H A S T I N G S C E N T E R R E P O R T 13May- June 2006

Avastin, a widely used colon cancer drug manufac-tured by San Francisco-based biotech companyGenentech, has proven a somewhat effective

treatment for lung and breast cancer when administeredat twice the normal dose. In February, the New YorkTimes carried a story about Avastin’s extraordinarily highprice when used in this alternative way: $100,000 for oneyear’s treatment—a figure fully twice the price of the nor-mal dose, even though producing the higher dose coststhe company little additional money. And the treatmentyields only an average gain in life expectancy of fivemonths—very modest relative to the cost.

What garnered media attention, however, was Genen-tech’s novel justification for the price: “the inherent valueof these life-sustaining technologies.”1 Rather than mak-ing the usual appeal to high research costs, the companycited the pricelessness of human life, implying a moralreason for the pricing decision. I will pass in silence overthe obviously self-serving disingenuousness of this ap-peal. The fact is that many in our society—and perhaps asubstantial majority—think human life should bethought of as priceless. That assertion can be taken inthree very different ways.

We can take the pricelessness of human life to meanthat the social worth of an individual (their social status

or contribution to the national economy) should becompletely irrelevant when it comes to determining howmuch society should spend to save or prolong that indi-vidual’s life in the face of a life-threatening illness or acci-dent—a worthy moral principle. Likewise, we should af-firm the nonutilitarian view that the cost of saving eithera life or a life-year should not determine by itself whatwill count as a just allocation of limited health resourceswhen we cannot afford to save all the life-years medicaltechnology may salvage. But we need to reject the viewthat we have a moral obligation to spend any amount ofmoney to save all lives and life-years that medical tech-nology permits. The result of adopting this view wouldbe a gross distortion in our society’s health care prioritiesthat would not be just, compassionate, or prudent.

To see why this is true, let us look at some facts andreasonable projections. Health spending in the UnitedStates topped $1.8 trillion in 2004, roughly 16.3 percentof our gross domestic product (GDP), compared to 5.2percent in 1960. Projections to 2015 show us spendingmore than $4 trillion on health care then—almost 20percent of expected GDP.2 Medicare spending in 2005was about $330 billion. With deployment of the pre-scription drug benefit in 2006, spending will be about$424 billion—a price tag expected to rise by 2014 to$747 billion. Over the ten-year period ending in 2015,Medicare spending will exceed $4 trillion.3 These figuresare socially and politically problematic, especially in lightof growing federal deficits.

Health policy analysts generally agree that emergingmedical technologies drive escalating health costs.4 Yetthey and the public feel medical innovation should notbe slowed or stopped—a conclusion I endorse as well.However, if we couple this belief with that third sense ofthe pricelessness of human life, the results are morallyand economically disastrous.

� � �

The problems posed by pricey medical innovationcombined with a belief in the pricelessness ofhuman life began with the passage of the 1972

End Stage Renal Disease (ESRD) amendments to the

ESSAYS

The Costs of Caring: Who Pays?Who Profits? Who Panders?

B Y L E O N A R D M . F L E C K

Leonard M. Fleck, “The Costs of Caring: Who Pays? Who Profits? WhoPanders?” Hastings Center Report 36, no. 3 (2006): 13-17.

14 H A S T I N G S C E N T E R R E P O R T May- June 2006

Medicare program. Those amendments created a programthat would pay for renal dialysis or transplant for virtually anyU.S. citizen in kidney failure. The program was motivated bythe fact that thousands of patients died every year in the late1960s because they could not afford the cost of dialysis(roughly $90,000 a year per patient in 2005 dollars). Therhetoric at the time was that no one should be denied accessto effective, life-sustaining medical technology simply becausethey could not afford it, and society pressured Congress topass the program quickly.5 Further, Congress believed thatthis was a unique medical technology, rather than the head-waters of a torrent, and it expected the program to top out intwenty years at a half billion dollars per year, meaning futurecosts would be reasonable. Unfortunately, Congress waswrong on all these points.

The dialysis program in 2005 cost about $20.5 billion andsustained the lives of 390,000 patients. Current projectionsshow the program costing about $28.3 billion for 520,000patients by 2010. Another 180,000 former dialysis patientswill be kept alive with even more expensive transplants.6 ThatCongress got the numbers wrong is excusable; that it en-dorsed the rhetoric of the pricelessness of human life is intol-erable. The moral framework that should have guided theseearly decisions was that of health care justice.

Where was the moral mistake? The ESRD program paidfor kidney transplants as well as dialysis. These were the earlyyears of transplants; organ rejection problems meant successwas spotty. By the late 1970s cyclosporin resolved that prob-lem but created the question of whether to use public fundsfor other costly organ transplants. Why should renal patientsbenefit from public largesse, but not patients needing a heartor a liver? Aren’t their lives just as priceless?7

Hemophiliacs have essentially the same predicament asrenal patients: they need Factor VIII—a drug with annualcosts in excess of $100,000—to halt bleeding episodes. Butthe government did not create any special program to savetheir lives. We can imagine the rationale constructed to de-fend this lack of response: the needs of hemophiliacs are tooheterogeneous; not every bleeding episode is life-threatening;too many complex choices would have to be made among“deserving” and “nondeserving” episodes of bleeding. Gov-ernment programs need bright lines. This, however, repre-sents a second moral mistake.

Kidney failure draws the bright line that governmentseeks: dialysis works, sustaining lives otherwise doomed.However, that bright line obscures the morally problematic“ragged edge.”8 The fastest growing segment of the dialysispopulation are those over age seventy-five.9 This is not to sug-gest that age by itself should disqualify anyone from dialysis.The real problem is that dialysis is being used more often tosustain very marginal prolongations of life at very great cost.Behind this is a subtle version of the pricelessness of humanlife claim.

Dialysis is a paradigm case of “rescue” medicine. Individu-als faced with imminent death are given back the rest of theirlives—for some, another fifteen to twenty years. But if human

life is really priceless, then all prolonging of life through med-icine is equally worthy of being supported, whether the gainis ten years or ten days. Likewise, if human life is priceless,then the quality of sustained life is irrelevant. Patients in theend stages of dementia or in a persistent vegetative statewould have as much moral claim on prolonging their lives bydialysis as anyone else.10 In short, the morally problematic syl-logism is this: If dialysis is effective in saving life and we havechosen to fund it, then every instance must be covered—norationing decisions can be made.

� � �

The broader consequences of this thinking require ourattention. Literally dozens of extraordinarily expensivecancer drugs and medical devices are now entering the

market. All of them are “effective”—they prolong life. But theadded time is often measurable in weeks and months, notyears, which yields very high cost-effectiveness ratios—literal-ly hundreds of thousands of dollars per quality-adjusted lifeyear (QALY). Aggregate costs for all these treatments canquickly mount into the tens of billions of dollars. This hastwo morally problematic systemic consequences: employersdrop or drastically restrict health insurance,11 and health pri-orities get skewed in ways that are both unjust and uncaring.

As the cost of health insurance to employers mounts bydouble digits, more drop coverage as an employment benefitor increase copay requirements substantially. In either case,the economically disadvantaged have less access to neededhealth care than those who remain well insured and wellpaid.12 More problematic still is the social hypocrisy fosteredunder these circumstances: Medicare or private health insur-ance plans that pay for $100,000 anticancer drugs likeAvastin can affirm their endorsement of the pricelessness ofhuman life (for those with the relevant insurance coverage)while drawing our gaze away from the uninsured or underin-sured, as well as those who cannot afford the 20 percent copaytheir plan demands.

The other systemic consequence associated with thesecostly and marginally beneficial life-prolonging medical inter-ventions—the distortion of health care priorities—is illustrat-ed by a timely example from the United Kingdom. A major“grass roots” effort is underway there to compel the NationalHealth Service (NHS) to fund access to the drug Herceptin asa first-line treatment for HER-2 positive women with breastcancer.13 This drug costs about seventy thousand dollars for afull course of treatment. Approximately 25 percent of womenwith breast cancer are HER-2 positive, which means they aremore vulnerable to a rapid and fatal metastatic process if thecancer is not defeated on initial diagnosis. Herceptin has beenused as a second-line treatment for several years now, yieldingonly an average of five extra months of survival; but in late2005, dramatic reports asserted that Herceptin increased sur-vival by almost 50 percent when used as a first-line therapy.14

This way of reporting results certainly implies that Her-ceptin must be morally analogous to dialysis in its capacity to

H A S T I N G S C E N T E R R E P O R T 15May- June 2006

save and sustain lives. However, a relative risk ratio is being re-ported, rather than an absolute risk ratio. Women not treatedwith Herceptin had a 17 percent chance of a recurrence of thecancer within one year. Herceptin reduced that figure byslightly less than 50 percent. That means that seventeenwomen would need to be treated to achieve that outcome at atotal cost of about a million dollars to prevent a recurrence ofbreast cancer in that population within one year.15 The infer-ence we should not draw at this point (because there is no sci-entific evidence to support it) is that we have “saved theirlives”—i.e., completely curedthem of their cancer—yet this isexactly the impression mostlaypersons will have.

In the United Kingdom thereare about sixty thousand peoplewith ankylosing spondylitis (AS),a very painful and debilitatingform of inflammatory arthritislargely striking young men. A rel-atively new therapy called anti-TNF can dramatically improvetheir overall functioning, but at ayearly cost of about twenty thou-sand dollars per patient. Mosttrusts in the NHS will not pay forthe drug because it costs toomuch,16 yet it costs far less thanHerceptin. Why favor Herceptin?Herceptin “saves lives.”

Here we have an example ofthe moral insidiousness of a back-ground belief in the pricelessness of human life. We can read-ily tolerate relievable human suffering in order to save verymarginal portions of lives afflicted with a near-certain termi-nal condition. And as the number of these extraordinarily ex-pensive, marginally beneficial interventions proliferate underthe crusading banner of the pricelessness of human life, ourhealth care priorities will likely become even more ethicallyaskew.

In one recent news story, Dr. Leonard Saltz of MemorialSloan-Kettering is reported as saying that drugs to treat coloncancer cost five hundred dollars ten years ago, while today’scost per case is about $250,000 for a thirteen-month gain inlife expectancy.17 The aggregate yearly expense of fifty thou-sand colon cancer cases treated this way would be $12.5 bil-lion. Similarly, if each of the 155,000 individuals who die oflung cancer in the United States each year were given the drugAvastin at $100,000 each in order to achieve an average lifegain of two months, $15.5 billion would be added to the costof U.S. health care.18 This represents a cost per life-year savedof $600,000. We surely ought to wonder whether we couldsave lives and misery among the uninsured and underinsuredfor a small fraction of that.

There are about 450,000 individuals in the United Stateswho fall victim each year to sudden cardiac death. In theory

most of those deaths could be prevented if each had an im-plantable cardiac defibrillator (ICD) at a cost of about$40,000 per device. We implanted 162,000 ICDs in 2005 inthe United States.19 Projections to the year 2010 suggest wemight be implanting as many as 600,000 per year at a cost of$24 billion per year.20 But criteria for the appropriate use ofthese devices are very crude, and consequently, they never firefor 81 percent of people who have had them for five years.21

This looks like a major misallocation of health care resources,unless our starting premise is the pricelessness of human life.

One recent cost-effective-ness analysis suggests weshould be implanting nomore than fifty thousandICDs per year, which willyield a cost per QALY ofabout $37,000.22 As thingsare now, we purchase many ofthese QALYs at about$367,000 each.23 Again, mypoint is not that cost-effec-tiveness analysis should be thefoundation of moral judg-ments, but that it is a relevantconsideration needed tocounter the morally distortinginfluence of the pricelessnessof human life.24

Consider the FDA’s recentapproval of a T-wave alternanstest.25 This test can identify asubstantial portion of individ-

uals who should not be candidates for an ICD, though theymeet current criteria. It should reduce the candidate pool byabout 33 percent—fifty-five thousand individuals per year atcurrent implantation rates. However, the test is wrong in 1.2percent of cases, which means five to six hundred will die sud-denly from fatal arrhythmia preventable with an ICD im-plant. The cost of saving their lives would be about $3 billion.If we have a moral obligation not to put a price on humanlife, then we have to spend this money. This is a badly dis-torted moral judgment.

� � �

In a recent survey of oncologists, 80 percent of them en-dorsed the idea of spending seventy thousand dollars onvery expensive cancer drugs to give a patient two months

more life.26 I cannot imagine the moral argument to justifysuch a proposal. It would certainly not be rooted in any con-ception of either justice or social beneficence. If the appeal isto the pricelessness of life, then we ought to be honest in ad-mitting that only very well-insured lives are priceless. The restseem to be deeply discounted.

I have argued elsewhere that the need for health care ra-tioning is inescapable, and that rationing decisions are more

Dozens of expensive cancerdrugs and medical devices

are entering the market. Allof them are “effective”—theyprolong life. But the addedtime is often measurable in

weeks and months.

16 H A S T I N G S C E N T E R R E P O R T May- June 2006

likely to be just and caring if they are self-imposed through aprocess of rational democratic deliberation.27 This is not aRawlsian thought experiment; it is quite achievable (thoughnot easily) in the real world. But a sustained and comprehen-sive deliberative process is necessary—one two-hour town hallmeeting will not do. The public must receive a fair represen-tation of the problem of health care cost escalation, the role ofadvancing medical technology in driving those costs higher,and the often very marginal benefits we collectively purchasewith our health care dollars, as well as the opportunity tothink through trade-offs and priority setting for a very largerange of health care needs in the framework of a set budget. Ibelieve the public is capable ofthis sort of thinking.

Oncologists can endorse ex-pensive cancer drugs becausethey think only about cancer.They can ignore (though theyshould not) the question ofwhat it means to be a just andcaring society when we haveonly limited resources to meetvirtually unlimited health needs.But the broader healthy publicdoes have to struggle with thequestion of whether they can ingood conscience deny an expen-sive arthritis intervention to af-flicted individuals and imposefive very painful and dysfunc-tional years on them in order tofund Avastin and provide twomore months of life to someonewith lung cancer. In such publicconversations, properly man-aged, we develop our capacitiesfor public reason and sharedmoral understandings, whichwe need now more than ever be-fore.28

1. A. Berenson, “A Cancer DrugShows Promise, at a Price That ManyCan’t Pay,” New York Times, Feb. 15,2006.

2. S. Heffler et al., “U.S. Health Spending Projections for 2004-2014,” Health Affairs 24, Supplement 1 (2005): W5-74-85.

3. Ibid., at W5-78.4. D. Goldman et al., “Consequences of Health Trends and Medical

Innovation for the Future Elderly,” Health Affairs 24, Supplement 2(2005): W5-R5-17.

5. R. Rettig, “Historical Perspective,” in Ethics and the Kidney, ed. N.Levinsky (New York: Oxford University Press, 2002), 3-24.

6. J. Xue et al., “Forecast of the Number of Patients with End-StageRenal Disease in the United States to the Year 2010,” Journal of theAmerican Society of Nephrology 12 (2001): 2753-58.

7. Actually, “political” considerations likely prevented a quick answerto this question. With kidney disease, everyone would be saved either by

renal dialysis or transplant. But with other major organ transplants in-volving absolute shortages, the government would in effect be choosingwhich lives to save, then paying to save them—a visible and painful sortof rationing to which virtually no politician wanted to be linked.

8. D. Callahan, What Kind of Life? The Limits of Medical Progress(New York: Simon and Schuster, 1990), especially chapter two, “On theRagged Edge, Needs, Endless Needs,” 31-68.

9. L.A. Szczech and I. Lazar, “Projecting the United States ESRDPopulation: Issues Regarding Treatment of Patients with ESRD,” KidneyInternational 66, Supplement 90 (2004): S3-S7. Figure 2 shows agrowth in the dialysis population over age 75 from three hundred permillion in 1981 to fifteen hundred per million in 2001.

10. There is something of an egalitarian thrust to this moral judg-ment, but I think this is as much an irrational egalitarianism as the

pricelessness assertion is an irrationalform of respect for human life. Bothresult in injustices that should not betolerated.

11. T. Gilmer and R. Kronick, “It’sthe Premiums, Stupid: Projections ofthe Uninsured Through 2013,”Health Affairs 24, Supplement 1(2005): W5-143-51. They project anuninsured population of fifty-six mil-lion by 2013.

12. See E.L. Book, “Health Insur-ance Trends Are Contributing toGrowing Health Care Inequality,”Health Affairs 24, Supplement 3(2005): W5-R5-77-79.

13. S. Boseley, “The Selling of aWonder Drug,” The Guardian,March 29, 2006. “Grass roots” is inscare quotes because drug advertisingto consumers is not permitted in theUnited Kingdom. Companies such asRoche (maker of Herceptin) recruitpatients as “activists” for their causeto pressure the NHS to fund thesedrugs. This article reports the story ofLisa Jardine, who was annoyed byRoche’s persistent efforts to recruither for what were really their market-ing operations.

14. E.H. Romond et al.,“Trastuzumab Plus AdjuvantChemotherapy for Operable HER2-Positive Breast Cancer,” New EnglandJournal of Medicine 353 (2005):1673-84; M.J. Piccart-Gebhart et al.,“Trastuzumab after Adjuvant

Chemotherapy in HER2-Positive Breast Cancer,” New England Journalof Medicine 353 (2005): 1659-72.

15. A. Elliott, “Pros and Cons of Herceptin,” The Guardian, April 3,2006.

16. L. Moss, “Arthritis Patients Being Denied High-Cost Treatment,”Press Association Newsfile, January 3, 2006, accessed through Lexis-Nexis.

17. C. Arnst, “Going Broke to Stay Alive: Rising Prices for CancerTreatments Are Making Patients—and Doctors—Balk,” Business Week,January 30, 2006: 36.

18. “Cancer Drugs of the Not-Too-Distant Future,” May 6, 2002,accessed at http://www.cnn.com.

19. A. Agnvall, “Learning Fractions: Experts Debate Wider Testing ofHeart Risk Factor,” Washington Post, March 7, 2006.

Oncologists can endorse expensive cancer drugs because they think only

about cancer. The broaderhealthy public has to strugglewith whether to impose five

dysfunctional years onarthritis sufferers in order toprovide two more months of

life to someone with lung cancer.

H A S T I N G S C E N T E R R E P O R T 17May- June 2006

20. S. Pauker, N. Ester, and D. Salem, “Preventing Sudden CardiacDeath: Can We Afford the Benefit?” Annals of Internal Medicine 142(2005): 664-66. See also Z. Goldberger and R. Lampert, “ImplantableCardioverter-Defibrillators: Expanding Indications and Technologies,”Journal of the American Medical Association 295 (2006): 809-18.

21. A. Gehi, D. Haas, and V. Fuster, “Primary Prophylaxis with theImplantable Cardioverter-Defibrillator: The Need for Improved RiskStratification,” Journal of the American Medical Association 294 (2005):958-60. They conclude, “There is clearly much room for improvementin patient selection and thus more cost-effective management.” See alsoG.H. Bardy et al., “Amiadarone or an Implantable Cardioverter-Defib-rillator for Congestive Heart Failure,” New England Journal of Medicine352 (2005): 225-37.

22. S. Pauker et al., “Preventing Sudden Cardiac Death.”23. S. Al-Khatib et al., “Clinical and Economic Implications of the

Multicenter Automatic Defibrillator Implantation Trial-II,” Annals ofInternal Medicine 142 (2005): 593-600. The reader is reminded thatthere is no cost-effectiveness for “the device itself.” Cost-effectiveness de-pends entirely upon how the device is used with various patient popula-tions. Most candidates for the device have serious heart disease that putsthem at greatly elevated risk of death from associated complications. Ifpatients have an ICD implanted and die from their heart disease withinthree years of implantation (the device never having fired), then thecost-effectiveness for that cohort will be $367,000 per QALY. The prac-tical implication is that we should not implant ICDs in patients withless than a three-year life expectancy, though obviously some will die asa result of an arrhythmia otherwise preventable with the ICD.

24. Here I am endorsing the fine work of Peter Ubel in Pricing Life:Why It’s Time for Health Care Rationing (Cambridge, Mass.: MIT Press,2000).

25. S. Heuser, “Medicare to Pay for Heart Test,” Boston Globe, March22, 2006. See also D.M. Bloomfield et al., “Microvolt T-Wave Alter-nans and the Risk of Death or Sustained Ventricular Arrhythmias in Pa-tients with Left Ventricular Dysfunction,” Journal of the American Col-lege of Cardiology 47 (2006): 456-63. This work was brought to my at-tention by my second-year medical students as part of their IntegrativeExercise policy paper. My thanks to Charles Carter, Tim Haffey, RichHall, Nick Kuhl, Linda Murray, Shrishail Nashi, Ho-Jin Ra, AngelaSpry, and Ebon Wallace-Talifarro.

26. E. Nadler, B. Eckert, and P. Neumann, “Do Oncologists BelieveNew Cancer Drugs Offer Good Value?” The Oncologist 11 (2006): 90-95.

27. L.M. Fleck, “Last Chance Therapies: Can a Just and Caring So-ciety Do Health Care Rationing When Life Itself Is at Stake?” Yale Jour-nal of Health Policy, Law, and Ethics 2 (2002): 255-98. See also my essay“Just Caring: Oregon, Health Care Rationing, and Informed Democra-tic Deliberation,” Journal of Medicine and Philosophy 19 (1994): 367-88.

28. To iterate, the very opposite of a “properly managed” conversa-tion is one that has been surreptitiously manipulated by pharmaceuticalcompanies, as has happened in the United Kingdom with the Herceptinissue (see ref. 13 above). My thanks to my partner, Jean Edmunds, forher skillful researching of British sources.

Avastin, Genentech’s monoclonal antibody that it pro-poses to offer at twice the dose (and twice the pricetag) to treat breast and lung cancer, has already made

billions of dollars for the company through its original use—treating colon cancer. Now, with a potential pool of hundredsof thousands more patients, financial analysts predict itsUnited States sales alone could grow nearly sevenfold to $7billion by 2009.1

Extremely expensive drugs are hardly new. The pharma-ceutical companies have long argued that these prices are jus-tified by the extraordinarily high costs of getting new drugs tomarket. They typically estimate those costs at $800 million,which is said to reflect both the high costs of the large clinicaltrials required by the FDA to establish safety and efficacy, aswell as the fact that only a small minority of potential newdrugs ultimately makes it to market. The patent system forpharmaceuticals is designed to encourage research and devel-opment of new drugs by protecting the returns from success-ful drugs. Since the marginal costs of producing new drugslike Avastin are typically tiny in comparison to their patent-protected prices, pharmaceutical companies could not justifythe very large costs of research and development unless thepatents prevented other companies from producing and sell-ing them at those marginal costs.

Drug prices raise many controversial issues. Is the $800million figure typically cited by the industry accurate, or arethe real costs substantially less? Does the patent system pri-marily encourage the development of biologically new com-pounds, or does it instead promote so-called “me-too” drugs?Why do—and should—Americans pay substantially higherprices for drugs than citizens of other developed countries?Such economic, legal, and political issues are of great impor-tance, but they apply across the industry broadly.

B Y D A N W. B R O C K

How Much Is MoreLife Worth?

Dan W. Brock, “How Much Is More Life Worth?” Hastings Center Report 36, no.3 (2006): 17-19.