1 Ctrs. for Medicare & Medicaid Servs., National Health Expenditure Projections 2009-2019, tb.1 (2010), http://www.cms.hhs.gov/nationalhealthexpenddata/downloads/proj2009.pdf [hereinafter Projections].

2 Id. at tbl.3.

3 Organization for Economic Co-Operation and Development (OECD), OECD Health Data 2009 (2010), http://stats.oecd.org/Index.aspx?DatasetCode=HEALTH [hereinafter OECD Health Data].

4 Id.

5 Projections, supra note 1, at tbl.1.

6 Id., at tbl.3 (all figures adjusted for inflation).

7 See, e.g., Einer, Elhauge, Allocating Health Care Morally, 82 Cal. L. Rev. 1449, 1459 (1994)Google Scholar (referring to GNP, rather than GDP); Volker H. Schmidt, Models of Health Care Rationing, 52 Current Soc. 969, 971 n.6 (2004).

8 Peter Ferrara, Repeal Health Care Fascism, Am. Spectator, Feb. 25, 2009, at 1, http://spectator.org/archives/2009/02/25/repeal-health-care-fascism/.

9 See infra text accompanying notes 79, 127-128; see also, e.g., Stephanie Condon, Scott Brown Win Shakes Up Health Care Fight, C.B.S. NEWS, Jan. 20, 2010, http://www.cbsnews.com/8301-503544_162-6119035-503544.html (noting that Massachusetts’ 2010 election of Republican Scott Brown to the U.S. Senate to fill the late Senator Ted Kennedy's seat was “a repudiation of President Obama's health care reform package” and that the significance of his election “was difficult to overstate.”); Complaint, Commonwealth of Virginia v. Sibelius, No.3:10-cv-188 (E.D. Vir. filed Mar. 23, 2010); Complaint, State of Florida v. U.S. Dep’t. of Health & Human Servs., No. 3:10-cv-91 (N.D. Fla. filed Mar. 23, 2010).

10 Physicians receive payment for services rendered regardless of whether the services were actually necessary or even marginally helpful. This creates an incentive to treat even where treatment is excessive or wasteful.

11 I acknowledge that, notwithstanding these concerns, there are good reasons to trust physicians and that granting them the responsibility to ration care might yield some significant benefits. By focusing only on the reasons not to trust physicians, this paper suggests that there are no legitimate responses to the stated concerns–an unfortunate consequence necessitated by the limited scope of this paper. On balance, I believe that the arguments in favor of trusting physicians ultimately fail and that a grant of great authority to physicians is likely to result in an allocation at odds with public policy. Further analysis is beyond the scope of this paper.

12 “Social costs,” refer to government subsides and the increased costs occasioned by increased consumption and thus higher prices for everyone.

13 A 2005 survey revealed that of the 108 million Americans aged nineteen through sixtyfour who have private health insurance, 100 million of them received their insurance through an employer. Sara R. Collins, Squeezed: Why Rising Exposure to Health Care Costs Threatens the Health and Financial Well-Being of American Families 2 (2006), available at, http://www.commonwealthfund.org/ (enter the report title into the search field) [hereinafter Squeezed]. It is unclear what percentage of the 100 million hold insurance contracts with private insurance companies and what percent have contracts for coverage with employer self-funded plans.

14 Health insurance contracts, even if the product of negotiation, are often stipulated to the employee. Employees are generally presented with one or two such contracts and told to ‘take it or leave it.’ In the United States, federal tax exclusions create significant incentives to ‘take it,’ despite that the contract might be inefficient and insufficient. One recent survey found that an amazing 46% of people aged nineteen to sixty-four with employer based insurance received just one insurance “option.” For those employed in firms of less than twenty people, that number rises to 75%. Squeezed, supra note 13, at 6.

15 Note that while the employer negotiates the contract and often pays part or all of the insurance premium, the employee is actually the financially responsible party. The employer reduces the wages that it would be willing to pay the employee to cover the costs of insurance. See Ezekiel J. Emanuel & Victor R. Fuchs, Who Really Pays for Health Care?: The Myth of “Shared Responsibility”, 299 JAMA 1057, 1057 (2008).

16 The employer is less interested in the long-term health of the employee and so may be less motivated to bargain for preventative medical care. The employer is also less interested in end-of-life care and high-cost catastrophic care (patients needing such care are probably less likely to return to work or to be productive following treatment and recovery).

17 The term “ration” in this paper is defined broadly, as Part II will illustrate.

18 Schmidt, supra note 7, at 969.

19 At a high degree of abstraction, this approximates the scheme initially adopted by the United Kingdom. See discussion infra Part II.C.

20 See supra notes 1-7 and accompanying text.

21 The ability of medical providers to limit access to care via ad hoc rationing is virtually plenary because it occurs without detection by the patient and cannot easily be regulated or controlled by traditional government mechanisms. For example, physicians might provide less-effective, and less-expensive, care (“rationing by dilution”). Schmidt, supra note 7, at 972. They might discharge patients from the hospital earlier than otherwise desirable or find other ways of minimizing the length of treatment (“rationing by termination”). Id. They might even simply decide not to provide a given option due to cost considerations without even making that option known to the patient (“bedside rationing”). Mark A. Hall, Rationing Health Care at the Bedside, 69 N.Y.U. L. Rev. 693, 709 (1994). For an anecdotal example of bedside rationing, which resulted in the death of an author's mother, see M. Gregg Bloche, Trust and Betrayal in the Medical Marketplace, 55 Stan. L. Rev. 919, 919 (2002). Hospitals might restrict access to patients by telling them that other, less-expensive, facilities are better equipped to provide care (“rationing by redirecting”). Schmidt, supra note 7, at 972. Additionally, they might triage their patients to provide care first to those in less-urgent need (“utilitarian rationing”). Id. See also Iserson & Moskop, infra note 52, at 279. Or they might artificially limit the number of devises available, thus artificially limiting access (“rationing by quantity”). See Nicolas S. Ward & Mitchell M. Levy, Rationing and Critical Care Medicine, 35 Critical Care Med. S102, S102 (2007).

22 Pegram v. Herdrich, 530 U.S. 211, 220 (2000).

23 Id. at 221.

24 Id. at 234.

25 Thinking in terms of four discrete paradigmatic structures is analytically helpful as it keeps us cognizant of the universe of possibilities. Welfare economics, for example, is often limited to a discussion about the various methods of maximizing utility. That fact often obscures other important objectives because the inquiry initiates at a level that is too narrow. We might get better policy by first inquiring which of the paradigms are best suited for addressing the issue at hand.

26 Generally, the market paradigm appeals to market efficiency as a means of determining how health care dollars ought to be spent.

27 The professional paradigm places the onus of medical allocation on the medical profession; the theory being that appeal to their expertise and professional judgment, coupled with the self-imposed ethical Hippocratic-type limitations on medical decision-makers, is an effective and neutral means of achieving equitable rationing. Considering the various conflicting interests that physicians face in the normal course of their practice, an adoption of the professional paradigm in its pure form is probably unwise. See supra note 11 and accompanying text.

28 The political paradigm looks to political accountability as a means of controlling costs. This approach, taken in its pure form, seems particularly ill advised as it is at least as likely that the political process would be used to increase, rather than decrease, acute-care medical spending. This is so because it implicates the “public choice” problem of political decisionmaking. When medical providers expend great government resources to save one patient, they do so at the detriment of every taxpayer. The beneficiaries of additional care (the individuals in need of immediate care) are motivated and able to organize (at least, their families are) while those who are at risk (the public at large) are diffuse, anonymous, and relatively disinterested. It is not surprising that the political paradigm has had little appeal. But see generally Elizabeth, C. Price, The Evolution of Health Care Decision-Making: The Political Paradigm and Beyond, 65 Tenn. L. Rev. 619 (1998)Google Scholar (arguing that the political paradigm was in fact the primary mechanism of health care delivery in the United States at the time she wrote her paper).

29 Elhauge, supra note 7, at 1452.

30 Id. at 1453.

31 Elhauge's principal argument was that the moral paradigm, while attractive in many ways, is both impractical and indeterminate in the context of health care, and is thus not a viable option absent external limitations. Elhauge, supra note 7, 1457-65. While I find his arguments convincing as they are articulated, they rest on this unnecessarily narrow conception of the moral paradigm.

32 Richard D. Lamm, Rationing of Health Care: Inevitable and Desirable, 140 U. Pa. L. Rev. 1511, 1512 (1992). Some other important works on the topic include Elhauge, supra note 7, Leonard, M. Fleck, Just Health Care Rationing: A Democratic Decisionmaking Approach, 140 U. PA. L. REV. 1597 (1992)Google Scholar, Hall, supra note 21, and Robert M. Veatch & Carol Mason Spicer, Medically Futile Care: The Role of the Physician in Setting Limits, 18 Am. J.L. & Med. 15 (1992).

33 Lamm, supra note 32, at 1518.

34 Many call this “allocation” rather than “rationing.” See Michael D. Reagan, Health Care Rationing and Cost Containment Are Not Synonymous, 9 Pol’y Studies Rev. 219, 223 (1989). I have argued previously that the distinction between the two is subterfuge because “[b]oth processes center upon a decision to deny proper and necessary medical treatment to real people, even if at the time of the decision those people are anonymous and the decision is largely theoretical.” Meir Katz, Note, Bioterrorism and Public Law: The Ethics of Scarce Medical Resource Allocation in Mass Casualty Situations, 21 Geo. J. Legal Ethics 795, 795 n.2 (2008). Accordingly, I make no attempt to distinguish between those terms in this paper.

35 U.S. Dep't of Health & Human Serv., HHS Pandemic Influenza Plan, available at http://www.hhs.gov/pandemicflu/plan/appendixd.html (last visited Jan. 28, 2009); see also Katz, supra note 34, at 803-05.

36 See infra Part III and the discussion on John Taurek, starting with text accompanying note 135.

37 Lamm, supra note 32, at 1518.

38 Id. at 1511.

39 See id. at 1518.

40 Robin M. Weinick et al., Who Can't Pay for Health Care?, 20 J. Gen. Internal Med. 504, 505-06, tbl.1 (2004). Moreover, 5.8% of respondents between 200% and 400% of the federal poverty level reported delaying or refraining from treatment due to cost. Also surprising is the percentage of college graduates who reported delaying or refraining from treatment: 4.6%. Additionally, 6.1% of all respondents and 5.2% of college graduates did not fill prescription medicine in the twelve months prior to the survey also due to cost. Id.

41 See Ali R. Rahimi, Finacial Barriers to Health Care and Outcomes After Acute Myocardial Infarction, 297 JAMA 1063, 1065 tbl.1 (2007). The table reports that 5.7% of those with financial barriers had an annual income over $70,000. It also reports that there were 442 respondents claiming financial barriers. 5.7% of 422 respondents is approximately 25.19 people. The table also indicates that 19.6% of all respondents received an income over $70,000, which works out to approximately 489.61 people. Dividing 25.19 by 489.61 equals approximately 0.051 (or 5.1%). Id.

42 The data comes from telephone surveys, perhaps limiting somewhat our ability to extrapolate from them conclusions regarding the entire population. See Weinick et al., supra note 40, at 507; Rahimi, supra note 41, at 1071.

43 See Rahimi, supra note 41, at 1063, 1069-70 (emphasis added) (citing a joint study by USA Today, Kaiser Family Foundation, and Harvard School of Public Health).

44 Sara R. Collins et al., Losing Ground: How the Loss of Adequate Health Insurance is Burdening Working Families 15-16, figs. 17-18 (2008), available at, http://www.commonwealthfund.org/Content/Publications/Fund-Reports/2008/Aug/Losing-Ground--How-the-Loss-of-Adequate-Health-Insurance-Is-Burdening-Working-Families--8212-Finding.aspx (click “Fund Report”) [hereinafter Losing Ground].

45 We can only assume that decisions to forgo needed medical treatment positively correlate with decreasing wealth. Data from the most recent American recession (in the early 1990s) is likely unhelpful because that recession was much shorter, narrower, and less severe than the 2008-2009 recession. The most recent comparable recession was in the early 1980s. Data from that recession, if available, would also likely be unhelpful for that recession predates significant changes to the health care sector that are beyond the scope of this work. Additionally, because health care inflation out-paces general inflation, see Sean, Keehan et al., Health Spending Projections Through 2017: The Baby-Boom Generation is Coming to Medicare, 27 Health Aff. w145, w146 ex. 1 (2008)Google Scholar, data on health economics that is not recent is less likely to yield conclusions that are still relevant, unless that data is adequately adjusted for the current rate of health care inflation.

46 Lamm, supra note 32, at 1518-19 (quoting Jeffrey, C. Merrill & Alan, B. Cohen, The Emperor's New Clothes: Unraveling the Myths About Rationing, 24 INQUIRY 105, 106-07 (1987))Google Scholar.

47 See id.

48 A section of the American Recovery and Reinvestment Act of 2009 (“Act”) (commonly known as the “stimulus bill”) created the “Federal Coordinating Council for Comparative Effectiveness Research” (“Council”). American Recovery and Reinvestment Act of 2009 § 804, 42 U.S.C.A. § 299b-8 (West 2009). The Act stipulates that the Council's role is advisory only, explicitly declaring that the Council has no power to “mandate coverage, reimbursement, or other policies for any public or private payer” and that none of the Council's reports “shall be construed as mandates or [binding] clinical guidelines.” Id. Rather, the purpose of the Council is to “foster optimum coordination of comparative effectiveness and related health services research conducted or supported by relevant Federal departments and agencies.” Id. To that end, the Act appropriates $1.1 billion for “comparative effectiveness research.” American Recovery and Reinvestment Act of 2009, Pub. L. No. 111-5, div. A, tit. VIII, Healthcare Research and Quality. 111 H.R. 1, 62-63 (2009). The presumptive goal of that research is to reduce health care spending by rendering spending decisions more cost-efficient.

49 See infra text accompanying notes 72-77.

50 Consent is true in theory alone. See supra notes 13-16 and infra notes 72-77 and accompanying text.

51 These problems are addressed infra Part II.C-II.D.

52 Rationing by prioritization has a long history in war theaters. It was instituted no later than the eighteenth century by French military surgeons. Kenneth, V. Iserson & John, C. Moskop, Triage in Medicine, Part I: Concept, History, and Types, 49 Annals Emergency Med. 275, 276-77 (2007)Google Scholar. During World War II, for example, American physicians rationed scarce penicillin by supplying it to patients with gonorrhea, rather than patients with war wounds. The United States adopted a policy of treating those who could most quickly and “with the least expenditure of time and resources” return to the battlefield and assist directly in the war effort. Id. at 277.

53 These situations are generally known as situations of “medical futility.” I have written about medical futility at length elsewhere. See Meir Katz, When is Medical Care “Futile”? The Institutional Competence of the Medical Profession Regarding the Provision of Life-Sustaining Medical Care (unpublished manuscript, on file with author).

54 Introduced supra Part II.A.

55 See Malcolm Gladwell, The Moral-Hazard Myth: The Bad Idea Behind Our Failed Health-Care System, New Yorker, Aug. 29, 2005, available at http://www.newyorker.com/archive/2005/08/29/050829fa_fact. It is extremely well documented that insufficient medical insurance yields under-consumption. A few representative examples follow. A 2005 study found that 20% of adults with insurance deductibles of $1000 or more delayed or declined to receive recommended cancer screening tests, compared with just 5% of adults with deductibles under $500. (p R 0.05) (There was no correlation between high insurance deductibles and the decision to forgo lower-cost medical procedures, such as blood pressure screenings and dental exams, suggesting that the decision to forgo cancer screenings was primarily a financial one.) Squeezed, supra note 13, at 15; see also id. at 32 n.1. A 2007 study found adults aged nineteen through sixty-four with one or more of four named chronic conditions ((1) high blood pressure, (2) heart disease, (3) diabetes or (4) asthma, emphysema, or other lung disease) were more likely to visit the emergency room or seek admission into a hospital if they were uninsured. Specifically, 19% of patients with adequate medical insurance needed to visit the emergency room or be admitted to a hospital, compared with 43% of those who reported being uninsured at any time in the twelve months prior to the survey. Rising Health Care Costs: Implications for the Health and Financial Security of U.S. Families: Hearing on High Health Care Costs: A State Perspective? Before the S. Comm. on Finance, 110th Cong. 12-13, 28 fig.15 (2008) (statement of Sara R. Collins, Assistant Vice President, The Commonwealth Fund), http://www.commonwealthfund.org/Content/Publications/Testimonies/2008/Oct/Testimony--Rising-Health-Care-Costs--Implications-for-the-Health-and-Financial-Security-of-U-S--Fami.aspx (click “Testimony”) [hereinafter Commonwealth Testimony]. Another study found that 81% of insured diabetics reported that they were able to keep their diabetes under control as compared with just 63% of the uninsured. Further, 41% of insured blood pressure patients reported control over their blood pressure as compared with just 21% of the uninsured. Id. at 29 fig.16.

56 See, e.g., Ezekiel, J. Emanuel, What Cannot be Said on Television About Health Care, 297 JAMA 2131, 2131 (2007); Tom Daschle, Critical: What we Can Do About the Health-Care Crisis 3-42 (2008). The Emanuel paper stated that the American infant mortality rate is 0.57%. Emanuel, supra at 2131. I am not confident that this figure is accurate. The Organization for Economic Cooperation and Development reported that the infant mortality rate in 2005 was 0.69%. OECD Health Data 2008, supra note 3.

57 There is indeed a considerable body of research on the effects of uninsurance on health. See, e.g., Rahimi, et al., supra note 43, at 1069 (finding that certain cardiac patients without health insurance faced “worse quality of life, and poorer overall physical and mental function,” were 50% more likely to be readmitted to a hospital and were 70% more likely to be readmitted to a hospital with cardiac complications); David W. Baker, et al., Loss of Health Insurance and the Risk for a Decline in Self-Reported Health and Physical Functioning, 40 MED. CARE 1126 (2002)Google Scholar; David W. Baker, et al., Lack of Health Insurance and Decline in Overall Health in Late Middle Age, 345 NEW ENG. J. MED. 1106 (2001); Peter Franks, et al., Health Insurance and Mortality: Evidence from a National Cohort, 270 JAMA 737 (1993); Jack Hadley, Sicker and Poorer: The Consequences of Being Uninsured, 60 Med. Care Res. Rev. 3S (2003) [hereinafter Consequences]; Institute of Medicine, Care Without Coverage (2002); Helen Levy & David Meltzer, What do we Really Know About Whether Health Insurance Affects Health, in Health Policy and the Uninsured (2005); Paul D. Sorlie, et al., Mortality in the Uninsured Compared With That in Persons With Public and Private Health Insurance, 154 Arch Internal Med. 2409 (1994). Much of it, however, is imperfect due to selection bias between those with and those without insurance. In other words, “prior health may have affected both the general measures of the subsequent health change and the baseline health insurance status,” thus undermining the claim that observed correlation between uninsurance and poor health indicated causation. The articles that I have described in the body of the paper are less susceptible to that criticism. Jack Hadley, Insurance Coverage, Medical Care Use, and Short-term Health Changes Following an Unintentional Injury or the Onset of a Chronic Condition, 297 JAMA 1073, 1074 (2007) [hereinafter Unintentional Injury].

58 I discuss why I find them persuasive supra note 57.

59 John, Hsu et al., Unintended Consequences of Caps on Medicare Drug Benefits, 354 New Eng. J. Med., 2349, 2350 (2006)Google Scholar.

60 Research showed a significant difference in noncompliance between the two groups. Those with the $1000 cap failed to comply with their drug treatment plan about 5% more often. See id. at 2354 tbl.2. Those with the cap saw a 4% increase in emergency room visits and a 2% increase in non-elective hospital admissions. Id. at 2355 tbl.3. These figures, and those in the body of the text, are all statistically significant and were subject to regression analysis.

61 3.05% of those not subject to the $1000 cap died each year. That number increased to 3.73% for those with the cap. Id. The difference between those figures is 0.68%. The 95% confidence interval is 0.30-1.07%. Id. at 2356

62 Patients with the cap saw a 9% cost increase due to greater use of the emergency room visits and a 14% increase from increased non-elective hospital admissions. Id.

63 Unintentional Injury, supra note 57, at 1080. The study clearly demonstrated that the uninsured were less likely than the insured to receive recommended follow-up care. Id. at 1077. 92.1% of the insured were able to stay out of the emergency room as compared with 86.9% of the uninsured (p value < .001). Id. at 1079 tbl.4. Also of interest was the observation that among the insured patients, the number of patients who reported doing “significantly worse” after 3.5 months of treatment was greater than the number reporting the same after 7 months (from 10.1% down to 9.7%). Id. at 1080. Among the uninsured patients, the trend reversed (from 12.3% up to 13.2%), suggesting that the health of uninsured patients is more likely to deteriorate over time due to insufficient treatment. See id. The claim that uninsured patients do worse over time has been corroborated: “[R]esearch suggests that the progression from good to poor health resulting from lack of health insurance is cumulative and gradually leads to higher mortality rates for uninsured individuals over time.” Id. (emphasis added).

64 Presumably, some uninsured patients can afford the costs of treatment. Indeed, many of the uninsured patients in this sample did pay for their care. The implicit assumption is that a population of uninsured patients is more likely on average to be unable to pay for necessary medical care than is a population of insured patients.

65 See Gladwell, supra note 55 and accompanying text.

66 Erik Eckholm, To Lower Costs, Hospitals Try Free Basic Care for Uninsured, N.Y. Times, Oct. 25, 2006, at A1.

67 There are notable exceptions. For example, the Emergency Treatment and Active Labor Act (EMTALA), requires that emergency rooms “stabilize” a patient that presents himself before the emergency room, regardless of his ability to pay. 42 U.S.C.A. § 1395dd (West 2008). For background on EMTALA, see generally Alicia K. Dowdy et al., The Anatomy of EMTALA: A Litigator's Guide, 27 St. Mary's L. J. 463 (1996), and Tiana Mayere Lee, An EMTALA Primer: The Impact of Changes in the Emergency Medicine Landscape on EMTALA Compliance and Enforcement, 13 Annals Health L. 145 (2004).

68 The provision of full compensation does not upset the incentives that this scheme attempts to create. While providers face no direct economic loss (assuming that the government actually makes the payments that it promises), they lose access to their capital for a time. Government repayment will not be instantaneous. The delay in the repayment process is the source of the incentive here. The more medical providers can reduce their costs, the less money they will have to lay out in anticipation of government reimbursement.

69 This argument operates under the assumption that any care not provided by the private sector would otherwise eventually be provided by the government, typically via Medicare or Medicaid in the American system. If, in fact, the patient would have otherwise persisted (or perished) without the relevant medical treatment, this economic justification does not apply.

70 See infra Part II.C.

71 To the extent that insurance contracts conform to certain guidelines, broad grants to the insurance company permitting the insurer to use its discretion have been fairly consistently upheld by the courts. See Firestone Tire & Rubber Co. v. Bruch, 489 U.S. 101, 113 (1989); Timothy, Stolzfus Jost, Health Care Rationing in the Courts: A Comparative Study, 21 HASTINGS INT’L & COMP. L. REV. 639, 706 (1998)Google Scholar. But see Rush Prudential HMO, Inc. v. Moran, 536 U.S. 355, 386 (2002) (finding that ERISA does not preempt a state statute designed to limit the discretion of insurance companies).

72 As a window into the literature, the following dialogue is interesting: Hall, supra note 21; Mark A. Hall, Law, Medicine, and Trust, 55 Stan. L Rev. 463 (2002); Bloche, supra note 21; Mark, A. Hall, Ideology and Trust: A Reply to Bloche, 55 Stan. L. Rev. 955 (2002)Google Scholar. Indeed, many have argued that the contract is central to modern health law. See generally, e.g., Nan D. Hunter, Risk Governance and Deliberative Democracy in Health Care, 97 GEO. L.J. 1 (2008) (arguing that risk governance is the central principle in modern health law).

73 See supra note 13 and accompanying text.

74 See supra notes 13-16 and accompanying text.

75 See Aetna Health Inc. v. Davila, 542 U.S. 200, 204 (2004). For a brief history leading up to the enactment of ERISA and the subsequent development of its preemption provisions, see Hunter, supra note 72, at 24-42.

76 Perhaps this insurance model governed by a strict contractual paradigm made sense in a world in which insurance payment or non-payment did not govern treatment decisions. In 1974, when ERISA was enacted, the dominant form of physician compensation was a “fee-forservice” arrangement. “Under this model of health care delivery, a plan beneficiary … would visit the doctor of her choice, receive treatment, and then send the bill to her health insurer. If the insurer improperly refused to pay, the beneficiary could be made whole by commencing suit to recover” her costs. Andrews-Clarke v. Travelers Insurance, 984 F. Supp. 49, 58 (D. Mass. 1997). That model is no longer the dominant model of physician reimbursement. Under typical modern managed care systems, the insurance company serves as a gatekeeper prior to treatment. As a result, “the wrongful denial of benefits by an insurer – whether intentional, or the result of negligent medical decisions made during the utilization review process – will sometimes result in the beneficiary never receiving the treatment that she requires.” Id at 59. One federal court sitting in 1997 made the above observations pertaining to the “repeated[] and arbitrar[y]” denial of medical treatment that the plaintiff's husband “so desperately required” and without which he “suffered horribly, and ultimately died needlessly at age fortyone.” Id. at 52. That court, after concluding that it had “no choice” but to authorize the removal of the case from a state court to federal court and then to “slam the courthouse doors in her face and leave her without any remedy,” id. at 53, made the following rhetorical observation:

Although the alleged conduct of Travelers and Greenspring [the defendants] in this case is extraordinarily troubling, even more disturbing to this Court is the failure of Congress to amend a statute that, due to the changing realities of the modern health care system, has gone conspicuously awry from its original intent.

Does anyone care?

Do you?

Id. at 65. Over ten years have past since Judge William Young wrote this stinging criticism of the congressional failure to amend ERISA. Since then, the power of ERISA preemption has increased without congressional intervention. See Hunter, supra note 72, at 24-28.

77 This analogy to contracts of adhesion is imperfect because where the contract is for insurance from a private insurance company (as opposed to a self-funded insurance plan, which is sometimes offered by very large employers), the contract has been negotiated by the private insurance company and the employer. The difficulty is that neither party in the negotiation properly represents the interests of the employee. Collective bargaining mitigates that problem somewhat by permitting a union to stipulate to an employer the contractual terms it finds acceptable. Still, the analogy to contracts of adhesion is acceptable because ultimately, the employee has very little say on the terms of the contract and has no individual power or ability to negotiate its terms. While the employee might in theory be able to request increased wages in lieu of health benefits, it is doubtful that most employees are aware of that option or that most employers will increase the wages by an amount sufficient to offset the lost value to the employee. Additionally, plans offered on the private market might be more expensive on balance. See Squeezed, supra note 13, at 7. In other words, the value to the employee of this imperfect insurance contract is greater than the value of anything the employee might receive by turning the contract down. That is precisely the dynamic in a contract of adhesion. In any event, it is clear that assent to the contract does not represent a statement by the employee that he believes the specific terms of the contract are efficient or even in his ultimate interests.

78 See supra text accompanying note 70.

79 See supra text accompanying notes 8-9 and infra text accompanying notes 127-128.

80 I discuss the need for standards rather than bright-line rules below. See infra text accompanying note 149.

81 Chrisopher, Newdick, Resource Allocation in the National Health Service, 23 AM. J.L. & MED. 291, 291-92 (1997)Google Scholar (quoting Ministry of Health, Department of Health for Scotland, A National Health Service (1944)).

82 Id. at 292.

83 Id.

84 Id. at 305.

85 See generally Martin, Gorsky, The British National Health Service 1948–2008: A Review of the Historiography, 21 Soc. Hist. Med. 437 (2008)Google Scholar.

86 Id. at 438. Although NHS is driven more by market forces than it was at its creation, the agency is also driven more by appeal to “scientific research and evidence” on the value for cost of various forms of medical treatment, and less by “decisions based on opinion or current practice.” Simon Walker, et al., The Role of NICE Technology Appraisal in NHS Rationing, 81- 82 Brit. Med. Bull. 51, 51-52 (2007). It is thus an unusual market; it utilizes market forces but is governed by rules that, presumably, could never develop in a well-functioning market. It is, at best, unclear why consumers would opt for a system governed by scientific research and evidence to the extent that scientific research and evidence yielded conclusions not in their short-term personal interest. For such a system to develop under well-functioning markets, we would have to assume, counterfactually, that medical decisions governed by scientific data regarding value for cost actually provided greater short-term benefit to consumers than could decisions governed by popular opinion (the collective opinions of the consumers).

87 Supply shortages in centralized systems such as the NHS might be exacerbated by the failure to sufficiently ration care, but are actually entirely predictable and unavoidable in any market in which a single party (or, as in the case of the NHS, multiple quasi-public actors all working for a single party) is the sole provider of goods in that market. A clear explanation of such “monosponies” in the labor market is provided by William M. Boal & Michael R. Ransom, Monopsony in American Labor Markets (2002), http://eh.net/encyclopedia/article/boal.monopsony.

88 See National Health Service, 18 Weeks, http://www.18weeks.nhs.uk/Content.aspx (last visited Aug. 23, 2009).

89 National Health Service, I’m a Patient, http://www.18weeks.nhs.uk/Content.aspx?path=/What-is-18-weeks/patient (last visited Aug. 23, 2009)

90 One particular hospital attempted to address their problem by instituting a rule requiring all patients presenting themselves in the emergency room to be seen by a triage nurse within four hours. One day, the hospital fell behind on its goal and required patients to wait up to six hours. Stewart Fleming was sent by his physician to that emergency room along with a note demanding “immediate” care. It turned out that his body was being ravaged by an aggressive virus and he had very little time left. Stewart Fleming and his wife sat in their ambulance, watching Stewart's body slowly deteriorate, not knowing what to do or whom to turn. By the time the mandatory six hours had expired, so had Stewart Fleming. Lynsey Haywood, Dying Father Ignored for 6hrs, The Sun, Dec. 29, 2008, http://www.thesun.co.uk/sol/homepage/news/article2077919.ece.

91 See infra text accompanying notes 102-106.

92 Walker, supra note 86, at 52. “NICE was initially established in England and Wales to help the NHS meet three continuing objectives: (i) to improve continually the overall standards of care; (ii) to reduce unacceptable variation in clinical practice; and (iii) to ensure the best use of resources so that patients receive the greatest benefit.” Id. The purpose of NICE is commonly articulated using language similar to the following: “[T]o make decisions, which support an efficient use of NHS resources – that is, the maximization of population health from available resources.” Id. at 54 (emphasis added); see also Keith Syrett, Nice Work? Rationing, Review and the ‘Legitimacy Problem’ in the New NHS, 10 Med. L. Rev. 1, 11-12 (2002). That last line deserves repetition for it is both the purpose of and the primary problem with centralized rationing by quantity: The most “efficient use” of taxpayer resources is, apparently, to maximize “population health.” I find this assumption very troubling. It is not obviously true that the “most efficient” use of resources is to maximize total health where doing so almost certainly means that many are made worse-off in the process. Maximization of total health is very different from maximization of wealth; efforts in the UK to conflate the two should not be taken as anything other than either simple confusion or a politicized attempt to obfuscate the realities of the NHS experiment. There are no theoretical limits on the wealth that a person can achieve. Accordingly, there are no limits on the wealth that a society can achieve. The same is not true regarding aggregate health. A person cannot be more than 100% healthy. So too, a society cannot be more than 100% healthy. If so, maximization of societal health cannot be achieved by improving the lot of the most-healthy; the improvement of individual health experienced for each increasing unit of resources decreases exponentially as an individual's health approaches 100%. Rather, a society trying to maximize aggregate health must focus materially all attention on the least-healthy (assuming that they have reasonable hope of recovery). Given limited resources, this approach necessitates the promulgation of rules that ignore and possibly even harm (if indirectly) the most healthy in society. This is a social engineering that results in harm to innocent individuals for doing nothing wrong other than having above-average health.

93 NICE does not actually impose regulatory limits. It simply makes recommendations to NHS. Medications and treatments that are recommended by NICE must be made available to patients. Those medications and treatments not recommended by NICE are not thereby banned, but they are far less likely to be available. See Clare, Sellars & Amanda, Easey, First Successful Legal Challenge to NICE Guidance, 3 J. Intell. Prop. L. & Prac. 692, 692 (2008)Google Scholar.

94 See supra Part II.C.

95 Nick Triggle, Why Some Drugs are Not Worth It, BBC NEWS, Nov. 9, 2005, http://news.bbc.co.uk/1/hi/health/4420584.stm.

96 Id.

97 Id.

98 Inescapable Trade-Offs: Weighing up the Costs as Well as the Benefits from New Medicines is Vital, Economist, Feb. 25, 2006, at 62.

99 Triggle, supra note 95.

100 See infra text accompanying note 110 and subsequent paragraphs

101 Christopher Newdick, Judicial Review: Low-Priority Treatment and Exceptional Case Review, 15 Med. L. Rev. 236, 238 (2007) (quoting R. (Ann Marie Rogers) v. Swindon Primary Care Trust and the Sec’y of State, [2006] EWCA (Civ) 392, [79]-[81] (Eng.)).

102 For a concise look at judicial review in the NHS, see generally id.

103 R v. Central Birmingham Hosp. Auth., Ex parte Collier, unreported decision (discussed in Newdick, supra note 81, at 302); Christopher, Newdick, The NHS in Private Hands? Regulating Private Providers of NHS Services, 3 Law & Med.: Current Legal Issues, 1, 10 (2000)Google Scholar; Judith Hedrick, Legal Aspects of Child Health Care 11 (1996).

104 Newdick, supra note 81, at 302. It appears that a one-month wait for life-saving open heart surgery was considered normal and acceptable in the United Kingdom in 1987. See id.

105 Id.

106 Id.

107 My specific interest here is to discuss the problem that Canada faced with excessive waiting times and its decision to permit patients to purchase medical services on a private market, which contradicts the Canadian philosophy on health care delivery.

108 Oregon's centralized medical care system prioritizes certain medical procedures as a means of limiting supply. Procedures higher on the prioritization list are more likely to be funded. Procedures below a certain point on the list, 503 in 2009, are excluded from state coverage. The list that Oregon developed is fascinating. They place a severe head injury at 101, injury to internal organs at 88, ruptured spleen at 178, and a deep open wound in the neck at 91. Number 3 on the list is preventative services from birth until age ten and number 4 is preventative services for after age ten. Treatment for drug abuse is 5 and tobacco dependence is 6. Contraception management and sterilization are 7. Abortion ranks 41 and treatment for sexually transmitted diseases is 56. Linda Gorman, Nat’l Ctr. for Policy Analysis, Rationing Care: Oregon Changes Its Priorities, (2009), available at http://www.ncpa.org/pub/ba645/. Whether you believe abortion is a socially desirable medical procedure or not, abortion cannot reasonably be ranked ahead of severe head injuries, injuries to internal organs, and deep open neck wounds! This list is explainable only by appeal to political considerations. Needless to say, there are a great many more procedures that rank below abortion and should not. For the complete 2009 list, see Heath Service Commission, Prioritized List of Health Services (2009), available at http://www.oregon.gov/OHPPR/HSC/docs/Jan09Plist.pdf. The 2010 priorities are virtually identical. Heath Service Commission, Prioritized List of Health Services (2010), available at http://www.oregon.gov/OHPPR/HSC/docs/Jan10Plist.pdf.

109 Kathy Kinlaw & Robert Levine, Ctrs. for Disease Control & Prevention, Ethical Guidelines in Pandemic Influenza 7 (2007). The CDC's position is somewhat ironic. The federal agency that is dedicated to “prevent and control infectious and chronic diseases, injuries, workplace hazards, disabilities, and environmental health threats,” Ctrs. For Disease Control & Prevention, Our History - Our Story, http://www.cdc.gov/about/history/ourstory.htm. (last visited Mar. 6, 2010), in the United States finds the method of prioritization in most common use in the United States to be inappropriate considering its inequities.

110 “Quality Adjusted Life Years.” See supra text accompanying notes 94-100.

111 Ezekiel J. Emanuel & Alan Wertheimer, Who Should Get Influenza Vaccine When Not All Can?, 312 SCI. 854 (2006).

112 Id. at 855.

113 Id.

114 The theory behind QALY starts with a very strong normative judgment about which there is a significant divide. It assumes that human life has defined value that is capable of entering into mathematical calculation. If we assume, instead, that the value of human life is infinite, we would necessarily reach different results. I consider this problem more explicitly in Part III of this paper. QALY makes an additional highly questionable normative judgment. It suggests that some years of human life are more valuable than other years of human life on the basis of an undefined and amorphous standard called “quality of life.” The assumption is that a year spent suffering in a hospital room on a respirator is substantially without value. That argument presumes that the purpose and value of human life is known to health policy-makers and is frustrated when a person is in a state of suffering. Presumably, these policy-makers are assuming either that that human life is defined primarily by cognitive ability or that the purpose of life is to experience pleasure. The moral overtones to this decision are blatantly obvious. Very few sincere people believe that it is okay to harvest the organs of the comatose and almost no one lives his life exclusively to maximize pleasure. If humanity were defined by cognitive ability, our value would be numerically assessable by looking at IQ or SAT scores. If we really believed that life existed to maximize pleasure, we would be terribly unproductive as a society because we would all be too busy fulfilling our base desires. If indeed there is something more to life than cognitive ability and pleasure, is it so obvious that the “something more” is not achieved or advanced during periods of intense pain and suffering? Can it even be said conclusively and definitively that there is no value to pain and suffering in its own right?

Some might argue that while these are strong normative positions that cannot be defended, they are not controversial. We place dollar figures on human beings all the time out of necessity – for example, we grant specific monetary awards in wrongful death cases. It might appear that, given our long history of expressly valuing human life, it is appropriate to do so in a QALY analysis for the purpose of controlling health care costs. Appearances can be deceiving. Courts assign value to life in wrongful death cases, for example, because they have no other choice. For a court to declare human life of infinite value and thereby deny a wrongful death award on the grounds that an appropriate award cannot be determined by the court is functionally the same as ruling that human life is without value. Courts, by their nature, must decide such questions out of necessity – it is their responsibility – because the decision not to decide makes one party the winner in litigation and the other party the loser. Efforts to reduce the price of health care, while also necessary, do not fall within the jurisdiction of the courts. For the court, then, the necessity of determining wrongful death judgments is greater than the necessity of controlling medical consumption. While necessity does exist for policy-makers, it is not necessary to arrive at policy by first issuing myriad questionable normative decisions about human life and then assigning a dollar figure to that life is one way of controlling consumption. While that is one way to deal with the problem of rationing, it is not the only way and should not be performed absent true necessity. We must recognize that we are probably quite inept at determining the value of human beings. Sensitive to that shortcoming, we should refrain from doing it wherever feasible. Our inability to do this well necessitates a certain conservatism that seems utterly absent from the QALY approach.

115 I provide a list of eight such factors at the conclusion of Part II.A.

116 QALY, in particular, is quite controversial. It seems highly unlikely that NICE or any other governmental body can credibly argue that it chose QALY singularly because it, and only it, is the most efficacious and just means of allocating resources.

117 See supra note 114.

118 Most of these questions are being debated in the literature; there are obvious answers to none of them.

119 Researchers analyzing a hypothetical virulent strain of influenza determined that “[v]accinating [eighty] percent of the children has nearly as high an overall effectiveness as vaccinating [eighty] percent of the entire population,” suggesting that in some sense it might be preferable to conserve resources and just vaccinate the children. Ira M. Longini, et al., Containing Pandemic Influenza with Antiviral Agents, 159 Am. J. Epidemiology 623, 627 (2004).

120 See Norman Daniels & James Sabin, Limits to Health Care: Fair Procedures, Democratic Deliberation, and the Legitimacy Problem for Insurers, 26 Phil. & Pub. Aff. 303, 319-320 (1997).

121 See id.

122 See supra text accompanying note 30.

123 Elhauge, supra note 7, at 1457.

124 See the discussion on John Taurek, starting with text accompanying note 135.

125 This conclusion is related to, but not necessitated by, the assumption that human life is infinitely valuable. If human life is infinitely valuable, the value of each year, hour, day, and second of human life is infinite as well. The value of the integral parts of something with infinite value is also infinite, assuming that the rules defining the whole and the parts are not different. In other words, while it is possible to conclude that the value of human life is infinite while the value of a human arm is $3704.23 because the terms that define the value of an arm are not the same as the terms that define human life, it is not possible to conclude that the value of human life is infinite but the value of one year of life is finite. The value of a year of life is merely a component part of the value of life and is defined by the same terms in every respect. Continuing the analysis, if a year, day, hour, or second of human life is of infinite value, it is at best not obvious that the value of extending patient X's life by three years is greater than the value of extending patient Y's life by three seconds because both are infinite.

This does not mean that the question is indeterminate. Not all infinities are equal. It might well be true that the three years are greater than the three seconds, but that result does not flow from simple arithmetic. To make that determination, we must have some means to assess the difference that precedes mathematical analysis. See infra note 141.

126 Certainly, not all rationing schemes that do not conform to the scheme proposed by this Article can be classified as “immoral” and I do not mean to suggest that they are all guilty of “[r]ationing without respect and care for those disadvantaged.” The point is that one cannot turn to the inevitability of rationing as an excuse to adopt an approach that, by hypothesis, fails to afford patients their due respect and care. Indeed, the entire point of the moral paradigm is to place moral reasoning into the rationing debate. As I will explain, I believe this can be done without first specifying what is to be deemed “moral” and what “immoral.”

127 See supra text accompanying notes 8-9, 79.

128 See id.

129 Examples used throughout this paper tend to focus on death or possible death. That is for illustrative reasons only. It is easier to articulate the problems of rationing when the result of a poor decision is extreme and irreversible. However the arguments presented in this paper, as articulated in this paragraph, are not limited to such cases. For example, it should be easy to see how a situation of severe suffering, even if only temporary, could be subject to the analysis provided in this paper.

130 Goldberg v. Kelly, 397 U.S. 254, 261, 264 (1970).

131 Mathews v. Eldridge, 424 U.S. 319, 326, 335 (1976). The three-factor test involves a consideration of (1) “the private interest that will be affected by the official action,” (2) “the risk of an erroneous deprivation of such interest through the procedures used, and the probable value, if any, of additional or substitute procedural safeguards,” and (3) “the Government's interest, including the function involved and the fiscal and administrative burdens that the additional or substitute procedural requirement would entail.” Id. at 335. As I have demonstrated throughout, the private interest is enormous. The risk of “erroneous deprivation” is dependant upon a definition of “erroneous.” A fair interpretation of the Eldridge decision would define a deprivation of health care without regard to objectively compelling moral objections as “erroneous.” The Government's interest in reducing costs is strong and the fiscal and administrative burdens associated with the additional procedural protections that I propose are significant. While perhaps not readily intuitive, I believe that my proposal will actually serve to reduce costs for reasons articulated in Parts IV and V.

132 They argued that there is no conceptual difference between depriving an infant of a toy and killing that infant. Infants lack cognitive ability and, in their view, lack an important measure of humanity. In Glover's words:

“The objection to killing provides no argument against infanticide, for newborn babies have no conception of death so they cannot have any preference for life over death.” The objection to infanticide is at most no stronger than the objection to frustrating a baby's current set of desires, say by leaving him to cry unattended for a longish [sic] period.

Ramesh Ponnuru, The Party of Death: The Democrats, the Media, the Courts, and the Disregard for Human Life 175-87 (2006) (quoting Glover on page 181); See Wesley, J. Smith, Does Human Life Have Intrinsic Value Merely Because it is Human?, 13 Trinity L. Rev. 45 (2006)Google Scholar; Peter Singer, Practical Ethics 170-71 (2d ed., 1993).

133 See Ponnuru, supra note 132.

134 An earlier version of this article received criticism for failing to explain the logical basis of the apparent assumption by Taurek, infra note 135 and accompanying text, and others that life has infinite value. Frankly, I find this criticism surprising. I am not adopting Taurek in toto and my arguments in this Article are in no way dependant upon an assumption of infinite value. But even if one were to erroneously assume that my arguments are entirely dependant upon the infinite value presumption, that in no way implies an obligation to defend that presumption. While it may be possible to provide logical proof for the claim of infinite value (a matter well beyond the scope of this Article), the source of this presumption is very likely religious or highly theoretical and/or philosophical and exist in a realm outside of logical discourse. A person can believe (or perhaps even know) something to be true without being able to explain his views logically. In the context of an academic article, presumptions as such are valid, as long as they are properly identified as presumptions. The point I intend to make is not that life is infinitely valuable but that many people believe that it is so and the fact that they do draws them towards interesting conclusions that deserve consideration.

Perhaps my critic meant only that if there is no logical basis for their conclusions, and that they rest on an assumption that cannot be disproven, we need not address these conclusions or build policy around them. To him I ask: What is your logical basis for concluding that these assumptions are wrong? If, for example, those making the presumption of infinite value are making a religious or theological argument, a logical proof against them will have to start by logically disproving their religion and the sources of their revelation. I request instead that we inject a little intellectual humility into the debate, recognize that the positions of people with whom we disagree remain relevant notwithstanding our disagreements, and, more importantly, remain tolerant of those views that we do not understand. See generally Nomi Maya Stolzenberg, ‘He Drew a Circle that Shut Me Out’: Assimilation, Indoctrination, and the Paradox of a Liberal Education, 106 Hav. L. Rev. 581 (1993) (defining tolerance).

135 John M. Taurek, Should the Numbers Count?, 6 Phil. & Pub. Aff. 293, 294 (1977).

136 Id. at 299.

137 Id. at 300.

138 That is, we would not require him to surrender the drug on the basis that doing so would save more people than it is able to save while in his possession. Id.

139 Id. at 303, 306.

140 Derek, Parfit, Innumerate Ethics, 7 PHIL. & PUB. AFF. 285, 301 (1978)Google Scholar.

141 This observation can be best demonstrated in the language of mathematics. The area under the curve y=x² is infinite (x r 0). The area under the curve y=(x/2)² is also infinite (x r 0). Nevertheless, at any point on the curve, x² will be greater or equal to (x/2)². For example, if we set x at 50, x² is 2500 and (x/2)² is 625. Similarly, if we set x at 0.5, x² is 0.25 and (x/2)² is 0.0625. Both curves are infinite and yet one is smaller than the other is. See also supra note 125.

142 See generally id.

143 Taurek, supra note 135, at 307 (emphasis added).

144 U.S. Census Bureau, U.S. & Worls Population Clocks, http://www.census.gov/main/www/popcld.html (last visited Feb. 12, 2010).

145 See supra note 134.

146 See supra text accompanying notes 8-9.

147 See Kyron, Huigens, The Dead End of Deterrence, and Beyond, 41 Wm. & Mary L. Rev. 943, 1033 (2000)Google Scholar.

148 Id.

149 See supra text accompanying note 80.

150 See Thomas v. Union Carbide Agr. Prod. Co., 473 U.S. 568, 587 (1985).

151 I refer to Ex parte Collier. See supra text accompanying note 103.

152 See supra note 97 and accompanying text.

153 See supra text accompanying notes 132-144.

154 Cf. Kolender v. Lawson, 461 U.S. 352, 357-59 (1983) (particularly discussing penal statutes); Smith v. Goguen, 415 U.S. 566, 575-76 (1974) (“Where inherently vague statutory language permits such selective law enforcement, there is a denial of due process.”); U.S. v. Reese, 92 U.S. 214, 221 (1875) (“It would certainly be dangerous if the legislature could set a net large enough to catch all possible offenders, and leave it to the courts to step inside and say who could be rightfully detained, and who should be set at large.”).

155 U.S. Const. amend. V. (“No person shall be … deprived of life, liberty, or property, without due process of law … .”); U.S. Const. amend XIV, § 1 (”… nor shall any State deprive any person of life, liberty, or property, without due process of law … .”).

156 This screening process implicates some institutional design questions that I address infra note 183 and accompanying text.

157 37 C.F.R § 1.56 (2000).

158 Praxair, Inc. v. ATMI, Inc., 543 F.3d 1306, 1313, 1328 (Fed. Cir. 2008).

159 Id. at 1313 (quoting Cargill, Inc. v. Canbra Foods, Ltd., 476 F.3d 1359, 1363 (Fed.Cir.2007)).

160 See id. at 1315, 1318 (inferring intent from the degree of materiality of the information, actual or constructive knowledge, and the absence of a good faith explanation for failure to disclose).

161 Penal remedies would very likely have to proceed through the criminal court system with trial by jury and the other protections granted to criminal defendants. A further discussion on criminal procedure and constitutional law is beyond the scope of this paper.

162 See supra note 108 and accompanying text.

163 Supra note 108.

164 The following sentences provide a partial reply to the criticism that this process is perverse and may serve to demean patients. I recognize that my response may not be satisfying to all readers. I am writing here to remind those readers that under my proposal, the entire process is optional. If the rationing system in place determines that a particular patient is to be denied treatment, that patient will then be in the position to choose between embarking on these appellate procedures or accepting her denial of treatment. Giving to those patients who are willing to fight through the proposed procedures for the chance to convince a panel that they ought to be entitled to treatment the option to do so is far better than denying all parties that option because the process seems demeaning to some.

165 This is a modification of John Taurek's argument. See supra text accompanying note 143.

166 Clearly, where the provision of care costs less than the denial of care, care should be mandated. See supra notes 65-67 and accompanying text for a fuller discussion of this phenomenon (discussing the treatment of Dee Dee Dodd).

167 Note that the objective here is to use the purchase of insurance as a proxy to determine the true preferences of the patient at the time she purchased her insurance. If various circumstances render the proxy unconvincing, such as where the insurance policy was provided by an employer and the patient did not know that she was supposed to read it or that she had the option to supplement the insurance with an additional policy, it is irrelevant to this discussion. Further, the fact that the insurance proved ex post to be insufficient is not relevant. The argument is simply that patients should be expected to make reasonable guesses given the information they had at the time they entered into their insurance contracts. If the patient purchased a policy that was subjectively reasonable given her anticipated needs as she understood them at the time of purchase, the inadequacy of the policy is not a meaningful proxy.

168 See id.

169 “Congress shall make no law … prohibiting the free exercise [of religion].” U.S. Const. amend. I.

170 See Thomas v. Review Bd., 450 U.S. 707, 714-16 (1981) (“[R]eligious beliefs need not be acceptable, logical, consistent, or comprehensible to others in order to merit First Amendment protection… . Courts should not undertake to dissect religious beliefs because … [they] are not articulated with the clarity and precision that a more sophisticated person might employ… . Particularly in this sensitive area, it is not within the judicial function and judicial competence to inquire [which of two people] more correctly perceived the commands of their common faith. Courts are not arbiters of scriptural interpretation.”).

171 DEPARTMENT OF DEFENSE, INSTRUCTION NO. 1300.06 at 4-6, 16 (May 5, 2007) (imposing on the petitioner a burden of producing “clear and convincing evidence” regarding the content and sincerity of belief, asking the petitioner to explain how his beliefs affect his actions, and asking the petitioner to demonstrate the consistency and depth of his convictions), available at http://www.dtic.mil/whs/directives/corres/pdf/130006p.pdf.

172 The possibility that this patient – a patient who previously adopted a moral system that values human life according to cognitive ability or the ability to experience pleasure, but now stands before the board begging for the opportunity to access treatment that will be painful and leave her with little ability to function as she did before – has simply changed her mind has not escaped me. Indeed, the fact that this patient is before the board asking for a course of treatment that seems to contradict her moral code suggests that she might have had a genuine change of heart as death began to stare her in the face. It is also possible that she does not trust her doctors and thinks that if she gets the treatment that she requests, she will have a complete recovery in a period of months. Or perhaps stress prevents her from thinking clearly and she persists in a state of confusion. It will often be impossible to discern whether her requests accurately mirror her true preferences at the time she makes her request. While this is a significant problem, it does not justify the current system in which medical resources are allocated without any attempt to discern a patient's moral, religious, ethical, emotional, irrational, or other subjective positions. When it is difficult to determine what those preferences or beliefs are, perhaps because there is reason to believe that they have changed over time, the board will be in the difficult position of using the information that it has and coming to a conclusion.

173 Whether the court be formed as an Article I court, see generally N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50 (1982), or as an Article III court is a question that I reserve. I believe that either construct will satisfy the demands of the United States Constitution provided that questions outside this court's jurisdiction (most significantly, underlying questions of law) remain under the jurisdiction of the standard Article III courts. If this “adjudicatory board” is formed as an Article III court, presumably the only potential bar to jurisdiction is whether the equitable questions before the court properly “aris[e] under” the Federal Constitution. U.S. CONST. art. III, § 2. Under Osborn v. Bank of the U. S., 22 U.S. 738 (1824), there seems to be no doubt that the arising under test is met. Cf. Carlos, M. Vazquez, The Federal ‘Claim’ in the District Courts: Osborn, Verlinden, and Protective Jurisdiction, 95 CAL. L. REV. 1731 (2007)Google Scholar. If the “adjudicatory board” is formed as an Article I court, it would have to survive the tests created by the Supreme Court to determine whether jurisdiction properly rests with the proposed Article I court. To explore those tests, which are well beyond the scope of this paper, see generally Granfinanciera, S.A. v. Nordberg 492 U.S. 33 (1989); Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833 (1986); N. Pipeline Const. Co., 458 U.S. 50; James E. Pfander, Article I Tribunals, Article III Courts, and the Judicial Power of the United States, 118 Harv. L. Rev. 643 (2004).

174 Shortly, I will more explicitly discuss the question of who ought to pay.

175 See supra Part IV.B.3.

176 See supra note 1 and accompanying text.

177 Supra Part IV.B.3.

178 See supra note 173 and accompanying text.

179 For example, if a patient is denied access by his insurance company, the board might explicitly consider the moral relevance of the insurance contract. If it is factually true that the patient was not involved in the contract negotiations and merely accepted it from his employer, and if it seems clear to the board that no patient negotiating the contract would have agreed to a particular exclusion (or if the custom in the industry is to grant coverage for the excluded item), the board would reasonably find the insurance company at least partially liable. The theory would be that the insurance company engaged in bad faith by presenting a contract that no one would want, knowing that it would not be read.

180 There is virtually no question that an Article III court can constitutionally be granted these powers, provided that other questions not within the jurisdiction of this Article III court are within the jurisdiction of other Article III courts. If the “adjudicatory board” be structured as an Article I court, I am not aware of a case directly on-point. There is broad authority for the proposition that administrative findings of fact are not subject to appeal provided that constitutional Due Process protections (beyond the scope of this paper) are properly observed. See Crowell v. Benson, 285 U.S. 22, 92-94 (1932) (holding that administrative factual findings are to be granted deference); Sheldon v. Sill, 49 U.S. 441, 448-49 (1850) (holding that Congress has plenary power to limit the jurisdiction of the lower federal courts); supra note 173. The types of questions presented before this court would plainly be questions of fact rather than questions of law.

181 See supra notes 8-24; see also generally Daschle, supra note 56.

182 See supra Parts III & IV.A.

183 I introduced the screening process supra note 156 and accompanying text.

184 This is not to suggest that emotional costs are irrelevant. Indeed, high emotional costs likely implicate many of the factors discussed in Part IV.B. of this paper. Emotion alone, however, should not be dispositive primarily because emotion will be present in nearly every case that comes before the board. A larger panel is less likely to be swayed by emotion and thus better equipped to make wise and just decisions.

185 Individual jurors suffer from prejudice that we like to assume is not present among adjudicators. Where prejudice exists, larger numbers are desirable to ensure less bias in the results. Additionally, jurors, unlike adjudicators, are often not permitted to take notes during deliberations and must thus rely more heavily on their memories. Greater numbers are likely to result in better recall of the facts. See Ballew v. Georgia, 435 U.S. 223, 232-35 (1978). Perhaps most importantly, at least for jury panels of six and for all federal criminal juries, the Supreme Court has ruled that the Federal Constitution requires unanimity. See Apodaca v. Oregon, 406 U.S. 404, 410-12 (1972); Johnson v. Louisiana, 406 U.S. 356, 363 (1972). This fact dramatically alters the nature of negotiations among jurors and strains the attempt to compare jurors and adjudicators. See Ballew, 435 U.S. 223, 432-35.

186 See Ballew, 435 U.S. at 234-37. Ballew quoted some studies showing that large jury panels were more likely to acquit than were smaller panels and that the fear of wrongful acquittals counseled against making the panels excessively large. One such study found that the optimal panel would be a jury of between six and eight members. Id. at 234. I do not believe this to be relevant for the analysis appears to turn on the unique features of jury deliberation. See id. at 236.

187 See supra notes 173 & 180.

188 See U.S. Const. art. I. § 2, cl. 5; U.S. Const. art. I. § 3, cl. 6-7.

189 U.S. Const. art. II, § 2, cl. 2.

190 Presumably, the adjudicators need not be practicing attorneys. Perhaps it is not even necessary that they be a member of any state bar. Successfully completing three years at an accredited law school likely provides the background knowledge necessary to meet the requirements described in the text.

191 See supra paragraph following note 179.

192 A necessity if the court be formed as an Article III court. See supra note 173.

193 Daschle, supra note 56, at 169-80.

194 Judges of Article III courts must have life tenure. U.S. Const. art. III, § 1. If the adjudicatory board be formed as an Article III court, the judges admitted for terms shorter than life tenure would be appointed for an office other than “judge.” Accordingly, only onequarter of the court would be staffed by actual judges. This creates some interesting questions on the authority of non-judges to sit in an adjudicatory fashion alongside judges. To my knowledge, such questions have never been addressed. I suspect that the court would be properly constituted provided that at least one judge, who will be nominally responsible for the case, sit on every case and that the statute granting the court jurisdiction be drafted in a manner that tolerates such a scheme.