OLMSTEAD V. L. C. (98-536) 527 U.S. 581 (1999)
138 F.3d 893, affirmed in part, vacated in part, and remanded.
[ Ginsburg ]
[ Stevens ]
[ Kennedy ]
[ Thomas ]
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Kennedy, J., concurring in judgment


No. 98—536



[June 22, 1999]

Justice Kennedy, with whom Justice Breyer joins as to Part I, concurring in the judgment.


Despite remarkable advances and achievements by medical science, and agreement among many professionals that even severe mental illness is often treatable, the extent of public resources to devote to this cause remains controversial. Knowledgeable professionals tell us that our society, and the governments which reflect its attitudes and preferences, have yet to grasp the potential for treating mental disorders, especially severe mental illness. As a result, necessary resources for the endeavor often are not forthcoming. During the course of a year, about 5.6 million Americans will suffer from severe mental illness. E. Torrey, Out of the Shadows 4 (1997). Some 2.2 million of these persons receive no treatment. Id., at 6. Millions of other Americans suffer from mental disabilities of less serious degree, such as mild depression. These facts are part of the background against which this case arises. In addition, of course, persons with mental disabilities have been subject to historic mistreatment, indifference, and hostility. See, e.g., Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432, 461—464 (1985) (Marshall, J., concurring in judgment in part and dissenting in part) (discussing treatment of the mentally retarded).

Despite these obstacles, the States have acknowledged that the care of the mentally disabled is their special obligation. They operate and support facilities and programs, sometimes elaborate ones, to provide care. It is a continuing challenge, though, to provide the care in an effective and humane way, particularly because societal attitudes and the responses of public authorities have changed from time to time.

Beginning in the 1950’s, many victims of severe mental illness were moved out of state-run hospitals, often with benign objectives. According to one estimate, when adjusted for population growth, “the actual decrease in the numbers of people with severe mental illnesses in public psychiatric hospitals between 1955 and 1995 was 92 percent.” Brief for American Psychiatric Association et al. as Amici Curiae 21, n. 5 (citing Torrey, supra, at 8—9). This was not without benefit or justification. The so-called “deinstitutionalization” has permitted a substantial number of mentally disabled persons to receive needed treatment with greater freedom and dignity. It may be, moreover, that those who remain institutionalized are indeed the most severe cases. With reference to this case, as the Court points out, ante, at 7—8, 17—18, it is undisputed that the State’s own treating professionals determined that community-based care was medically appropriate for respondents. Nevertheless, the depopulation of state mental hospitals has its dark side. According to one

“For a substantial minority. . . deinstitutionalization has been a psychiatric Titanic. Their lives are virtually devoid of ‘dignity’ or ‘integrity of body, mind, and spirit.’ ‘Self-determination’ often means merely that the person has a choice of soup kitchens. The

‘least restrictive setting’ frequently turns out to be a cardboard box, a jail cell, or a terror-filled existence plagued by both real and imaginary enemies.” Torrey, supra, at 11.

It must be remembered that for the person with severe mental illness who has no treatment the most dreaded of confinements can be the imprisonment inflicted by his own mind, which shuts reality out and subjects him to the torment of voices and images beyond our own powers to describe.

It would be unreasonable, it would be a tragic event, then, were the Americans with Disabilities Act of 1990 (ADA) to be interpreted so that States had some incentive, for fear of litigation, to drive those in need of medical care and treatment out of appropriate care and into settings with too little assistance and supervision. The opinion of a responsible treating physician in determining the appropriate conditions for treatment ought to be given the greatest of deference. It is a common phenomenon that a patient functions well with medication, yet, because of the mental illness itself, lacks the discipline or capacity to follow the regime the medication requires. This is illustrative of the factors a responsible physician will consider in recommending the appropriate setting or facility for treatment. Justice Ginsburg’s opinion takes account of this background. It is careful, and quite correct, to say that it is not “the ADA’s mission to drive States to move institutionalized patients into an inappropriate setting, such as a homeless shelter . . . .” Ante, at 20.

In light of these concerns, if the principle of liability announced by the Court is not applied with caution and circumspection, States may be pressured into attempting compliance on the cheap, placing marginal patients into integrated settings devoid of the services and attention necessary for their condition. This danger is in addition to the federalism costs inherent in referring state decisions regarding the administration of treatment programs and the allocation of resources to the reviewing authority of the federal courts. It is of central importance, then, that courts apply today’s decision with great deference to the medical decisions of the responsible, treating physicians and, as the Court makes clear, with appropriate deference to the program funding decisions of state policymakers.


With these reservations made explicit, in my view we must remand the case for a determination of the questions the Court poses and for a determination whether respondents can show a violation of 42 U.S.C. § 12132’s ban on discrimination based on the summary judgment materials on file or any further pleadings and materials properly allowed.

At the outset it should be noted there is no allegation that Georgia officials acted on the basis of animus or unfair stereotypes regarding the disabled. Underlying much discrimination law is the notion that animus can lead to false and unjustified stereotypes, and vice versa. Of course, the line between animus and stereotype is often indistinct, and it is not always necessary to distinguish between them. Section 12132 can be understood to deem as irrational, and so to prohibit, distinctions by which a class of disabled persons, or some within that class, are, by reason of their disability and without adequate justification, exposed by a state entity to more onerous treatment than a comparison group in the provision of services or the administration of existing programs, or indeed entirely excluded from state programs or facilities. Discrimination under this statute might in principle be shown in the case before us, though further proceedings should be required.

Putting aside issues of animus or unfair stereotype, I agree with Justice Thomas that on the ordinary interpretation and meaning of the term, one who alleges discrimination must show that she “received differential treatment vis-à-vis members of a different group on the basis of a statutorily described characteristic.” Post, at 1—2 (dissenting opinion). In my view, however, discrimination so defined might be shown here. Although the Court seems to reject Justice Thomas’ definition of discrimination, ante, at 13, it asserts that unnecessary institutional care does lead to “[d]issimilar treatment,” ante, at 16. According to the Court, “[i]n order to receive needed medical services, persons with mental disabilities must, because of those disabilities, relinquish participation in community life they could enjoy given reasonable accommodations, while persons without mental disabilities can receive the medical services they need without similar sacrifice.” Ibid.

Although this point is not discussed at length by the Court, it does serve to suggest the theory under which respondents might be subject to discrimination in violation of §12132. If they could show that persons needing psychiatric or other medical services to treat a mental disability are subject to a more onerous condition than are persons eligible for other existing state medical services, and if removal of the condition would not be a fundamental alteration of a program or require the creation of a new one, then the beginnings of a discrimination case would be established. In terms more specific to this case, if respondents could show that Georgia (i) provides treatment to individuals suffering from medical problems of comparable seriousness, (ii) as a general matter, does so in the most integrated setting appropriate for the treatment of those problems (taking medical and other practical considerations into account), but (iii) without adequate justification, fails to do so for a group of mentally disabled persons (treating them instead in separate, locked institutional facilities), I believe it would demonstrate discrimination on the basis of mental disability.

Of course, it is a quite different matter to say that a State without a program in place is required to create one. No State has unlimited resources and each must make hard decisions on how much to allocate to treatment of diseases and disabilities. If, for example, funds for care and treatment of the mentally ill, including the severely mentally ill, are reduced in order to support programs directed to the treatment and care of other disabilities, the decision may be unfortunate. The judgment, however, is a political one and not within the reach of the statute. Grave constitutional concerns are raised when a federal court is given the authority to review the State’s choices in basic matters such as establishing or declining to establish new programs. It is not reasonable to read the ADA to permit court intervention in these decisions. In addition, as the Court notes, ante, at 6—7, by regulation a public entity is required only to make “reasonable modifications in policies, practices, or procedures” when necessary to avoid discrimination and is not even required to make those if “the modifications would fundamentally alter the nature of the service, program, or activity.” 28 CFR § 35.130(b)(7) (1998). It follows that a State may not be forced to create a community-treatment program where none exists. See Brief for United States as Amicus Curiae 19—20, and n. 3. Whether a different statutory scheme would exceed constitutional limits need not be addressed.

Discrimination, of course, tends to be an expansive concept and, as legal category, it must be applied with care and prudence. On any reasonable reading of the statute, §12132 cannot cover all types of differential treatment of disabled and nondisabled persons, no matter how minimal or innocuous. To establish discrimination in the context of this case, and absent a showing of policies motivated by improper animus or stereotypes, it would be necessary to show that a comparable or similarly situated group received differential treatment. Regulations are an important tool in identifying the kinds of contexts, policies, and practices that raise concerns under the ADA. The congressional findings in 42 U.S.C. § 12101 also serve as a useful aid for courts to discern the sorts of discrimination with which Congress was concerned. Indeed, those findings have clear bearing on the issues raised in this case, and support the conclusion that unnecessary institutionalization may be the evidence or the result of the discrimination the ADA prohibits.

Unlike Justice Thomas, I deem it relevant and instructive that Congress in express terms identified the “isolat[ion] and segregat[ion]” of disabled persons by society as a “for[m] of discrimination,” §§12101(a)(2), (5), and noted that discrimination against the disabled “persists in such critical areas as . . . institutionalization,” §12101(a)(3). These findings do not show that segregation and institutionalization are always discriminatory or that segregation or institutionalization are, by their nature, forms of prohibited discrimination. Nor do they necessitate a regime in which individual treatment plans are required, as distinguished from broad and reasonable classifications for the provision of health care services. Instead, they underscore Congress’ concern that discrimination has been a frequent and pervasive problem in institutional settings and policies and its concern that segregating disabled persons from others can be discriminatory. Both of those concerns are consistent with the normal definition of discrimination–differential treatment of similarly situated groups. The findings inform application of that definition in specific cases, but absent guidance to the contrary, there is no reason to think they displace it. The issue whether respondents have been discriminated against under §12132 by institutionalized treatment cannot be decided in the abstract, divorced from the facts surrounding treatment programs in their State.

The possibility therefore remains that, on the facts of this case, respondents would be able to support a claim under §12132 by showing that they have been subject to discrimination by Georgia officials on the basis of their disability. This inquiry would not be simple. Comparisons of different medical conditions and the corresponding treatment regimens might be difficult, as would be assessments of the degree of integration of various settings in which medical treatment is offered. For example, the evidence might show that, apart from services for the mentally disabled, medical treatment is rarely offered in a community setting but also is rarely offered in facilities comparable to state mental hospitals. Determining the relevance of that type of evidence would require considerable judgment and analysis. However, as petitioners observe, “[i]n this case, no class of similarly situated individuals was even identified, let alone shown to be given preferential treatment.” Brief for Petitioners 21. Without additional information regarding the details of state-provided medical services in Georgia, we cannot address the issue in the way the statute demands. As a consequence, the judgment of the courts below, granting partial summary judgment to respondents, ought not to be sustained. In addition, as Justice Ginsburg’s opinion is careful to note, ante, at 19, it was error in the earlier proceedings to restrict the relevance and force of the State’s evidence regarding the comparative costs of treatment. The State is entitled to wide discretion in adopting its own systems of cost analysis, and, if it chooses, to allocate health care resources based on fixed and overhead costs for whole institutions and programs. We must be cautious when we seek to infer specific rules limiting States’ choices when Congress has used only general language in the controlling statute.

I would remand the case to the Court of Appeals or the District Court for it to determine in the first instance whether a statutory violation is sufficiently alleged and supported in respondents’ summary judgment materials and, if not, whether they should be given leave to replead and to introduce evidence and argument along the lines suggested above.

For these reasons, I concur in the judgment of the Court.