Legal Regulation of the Use of Race in Medical Research

Lillquist E. and Sullivan C. A. , “The Law and Genetics of Racial Profiling in Medicine,” Harvard Civil Rights-Civil Liberties Law Review 39 (2004): 391–483.

See, e.g., Wood A. J. J. , “Racial Differences in the Response to Drugs – Pointers to Genetic Differences,” New England Journal of Medicine 344 (2001): 1393–96; Schwartz R. S. , “Racial Profiling in Medical Research,” New England Journal of Medicine 344 (2001): 1392–93; Burchard E. G. , “The Importance of Race and Ethnic Background in Biomedical Research and Clinical Practice,” New England Journal of Medicine 348 (2003): 1170–71; Cooper R. S. , “Race and Genomics,” New England Journal of Medicine 348 (2003): 1166–67; Risch N. , “Categorization of Humans in Biomedical Research: Genes, Race and Disease,” Genome Biology 3 (2002): 1–12.

But, see Lillquist and Sullivan , supra note 1.

Noah B. , “Racial Disparities in the Delivery of Health Care,” San Diego Law Review 39 (1998): 135–178.

See Lillquist and Sullivan , supra note 1, at 398.

There is a debate over whether the term “subject” or “participant” is more appropriate. Coleman C. , “Duties to Subjects in Clinical Research,” Vanderbilt Law Review 58 (2005): 387–449, at 387, n. 1.

For a similar definition, see Office of Human Research Protections, Department of Health and Human Services, Institutional Review Board Guidebook, Chapter IV, available at <http://www.hhs.gov/ohrp/irb/irb_chapter4.htm#f3> (last visited June 14, 2006).

Peters A. , “Exposure to Traffic and the Onset of Myocardial Infraction,” New England Journal of Medicine 351 (2004): 1721–30.

That is, research done in a living organism. See The American Heritage Dictionary (New York: Dell Publishing, 4th ed., 2001): at 449.

For a discussion of the use of such mice research, see Liggett S. B. , “Genetically Modified Mouse Models for Pharmacogenomic Research,” Nature Reviews Genetics 5 (2004): 657–63.

That is, research done outside a living organism. See The American Heritage Dictionary, supra note 9, at 449.

Liggett , supra note 10, at 657.

Oguma T. , “Role of Prostanoid DP Receptor Variants in Susceptibility to Asthma,” New England Journal of Medicine 351 (2004): 1752–63.

See Institutional Review Board Guidebook, supra note 7.

Lillquist and Sullivan , supra note 1, at 461–62.

See, e.g., Oguma , supra note 13, at 1753 (“Ancestral origin was determined by self-report”). Interestingly, researchers outside the United States often do not seem inclined to collect such information. See, e.g., Peters , supra note 8 (no report of race in study performed in Germany); Thwaites G. E. , “Dexamethasone for the Treatment of Tuberculous Meningitis in Adolescents and Adults,” New England Journal of Medicine 351 (2004): 1741–51 (no report on race in study performed in Vietnam).

See Risch , supra note 2, at 6.

This was in part the basis of opposition to a mixed race category in the Census: some black political leaders believed that, if forced to choose, individuals of mixed race would choose to be identified as black, thus increasing the numbers and power of that group. See Karst K. L. , “Myths of Identity: Individual and Group Portraits of Race and Sexual Orientation,” University of California Los Angeles Law Review 43 (1995): 263–369, at 330.

See Mezey N. , “Erasure and Recognition: The Census, Race and the National Imagination,” Northwestern University Law Review 97 (2003): 1701–68, at 1752 (“Traditional civil rights groups who testified before Congress on the issue uniformly opposed a single multiracial category. Their motivation appeared to be twofold. First and foremost, they sought to protect the fragile advances that anti-discrimination laws have made toward racial equality, and this entailed protecting the prevailing race categories and the number of minorities who identified with them. A related but distinct strategy also becomes evident in their testimony: They sought to police the boundaries of racial identity in effort to keep those who might identify themselves as multiracial from defecting.”) The history of the racial categories in the Census is a fascinating study of the social construction of race in the most literal sense. See generally, Hickman C. B. , “The Devil and the One Drop Rule: Racial Categories, African-Americans, and the U.S. Census,” Michigan Law Review 95 (1997): 1161–1265.

Mezey , supra note 19, at 1760.

Whether race should be inquired into by a check-the-box approach (with the concomitant question of which boxes are available to check) or by a more open-ended question is worthy of considerable attention. The use of a Hispanic or Latino category is particularly troubling from an ancestry perspective. The appropriate approach might be that of the Census, which “presently categorizes Hispanics as an ethnic group, as opposed to a racial group, and defines Hispanics as ‘persons of Mexican, Puerto Rican, Cuban, Central or Southern American or other Spanish culture or origin, regardless of race.’ Because the Hispanic category is not a racial category, persons who identify with the Hispanic ethnicity must also identify with one of the five racial categories.” Powell J. A. , “A Minority-Majority Nation: Racing the Population in the Twenty-First Century,” Fordham Urban Law Journal 29 (2002): 1395–1415, at 1407.

Ford C. , “The Determination of ‘Race’ in Race-Conscious Law,” California Law Review 82 (1994): 1231–1285.

Burchard , supra note 2, at 1173–74. “Accuracy,” of course, merely means that the individual's genetic heritage is from ancestors who predominately came from a particular geographic location; it does not suggest genetic “purity.”

Lillquist and Sullivan , supra note 1, at 407–08.

Id. As we note there, American law tends to be ambiguous about how broadly or narrowly a race should be defined; what tends to be consistent is the emphasis on ancestry over physical traits, although there are exceptions.

Kitcher P. , In Mendel's Mirror: Philosophical Reflections on Biology (New York: Oxford University Press, 2003): 239–45.

Race, Ethnicity, and Genetics Working Group, “The Use of Racial, Ethnic, and Ancestral Categories in Human Genetics Research,” American Journal of Human Genetics 77 (2005): 519–32.

Lillquist and Sullivan , supra note 1, at 430–36.

Whaley A. L. , “Ethnicity/Race, Ethics, and Epidemiology,” JAMA 95 (2003): 736–42.

This assumption may or may not be true. If the assumption is wrong, then this particular use of race is unjustified.

For similar definitions, see Gerstman B. B. , Epidemiology Kept Simple, 2nd ed. (Hoboken, NJ: Wiley-Liss, 2003): at 2.

At least initially, it may be more important to determine occurrence rather than causation. Thus, race may be useful whether the goal is to identify allele variances or social causes of disease. When the inquiry shifts to causation, however, it will typically be critical to distinguish between genetic and other factors.

Whaley , supra note 29.

Cf. Bhopal R. , “Is Research Into Ethnicity and Health Racist, Unsound, or Important Science?” British Medical Journal 314 (1997): 1751–56, at 1753 [hereinafter Bhopal, Research into Ethnicity] (noting the need for epidemiology to cease simply demonstrating racial differences and to instead focus on causal mechanisms). There is an alternative way to justify the use of race in epidemiological research, which is the provision of services. Once we know that one racial group has an increased incidence of a disease, we can target treatment efforts at that group. See Bhopal R. , “Ethnicity as a Variable in Epidemiological Research,” British Medical Journal 309 (1994): 327–20, at 327–28 [hereinafter Bhopal, Ethnicity as a Variable] (noting use of ethnicity in the provision of services). Elsewhere, we have addressed the use of race in the provision of services to racial groups. See Lillquist and Sullivan , supra note 1, at 469–74. There, we argued against the use of race in the provision of screening tests and against outreach efforts that are targeted only at a particular racial group. We also noted that race-based screening appears to violate existing legal restrictions, see id., at 457–59, but that race-based outreach efforts, even when targeted only at a particular racial group, may indeed pass legal muster, see id., at 455–56. For purposes of this article, we will deem these uses generally impermissible, and therefore not a way in which the use of race in epidemiology can be justified.

Burchard , supra note 2, at 1171.

Id., noting that certain “black Americans” are referred for renal transplantation and cardiac catheterization at lower rates than “white Americans”.

We, of course, are not trying to say that all differences are the result of treatment usage rates. As Burchard and his colleagues note, there are cases where, even controlling for many possible treatment variables, differences in disease occurrence persist. Id.

See Bhopal , Research into Ethnicity, supra note 34, at 1753.

Kalinowski L. , “Race-Specific Differences in Endothelial Function: Predisposition of African-Americans to Vascular Disease,” Circulation 109 (2004): 2511–17, at 2512.

Id., at 2515.

See generally, supra note 2.

Risch N. , “Geographic Distributions of Disease Mutations in Ashkenazi Jewish Population Supports Genetic Drift Over Selection,” American Journal of Human Genetics 72 (2003): 812–22.

See generally, Bamshad M. , “Deconstructing the Relationship Between Genetics and Race,” Nature Reviews Genetics 5 (2004): 598–609.

Bamshad M. , “Genetic Influences on Health: Does Race Matter?” JAMA 294 (2005): 937–46, at 944.

See Oguma , supra note 13, at 1752.

The limitation of the study to white and black patients also ensured that large population groups were excluded.

See Oguma , supra note 13, at 1752.

See Zhang J. , “A Comparison of Medical Management with Misoprostol and Surgical Management for Early Pregnancy Failure,” New England Journal of Medicine 353 (2005): 761–69.

See id., at 763, 765, table 1.

Carson P. , “Racial Differences in Response to Therapy for Heart Failure: Analysis of the Vasodilator Heart Failure Trials,” Journal of Cardiac Failure 5 (1999): 178–87.

See Taylor A. L. , “Combination of Isosorbide Dinitrate and Hydralazine in Blacks with Heart Failure,” New England Journal of Medicine 351 (2004): 2049–57, at 2050.

We have discussed elsewhere a third federal statutory scheme, Title II of the Civil Rights Act of 1964. Lillquist and Sullivan, supra note 1, at 446. Because for our purposes Title II generally overlaps with Title VI and the Equal Protection Clause, which we will discuss below, we ignore it in this piece.

For example, Adarand Constructors v. Pena , 515 U.S. 200 (1995) (benign purpose does not avoid strict scrutiny of racial classification).

Rendell-Baker v. Kohn , 457 U.S. 830 (1982); Davis v. Regis College , 830 P.2d 1098, 1101 (Colo. Ct. App. 1992) (“Regis' receipt of federal funds for research and energy assistance grants…without other indicators of government involvement, is insufficient to establish state action.”). See generally, Byrne J. P. , “Academic Freedom: A Special Concern of the First Amendment,” Yale Law Journal 99 (1989): 251–340, at 301. (“Thus, the state action doctrine mandates judicial enforcement of constitutional liberties against institutional infringements for half the nation's academics and denies it to the other half for reasons which, if desirable at all, are very far removed from the realities of academic life.”)

See, e.g., Shaw v. Hunt, 517 U.S. 899 (1996); Miller v. Johnson, 515 U.S. 900 (1995); Korematsu v. United States , 323 U.S. 214 (1944).

438 U.S. 265, 310 (1978).

See 42 U.S.C. § 2000a (2000).

See 42 U.S.C. § 2000d (2000).

For example, Johnson v. Ry. Express Agency, 421 U.S. 454, 459–60 (1975) (“It is well settled among the federal Courts of Appeals – and we now join them – that § 1981 affords a federal remedy against discrimination in private employment on the basis of race.” [footnote omitted]).

Grutter v. Bollinger , 539 U.S. 306, 343 (2003) (because its Equal Protection Clause claims fail, “petitioner's statutory claims based on Title VI… fail,” quoting the opinion of Justice Powell in Regents of Univ. of Cal. v. Bakke, 438, U.S. 265, 287 (1978) that “Title VI…proscribes only those racial classifications that would violate the Equal Protection Clause of the Fifth Amendment”).

Title VI provides: “No person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.” 42 U.S.C. § 2000d.

This is because “program” is defined to include “all of the operations of” a number of entities, including “a college, university, or other postsecondary institution.” 42 U.S.C. § 2000d-4a.

See, e.g., Ferrill v. Parker Group , 168 F.3d 468, 473 (11th Cir. 1999). (“Because § 1981 proscribes discrimination solely on the basis of race, and the BFOQ [bona fide occupational qualification] defense [in Title VII] does not apply to racial discrimination, the BFOQ defense is never available to the § 1981 defendant” [emphasis added].)

The Supreme Court's approval of affirmative action plans under Title VII (see, e.g., Johnson v. Transp. Agency , 480 U.S. 616, 641–42 (1987); United Steelworkers v. Weber , 443 U.S. 193, 209 (1979)), has been carried over into § 1981. See, e.g., Schurr v. Resorts Int'l Hotel , 196 F.3d 486, 498–99 (3d Cir. 1999). (“While a valid affirmative action plan serves as a defense to an action under section 1981, the standard for evaluating the validity of a plan is identical to the standard developed in Title VII cases.”)

For example, Boy Scouts of America v. Dale 530 U.S. 640 (2000), invalidated a state public accommodation law to the extent it would require a nonprofit organization to admit individuals whose presence would affect in a significant way the organization's ability to advocate its viewpoint.

Hamburger P. , “The New Censorship: Institutional Review Boards,” University of Chicago, Public Law Working Paper No. 95 (May 2005) (arguing that institutional review boards are the equivalent of licensing publications).

United States v. O'Brien , 391 U.S. 367 (1968). See generally, McDonald B. , “Government Regulation or Other ‘Abridgements’ of Scientific Research: The Proper Scope of Judicial Review Under the First Amendment,” Emory Law Journal 54 (2005): 979–1091. In contrast, Professor Hamburger argues that courts will find governmental regulations that facially control conduct to constitute regulations of speech when they discriminate against certain content and, more generally, when they fail a balancing test.

Nordlinger v. Hahn , 505 U.S. 1, 10 (1992). (“The Equal Protection Clause does not forbid classifications. It simply keeps governmental decision makers from treating differently persons who are in all relevant respects alike.”) See also, Saunders M. L. , “Equal Protection, Class Legislation, and Colorblindness,” Michigan Law Review 96 (1997): 245–337, at 314–16 (“a state can classify persons in the lay sense – that is, divide, sort, or arrange them into classes – without subjecting those classes to differing benefits or burdens. For example, it can sort persons into classes for informational purposes, as it does when it gathers and organizes census or other demographic data. Such a classification does not, in and of itself, implicate the Equal Protection Clause, for it does not single out any class of persons for special benefits or burdens”).

If such research were directly linked to a questionable use, the research itself might be more problematic. A handful of cases invalidate racial data-keeping by the government. For example, Anderson v. Martin, 375 U.S. 399 (1964) invalidated racial designations in elections, and, in a case closer to the medical data context, Hamm v. Virginia State Board of Elections , 230 F. Supp. 156 (E.D. Va. 1964), aff'd sub nom. Tancil v. Woolls, 379 U.S. 19 (1964) (mem.) struck down a Virginia law that required public records regarding voting and property taxes be maintained with racial designations, although the plaintiffs were not discriminated against in any way. The same court, however, upheld racial designations in divorce records. Other courts have struck down racial collection of information, but usually in the context of a threatened or intended use against one race. See, e.g., Hall v. Pa. State Police , 570 F.2d 86 (3d Cir. 1978) (finding that a police photography program targeted at black bank customers was impermissible). Generally speaking, when the data collection has not been viewed as likely to be used in impermissible ways, racial collection and retention of data has been permitted, at least when a legitimate use can be envisioned. See Caulfield v. Bd. of Educ. , 583 F.2d 605 (2d Cir. 1978) (affirming the denial of a preliminary injunction to prevent collection of racial data in connection with federal enforcement of Title VI against a school district); United States v. New Hampshire , 539 F.2d 277 (1st Cir. 1976) (noting that hypothetical misuse of racial data collected by the federal government did not justify finding unconstitutional a statute requiring such data to be provided).

Even the Racial Privacy Initiative in California, which failed, would have exempted “otherwise lawful classification of medical research subjects and patients” from its general proscription of race classifications in government record keeping. See California Secretary of State, Official Voter Information Guide 45, § 32(a) (2003); see also Proposition 54 to amend Cal. Const. art. I (2003), available at <http://vote2003.ss.ca.gov/voterguide/english.pdf> (last visited June 15, 2006).

Oguma , supra note 13, at 1753.

“Consideration” is normally required to make a promise enforceable, i.e., to make a contract. See §§ 17, 71, Restatement (Second) of Contracts (St. Paul, MN: American Law Institute Publishers, 1981).

The notion that the Equal Protection Clause should be applied from an expressivist perspective is far more supported by commentators than by cases. See Lillquist and Sullivan , supra note 1, at 398–99; see generally Bowser R. , “Racial Profiling in Health Care: An Institutional Analysis of Medical Treatment Disparities,” Michigan Journal of Race and the Law 7 (2001): 79–133. However, government or government-sponsored research that suggests that one race is less important than others is very problematic.

See supra note 70.

Lillquist and Sullivan , supra note 1, at 474, n.450.

Stolberg S. G. , “Shouldn't a Pill Be Colorblind?” New York Times, May 13, 2001, section 4, at 1. The exclusion of non-minorities from BiDil trials would be especially ironic in light of the history of BiDil's development. Jonathan D. Kahn argues that BiDil is an “ethnic drug” today because of the interventions of law and commerce rather than because of any biomedical considerations; he concludes that the story implicates federal agencies in inappropriately giving the state's imprimatur to using race as a biological category. Kahn J. , “How a Drug Becomes ‘Ethnic’: Law, Commerce, and the Production of Racial Categories in Medicine,” Yale Journal of Health Policy, Law & Ethics 4 (2004): 1–46. He traces the development from “a drug for everyone, with no ethnic marketing,” id., at 4, to its present focus on African-Americans as stemming from a re-analysis of the trial data after the FDA had disapproved the drug's use in the general population. Id., at 15–16.

Saul S. , “FDA Approves a Heart Drug for African-Americans,” New York Times, June 24, 2005, Late Edition-Final, section C, column 1.

See supra text accompanying notes 51–52.

One could argue that there is no consideration where participants are not paid for their services, but the opportunity to participate may itself be of value to at least some subjects and the free medication provided may operate as consideration in any event.

See Carson , supra note 51.

Id., at 186.

See Kahn , supra note 79, at 481.

Presumably, the terms of the contract are the subject's participation in return for whatever medications (including the possibility of a placebo) and consideration for time spent.

As with every other suit under § 1981, the plaintiff would seek either damages for refusal of the other party to enter into a contract or damages for the expectation interest denied. Admittedly, quantifying the expectation interest might be difficult in these cases.

NIH requires that researchers “address” the inclusion of such groups by describing the composition of the proposed study population in terms of “racial/ethnic group,” and it directs researchers to “provide a rationale for selection of such subjects.” National Institutes of Health, “Policy and Guidelines on the Inclusion of Women and Minorities as Subjects in Clinical Research,” last modified October 11, 2001, available at <http://grants.nih.gov/grants/funding/women_min/guidelines_amended_10_2001.htm> (last visited June 15, 2006). The most detailed guidance is given in connection with Phase III drug studies, where inclusion is generally presumed, but not required when no significant differences between races have been demonstrated in prior studies. Id.

Id. “If the data from prior studies strongly support no significant differences of clinical or public health importance in intervention effect based on sex/gender, racial/ethnic and/or relevant subpopulation comparisons, then sex/gender and race/ethnicity will not be required as subject selection criteria. However, the inclusion and analysis of sex/gender and/or racial/ethnic subgroups is still strongly encouraged.”

See generally, supra note 2.

For one of many discussions justifying the difference between moral and legal prohibitions, see Hart H. L. A. , The Concept of Law (New York: Oxford University Press, 1961): at 181.

Roberts D. E. , “Legal Constraints on the Use of Race and Ethnicity in Biomedical Research,” Journal of Law, Medicine & Ethics 34, no. 3 (2006): 526–534.

See, e.g., Stevens J. , “Racial Meanings and Scientific Methods: Changing Policies for NIH-Sponsored Publications Reporting Human Variation,” Journal of Health Politics, Policy and Law 28 (2003): 1033–87 (calling for an end to NIH funding for genetic studies of associations of traits with race or ethnicity); Kahn , supra note 79 (calling for restrictions on when federal agencies permit use of race as a biological or genetic category).

National Institutes of Health, Policy and Guidelines on the Inclusion of Women and Minorities as Subjects in Clinical Research, available at <http://grants.nih.gov/grants/funding/women_min/guidelines_amended_10_2001.htm> (last visited June 15, 2006).

U.S. Department of Health and Human Services, Office of Information Resource Management, HHS Policy for Improving Race and Ethnicity Data, available at <http://www.hhs.gov/oirm/infocollect/nclusion.html> (last visited June 15, 2006).

21 CFR §312.33(a)(2)(2006).

Roberts, supra note 92.

See, e.g., Vaccarino V. , “Sex and Race Differences in the Management of Acute Myocardial Infarction, 1994 through 2002,” New England Journal of Medicine 353 (2005): 671–82, at 681(noting partial funding through grants by NIH).

See, e.g., id.; Jha A. K. , “Racial Trends in the Use of Major Procedures Among the Elderly,” New England Journal of Medicine 353 (2005): 683–91, at 691 (funded in part by National Institute of Aging); Trivedi A. N. , “Trends in the Quality of Care and Racial Disparities in Medicare Managed Care,” New England Journal of Medicine 353 (2005): 692–700, at 700 (funded in part by the Health Resources and Services Administration and the Agency for Healthcare Research and Quality).

See, e.g., Stevens , supra note 93 (arguing against such funding).

We are not aware of a situation in which the government has done this yet. The best example of such a trial of which we are aware is, of course, the A-HeFT, but that trial appears to have operated without any governmental financing. See Taylor , supra note 52, at 2055.

See Kahn , supra note 79, at 31–32; see also U.S. Patent Nos. 6,784,177 (issued Aug. 31, 2004) & 6,465,463 (issued October 15, 2002).

NitroMed application, page 12.

See Kahn , supra note 79, at 14.

See, e.g., Kaufman J. S. and Cooper R. S. , “Commentary: Consideration for Use of Racial/Ethnic Classification in Etiologic Research,” American Journal of Epidemiology 154 (2001): 291–98 (suggesting limitation on usefulness).

See, e.g., id., at 296 (recommending the continuation of both “surveillance” research and research into the interaction between patients and health care providers).

It is possible that some might disagree with this conclusion. Given that the existence of racial disparities has been well documented (see Lurie N. , “Editorial: Health Disparities – Less Talk, More Action,” New England Journal of Medicine 353 (2005): 727–29, at 727), there is a non-trivial argument that the investment of more resources into such research is misplaced. This would then suggest that, at a minimum, the federal government ought to cease funding such research. We believe, however, that such a move would be short-sighted. Continuing to monitor the existence of racial differences is necessary in order to determine whether efforts to eradicate such differences are successful and to help choose among potential strategies. See id., at 728–29.

See Kaufman and Cooper , supra note 105, at 294–95.

Jones C. P. , “Invited Commentary: ‘Race,’ Racism and the Practice of Epidemiology,” American Journal of Epidemiology 154 (2001): 299–304.

Id., at 302–03.

Kaufman and Cooper , supra note 105, at 296, table 3.

See, e.g., Bhopal , Research into Ethnicity, supra note 34, at 1755.

See supra text accompanying notes 26–30.

See supra note 2.

For a recent cataloging of such alleles, see Bamshad , supra note 45.

The correlation between geography and ancestry is not perfect because people sharing a similar ancestry may find themselves in quite distinct geographic locations.

See id., at 939.

Lillquist and Sullivan , supra note 1, at 434.

See id., at 940 (noting that 30% of European Americans have less than 90% European ancestry).

The common examples are sickle-cell anemia and Tay-Sachs, but there are others.

Cooper , supra note 2.

Lillquist and Sullivan , supra note 1, at 408–09.

See Bamshad , supra note 44, at 607.

Bamshad , supra note 45, at 944–45.

See supra text accompanying, at note 18.

Stevens , supra note 93, at 1075.

Bamshad , supra note 45, at 944.

Id. (citing Knight J. C. , “A Polymorphism that Affects OCT-1 Binding to the TNF Promoter Region Is Associated with Severe Malaria,” Nature Genetics 22 (1999): 145–50.

Mercier B. , “Complete Detection of Mutations in Cystic Fibrosis Patients of Native American Origin,” Human Genetics 94 (1994): 629–32.

See Lillquist and Sullivan , supra note 1, at 431, n. 236.

See Estivil X. , “Geographic Distribution and Regional Origin of 272 Cystic Fibrosis Mutations in European Populations,” Human Mutation 10 (1997): 135–54.

Id., at 136–37, table 1; Kere J. , “Cystic Fibrosis in a Low-Incidence Population: Two Major Mutations in Finland,” Human Genetics 93 (1994): 162–66.

Kahn , supra note 79, at 44.

Cf. id. (suggesting that “[p]roviding an effective therapy for heart failure in African-Americans would likely be a compelling interest”).

Professor Kahn , we believe, agrees with us on this score. E-mail from Jonathan D. Kahn to Erik Lillquist (Jan. 17, 2006) (copy on file with authors).

Bamshad , supra note 45, at 944.

If the hypothesis instead is that the biochemistry has been affected not by genetics but by the environment, then it would be improper to use race as a proxy for ancestry. Instead, researchers in such a case should be focused on the social construction of the subject's race – both how the person identifies herself and how she is perceived – in order to draw any conclusions about the effect of environmental factors through the use of race. Here again, we would suggest that the government should only fund such research where it meets the strict scrutiny test, and we believe that this would rarely occur; just as ancestry can generally be directly ascertained without the use of race, in most cases, environmental factors can be discussed without resort to race.

The sole exception to this is where the hypothesis is that the difference is caused by racism. In the context of clinical research, however, we generally believe that, at least with pharmaceuticals, it is highly unlikely that a drug will or will not work for an individual because of racism. Of course, research has already shown that members of some groups appear to receive less or different treatments as a result of racism. See supra note 99. But uncovering such differences is mainly the work of what we have referred to as epidemiological research, not clinical research.

For a similar view, see Kahn , supra note 79, at 42.

Lillquist and Sullivan , supra note 1, at 399–400.

A whole field, pharmacogenomics, has arisen to account for “the effects of genetic variability on drug toxicity and efficacy.” Homepage of the Pharmacogenomics Journal website, available at <http://www.nature.com/tpj/index.html> (last visited June 15, 2006).

See Lillquist and Sullivan , supra note 1, at 461–62.

Kahn , supra note 79, at 44.

Lillquist and Sullivan , supra note 1, at 475.

Id., at 475–76; see also Cho M. K. , “Racial and Ethnic Categories in Biomedical Research: There's No Baby in the Bathwater,” Journal of Law, Medicine & Ethics 34, no. 3 (2006):497–499 (also noting variations in CYP2D6 allelic frequencies).

See supra text accompanying notes 125–126.

See Lillquist and Sullivan , supra note 1, at 478.

See Bloche M. G. , “Race-Based Therapeutics,” New England Journal of Medicine 351 (2004): 2035–37, at 2036–37.