《卫生经济学》课程参考文献：10.Victor R. Fuchs and Ezekiel J.Emanuel_Health Care Reform：Why What When What it might take to effect comprehensive change 2005

HEALTH CARE REFORM Health care Reform why What? When? What it might take to effect comprehensive change by victor R. Fuchs and Ezekiel J. Emanuel ABSTRACT: Dissatisfaction with the U. S health care system is widespread, but no consen- sus has emerged as to how to reform it. The principal methods of finance-employer-based insurance, means-tested insurance and Medicare-are deeply and irreparably flawed Policymakers confront two fundamental questions: Should reform be incremental or com- prehensive? And should priority be given to reforming the financing system or to improving organization and delivery? We consider here several proposals for incremental reform and three for comprehensive reform individual mandates with subsidies, single payer, and uni- versal vouchers. Over the long term, reform is likely to come in response to a major war, de- pression, or large-scale civil unrest Changes are Necessary, but what they Ought to be, what they will be, and how, and when to be produced -John Jay(787) HESE WORDS. WRITTEN ON THE EVE OF THE COnstitutional Conve tion, aptly describe U.S. health care in 2005. Most observers agree that re- form is necessary, but whether that reform should be incremental or com prehensive and whether changes in finance or changes in organization and delivery should receive priority are questions that must be answered. What are the options before the country? What are the obstacles to reform? What are the conditions that might make reform possible? Before addressing these questions, however, it is useful to review briefly why reform deserves urgent attention Why The United States Needs Health Care Reform The palpable symptoms of our"sick health care system"are described almost daily in popular print and broadcast media. Almost one of every six Americans have no health insurance, health care spending is escalating rapidly, administra- tive costs are excessive, and medical errors(including overuse and underuse of medications and procedures )are rampant. Less frequently discussed, but of fun- Victor Fuchs (fuchs @newage. stanford. edu) is the Henry Kaiser ] r. Professor Emeritus at Stanford University in Stanford, California, and a research associate at the National Bureau of Economic Research, also in Stanford Ezckiel Emanuel is chair of the Department of Clinical Bioethics, Warren G. Magnuson Clinical Center, National Institutes of health, in Bethesda, Maryland. HEALTH AFFAIRS Volume 24. Number 6 139 DO1 10 1377/hlthaff 24.6.1399 o2005 Project HOPE-The People-to-People Health Foundation,Inc

HEALT H CAR E REEOR M Health Care Reform: Why? What? When? What it might take to effect comprehensive change. by Victor R. Fuchs and Ezekiel J. Emanuel ABSTRACT: Dissatisfaction with the U.S. health care system is widespread, but no consen￾sus has emerged as to how to reform it. The principal methods of finance—employer-based insurance, means-tested insurance, and Medicare—are deeply and irreparably flawed. Policymakers confront two fundamental questions: Should reform be incremental or com￾prehensive? And should priority be given to reforming the financing system or to improving organization and delivery? We consider here several proposals for incremental reform and three for comprehensive reform: individual mandates with subsidies, single payer, and uni￾versal vouchers. Over the long term, reform is likely to come in response to a major war, de￾pression, or large-scale civil unrest. changes are Necessary, but what they Ought to be, what they will be, and how, and when to be produced are arduous questions. —John Jay (1787) T HESE WORDS, WRITTEN ON THE EVE OF THE Constitutional Conven￾tion, aptly describe U.S. health care in 2005.^ Most observers agree that re￾form is necessary, but whether that reform should be incremental or com￾prehensive and whether changes in finance or changes in organization and delivery should receive priority are questions that must be answered. What are the options before the country? What are the obstacles to reform? What are the conditions that might make reform possible? Before addressing these questions, however, it is useful to review briefly why reform deserves urgent attention. Why The United States Needs Health Care Reform The palpable symptoms of our "sick health care system" are described almost daily in popular print and broadcast media. Almost one of every six Americans have no health insurance, health care spending is escalating rapidly, administra￾tive costs are excessive, and medical errors (including overuse and underuse of medications and procedures) are rampant.^ Less frequently discussed, but of fun￾Victor Fuchs (fuchs@ncwagc3.stanford.edu) is the Henry]. Kaiser ]r Professor Emaitus at Stanford University in Stanford, California, and a research associate at the 'National Bureau of Economic Research, also in Stanford. Ezekiel Emanuel is chair ofthe Department of Clinical Bioethics, Warren G. Magnuson Clinical Center, National Institutes of Health, in Bethesda, Maryland. HEALTH AFFAIRS - Volume 24, Number 6 1399 DOI 10.1377/hltha£f.24.6.1399 C2005 Project HOPE-The People-to-Peopk Health Foundation, Inc

RETHINKING REFORM damental importance, are systemic problems in the financing of health care and in the organization and delivery of care Not discussed in this paper, because of space limitations and because their relation to health care is usually indirect, are many problems that affect population health, such as cigarette smoking, diet, exercise, air pollution, and road safety a Health care finance. Employer-based insurance. Ever since World War II, the cornerstone of U.S. health care finance has been employer-based insurance. Today such insurance still covers approximately 55 percent of the population, but with de clining coverage and loss of community rating, its role as quasi-social insurance has greatly eroded in recent decades. Competitive pressures on U.S. firms have increased as a result of antitrust actions, deregulation of many industries, and the inroads of foreign competition. Many fewer U.S. firms now enjoy steady monopoly profits that hey can draw on to subsidize health insurance for their worker For example, forty years ago the largest private employer was AT&T, a regu lated monopoly with guaranteed profits If health insurance premiums rose, they could easily be passed on to telephone subscribers. Moreover, aT&T was underno pressure to force older and sicker workers to pay more than their peers to compen sate for their higher use of care. Today the largest private employer is Wal-Mart, which despite its size faces intense competition daily from a host of other retail outlets. When they offer health insurance, it must come out of their workers' wages; for minimum-wage employees, this is not possible, so it often will mean loss of jobs The decline of unions to about one in twelve private-sector workers has also re duced the role of employer-based insurance as quasi-social insurance In indus tries comprising many small firms, such as residential construction, strong unions organize industrywide labor-management health insurance plans, which typi cally allow for sizable cross-subsidies among firms and among individual employ- ees within firms. In the past, tens of millions of workers were insured through such plans, but the disappearance of unions from most of the private sector tends to diminish the importance of this type of coverage Today most beneficiaries of employer-based insurance are in plans in which the firms are self-insured or are experience-rated (which means that they are self insured with a year lag). Community-rated premiums have disappeared, thus eliminating cross-subsidies among firms and industries, and the move toward large deductibles and health savings accounts(HSAs)will reduce cross-subsidies among workers within individual firms. 4 Employer-based insurance involves high administrative costs for the more than 1,000 insurance companies seeking contracts with millions of employers and for providers seeking reimbursement. Moreover, the large costs incurred by employ ers when negotiating contracts and administering benefits do not appear in the health care spending estimates. Employer-based insurance has other serious flaws: It distorts the labor-market decisions of workers and firms; it generates dis- 1400 November/ december 2

RETHINKIN G REFOR M damental importance, are systemic problems in the financing of health care and in the organization and delivery of care. Not discussed in this paper, because of space limitations and because their relation to health care is usually indirect, are many problems that affect population health, such as cigarette smoking, diet, exercise, air pollution, and road safety. • Health care finance. Employer-based insurance. Ever since World War II, the cornerstone of U.S. health care finance has been employer-based insurance. Today such insurance still covers approximately 55 percent of the population, but with de￾clining coverage and loss of community rating, its role as quasi-social insurance has greatly eroded in recent decades. Competitive pressures on U.S. firms have increased as a result of antitrust actions, deregulation of many industries, and the inroads of foreign competition. Many fewer U.S. firms now enjoy steady monopoly profits that they can draw on to subsidize health insurance for their workers. For example, forty years ago the largest private employer was AT&T, a regu￾lated monopoly with guaranteed profits. If health insurance premiums rose, they could easily be passed on to telephone subscribers. Moreover, AT&T was under no pressure to force older and sicker workers to pay more than their peers to compen￾sate for their higher use of care. Today the largest private employer is Wal-Mart, which despite its size faces intense competition daily from a host of other retail outlets. When they offer health insurance, it must come out of their workers' wages; for minimum-wage employees, this is not possible, so it often will mean loss of jobs. The decline of unions to about one in twelve private-sector workers has also re￾duced the role of employer-based insurance as quasi-social insurance.^ In indus￾tries comprising many small firms, such as residential construction, strong unions organize industrywide labor-management health insurance plans, which typi￾cally allow for sizable cross-subsidies among firms and among individual employ￾ees within firms. In the past, tens of millions of workers were insured through such plans, but the disappearance of unions from most of the private sector tends to diminish the importance of this type of coverage. Today most beneficiaries of employer-based insurance are in plans in which the firms are self-insured or are experience-rated (which means that they are self￾insured with a year lag). Community-rated premiums have disappeared, thus eliminating cross-subsidies among firms and industries, and the move toward large deductibles and health savings accounts (HSAs) will reduce cross-subsidies among workers within individual firms."* Employer-based insurance involves high administrative costs for the more than 1,000 insurance companies seeking contracts with millions of employers and for providers seeking reimbursement. Moreover, the large costs incurred by employ￾ers when negotiating contracts and administering benefits do not appear in the health care spending estimates. Employer-based insurance has other serious flaws: It distorts the labor-market decisions of workers and firms; it generates dis- 1400 November/December 2005

HEALTH CARE REFORM continuities in coverage; and, because premiums are exempt from income taxes,it provides a greater subsidy for high-wage than for low-wage workers Means-tested insurance. Since 1965 the financing system based on employer-based insurance has been shored up by means-tested insurance, such as Medicaid, and by Medicare. These programs, which together cover almost 30 percent of the pop- ulation, are also deeply flawed, but for different reasons. Means-tested insurance requires costly determination of eligibility, imposes high marginal tax rates on re- cipients because the subsidies fall or disappear as income rises, encourages eva sion of reported income, and generates discontinuities of coverage as recipients move into and out of eligibility. State governments, which share in the cost of these federally mandated programs, are finding it increasingly difficult to fund them. As a result, education and other state programs face budget cuts at a time when they need increases Medicare is popular with most beneficiaries and with many physicians and hospitals, but it has two big flaws On the benefit side, it is an open-ended entitle- ment that does not consider the costs of technologies relative to their benefits. Re- imbursement of physicians by fee-for-service also contributes to escalating ex penditures. Medicare expenditures are difficult to predict or control. For example, from 2003 to 2004 the bill for physician services jumped 15 percent. Ac cording to Medicare analysts, this was the result of longer and more intensive of fice visits, more laboratory tests, more frequent and complex imaging procedures, and greater use of in-office prescription drugs. With implementation of the new drug benefit in 2006, Medicare alone will consume 3. 3 percent of U.S. gross do- mestic product(GDP)(7.5 percent in 2035), and the Medicare Hospital Insurance Trust Fund is predicted to be depleted by 2019. Short of massive increases in taxes or a slowdown in spending growth, Medicare is headed for financial failure a Organization and delivery of care. Lack of IT. In addition to and in part be ause of flaws in the financing system, there are serious problems with the way med ical care is organized and delivered. More than half of all physicians are in sm practices; thus, they are poorly positioned to take advantage of advances such as in formation technology(IT), which could contribute to more efficient and effective care. Although large organizations such as the Palo Alto Medical Foundation have transformed their records system(without government subsidy), to the benefit of their patients, physicians, and the organization, for many physicians in small prac- tices, the cost of electronic medical records(EMRs)outweigh their benefits Quality control. Inadequate quality control is another problem. Although there re frequent statements of concern about quality, Donald Berwick, a leading ex- pert, recently expressed frustration with the slow progress that has been made. A critical examination of research on methods of quality improvement points out that even though "quality problems are widespread. reasons for these problems remain unclear. 0 The authors of this statement, Kaveh Shojania and jeremy Grimshaw, suggest several possible explanations: Providers may not know what HEALTH AFFAIRS Volume 24. Number 6 40l

HEALT H CAR E REFOR M continuities in coverage; and, because premiums are exempt from income taxes, it provides a greater subsidy for high-wage than for low-wage workers.^ Means-tested insurance. Since 1965 the financing system based on employer-based insurance has been shored up by means-tested insurance, such as Medicaid, and by Medicare. These programs, which together cover ahnost 30 percent of the pop￾ulation, are also deeply flawed, but for different reasons. Means-tested insurance requires costly determination of eligibility, imposes high marginal tax rates on re￾cipients because the subsidies fall or disappear as income rises, encourages eva￾sion of reported income, and generates discontinuities of coverage as recipients move into and out of eligibility.* State governments, which share in the cost of these federally mandated programs, are finding it increasingly difficult to fund them. As a result, education and other state programs face budget cuts at a time when they need increases. Medicare is popular with most beneficiaries and with many physicians and hospitals, but it has two big flaws. On the benefit side, it is an open-ended entitle￾ment that does not consider the costs of technologies relative to their benefits. Re￾imbursement of physicians by fee-for-service also contributes to escalating ex￾penditures. Medicare expenditures are difficult to predict or control. For example, from 2003 to 2004 the bill for physician services jumped 15 percent. Ac￾cording to Medicare analysts, this was the result of longer and more intensive of￾fice visits, more laboratory tests, more frequent and complex imaging procedures, and greater use of in-office prescription drugs.'' With implementation of the new drug benefit in 2006, Medicare alone will consume 3.3 percent of U.S. gross do￾mestic product (GDP) (7.5 percent in 2035), and the Medicare Hospital Insurance Trust Fund is predicted to be depleted by 2019.^ Short of massive increases in taxes or a slowdown in spending growth. Medicare is headed for financial failure. • Organization and delivery of care. Lack of IT. In addition to and in part be￾cause of flaws in the financing system, there are serious problems with the way med￾ical care is organized and dehvered. More than half of all physicians are in small practices; thus, they are poorly positioned to take advantage of advances such as in￾formation technology (IT), which could contribute to more efficient and effective care. Although large organizations such as the Palo Alto Medical Foundation have transformed their records system (without government subsidy), to the benefit of their patients, physicians, and the organization, for many physicians in small prac￾tices, the cost of electronic medical records (EMRs) outweigh their benefits. Quality control. Inadequate quality control is another problem. Although there are frequent statements of concern about quality, Donald Berwdck, a leading ex￾pert, recently expressed frustration with the slow progress that has been made.' A critical examination of research on methods of quality improvement points out that even though "quality problems are widespread...reasons for these problems remain unclear."'" The authors of this statement, Kaveh Shojania and Jeremy Grimshaw, suggest several possible explanations: Providers may not know what HEALTH AFFAIRS - Volume 24, Number 6 1401

RETHINKING REFORM experts recommend; they may know but disagree with the experts; the support systems needed to comply with the recommendations may be absent; or financial incentives may be misaligned. As has been observed in many contexts, it is diffi- cult to get people to understand something when their income depends on their t understanding it Cost-benefit trade-o ffs. Another major problem with care organization and deliv ery is insufficient attention to benefit-cost trade-offs. Most physicians are consci- entious about evaluating the benefits versus the risks of any intervention, but many fewer will consider costs, except perhaps for uninsured patients. Physicians often do not know what the costs are; they may believe that their responsibility is to deliver the best care that is technically possible, regardless of costs; or as Shojania and Grimshaw write, "Financial incentives may be misaligned. "ll Over the years, there has been a plethora of recommendations as to how physicians can contribute to more cost-effective care, but the fact that they have not been impl mented on a wide scale suggests that the problem is systemic in nature. Some ex perts advocate a complete overhaul of the financing of health care to give physi cians the information, opportunity, and incentive to deliver cost-effective care What Kind of reform? Policymakers seeking to reform health care face two fundamental questions Should reform be incremental or comprehensive? And should reform focus first on the financing of care or on the organization and delivery of care? a Incremental reform. Employer mandates. Most incremental reform proposals fo- cus on financing-in particular, on reducing the number of uninsured people. Some reformers seek to increase coverage by mandating that all employers above some specified size offer their workers health insurance; these mandates may or may not pe accompanied by subsidies or tax credits to the firms. One economic rationale for employer mandates is that the cost of care for uninsured workers is often passed along to the insured through taxes and other mechanisms. Mandates are, in part,an attempt to eliminate" free riders. " Their possible negative effects include loss of er ployment, especially for workers who are at or near the minimum wage, which makes it impossible for firms to pass on the costs by reducing wages. Firms that are below the specified minimum size that trigger the mandate could be discouraged from expanding or could resort to using more part-time or temporary workers toes cape the mandate. When mandates are legislated at the state rather than the federal level, firms have an incentive to move to a state that does not have a mandate, even though economic efficiency would be greater in their original location Subsidies. Another incremental approach would provide subsidies for the unin sured, usually through tax credits, to purchase insurance in the individual mar- ket. The main advantage claimed for this approach is that it increases freedom of choice and does not intrude directly on the labor market. It could have indirect ef- fects such as reducing the numbers of workers covered by employers. If subsidies 1402 ember/December 2005

RETHINKIN G REFOR M experts recommend; they may know but disagree with the experts; the support systems needed to comply with the recommendations may be absent; or financial incentives may be misaligned. As has been observed in many contexts, it is diffi￾cult to get people to understand something when their income depends on their not understanding it. Cost-benefit trade-offs. Another major problem with care organization and deliv￾ery is insufficient attention to benefit-cost trade-offs. Most physicians are consci￾entious about evaluating the benefits versus the risks of any intervention, but many fewer wall consider costs, except perhaps for uninsured patients. Physicians often do not know what the costs are; they may believe that their responsibility is to deliver the best care that is technically possible, regardless of costs; or as Shojania and Grimshaw write, "Financial incentives may be misaligned."" Over the years, there has been a plethora of recommendations as to how physicians can contribute to more cost-effective care, but the fact that they have not been imple￾mented on a wdde scale suggests that the problem is systemic in nature. Some ex￾perts advocate a complete overhaul of the financing of health care to give physi￾cians the information, opportunity, and incentive to deliver cost-effective care. What Kind Of Reform? Policymakers seeking to reform health care face two fundamental questions: Should reform be incremental or comprehensive? And should reform focus first on the financing of care or on the organization and delivery of care? • Incremental reform. Employer mandates. Most incremental reform proposals fo￾cus on financing—in particular, on reducing the number of uninsured people. Some reformers seek to increase coverage by mandating that all employers above some specified size offer their workers health insurance; these mandates may or may not be accompanied by subsidies or tax credits to the firms. One economic rationale for employer mandates is that the cost of care for uninsured workers is often passed along to the insured through taxes and other mechanisms. Mandates are, in part, an attempt to eliminate "free riders." Their possible negative effects include loss of em￾ployment, especially for workers who are at or near the minimum wage, which makes it impossible for firms to pass on the costs by reducing wages. Firms that are below the specified minimum size that trigger the mandate could be discouraged from expanding or could resort to using more part-time or temporary workers to es￾cape the mandate. When mandates are legislated at the state rather than the federal level, firms have an incentive to move to a state that does not have a mandate, even though economic efficiency would be greater in their original location. Subsidies. Another incremental approach would provide subsidies for the unin￾sured, usually through tax credits, to purchase insurance in the individual mar￾ket.^^ The main advantage claimed for this approach is that it increases freedom of choice and does not intrude directly on the labor market. It could have indirect ef￾fects such as reducing the numbers of workers covered by employers.'^ If subsidies 1402 November/December 2005

HEALTH CARE REFORM are geared to income, there are the additional disadvantages associated with de termining eligibility and disincentives imposed on people who might increase heir income Medicare and Medicaid. Two other approaches for reducing the number of unin sured people are to build on the Medicaid and Medicare programs For Medicaid, this would involve raising the income level for eligibility; for Medicare, it would involve lowering the age for eligibility. 4 Expansion of these programs would mag ify their existing advantages and their disadvantages. In addition, if nonpoor, working-age people become eligible for these public programs, there would prob ably be a decline in the number covered by employer-based insurance and even some decline in labor-force participation Health savings accounts. Some incremental reform proposals have objectives other than reducing the number of uninsured people. Consumer-directed health care subsidized by favorable tax treatment of HSAs, aims at making patients more cost-conscious, leading to usage reductions and possibly more price competition among providers. 6 It is also said that if costs to individuals vary with use, they will choose healthier behavior, such as stopping cigarette smok ing. Out-of pocket payments do give patients an incentive to use less care; whether they are able to make appropriate choices is much more doubtful. The AND Health Insurance Experiment showed that patients with a higher percent age of out-of-pocket expense use less care, but the proportion of care that experts deem "appropriate"did not vary with the extent of insurance coverage ere a re several reasons for thinking that HSAs, or large deductibles in gen eral, would not have as favorable an effect on utilization as advocates claim First a large fraction of health spending is accounted for by a small proportion of pa tients-patients whose spending levels will be far above their deductible. Second even for those who have not yet exceeded their deductible but expect to do so be fore the end of the year, any particular test, visit, or procedure will effectively be free because the patient's total outlay(the deductible)would be the same, regard less of whether or not they get the particular service. Third, a considerable amount of care is elective with respect to timing. People who have exceeded their deduct ible have a great incentive to undergo in the same year all of the tests and other procedures that they are contemplating because there will be no cost to them. Fi- nally, a deductible that might be reasonable for a high -wage worker would be un reasonable for one making much less. Thus, there will be pressure to have the de ductible vary with income, and that will give rise to other problems, including increasing the administrative costs of such plans Managed competition. Managed competition is another incremental reform pro- posal. Although in principle it can be applied to all health coverage, in recent years at improv efficiency of employer-based insur- ance. The leading proponent of managed competition, Alain Enthe oven eve that employers would see its advantages and voluntarily adopt it. 8 Some have, but HEALTH AFFAIRS Volume 24, Number 6

HEALT H CAR E REEOR M are geared to income, there are the additional disadvantages associated with de￾termining eligibility and disincentives imposed on people who might increase their income. Medicare and Medicaid. Two other approaches for reducing the number of unin￾sured people are to build on the Medicaid and Medicare programs. For Medicaid, this would involve raising the income level for eligibility; for Medicare, it would involve lowering the age for eligibility.''' Expansion of these programs would mag￾nify their existing advantages and their disadvantages. In addition, if nonpoor, working-age people become eligible for these public programs, there would prob￾ably be a decline in the number covered by employer-based insurance and even some decline in labor-force participation.'^ Health savings accounts. Some incremental reform proposals have objectives other than reducing the number of uninsured people. Consumer-directed health care, subsidized by favorable tax treatment of HSAs, aims at making patients more cost-conscious, leading to usage reductions and possibly more price competition among providers.'^ It is also said that if costs to individuals vary with use, they will have an incentive to choose healthier behavior, such as stopping cigarette smok￾ing. Out-of-pocket payments do give patients an incentive to use less care; whether they are able to make appropriate choices is much more doubtful. The RAND Health Insurance Experiment showed that patients with a higher percent￾age of out-of-pocket expense use less care, but the proportion of care that experts deem "appropriate" did not vary with the extent of insurance coverage.''' There are several reasons for thinking that HSAs, or large deductibles in gen￾eral, would not have as favorable an effect on utilization as advocates claim. First, a large fraction of health spending is accounted for by a small proportion of pa￾tients—patients whose spending levels will be far above their deductible. Second, even for those who have not yet exceeded their deductible but expect to do so be￾fore the end of the year, any particular test, visit, or procedure will effectively be free because the patient's total outlay (the deductible) would be the same, regard￾less of whether or not they get the particular service. Third, a considerable amount of care is elective with respect to timing. People who have exceeded their deduct￾ible have a great incentive to undergo in the same year all of the tests and other procedures that they are contemplating because there will be no cost to them. Fi￾nally, a deductible that might be reasonable for a high-wage worker would be un￾reasonable for one making much less. Thus, there will be pressure to have the de￾ductible vary with income, and that will give rise to other problems, including increasing the administrative costs of such plans. Managed competition. Managed competition is another incremental reform pro￾posal. Although in principle it can be applied to all health coverage, in recent years it has been aimed primarily at improving the efficiency of employer-based insur￾ance. The leading proponent of managed competition, Alain Enthoven, believed that employers would see its advantages and voluntarily adopt it.'^ Some have, but HEALTH AFFAIRS - Volume 24, Number 6 1403