POINT-COUNTERPOINT

How to Make Market Competition Work in Healthcare

Kerianne H. Holman, MD,*† and Rodney A. Hayward, MD*‡§

B rezis and Wiist write a provocative article and make several salient points regarding
the broken state of our current healthcare system.1 We take no issue with their telling
of the well-known tale of how vested interests manipulate markets so as to generate huge
profits while providing little benefit (or even causing harm) to the public. We would,
however, argue with their implication that profits and corporate self interest are the root
problems afflicting US healthcare.
Winston Churchill once famously noted that “democracy is the worst form of
government, except [for] all those other forms that have been tried.”2 The same, we would
argue, can be said about the economic system of capitalism. And just as democracy works
well only with an informed public under the rule of law, so markets work well only under
sufficient regulation to ensure transparency for consumers and fair competition among
suppliers. When government focuses on ensuring these two requisites of a functional
market, the opportunity to profit can be transformed from an instrument of the corrupt
practices lamented by Brezis and Wiist to the most powerful and effective tool we have
to further societal good3—including higher-quality and lower-cost medical care.
The problem is that under our current healthcare system, the ability to profit bears
almost no relationship to the value of care provided. At the same time, the lack of
accessible information regarding the real value of competing healthcare services precludes
productive competition, and is easily exploited by vested interests to convince the public
that more of the most technologically advanced healthcare is always better.4 Finally,
protectionism by the profession and microregulation by government combine to stifle
innovations in healthcare delivery that might otherwise bring down costs.5
None of these problems is primarily the fault of the pharmaceutical industry,
for-profit hospitals, or any others who profit under our current system—as they are simply
behaving in a rational way, taking advantage of the opportunity to profit where they may.
Rather, we are at fault—the American public, our elected officials, and the medical
profession itself—for our persistent tolerance of a dysfunctional system in which profits
can be gained whether or not value is provided.
So let us change the system. Rather than blaming profits and greed, let us focus on
finding ways to fix the broken US healthcare market— by realigning incentives to reward
value, increasing the accuracy and transparency of information available to insurers and
patients, and making room for innovation in models of healthcare delivery.
Below, we describe the most fundamental obstacles to value-driven healthcare
under our current system, and briefly propose solutions that might move us towards a more
functional healthcare market: one in which competition drives down cost, and providing
value is rewarded.

From *The Robert Wood Johnson Foundation Clinical Scholars Program, University of Michigan, Ann Arbor, MI; †Department of Surgery, University of
Michigan, Ann Arbor, MI; ‡Department of Internal Medicine, University of Michigan, Ann Arbor, MI; and §Department of Veterans Affairs, VA Health
Services Research & Development Center of Excellence, VA Ann Arbor Health Care System, Ann Arbor, MI.
Dr. Holman is a Robert Wood Johnson Foundation (RWJF) Clinical Scholar. The opinions and views expressed in this paper are those of the authors, and do
not represent those of the RWJF or the University of Michigan.
Reprints: Kerianne H. Holman, MD, The Robert Wood Johnson Foundation Clinical Scholars Program, University of Michigan Medical School, 1150 W.
Medical Center Drive, 6312 Medical Science Building I, Ann Arbor, MI 48109. E-mail: [email protected].
Copyright © 2011 by Lippincott Williams & Wilkins
ISSN: 0025-7079/11/4903-0240

240 | www.lww-medicalcare.com Medical Care • Volume 49, Number 3, March 2011

Medical Care • Volume 49, Number 3, March 2011 Market Competition in Healthcare

Physician Incentives Are Disconnected From But let us not delude ourselves. Payment reforms such
the Quality or Efficiency of the Care That They as these cannot succeed—and might even cause harm, by
Provide stimulating restrictions on needed care— unless we also re-
As healthcare policy-makers brainstorm over innova- form the very system under which healthcare is practiced. In
tive approaches to “bending the cost curve,”6 they should pay short, we must have a way of monitoring the true quality of
particular attention to the critical fact that physicians order or care provided by physicians, and assessing the effect of
provide most of the medical care for which our nation spends various payment methods not only on costs, but also on
so much. It should come as little wonder that our healthcare outcomes of care. In this regard, it must be noted that
spending is at unsustainable levels—without any measurable obtaining accurate information on quality of care for individ-
effect on quality7—in a system where physicians’ financial ual physicians and small group practices has proven difficult,
success generally has little or no relationship to the value of due to both small sample size per physician and the substan-
the care that they provide. tial costs involved in obtaining clinically detailed information
If we consider efficient, high-quality care to indicate (1) for each patient receiving care.17,18 As discussed further
providing the appropriate care at the right time while avoiding below, for these and other reasons, many have concluded that
unnecessary care, and (2) providing that care effectively—ie, in order to truly make progress in rewarding value in health-
with good technique, prudent resource utilization, quick- care, physicians should be incentivized to practice within
ness, and avoidance of undue complications—we might integrated healthcare systems.
quickly note that our largely fee-for-service healthcare A Lack of Comprehensible Information by
system fails to reward efficiency and quality at best, and
Which to Compare Providers and Health Plans
punishes them at worst. Caring physicians no doubt strive
to provide high-quality care within the current system; but Precludes Productive Competition Under Our
because our fee-for-service system primarily rewards vol- Current System
ume of care, there is an incentive to prescribe more and In a functional market, the quality of goods and ser-
more care, regardless of the value of that care.8 –10 Con- vices goes up and costs go down as suppliers compete with
versely, discerning providers who judiciously minimize each other to win consumer purchases. It is critical to note
low-value return visits, referrals, tests, or procedures tend that in order for competition among suppliers to work in this
to do less well financially and in reputation. When it comes way, consumers must possess both discernment and financial
to quality in execution, there is undoubtedly a sense of liability. Discernment enables consumers to reward suppliers
responsibility and personal pride that drives physicians to of high-quality goods or services through their informed
perform their work well. But again, in relation to where purchases. Financial liability makes their demand sensitive to
incentives lie, our fee-for-service reimbursement system price, which promotes efficiency by giving suppliers an
either does not distinguish among well- versus poorly- incentive to keep costs low.
executed care, or inadvertently provides higher reimburse- The problem in healthcare, of course, is that patients
ment for poorly-executed care that results in avoidable generally possess neither the opportunity for discernment, nor
complications and the need for even more medical care. sufficient financial liability. Given the complexity of medical
Unfortunately, the medical profession has often al- care and the difficulty in assessing an individual provider’s
lowed vested interests to influence care guidelines so as to quality, patients rarely have good information by which to
exacerbate misaligned incentives—as many of the so-called choose among providers or services, and are imminently
pay-for-performance “quality measures” accomplish little vulnerable to provider- and profession-induced demand.14,19
more than the promotion of low-value care and cherry- Patients want to stay well or get better, but beyond these
picking of healthier patient populations.4,11–15 Further, these generalities, they are largely subject to the profession’s rec-
“quality measures” only reinforce the public’s misguided ommendations, individually and collectively, on what partic-
impression that physicians with a lower threshold for provid- ular care is necessary. From this perspective, “valuable”
ing care— eg, those who more readily perform diagnostic medical care becomes whatever the provider at the point of
tests, or provide more intensive management for patients with care suggests. To make matters worse, given the unpredict-
minimal disease—are the best, most conscientious doctors, able nature of the demand for most medical care— one does
and help to sustain the prevalent societal impression that not generally know whether or when one will get sick or
more medical care is always better. injured— grouping of risk under insurance is a necessity,
Thankfully, both the medical profession and policy- such that patients are largely disconnected from any direct
makers have begun to recognize the problem of misaligned financial liability related to their care.
provider incentives. The recently-passed Affordable Care Act Of course, the problems of the uninformed healthcare
calls for methods of physician payment that might incentivize consumer and the “moral hazard” of health insurance have
value, as opposed to volume.16 Capitated payments for been long recognized. Many have attempted to address these
chronic disease care, for example, might encourage coordi- problems by increasing patients’ financial liability for their
nation of care and prudent resource allocation. Bundled care, with the idea that increased liability will drive patients
payments for acute care might incentivize vigilance to pre- to become more judicious in their healthcare purchases.
vent avoidable complications, and the streamlining of acute Healthcare savings accounts are an example of a recent effort
care processes. in this direction, as a form of insurance under which benefi-

© 2011 Lippincott Williams & Wilkins www.lww-medicalcare.com | 241

Holman and Hayward Medical Care • Volume 49, Number 3, March 2011

ciaries pay for most of the medical services they choose to the cost of caring for the sick, so as to hold down premium
pursue with their healthcare savings accounts funds.20 The costs for all.
problem, however, is that considerable evidence shows that Note that under a system of managed competition,
the public is neither comfortable with, nor adept at, making additional financial leverage could always be gained within
decisions about the value of individual components of med- insurance plans through variation of out-of-pocket costs.
ical care at the time of illness. Annual premium costs, Value-based insurance design (VBID), for example, is an
therefore, are probably a better leverage point for increasing insurance structure under which out-of-pocket costs depend
consumer financial liability to stimulate competition on on the degree of benefit offered by the medical service being
cost.21 provided. Highly beneficial services are fully covered, while
As mentioned earlier, for purposes of the “goods and progressively more discretionary services require progres-
services” being compared, competition in healthcare might sively greater copays.23 While we believe that VBID is a
best be accomplished through the comparison of integrated promising idea, we must admit that the complexity of its
healthcare systems, rather than of individual clinicians or real-world administration might prove prohibitive—practi-
elements of care. Aside from the large sample sizes needed to cally and politically—particularly given that the designation
accurately assess outcomes, there is the complicating reality of “necessary” versus ‘discretionary’ care can be complex,
that patients are not interchangeable widgets. Patients’ sever- frequently varying with the individual circumstances of the
ity of illness, comorbidities, health habits, and personal pref- patient receiving the care.24 –29 Nonetheless, the beauty of
erences all play a major role in dictating the patient outcomes managed competition is that it might stimulate insurance
and costs of their care. As a result, clinically detailed infor- plans to explore new methods—whether VBID or other-
mation is essential to distinguish whether outcomes and costs wise— of incentivizing high-value care and minimizing low-
vary due to patient factors, or to the care that they receive13— value care so as to both improve outcomes and hold down
and the costs of obtaining sufficiently detailed information to premium costs.
make reliable assessments for individual clinicians can be Aside from the political will necessary to implement a
prohibitive. Within a larger, organized system, the care pro- system of managed competition, arguably the most elusive
vided by groups of physicians might feasibly be monitored component would be accurate and transparent information on
through sampling, with a degree of accuracy and reliability which provider groups are providing beneficial medical care,
how well, and at what cost. As suggested above, the ability to
that remains elusive in the assessment of individual physi-
know what providers are doing at all would require the
cians or episodes of care.17,18
formation of organized healthcare delivery systems, within
For purposes of facilitating competition among health-
which processes, outcomes, and costs of care could be mea-
care systems, one approach that deserves revisiting is En-
sured. This may sound daunting, but it can and has been done
thoven’s idea of “managed competition.” Enthoven clearly
by some—including Kaiser Permanente, Geisinger Health
and eloquently describes the finer details of managed com-
System, Mayo Clinic, and no less than the Department of
petition elsewhere,22 but its primary novelty entails the pres- Veterans Affairs (VA).30 Ironically, the “socialized” VA
ence of governing bodies, or “sponsors,” who operate inde- Healthcare System has created one of the most successful
pendently of, and above, health insurance companies. These healthcare markets in the world— by stimulating competition
sponsors set the rules under which insurance plans must among its 21 networks through the monitoring and promul-
operate, so as to facilitate fair and productive competition at gation of care processes and outcomes, and by rewarding
the level of annual premiums. For example, sponsors would high performers on measures of quality and value.31 Under a
set requirements as to what broad areas of medical care must system of managed competition, sponsors could require that
be covered under a given tier of plan, so as to prevent product insurance networks include only those providers who operate
differentiation and facilitate price comparisons. Second—and within similar organized systems of care. While accountable
critically—these sponsors would provide transparent infor- care organizations are little more than an amorphous idea at
mation on which physician groups or healthcare systems are present,32 with the right incentives, they could become a
providing beneficial medical care, how well they are doing functional implementation project-in-progress.
so, and at what cost, so as to facilitate competition among Apart from the organization necessary to know what
providers as insurers and their beneficiaries use this informa- care is being provided, on the issue of what care should be
tion to direct their healthcare purchases. Third, sponsors designated as beneficial, we agree with Brezis and Wiist. Too
would collect and promulgate information on outcomes and often, evidence related to the benefit of medical treatments is
satisfaction under various insurance plans, so as to facilitate interpreted by professional societies whose members profit
informed consumer choice among plans. At the same time, to from the services being examined, or derived from propri-
prevent such problems as adverse selection and exclusion due etary studies funded and filtered by industry. Because of this,
to pre-existing conditions, sponsors would act as the conduit such evidence can often be distorted to serve vested inter-
by which patients enroll in any given plan. Lastly, there ests.24,33 At the same time, lobbying by professional guilds
would be a requirement for continuous coverage, combined for generous physician reimbursement admittedly has an
with an individual mandate to carry health insurance—so that effect on driving the overutilization of expensive interven-
healthy persons could not obtain a “free ride” by enrolling tions and treatments, even when evidence for benefit is
only once they became sick, and might meanwhile subsidize lacking.34 Finally, the combined effects of direct-to-con-

242 | www.lww-medicalcare.com © 2011 Lippincott Williams & Wilkins

Medical Care • Volume 49, Number 3, March 2011 Market Competition in Healthcare

sumer advertising, large payments to physicians by industry, advanced training.5 For example, becoming an electrocardio-
and lobbying by industry-supported medical experts them- gram reader or endoscopist might require a 2-year technical
selves often make it politically infeasible for a given provider certificate, with continued licensure based on periodic skill
or healthcare system not to provide unproven care, when and competence testing. A similar paradigm could be applied
these other forces have established such care as “needed” in for other technical tasks—not just as a means of replacing
the eyes of the profession and public. Physician payment physicians, but also as a means of expanding their produc-
reform, combined with insurer-driven competition to lower tivity as they oversee several tasks at once.
the costs of care, might go a long way towards mitigating
such overuse of low-value care for the sake of profit. But to In the Midst of Misaligned Incentives and a
accurately distinguish low-value from high-value care, we Lack of Comprehensible Information, the
first need better access to, and more objective vetting of, the Corporate Voice Speaks Loudly
evidence related to the benefit provided by various compo- It is inarguable and unsurprising that corporations,
nents of care—including medications, devices, screening including the pharmaceutical industry, act in accordance with
tests, procedures, and management approaches for various a primary goal of maximizing profits, rather than public
chronic diseases. To this end, we have previously presented a benefit or social welfare. We would point out that this
proposal for improving transparency and reducing the collu- behavior is prudent, rather than corrupt— because any corpo-
sion of vested interests around medical evidence and guide- ration that actually allowed the public good to substantively
lines for care.24 In brief, we have suggested that an indepen-
hurt its bottom line would undoubtedly be replaced by one
dent, private-public partnership be formed that strives to
that did not. As a more philosophical point, we believe that
objectively vet and report on the scientific evidence around a
social responsibility belongs to the role of individuals and
given component of medical care—working not only to
discern whether it offers any real benefit on average (yes/no), their democratically elected governments—not to corporate
but also to provide much-needed information on how bene- boards. Government certainly can, and should, enforce laws
ficial it is likely to be for individual patients with varying and regulations that discourage corporations from harming
levels of risk and disease. In short, we agree with Brezis and the public in an overzealous pursuit of profit (eg, by con-
Wiist that members of the healthcare industry should no straining pollution, collusion, willful deception, etc).4,37 But
longer be allowed to hide information regarding variations in we ought not to limit profit potential in an effort to redirect
their treatments’ safety and efficacy across the popula- corporate mores; the approach would not work anyway.
tion.25–29 However, merely making the data available will not While we would not disagree with Brezis and Wiist’s state-
be enough. Much work is needed on the part of health ment that “human minds can innovate without financial
services researchers to make such information cognitively incentives,”1 it usually takes much more than an innovative
accessible to the public and their insurers, and applicable to mind to develop effective new treatments in medicine. Con-
the formation of professional care guidelines. sider, for example, the enormous length of time and hundreds
of millions of dollars often required to develop a new med-
The Medical Profession, Aided by Regulators, ication, carry out the trials necessary to obtain FDA approval,
Fosters Protectionism and subsequently make the medication known and accessible
Independent of the misaligned incentives under which to patients and their doctors.38,39 In short, a sizeable profit
our healthcare system operates, we cannot deny the anticom- needs to light the end of the tunnel in order for any company
petitive practices of the medical profession itself. The number to risk such a large investment on the mere possibility of an
of providers in any given specialty, for example, is precisely effective new treatment.
controlled by the number of residency and fellowship posi- To be sure, our current healthcare system frequently
tions allowed by the profession and its accrediting boards.35 enables new treatments of unproven or little value to be
Undoubtedly, there are some aspects of medicine for which 4
hugely profitable— but that is only partly the fault of indus-
years of medical school, followed by a long residency and
try. While it is true that corporations expend enormous
interminable fellowship, are necessary. But there is little
amounts of energy and resources to influence public opinion
doubt that any bright person could learn to perform Pap
smears and colonoscopies, or to read mammograms, electro- and policy so as to make their businesses more profitable,
cardiograms, and routine chest x-rays, eg, with 6 months we—the public, the medical profession, and our elected
training—at least well enough to take care of the 90% that officials— have enabled those corporations to play such an
require nothing special. We use our clout as physicians, effective role in dictating our nation’s use of health care,
however, to preclude such technicians from market entry, and because we have failed to provide any accessible, alternative
protect our cabals against the interests of the public. source of influence: one that promulgates balanced informa-
To be sure, the impending “physician shortage,” pre- tion, and enables value-based decision-making. The health-
dicted due to hyper-specialization within the profession and a care system itself can become that source of influence—if
progressively aging population,36 may, in part, fix this prob- only we would realign provider incentives; increase the rigor
lem on its own by forcing creativity in medical staffing. and availability of evidence by which to make medical
Policy-makers might greatly facilitate this cost-saving oppor- decisions; and give the public tools, including transparent
tunity, however, by directing licensing requirements for tech- information and innovative insurance models, by which to
nical tasks based upon competence, rather than years of demand that providers compete on cost and quality.

© 2011 Lippincott Williams & Wilkins www.lww-medicalcare.com | 243

Holman and Hayward Medical Care • Volume 49, Number 3, March 2011

CONCLUSION 18. Hofer TP, Hayward RA, Greenfield S, et al. The unreliability of
individual physician “report cards” for assessing the costs and quality of
We ultimately agree with Brezis and Wiist that vested care of a chronic disease. JAMA. 1999;281:2098 –105.
interests have often been able to generate huge profits while 19. TG McGuire. Physician agency. In: AJ Culyer, JP Newhouse, eds.
providing little value under our current healthcare system. Handbook of Health Economics. Vol 1A. North Holland: Elsevier;
But while Brezis and Wiist suggest that we should reform the 2000:461–536.
players who are taking advantage of a broken system, we 20. Claxton G, Gabel J, Gil I, et al. What high-deductible plans look like:
findings from a national survey of employers, 2005. Health Aff (Mill-
say—let’s fix the broken system. We suggest that healthcare wood). 2005;Suppl Web Exclusives:W5– 434 – 41.
policies be aimed at realigning provider incentives, improv- 21. Liu S, Chollet D. Price and Income Elasticity of the Demand for Health
ing transparency, and facilitating fair competition on cost and Insurance and Health Care Services: A Critical Review of the Literature.
quality. Washington, DC: Mathematica Policy Research, Inc; 2006.
22. Enthoven AC. The history and principles of managed competition.
Health Aff (Millwood). 1993;12(suppl):24 – 48.
REFERENCES 23. Chernew ME, Rosen AB, Fendrick AM. Value-based insurance design.
1. Brezis M, Wiist W. Vulnerability of Health to Market Forces. Med Care. Health Aff (Millwood). 2007;26:w195–w203.
2010;49:232–239. 24. Quanstrum KH, Hayward RA. Lessons from the mammography wars.
2. Churchill W. The Official Report, House of Commons (5th Series), 11 N Engl J Med. 2010;363:1076 –1079.
November 1947, vol. 444, cc. 206 –207. 25. Hayward RA, Krumholz HM, Zulman DM. Optimizing statin treatment
3. Gilder G. Wealth and Poverty. 2nd ed. Richmond, CA: ICS Press; 1993. for primary prevention of coronary artery disease. Ann Intern Med.
4. Deyo RA, Patrick DL. Hope or Hype: The Obsession With Medical 2010;152:69 –77.
Advances and the High Cost of False Promises. New York, NY:
26. Rothwell PM. Treating Individuals: From Randomized Trials to Person-
AMACOM; 2005.
alized Medicine. Philadelphia, PA: Elsevier Health Sciences; 2007.
5. Conover CJ. Health Care Regulation: A $169 Billion Hidden Tax.
27. Kravitz RL, Duan N, Braslow J. Evidence-based medicine, heterogene-
Washigton DC: Cato Institute; 2004.
ity of treatment effects, and the trouble with averages. Milbank Q.
6. Antos J, Bertko J, Chernew M, et al. Bending the Curve: Effective Steps
2004;82:661– 687.
to Address Long-Term Health Care Spending Growth. Washington, DC:
28. Hayward RA, Kent DM, Vijan S, et al. Reporting clinical trial results to
Engelberg Center for Health Care Reform at Brookings; 2009.
inform providers, payers, and consumers. Health Aff (Millwood). 2005;
7. Baicker K, Chandra A. Medicare spending, the physician workforce, and
beneficiaries’ quality of care. Health Aff. 2004;7:W4 –184 –197. 24:1571–1581.
8. Gosden T, Forland F, Kristiansen IS, et al. Capitation, salary, fee-for- 29. Kent DM, Hayward RA. Limitations of applying summary results of
service and mixed systems of payment: effects on the behaviour of clinical trials to individual patients: the need for risk stratification.
primary care physicians. Cochrane Database Syst Rev. 2000;CD002215. JAMA. 2007;298:1209 –1212.
9. Chaix-Couturier C, Durand-Zaleski I, Jolly D, et al. Effects of financial incen- 30. Shih A, Davis K, Schoenbaum SC, et al. Organizing the US Health Care
tives on medical practice: results from a systematic review of the literature and Delivery System for High Performance. New York, NY: The Common-
methodological issues. Int J Qual Health Care. 2000;12:133–142. wealth Fund; 2008.
10. Gosden T, Pedersen L, Torgerson D. How should we pay doctors? A 31. Kizer KW. The “new VA”: a national laboratory for health care quality
systematic review of salary payments and their effect on doctor behav- management. Am J Med Qual. 1999;14:3–20.
ior. Q J Med. 1999;92:47–55. 32. Fisher ES, Shortell SM. Accountable care organizations: accountable for
11. Hayward RA. All-or-nothing treatment targets make bad performance what, to whom, and how. JAMA. 2010;304:1715–1716.
measures. Am J Manag Care. 2007;13:126 –128. 33. Kassirer JP. On the Take: How Medicine’s Complicity With Big Business
12. Hayward RA. Performance measurement in search of a path. NEJM. Can Endanger Your Health. New York, NY: Oxford University Press,
2007;356:951–953. Inc.; 2005.
13. Hayward RA. Access to clinically-detailed patient information: a fun- 34. Fenton JJ, Foote SB, Green P, et al. Diffusion of computer-aided
damental element for improving the efficiency and quality of healthcare. mammography after mandated Medicare coverage. Arch Intern Med.
Med Care. 2008;46:229 –231. 2010;170:987–989.
14. Welch HG. Should I Be Tested for Cancer? Maybe Not and Here’s Why. 35. Schroeder SA. Physician supply and the U.S. medical marketplace.
Berkeley and Los Angeles, CA: University of California Press; 2004. Health Aff (Millwood). 1992;11:235–243.
15. Nicholas LH, Osborne NH, Birkmeyer JD, et al. Hospital process 36. Dill MJ, Salsberg ES. The Complexities of Physician Supply and
compliance and surgical outcomes in Medicare beneficiaries. Arch Surg. Demand: Projections Through 2025. Washington, DC: AAMC Center
2010;145:999 –1004. for Workforce Studies; 2008.
16. United States. Cong. Senate. 111th Congress, 2nd Session. Title III of 37. Smith A. The Wealth of Nations. New York, NY: P.F. Collier & Son;
H.R. 3590, Patient Protection and Affordable Care Act 关introduced in the 1911.
U.S. Senate; 17 September 2009兴. Available at: http://www.healthcare. 38. DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new
gov/center/authorities/title_iii_improving_the_quality_and_efficiency.pd. estimates of drug development costs. J Health Econ. 2003;22:151–185.
Accessed November 11, 2010. 39. Adams CP, Brantner VV. Estimating the cost of new drug development:
17. Birkmeyer JD, Dimick JB. Understanding and reducing variation in is it really 802 million dollars? Health Aff (Millwood). 2006;25:420 –
surgical mortality. Annu Rev Med. 2009;60:405– 415. 428.

244 | www.lww-medicalcare.com © 2011 Lippincott Williams & Wilkins