I will refer to “assisted suicide” rather than “physician-assisted suicide.” Although I recognize that the only statute currently in force in the United States that allows assisted suicide, the Oregon Death with Dignity Act, Or. Rev. Stat. §§ 127.800 et seq. (1995), contemplates only the (limited) assistance of a physician, I believe it is likely that as time passes and additional jurisdictions allow assisted suicide, other health-care professionals, such as nurses, will be involved in many cases of legalized assisted suicide. In this connection, see the thoughtful article by Kjervik, D.K., “Assisted Suicide: The Challenge to the Nursing Profession,” Journal of Law, Medicine & Ethics, 24 (1996): 237–42. Therefore, I believe it may be somewhat misleading to refer exclusively to “physician-assisted suicide.” However, nothing in my argument turns on this semantic point, and those so inclined can freely substitute “physician-assisted suicide” for “assisted suicide” throughout my paper.CrossRefGoogle Scholar

Some of those who have advanced this argument include: Annas, G.J., “Physician-Assisted Suicide — Michigan's Temporary Solution,” N. Engl. J. Med., 328 (1993): 1573–76; Arras, J.D., “Physician-Assisted Suicide: A Tragic View,” in Battin, M.P., Rhodes, R. and Silvers, A., eds., Physician-Assisted Suicide: Expanding the Debate (New York: Routledge, 1998): at 279-300; Kamisar, Y., “Are Laws Against Assisted Suicide Unconstitutional?,” Hastings Center Report, 23, no. 3 (1993): 32-41; King, P.A. and Wolf, L.E., “Empowering and Protecting Patients: Lessons for Physician-Assisted Suicide from the African-American Experience,” Minnesota Law Review, 82 (1998): 1015-43; Pittman, L.J., “Physician-Assisted Suicide in the Dark Ward: The Intersection of the Thirteenth Amendment and Health Care Treatments Having Disproportionate Impacts on Disfavored Groups,” Seton Hall Law Review, 28 (1998): 776-896; Singer, P. and Siegler, M., “Euthanasia — A Critique,” N. Engl. J. Med., 322 (1990): 1881-83; Teno, J. and Lynn, J., “Voluntary Active Euthanasia: The Individual Case and Public Policy,” Journal of the American Geriatrics Society, 39 (1991): 827-30; Wolf, S.M., “Physician-Assisted Suicide, Abortion and Treatment Refusal,” in Weir, R.F., ed., Physician-Assisted Suicide (Bloomington, Indiana: Indiana University Press, 1997): 167-201. As indicated in the text of my article, arguably the most influential statement of this argument is set forth in the report of the New York State Task Force on Life and the Law, When Death Is Sought: Assisted Suicide and Euthanasia in the Medical Context (New York, 1994 especially at pages 120–25. Note that, as is obvious from the titles of their articles, Singer and Siegler and Teno and Lynn specifically address the wisdom of lawful “euthanasia” and not assisted suicide, but I think a fair reading of their arguments indicates that they would view disparate impact as a reason for resisting legalization of assisted suicide. Thus, Singer and Siegler's contention that in “a society where discrimination is common” patients “belonging to vulnerable groups … might be subtly coerced into ‘requesting’ euthanasia” (supra, at 1882) appears equally applicable to the context of assisted suicide.CrossRefGoogle Scholar

Arras, , supra note 2, at 285, and Singer and Siegler, supra note 2, at 1882, are examples of advocates who deploy the argument as but one arrow in their quiver, whereas Pittman, supra note 2, and Wolf, supra note 2, appear to place significant weight on it. Kamisar, supra note 2, at 37, 39, is unclear about the significance he would attach to it, in part because he states his case largely in the form of rhetorical questions such as, “In a suicide-permissive society, how often will the ‘right’ to commit suicide … be interpreted, especially by the most vulnerable, as the duty to do so?” Id. at 37.Google Scholar

New York State Task Force on Life and the Law, supra note 2, at 125.Google Scholar

For example, the majority opinion in the U.S. Supreme Court's recent decision in Washington v. Glucksberg, 521 U.S. 702, 731–32 (1997), makes favorable reference to this argument, and cites the report of the New York Task Force as support for this argument. However, it is unclear whether the majority opinion explicitly endorses this argument. The Supreme Court's reference is in the context of listing the various state interests that “could” be “rationally related” to a ban on assisted suicide. One of the listed interests is “protecting vulnerable groups — including the poor, the elderly and disabled persons — from abuse, neglect, and mistakes.” As constitutional scholars know, it is not difficult to establish a “rational” relationship between a statute and a putative state interest. On the other hand, the vacated panel opinion of the U.S. Court of Appeals for the Ninth Circuit in the same case, Compassion in Dying v. Washington, 49 F.3d 586, 592-94 (9th Cir. 1995), not only referred to the alleged state interest in protecting the “vulnerable,” but appeared to endorse the disparate impact argument, opining that legal assisted suicide would be a “catastrophe,” in part because the “poor and minorities would be especially open to manipulation in a regime of assisted suicide.”.Google Scholar

Annas, , supra note 2, at 1575. Interestingly, this contention by Annas occurs in an article largely devoted to related, but distinct issues (an analysis of Dr. Jack Kevorkian's conduct and Michigan's response thereto). However, I believe it is appropriate to highlight Annas's summary of the disparate impact argument not only because it is well-phrased, but also because in summarizing this argument without much (if any) discussion of why this argument is the “most powerful” argument against legalizing assisted suicide, Annas illustrates a problem that I discuss in the text of my article — namely that the premises of the disparate impact argument remain largely unexamined by its proponents.Google Scholar

See, for example, Cleeland, C.S. et al., “Pain and Its Treatment in Outpatients with Metastatic Cancer,” N. Engl. J. Med., 320 (1994): 592–96; Gornick, M.E. et al., “Effects of Race and Income on Mortality and Use of Services Among Medicare Beneficiaries,” N Engl. J. Med., 335 (1996): 791-99; Schulman, K.A. et al., “The Effect of Race and Sex on Physicians' Recommendations for Cardiac Catheterization,” N. Engl. J. Med., 340 (1999): 618-26. The article by King and Wolf, supra note 2, discusses a number of other similar studies.CrossRefGoogle Scholar

Sullivan, A.D., Hedberg, K. and Hopkins, D., “Legalized Physician-Assisted Suicide in Oregon, 1998–2000,” N. Engl. J. Med., 344 (2001): 605–07; Sullivan, A.D., Hedberg, K. and Fleming, D.W., “Legalized Physician-Assisted Suicide in Oregon —The Second Year,” N. Engl. J. Med., 342 (2000): 598-604. See also Meier, D.E. et al., “A National Survey of Physician-Assisted Suicide and Euthanasia in the United States,” N. Engl. J. Med., 338 (1998): 1193-2000. These studies indicate that at least as far as race and sex are concerned there is (as yet) no correlation between death through assisted suicide (or euthanasia) and being a member of a vulnerable group. In fact, the empirical information can be interpreted as suggesting a different type of problem, namely that requests for assistance in dying made by women are less likely to be heeded than those made by men. See Parks, J.A., “Why Gender Matters to the Euthanasia Debate,” Hastings Center Report, 30, no. 1 (2000): 30-36.CrossRefGoogle Scholar

In this article, I will not discuss how a statistically significant disproportionality is to be determined. However, I do note in passing that any assessment of disproportionality must take account of various relevant factors, including, most obviously, whether members of certain groups are more likely to be legitimate candidates for assisted suicide. For example, if we restrict assisted suicide to terminally ill patients, then given the correlation between terminal illnesses and old age, it is probable that proportionally more older individuals than younger ones would elect assisted suicide even under ideal conditions. Therefore, to determine whether there is a sufficiently significant disproportionality between older and younger individuals who choose assisted suicide, such that one can reasonably infer that the choice of older individuals is more likely to be the result of direct or indirect pressure, this disproportionality must be adjusted or “weighted.”.Google Scholar

And, of course, we routinely allow family members or other proxies to make decisions about withdrawal of life-preserving treatment for persons who are unconscious or incompetent even though the risks of a death that does not truly reflect the autonomous desires of the person receiving treatment are arguably greater than the risks associated with legalized assisted suicide. See Meisel, A., “Managed Care, Autonomy and Decision Making at the End of Life,” Houston Law Review, 35 (1999): 1393–436.Google Scholar

One exception is found in the previously cited article by King and Wolf, supra note 2. The authors recognize that assessing the risks of legalized assisted suicide by analyzing the disparate impact it may have on various groups poses certain conceptual problems. They state, for example: “Whether particular group members, in contrast to groups themselves, are at special risk of coercion, presents [a] … difficult question…. [I]ndividual members of a group that is regarded as vulnerable may not be susceptible to coercion or undue influence. Here the danger is that appeal to a shared experience may obscure the heterogeneity of group members.” Id. at 1019. Unfortunately, the authors do not develop this point.Google Scholar

One could make an argument that some members of protected groups — namely those who would be likely to have access to certain employment, housing, and educational opportunities even with improper barriers in place — are disadvantaged by removal of these barriers because of the increased competition for these opportunities. However, this ignores, among other things, the fact that the statutes that prohibit discrimination in these areas do not remove barriers just for one particular job, one particular school, or one particular dwelling, and a person's ability to compete for these opportunities will vary across different jobs, different schools, and different dwellings. Thus, in the long run, almost all members of protected groups will derive some benefit from the legal requirement that selection procedures for jobs, schools, and dwellings not have a disparate impact on statutorily protected groups.Google Scholar

See Griggs v. Duke Power Co., 401 U.S. 424 (1971), and Albemarle Paper Co. v. Moody, 422 US. 405 (1975).Google Scholar

Griggs, 401 U.S. at 427.Google Scholar

Id. at 429–30.Google Scholar

Id. at 431.Google Scholar

42 U.S.C. §§ 3601-3631 (2000). For an example of a case applying disparate impact analysis to a Fair Housing Act claim, see Gamble v. City of Escondido, 104 F.3d 300 (9th Cir. 1997).Google Scholar

42 U.S.C. § 2000d (2000). For an example of a case applying disparate impact analysis to a Title VI education discrimination claim, see Villanueva v. Carere, 85 F.3d 481 (10th Cir. 1996). Note that for Title VI, the statute “itself proscribes only intentional discrimination [but] regulations promulgated pursuant to Title VI prohibit actions that have a disparate impact on groups protected by the act.” Id. at 486 The Supreme Court's recent decision in Alexander v. Sandoval, 121 S. Ct. 1511 (2001), held that this difference between the statute and its implementing regulations entails that there is no private right of action to bring a disparate impact claim under Title VI. This nuance, although significant for some purposes, does not affect our analysis here.Google Scholar

20 U.S.C. §§ 1681-1688 (2000). For an example of a case applying disparate impact analysis to a Title IX education discrimination claim, see Horner v. Kentucky High School Athletic Ass'n, 206 F.3d 685 (6th Cir. 2000). Disparate impact claims are cognizable under Title IX, but compensatory damages are available only if an intentional violation is shown. Id. at 692.Google Scholar

For a justification of the use of the disparate impact concept in housing discrimination cases employing this reasoning, see Huntington Branch, NAACP v. Town of Huntington, 844 F.2d 926, 934–36 (2d Cir.), aff'd in part, 488 U.S. 15 (1988).Google Scholar

Washington v. Davis, 426 U.S. 229,239 (1976) (rejecting argument that selection procedure's disproportionate impact on black applicants for government jobs can establish a violation of the Equal Protection Clause of the Fourteenth Amendment or the Due Process Clause of the Fifth Amendment). See Ellis v. United Airlines, 73 F.3d 999 (10th Cir.), cert. denied, 517 U.S. 1245 (1996) (disparate impact theory of liability not applicable under Age Discrimination in Employment Act).Google Scholar

29 U.S.C. §§ 621-634 (2000).Google Scholar

See, for example, Mullin v. Raytheon Co., 164 F.3d 696 (1st Cir.), cert. denied, 528 U.S. 811 (1999), Ellis v. United Airlines, 73 F.3d 999 (10th Cir.), cert. denied, 517 U.S. 1245 (1996). For the opposing view, see Mangold v. California Public Utilities Commission, 67 F.3d 1470 (9th Cir. 1995). Note that the U.S. Supreme Court has recently granted certiorari in Adams v. Florida Power Corp., Case No. 01-584 (Dec. 3, 2001) and will consider whether disparate impact claims are cognizable under the ADEA.Google Scholar

Schuck, P.H., “The Graying of Civil Rights Law: The Age Discrimination Act of 1975,” Yale Law Journal, 89 (1979): 27–93, at 37.CrossRefGoogle Scholar

This is the statutory language now found in the United States Code at 42 U.S.C. § 2000e-2(k)(1)(A)(i) (2000). Prior to the Civil Rights Act of 1991, the “business necessity” defense was articulated only in case law.Google Scholar

The law reviews are replete with articles discussing the significance of the “business necessity” defense. For a good recent overview, see Grover, S.S., “The Business Necessity Defense in Disparate Impact Discrimination Cases,” Georgia Law Review, 30 (1996): 387–430.Google Scholar

In the area of education discrimination, there is a similar recognition that a procedure or plan that has a disparate impact may nonetheless be lawful if it is necessary to the legitimate goals of the educational institution. See Brine v. University of Iowa, 90 F.3d 271, 273 (8th Cir. 1996), cert. denied, 519 U.S. 1149 (1997).Google Scholar

New York State Task Force on Life and the Law, supra note 2, at 125.Google Scholar

Wolf, , supra note 2, at 177.Google Scholar

I am indebted to one of the anonymous reviewers of an earlier draft of this article for the suggestion that the disparate impact argument could be interpreted as being more concerned with “differential influence” than a “differential distribution” of deaths. As indicated by my use of quotation marks, I have borrowed these terms from the reviewer as well; these terms capture the distinction nicely.Google Scholar

Wolf, , supra note 2, at 179.Google Scholar

Id. at 180.Google Scholar

Id. at 168.Google Scholar

Id. at 183.Google Scholar

King, and Wolf, , supra note 2, at 1043.Google Scholar

Arras, , supra note 2, at 294.Google Scholar

New York State Task Force on Life and the Law, supra note 2, at 120.Google Scholar