and Schmucker's article (2017). Principal among these has been the federal government's interpretation of Title VI of the Civil Rights Act that limits its application to physicians. Title VI prohibits discrimination based on “race, color, or national origin” by “any program or activity receiving federal financial assistance.” Hospitals became subject to Title VI mainly due to Medicare and Medicaid funding, but the same has not been true for physicians. Federal agencies have consistently ruled that Medicare payments to physicians do not trigger application of Title VI because Part B of Medicare constitutes a “contract of insurance,” which Title VI excludes from its definition of “financial assistance” (Crossley 2003). The same argument would not necessarily apply to Medicaid funding, but enforcement agencies failed to declare that Title VI applies to physicians who receive Medicaid funding, and the issue has not been litigated in court.

As a result, it has generally been assumed that physicians are free to discriminate on any basis they like due to the absence of express statutory authority stating otherwise (Orentlicher, Bobinski, and Hall 2013). Walker v. Pierce (560 F.2d 609 [4th Cir. 1977]), is a court decision that illustrates this gap in the law. There, two black female patients covered by Medicaid sued a South Carolina obstetrician for insisting that they be sterilized after delivering their children. The physician explained that, regardless of race, he had a personal policy of not treating women who refused to be sterilized after their third child if they “were unable to financially support themselves.” The court ruled that there is “no reason why Dr. Pierce could not establish and pursue the policy he has publicly and freely announced. Nor are we cited to judicial precedent or statute inhibiting this personal economic philosophy.”

The Walker case illustrates the extent to which patients are not protected from what are clearly inappropriate reasons for a physician to deny care. Although the physician's policy was targeted at women, no law at the time prohibited gender discrimination by physicians. And, although his policy likely had a disproportionate impact on blacks, civil rights laws still did not clearly apply—for reasons noted in the following section. In short, patients are not protected from morally problematic denials of care by physicians in the absence of a statute granting protection.

Judicial expansion of civil rights protections has also been blocked by a key Supreme Court decision regarding disparate impact. In Alexander v. Sandoval (532 U.S. 275 [2001]), the Court ruled that private parties may sue only for “disparate treatment,” meaning intentional discrimination, but not for “disparate impact,” which covers actions that, although neutral on their surface with regard to race, disproportionately affect minorities. The Court did not foreclose enforcement agencies taking action based on disparate impact, but it ruled that any agency regulations using that approach do not, in themselves, allow private enforcement using a disparate impact theory, and the statute itself allows private enforcement only for intentional discrimination. To date, the relevant enforcement agencies have not pursued disparate impact approaches in the health care arena.

A prominent example of this refusal to extend the reach of Title VI is the rejection of challenges to the location of health care facilities. When hospitals serving inner-city populations, for instance, were closed or moved to more affluent suburbs, lawsuits challenged those location or closure decisions based on the racial composition of the affected populations (Lado 1994; Watson 1990). Courts have consistently rejected these challenges, however, ruling that, absent actual discriminatory intent, merely a disparate impact on minorities does not constitute prohibited discrimination as long as the decision has a plausible, nondiscriminatory justification. See, for instance, Bryan v. Koch (627 F.2d 612 [2d Cir. 1980]; allowing closure of a public hospital in Harlem, for budgetary reasons) and NAACP v. Wilmington Medical Center (657 F.2d 1322 [3d Cir. 1981]; finding no discriminatory intent in the decision to relocate an inner-city Wilmington, DE, hospital to the suburbs).

Another path untaken relates to racial disparities in treatment. Voluminous evidence documents that blacks, as a group, consistently fail to receive the same level of care provided to whites across a broad range of conditions and treatments. Seldom, however, is this due to overt discrimination. Instead, divergent treatment patterns result from subtle forms of implicit or unconscious bias, as well as a variety of other socioeconomic factors that are not necessarily attributable to the care provider. No court decisions have yet considered whether disparate treatment patterns violate civil rights laws, but legal scholars believe that any such challenges would almost certainly fail under current law, again due to the absence of actual discriminatory intent (Crossley 2003; Matthew 2015).

Potential Changes under the ACA

Potentially, the Affordable Care Act, if it survives, would remove, or at least lower, several of these roadblocks to more expansive civil rights enforcement. Section 1557 of the ACA is a sweeping antidiscrimination provision that applies existing bodies of federal antidiscrimination law—covering race, sex, age, and disability, among other conditions—to any “health program or activity” that receives federal funding. Included are not only health care facilities, but also physicians and insurers. Discrimination is prohibited not only in the specific activity or program that receives federal funding, but also in any and all health-related activities conducted by these recipients of federal funding.

In its implementing rules, the US Department of Health and Human Services (HHS) has adopted some expansive positions that, if upheld by the courts and not undone by the Trump administration, go substantially beyond previous civil rights law (Benge 2016). Most significantly, the rule “interprets Section 1557 as authorizing a private right of action for claims of disparate impact discrimination on the basis of any of the criteria enumerated in the legislation” (81 Fed. Reg. 31375, May 18, 2016). Essentially, this reads the ACA as reversing the Supreme Court's Alexander v. Sandoval decision noted above. It can be expected that a ruling this bold will be challenged in court, but courts usually defer to an agency's interpretation of its own governing statute, if reasonable. Here, the agency noted that, although some parts of civil rights law previously did not allow private claims for disparate impact, other parts did, such as the law governing sex or age discrimination. The agency reasoned that, because section 1557 adopts a uniform nondiscrimination standard, it would not make sense to allow disparate impact claims for some forms of discrimination but not others. A different administration could take a different view, however.

A second potential expansion relates to coverage of physicians. Here, the agency's current rule is more equivocal. Without meaningful explanation, the agency declined to change its existing position that Medicare Part B payments do not constitute federal “financial assistance.” This surprised some commentators who thought it was obvious that section 1557 reverses this long-standing omission. That omission was based in large part on the exclusion from Title VI of “contracts of insurance,” but section 1557 explicitly includes insurance contracts; indeed, one of its main purposes is to apply discrimination law to the many private insurers that the ACA now funds. The agency briefly noted the existence of this and other arguments against its exclusion of Part B payments, but opaquely stated that it “does not believe that this rule is the appropriate vehicle to modify the Department's position.”¹

Nevertheless, the agency noted various expansive ways in which section 1557 covers physicians through other sources of federal funding. Most notably, the agency declared that participation in Medicaid constitutes “federal financial assistance,” even though those funds are paid to physicians only by states. Federal enforcement agencies had not previously made a clear statement that Medicaid's form of indirect federal financing triggers federal discrimination law. Again, that ruling might be subject to challenge in court, or might be revisited if Medicaid is converted to more of a program of block grants to the states; if upheld, though, this position will substantially expand the potential scope of discrimination law enforcement in the future. Of greatest significance, combining the two areas of expansion noted here, it could become possible to argue that physicians violate discrimination law when, without good justification, they make different treatment recommendations for minorities than for whites, or for women than for men (Blake 2016; Steege 2011).

Disability Discrimination

Disability discrimination law is a second major area of civil rights legal protection. There are two central federal disability laws, § 504 of the Rehabilitation Act of 1973, which governs federally funded programs, and the Americans with Disabilities Act of 1990 (ADA), which governs state and local programs and key aspects of the private sector (employers and “public accommodations”). Courts have been circumspect, but not entirely quiescent, in using these laws to shape decisions in two broad arenas: (1) medical treatment in individual cases, and (2) decisions about resource allocation in designing benefits covered by health insurance.

In brief summary, these laws prohibit discrimination on the basis of disability when the person in question is qualified for the service or program despite the disability, or when any disqualifying aspects of a disability can be overcome through “reasonable accommodations.” Because disability is defined broadly in a manner that includes many medical conditions (such as having a chronic or infectious disease that limits a major life activity, Bragdon v. Abbott, 524 U.S. 624 [1998]), these laws have been used to challenge decisions relating to a variety of health conditions, made by both medical care providers and by health care insurers (public and private).

Like Title VI of the Civil Rights Act, the original disability law (the Rehabilitation Act of 1973) applied to only federally funded programs, and thus did not reach physicians. The ADA, however, overcame this limitation in 1990 by extending protections to all businesses and services available to the general public, which includes physicians in private practice. The ADA also covers both public and private insurers. Thus, this body of law has received more judicial development than has the more limited protection against racial discrimination.

Treatment Decisions

U.S. v. University Hospital (729 F.2d 144 [2d Cir. 1984]) is an early leading decision that wrestled with the extent to which these laws apply to medical treatment decisions. It involved an infant born with spina bifida—a severe birth defect in which the baby has a fully exposed spinal cord, an unusually small head, and excessive cranial fluid. At the time, the condition was often fatal and, even now, those who survive usually suffer from severe neurological and cognitive impairments. Based on prognosis, the parents declined aggressive treatment. The Reagan administration became interested in the case and requested records from the hospital. When it refused, legal proceedings ensued.

The federal circuit court upheld the hospital's refusal to submit to federal investigation, ruling that its decision not to overrule the parents' refusal of treatment did not constitute potential disability discrimination. The core rationale was that, even though Baby Jane Doe was obviously disabled, disability discrimination law scrutinizes denial of medical service only when that is done because of another disabling condition that is unrelated to the condition for which treatment is being sought. For instance, disability law would apply if a patient were denied a liver transplant simply because she is blind, but not if a liver transplant were denied to a chronic alcoholic or someone with a severe immune system disorder (even though these conditions are recognized disabilities).

The University Hospital court was concerned that discrimination on the basis of disability is much more complicated an issue than discrimination on the basis of race. Race is essentially never an acceptable basis on which to deny services, but in medicine it obviously is often acceptable and even desirable to make decisions based on the patient's full set of medical conditions. As the court noted, it is not easy to decide whether a denial of treatment is unfairly discriminatory or is based on legitimate consideration of the patient's disability: “It would invariably require lengthy litigation primarily involving conflicting expert testimony to determine whether a decision to treat, or not to treat, or to litigate or not to litigate, was based on a ‘bona fide medical judgment,’ however that phrase might be defined” (U.S. v. University Hospital [729 F. 2d 144]).

Thinking about these concerns, courts could take one of two positions. They could simply declare that disability discrimination law does not apply at all to medical treatment decisions, but that would not square easily with the statute's plain language or its legislative history. Instead, the University Hospital court limited application of these laws to “only where the individual's handicap is unrelated to, and thus improper to consideration of, the services in question.” The distinction between conditions that are independent of the disability and those that are related is based on the ground that antidiscrimination law is designed to ensure that similarly situated persons receive the same treatment. When the need for treatment arises out of the disability, as when a severely disabled newborn needs a surgical procedure, there is no similarly situated person without the disability who needs the same surgical procedure.

The question shifts, then, to how exactly to interpret this concept of “relatedness.” Because the human body is a highly integrated organism and not compartmentalized into relatively independent parts, a disabling illness generally has a wide-ranging effect on that person's needs for, or potential benefit from, medical treatment. Consider the example of HIV infection. Such patients may have higher risks of complications from any surgical procedure, but it is often feasible to accommodate that higher risk. What should a court do, however, if a medical provider would prefer not to undertake those efforts?

Glanz v. Vernick (756 F. Supp. 632 [D. Mass. 1991]) is an illustrative decision. The doctor refused to perform a simple procedure to alleviate a patient's earaches, stating that the patient's HIV infection posed too great a risk—both to the patient and the surgeon. The court ruled that this position “requires an individualized inquiry and appropriate findings of fact” about the surgeon's stated justification. Although the court noted that there is “some merit to the argument that the court should defer to a doctor's medical judgment, . . . [a]ccepting this argument at face value . . . would completely eviscerate [the law's] function of preventing discrimination against the disabled in the healthcare context. A strict rule of deference would enable doctors to offer merely pretextual medical opinions to cover up discriminatory decisions. The evidentiary approach [required] . . . properly balances deference to sound medical opinions with the need to detect discriminatory motives.”

The Supreme Court endorsed this fact-based inquiry in Bragdon v. Abbott (524 U.S. 624 [1998]), a case in which a woman with HIV infection challenged her dentist's refusal to fill her cavity unless he performed the procedure in a hospital (she lacked insurance to pay for hospital-based dental work). The Court ruled that, in order for the dentist to insist on a hospital setting, he would need to show that office treatment posed “a significant risk to [his] health or safety . . . that cannot be eliminated by . . . provision of auxiliary aids or services.” In assessing whether the dentist's fear of HIV transmission was objectively reasonable, “the views of public health authorities, such as the U.S. Public Health Service, CDC, and the National Institutes of Health, are of special weight and authority. The views of these organizations are not conclusive, however. A health care professional who disagrees with the prevailing medical consensus may refute it by citing a credible scientific basis for deviating from the accepted norm.” On remand, the circuit court found in favor of the patient, based on evidence about the effectiveness of universal precautions to prevent transmission of HIV infection (Abbott v. Bragdon 163 F.3d 87 [1st Cir. 1998]).

Rather than refusing treatment outright, a provider might seek to refer a more difficult patient to a specialist whom the provider believes is better suited to meet the patient's needs. Referring patients poses similar issues about the reasonableness and evenhandedness of the provider's rationale, but the same court that decided the remand in the dentist case just mentioned later ruled that it would not closely scrutinize an obstetrician's decision to refer an HIV-infected pregnant woman to a specialized HIV pregnancy program for drug therapy designed to prevent transmission of HIV to the woman's child; see Lesley v. Chie (250 F.3d 47 [1st Cir. 2001]). Considering “the extent to which a court should defer to a physician's claim that he lacks the [necessary] experience, knowledge, or other prerequisites,” the court found no legal violation, writing that, to prevail, a patient would need to show “the decision to be devoid of any reasonable medical support.” “For example, a plaintiff may argue that her physician's decision was so unreasonable—in the sense of being arbitrary and capricious—as to imply that it was pretext for some discriminatory motive, such as animus, fear, or ‘apathetic attitudes.’”

This court went on to explain, however, that “instead of arguing pretext, a plaintiff may argue that her physician's decision was discriminatory on its face, because it rested on stereotypes of the disabled rather than an individualized inquiry into the patient's condition.” This effort to root out disability stereotypes contrasts with the Walker v. Pierce case above, which disturbingly upheld a physician's right to insist that two African American mothers be sterilized in order to accept them as Medicaid patients. When physicians adopt such racial stereotypes, they cannot currently be under federal discrimination law simply because the Civil Rights Act, unlike disability discrimination law, has not yet been applied to physicians in private practice.

Embedded in legal analyses of disability discrimination is the law's requirement to make “reasonable accommodations” that will avoid the need to refuse service. That concept too is challenging in the context of medical treatment decisions. Courts might say that if there are extra services that could improve a disabled person's ability to benefit, and the extra services are not too costly, they must be provided. Thus, if a person with a psychiatric illness needs a kidney transplant and psychiatric counseling would ensure that the patient complies with the medications and follow-up appointments, the clinic would have to provide the counseling.² But, what if the counseling does not appear to be working, or is only partially helpful?

A more expansive view of the law's purpose would emphasize the need to improve conditions for people with disabilities rather than simply avoiding discriminating against them. Thus, even though people with disabilities unavoidably may not benefit from medical treatment as much as other persons, making reasonable accommodations might mean that we do not disfavor a person merely because their disability causes them to realize less benefit from treatment than other persons. These flexible and judgmental concepts are difficult to apply, however, and courts are reluctant to do so in ways that interfere with prevailing medical standards of care. As a result, courts so far have not been active in using disability discrimination law to scrutinize medical treatment decisions.

Insurance Coverage

Another arena where disability discrimination law potentially applies is the design of health insurance coverage, either public or private. Here, we consider not just whether a particular patient can benefit from a specific treatment. That might be the issue under a garden variety “medical necessity” dispute. Beyond that, we also consider whether disability discrimination occurs when health insurance designs its coverage in a manner that excludes or limits a category or class of treatments. For instance, insurance might limit or exclude what or when it pays for mental health care or for organ transplants. Such decisions can be based on both considerations of cost—how expensive it is to cover such treatments—and on medical need and benefit—that is, whether there are higher priorities to which limited resources should be devoted.

On this front, there are two leading Supreme Court cases. Alexander v. Choate (469 U.S. 287 [1985]) rejected a disability discrimination challenge to Tennessee's decision to limit its Medicaid coverage for hospitalization to a maximum of two weeks a year. The limit was imposed to help offset a budget deficit. The challengers objected that this limit would have a disproportionate impact on people with disabilities, who need more extended care. Acknowledging that the law does not require a showing of discriminatory animus, still, the Court sought to avoid an application of the law that “would in essence require each recipient of federal funds first to evaluate the effect on the handicapped of every proposed action that might touch the interests of the handicapped, and then to consider alternatives for achieving the same objectives with less severe disadvantage to the handicapped. The formalization and policing of this process could lead to a wholly unwieldy administrative and adjudicative burden.”

The Court thus rejected the discrimination challenge, reasoning in part that the two-week limit is “neutral on its face [in that it] does not distinguish . . . on the basis of any test, judgment, or trait that the handicapped as a class are less capable of meeting or less likely of having. . . . [Instead, the] reduction in inpatient coverage will leave both handicapped and non-handicapped Medicaid users with identical and effective hospital services fully available for their use, with both classes of users subject to the same durational limitation.” The Court also rejected any notion that states must alter their Medicaid benefits “to meet the reality that the handicapped have greater medical needs. To conclude otherwise would be to find that the [law] requires states to view certain illnesses, that is, those particularly affecting the handicapped, as more important than others and more worthy of cure through government subsidization.”

In contrast, in Olmstead v. L.C. (527 U.S. 581 [1999]), the Court found a potential ADA violation where a state Medicaid plan covered long-term hospitalization for mental illness but not less restrictive community placement options. The Court reasoned that these patients could be “reasonably accommodated” if the state could expand its coverage without straining the budget for other mental health services. Going even further, Lovell v. Chandler (303 F.3d 1039 [9th Cir. 2002]) found an ADA violation where the state expanded eligibility for Medicaid up to three times the poverty level but excluded disabled participants from the increased eligibility parameters, even though the state said this was all it could afford.

Olmstead reached a different result than Choate mainly because the restriction in Olmstead targeted a class of patients essentially all of whom were disabled (those with mental illness serious enough to require hospitalization); thus, it was a case of discriminatory treatment rather than merely disparate impact. But what about programs that limit mental health treatment more generally or broadly? In Choate the Supreme Court cited approvingly a second circuit opinion holding that a broad limitation of mental health coverage does not constitute disability discrimination. The ostensible reason is that, although some mental illness is severe enough to be disabling, not all mental illness is such, and thus limiting all such treatment is neutral with regard to disability, even though a broad-based limitation might have a disproportionate impact on those with mental or emotional disabilities.³

Going even further, Modderno v. King (82 F.3d 1059 [D.C. Cir. 1996]) ruled that lesser coverage of mental health treatment is not disability discrimination because to rule otherwise would be “to invite challenges to virtually every exercise of the [insurer's or employer's] discretion with respect to the allocation of benefits amongst an encyclopedia of illnesses.”⁴ And, in Doe v. Mutual of Omaha Insurance Company (179 F.3d 557 [7th Cir. 1999]), the court upheld a specific cap on treatment of AIDS-related conditions that limited coverage to just $25,000, noting that, if it were to strike this cap, then some equally or more serious diseases could still be capped but others not, according arbitrarily to whether or not the disease happens to entail a disability. On the other hand, Henderson v. Bodine Aluminum (70 F.3d 958 [8th Cir. 1995]) ruled that denying coverage for expensive new therapy for breast cancer is potentially discriminatory where the plan covered this treatment for other cancers and there is evidence that it works for breast cancer.

In sum, disability discrimination law, like racial discrimination law, primarily targets differentiation that is expressly based on the protected trait, rather than decisions neutral on their face that have a disparate impact on the protected group. For racial discrimination, this is a major limitation in the law because policies rarely differentiate based on skin color, ethnicity, or the like. However, under disability discrimination law, cases of explicit differentiation arise more frequently because the criteria used to define disability are often ones that have relevance for medical decisions.

Reconciling Competing Goals

Clearly, the operative concepts here are difficult for courts to apply. Moreover, they appear to lead to potentially perverse results for broader health policy. According to one law professor (Crossley 1995), “The ADA is an inadequate and even inept tool for resolving whether we should tolerate cost-conscious [insurance] policies” because its concepts are so poorly suited for articulating and understanding the underlying social policy debate.

Recognizing large scale that, in a world of limited resources, some limits on coverage are inevitable, the fundamental issue is how such “rationing” should be done. Setting fewer limits means that more people will be entirely uninsured, leading to more disability or less treatment of disability overall. Setting some limits thus can make treatment more available generally, both for those with and without disabilities. But, if limits are set coarsely, without regard to medical condition, then such “meat ax” limits receive less or no scrutiny under disability law, because they are viewed as more neutral.

More refined criteria for setting limits, however, run a greater risk of encountering discrimination scrutiny, for the very reason that they include more medical specificity. If allocation criteria are specified that are unrelated to disabling conditions, then discrimination law does not come into play, but it is difficult, perhaps impossible, for rational allocation criteria in health care to entirely avoid addressing medical criteria of need, risk, and benefit without including some criteria that relate directly to disability broadly defined. The same can also be true for other legally protected criteria such as age or gender. When that occurs, courts face the daunting challenge of deciding when such criteria are and are not appropriate.

Against this backdrop of uncertainty, consider how much clarity there is in the ACA's legislative language that forbids designing “benefits in ways that discriminate against individuals because of their age, disability, or expected length of life,” or that are “subject to denial . . . on the basis of the individuals' . . . present or predicted disability, degree of medical dependency, or quality of life.” This appears aimed at prohibiting the government's use of more refined health policy concepts such as “quality-adjusted life years” (or QALYs) to define the standard set of “essential health benefits” that private insurers must cover (in the individual and small group markets). However, another ACA provision says that “nothing . . . shall be construed to prohibit . . . [health insurers] from carrying out utilization management techniques that are commonly used as of the date of enactment of this Act.” Undoubtedly, these concepts and their competing objectives will continue to challenge courts and administrative agencies for decades to come.

Constitutional Protections

So far, we have reviewed only statutory or regulatory protections. More fundamental are the civil rights protections built into the US Constitution. Here, the focus expands from protecting rights in health care to also protecting rights to health care. How patients are treated within the health care system is important, but perhaps more important is whether people have access to any kind of health care. Constitutional doctrine potentially addresses both dimensions of individual rights.

Constitution-Free Zone

Much of health law is a constitution-free zone, meaning that, so far as the federal Constitution is concerned, lawmakers are not required to protect or advance health, nor are they constrained, for the most part, in how they choose to go about doing so. This broad statement is subject to several very important qualifications—relating to procedural due process and to specially protected privacy interests, among other doctrines—but, these exceptions aside, the US Constitution is silent about health, which leaves lawmakers free to do just about anything they want or nothing at all. For a time, there was thought that Roe v. Wade (410 U.S. 113 [1973]), could expand into a more generalized constitutional freedom in important medical decision making, because that decision stressed the freedom of doctors and patients to exercise medical judgment without state interference when fetuses are nonviable. Subsequent abortion decisions, however, have phrased the protected right solely in terms of the woman's individual interest in avoiding procreation.

Most government positions on health care are easily justified under the Constitution because the Constitution has no general protection for individuals' “pursuit of health.” As constitutional lawyers put it, our Bill of Rights is a charter of negative, not positive, liberties, meaning that the government constitutionally owes us nothing as long as it leaves us alone (DeShaney v. Winnebago County Department of Social Services, 489 U.S. 189 [1989]). Wideman v. Shallowford Community Hospital (826 F.2d 1030 [11th Cir. 1987]) provides a good example of this reasoning. There, a county ambulance service took a woman in labor to its preferred hospital rather than to the hospital where the woman's obstetrician was to meet her, resulting in a delay in care that the woman claimed caused the death of her infant child. The court rejected any claim based on denial of constitutional rights because the county did not owe any obligation to transport her at all; therefore, it cannot be held responsible for transporting her to the wrong hospital.

This case was argued under the constitutional doctrine of “due process.” Also potentially relevant is the doctrine of “equal protection.” Without turning this summary into a constitutional treatise, it is helpful to sketch these basic constitutional concepts. In the early part of the last century, the Supreme Court took an “activist” approach to reviewing the constitutional validity of state economic and social regulation under the due process clause. In Lochner v. New York (198 U.S. 45 [1905]), the Court struck down a state's regulation of maximum work hours as a violation of the fundamental right to contract. However, the onslaught of programs in the late 1930s designed to ameliorate the Depression caused the Court to retreat by substituting a more deferential, rational basis standard of review of state economic regulation. Thereafter, the Lochner era of “substantive” or “economic” due process was thoroughly repudiated as a valid form of judicial review for most social regulation. It survived only for legislation that affects special categories of “fundamental” interests or liberties.

Health laws can also be challenged under the equal protection clause, since they inevitably draw lines or distinctions between permitted and impermissible activities or actors. Any such class-based legislation will be reviewed under one of three standards: (1) suspect classifications (those based on race, for example, or those that intrude on other fundamental rights) will be subjected to strict scrutiny; (2) quasi-suspect classifications (such as gender) will receive intermediate scrutiny; and (3) all other legislative classifications will be reviewed under the rational basis standard. Because most health laws fall within this last, broad generic category of social and economic legislation, they usually receive only light constitutional scrutiny. States, may, for instance, draw fine and contentious distinctions, such as funding some abortions but not others (Maher v. Roe, 432 U.S. 464 [1977]; Harris v. McRae, 448 U.S. 297 [1980]), or permitting palliative care for a patient who refuses life support at the same time that the state criminalizes physician-assisted suicide (Vacco v. Quill, 521 U.S. 793 [1997]; Washington v. Glucksberg, 521 U.S. 702 [1997]).

Courts have been especially deferential to a state's power to protect public health. They have upheld states' public health actions so long as they are not “arbitrary” or “unreasonable” or “unnecessary” to protect public health. Thus, for instance, courts have upheld the constitutionality of every form of professional and facility licensure—from banning alternative practitioners to barring inefficient facilities.⁵ Other, everyday health laws that have passed constitutional muster include those that require autopsies, that allow the removal of corneas for transplantation, and, most dramatically, that redefine the very essence of death, and therefore life (State v. Powell, 497 So. 2d 1188 [Fla. 1986]; State v. Schaffer, 574 P.2d 205 [1977]). Despite their obvious and sometimes profound impact on individual liberties, they require no extraordinary justification under prevailing constitutional analysis.

Instead of a sweeping right to pursue health, “fundamental rights” under current doctrine are limited to bodily integrity—that is, refusing unwanted treatment (e.g., Cruzan v. Director, Missouri Department of Health, 497 U.S. 261 [1990]), and certain specially protected “privacy” arenas such as procreation and parenting (Griswold v. Connecticut, 381 U.S. 479 [1965]; Planned Parenthood of Southeastern Pa. v. Casey, 505 U.S. 833 [1992]). In Washington v. Glucksberg (521 U.S. 702), the Court explained that, “in addition to the specific freedoms protected by the Bill of Rights, the ‘liberty’ specially protected by the due process clause includes the rights to marry, to have children, to direct the education and upbringing of one's children, to marital privacy, to use contraception, to bodily integrity, and to abortion. We have also assumed, and strongly suggested, that the due process clause protects the traditional right to refuse unwanted lifesaving medical treatment.” Also, persons subjected to state confinement have special constitutional protections (O'Connor v. Donaldson, 422 U.S. 563 [1975]), and health laws can implicate the First Amendment right to the free exercise of religion, the Fourth Amendment right to be free from unreasonable searches and seizures, and the right to “just compensation” if the government takes private property (Hall 2009).

Despite this tapestry of protected arenas, coercive health laws that impinge these freedoms are often justified because they serve a compelling public interest and are narrowly tailored to meet that interest. Accordingly, courts have repeatedly upheld invasions of these strong substantive protections in order to promote either individual or public health. With appropriate safeguards, the government may, for instance, require smallpox vaccinations (despite the inevitable risks), commit psychiatric patients to forced treatment, force-feed comatose patients who have not clearly refused such treatment, quarantine people with infectious disease, or intervene surgically to save a full-term fetus.⁶ In each instance, the justifications and analyses differ, and there are limits to what the government can require, but these precedents are notable for their permissive breadth.

Legislatures also gain considerable constitutional leeway when they condition government spending or privileges on obeying health policy requirements. Therefore, laws that might not be upheld standing alone are easily upheld if they are imposed as qualifications for receiving optional government benefits. On this basis, for instance, the federal government requires hospitals to treat emergency patients for free and it once forbade Planned Parenthood from discussing abortions (Burditt v. U.S. Department of Health and Human Services, 934 F.2d 1362 [5th Cir. 1991]; Rust v. Sullivan, 500 U.S. 173 [1991]). Similarly, there are no constitutional issues created by setting Medicare rates too low,⁷ and states can require doctors to accept Medicaid patients at reduced rates.⁸

Potential Constitutional Restrictions

Some conservative or libertarian justices and constitutional scholars call for fundamental change to this conventional constitutional regime. They would revive some version of the economic or substantive due process approach of the Lochner era by expanding the range of fundamental interests or tightening the justifications for restrictions of individual liberty. For instance, a panel of the DC Circuit sent shock waves through the health policy establishment with its 2006 decision in Abigail Alliance v. Eschenbach (445 F.3d 470 [D.C. Cir. 2006]), holding that the Food and Drug Administration (FDA) must make experimental drugs more readily available to terminally ill patients for whom there are no other therapeutic options. Two of the three judges reasoned that seeking medical treatment that might save one's life is a fundamental right, the restriction of which was not adequately justified here. The full court reversed this decision a year later, but the original decision still reverberates.

Legal scholars have also noted the Canadian Supreme Court's 2005 decision in Chaoulli v. Quebec (1 S.C.R. 791), striking down Quebec's ban on private health insurance that duplicates public coverage. Using reasoning under the Quebec Charter of Human Rights and Freedoms that broadly tracks US constitutional analysis under our Bill of Rights, the court reasoned similarly to the original Abigail Alliance decision that fundamental interests in pursuing health are at stake and that a sweeping ban on insurance is too broad.

Despite this noticeable undercurrent, substantial change in constitutional analysis of health care regulation is not likely to take hold in the United States any time soon. Only two of the DC Circuit Court's thirteen judges adhered to the original decision in Abigail Alliance, and the Canadian court's scrutiny under Quebec's Charter is much more aggressive than what one would expect under the US Constitution. As noted above, pursuing health has not been recognized as a fundamental right. Instead, US courts are inclined to characterize asserted rights in extreme or technical ways that tend to defeat their being constitutionalized. For instance, the medical-aid-dying case, Washington v. Glucksberg (521 U.S. 702), characterized the right at issue as receiving assistance in committing suicide, rather than choosing a humane manner to die. Similarly, the full court in Abigail Alliance, the cancer treatment case, characterized the right at stake as access to investigational drugs rather than pursuing all available means to avoid death. Despite inevitable shifts in the political and social views of the federal judiciary, there is not likely to be any fundamental reversal in these basic attitudes.

Those who might hope for more invigorated constitutional scrutiny of rights to health care should be cautious about the potential this might have to stymie government efforts to advance health policy goals. Constitutional rights can be a two-edged sword (or a two-direction shield). They not only can protect individuals from unfair treatment by government, they can also limit the range of actions that government may take. For instance, potential constitutional limitations arise under the First Amendment. In 2012, two different appellate courts ruled that the First Amendment limits requirements the FDA imposes in two prominent areas: (1) graphic warning labels on cigarette packages (R.J. Reynolds Tobacco Co. v. U.S. Food & Drug Administration, 696 F.3d 1205 [D.C. Cir. 2012]), and (2) restrictions on drug manufacturers' marketing of “off-label” uses of their products (United States v. Caronia, 703 F.3d 149 [2d Cir. 2012]).

These disputes have not yet reached the Supreme Court, but the potential for using the Constitution to restrict health policy is clearly seen in the Court's famous Affordable Care Act decision, NFIB v. Sebelius (132 S. Ct. 2566 [2012]). There, the law's “individual mandate” barely survived, but only because a majority of the Court ruled it is valid as a tax on a voluntary choice to be uninsured. On the other hand, a majority also ruled that the federal government may not require the purchase of insurance as a simple regulatory command. Whether that restriction impacts other important areas of health policy remains to be seen. But, even as a singular precedent, the Court's ACA case reminds us that, under long-prevailing doctrine, US constitutional law easily can do more to restrain than to promote equitable access to care.

Conclusion

This review demonstrates that courts have had only a limited role in shaping health equity in the United States. Across several bodies of law, the major achievement has been the rapid desegregation of hospitals in the 1960s, but that was attributable much more to administrative pressure under Medicare, as well as larger political and economic forces, than to judicial edicts under the Civil Rights Act. Otherwise, various civil rights protections, including disability discrimination laws, have had limited impact because the ability to seek judicial enforcement is strongest for overt, intentional discrimination, which is largely absent in health care in the modern era. The availability of judicial enforcement is much weaker for the forms of disparate impact that currently are more pervasive. And, constitutional rights have limited impact in health care beyond areas of special protections such as procreation.

It is open to debate whether or not these existing limitations in civil rights laws are inherent or might be subject to change. Wise health policy requires differentiating among patients, conditions, and treatments for various reasons. The core principle of equal protection is not only to treat like cases alike, but also to treat differences appropriately. Courts express reservations about using civil rights or constitutional law too aggressively to question professional and institutional decisions about which differentiations in health care are and are not appropriate.

Based on this reluctance, some legal and social scholars (e.g., Roberts 2013) believe that more progress toward health equity can be made by working outside of, rather than through, existing civil rights laws. Civil rights laws are well suited to protect against individual cases of invidious treatment. These laws are less well suited to producing more sweeping institutional and social changes, apart from overt segregation among patients, which has largely disappeared. Since the 1960s, broader health equity goals have been advanced much more by the introduction and expansion of health insurance and health care access programs that cover broad segments of the population than by enforcement of civil rights laws. For instance, when Medicare was first enacted, per capita medical spending for minorities was 26 percent less for hospital care, and 40 percent less for physician care, than spending for whites, but a generation later, per capita spending was substantially higher for minorities than for whites (Smith 2015).

This major correction came about not through civil rights laws, but by greatly increasing insurance coverage population-wide, which disproportionately benefits those who were most disadvantaged. Similarly, the Affordable Care Act's initial expansion of coverage and access, if sustained, would have much more positive impact on minorities than on the general population (Chen et al. 2016). In its first two years, the ACA produced insurance gains that were 50 to 100 percent greater among Hispanics, blacks, Native Americans, and Asians than among whites (Tavernise and Gebeloff 2016). Also, the ACA's prohibition of medical underwriting and coverage of preexisting conditions directly benefited people with disabilities (Rosenbaum, Teitelbaum, and Hayes 2011).

Realizing these greater gains from lifting all boats should not deter us from vigilance in continued enforcement of antidiscrimination laws against intentional discrimination, wherever that exists. Nevertheless, efforts to use these laws to change the systemic effects of facially neutral policies might ultimately produce fewer gains than continued efforts to improve health care quality and access across the board.

Acknowledgments

I am grateful to David Orentlicher for permission to base the section on “Disability Discriminiation—Treatment Decisions” partially on unpublished materials that he originally wrote for the teacher's manual for our casebook (Hall, Bobinski, and Orentlicher 2013).

Notes

References