Medicare spending and financing: a Primer, Potetz, Cubanski and Neuman

Hospital Costs 10.1377/hlthaff.2013.1327 HEALTH AFFAIRS 33, NO. 9 (2014): 1586–1594 ©2014 Project HOPE— The

ORDER A PLAGIARISM FREE PAPER NOW

People-to-People Health Foundation, Inc. doi: David U. Himmelstein (dhimmels@hunter.cuny.edu) is an internist; a professor at the School of Public Health and Hunter College, City University of New York (CUNY), in New York City; and a lecturer at Harvard Medical School. Miraya Jun was a research officer at the London School of Economics and Political Science (LSE), in the United Kingdom, at the time of this study. She is now an independent consultant to the LSE. Reinhard Busse is a professor of health care management at the Technische Universität Berlin–World Health Organization Collaborating Centre for Health Systems Research and Management, in Berlin, Germany. Karine Chevreul is the deputy director of the Paris Health Services and Health Economics Research Unit at the Assistance Publique– Hôpitaux de Paris (the Paris area’s University Medical Center) and deputy director of ECEVE (UMR 1123), a research team of the French National Institute of Medical Research, in Paris, France. Alexander Geissler is a senior research fellow in health care management at the Technische Universität Berlin, in Germany. Patrick Jeurissen is head of the Celsus Academy on Sustainable Healthcare, Nijmegen Medical Centre, Radboud University, in Nijmegen, the Netherlands. 1586 By David U. Himmelstein, Miraya Jun, Reinhard Busse, Karine Chevreul, Alexander Geissler, Patrick Jeurissen, Sarah Thomson, Marie-Amelie Vinet, and Steffie Woolhandler A Comparison Of Hospital Administrative Costs In Eight Nations: US Costs Exceed All Others By Far ABSTRACT A few studies have noted the outsize administrative costs of US hospitals, but no research has compared these costs across multiple nations with various types of health care systems. We assembled a team of international health policy experts to conduct just such a challenging analysis of hospital administrative costs across eight nations: Canada, England, Scotland, Wales, France, Germany, the Netherlands, and the United States. We found that administrative costs accounted for 25.3 percent of total US hospital expenditures—a percentage that is increasing. Next highest were the Netherlands (19.8 percent) and England (15.5 percent), both of which are transitioning to market-oriented payment systems. Scotland and Canada, whose single-payer systems pay hospitals global operating budgets, with separate grants for capital, had the lowest administrative costs. Costs were intermediate in France and Germany (which bill per patient but pay separately for capital projects) and in Wales. Reducing US per capita spending for hospital administration to Scottish or Canadian levels would have saved more than $150 billion in 2011. This study suggests that the reduction of US administrative costs would best be accomplished through the use of a simpler and less market-oriented payment scheme. A ll nations struggle with rising health care costs, but the United States remains a cost outlier. In 2010 it spent 17.6 percent of its gross domestic product on health care—far more than the next-highest spenders, the Netherlands (12.0 percent) and France and Germany (both 11.6 percent).1 Several factors help explain the US excess spending: greater use of high-tech interventions;2 more emphasis on specialty care and the underprovision of primary care; 3 higher drug prices;4 and higher physician fees.5 A few studies have noted US health insurers’ and providers’ outsize administrative costs, mostly in relation to Canadian costs.6–13 How- Health Affairs S ep t e m b e r 20 1 4 ever, no research has compared the administrative costs of hospitals across nations representing a broad spectrum of health care systems. Cross-national differences in accounting standards make such international comparisons challenging. To address this challenge, we assembled an international team of health policy experts to analyze hospital administrative costs for eight nations: Canada, England, Scotland, Wales, France, Germany, the Netherlands, and the United States. This article summarizes the findings of this research team and offers some lessons for policy makers who are searching for payment strategies that minimize administrative overhead. 33 : 9 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. Study Data And Methods Data Sources And Analysis To assess the impact of a range of payment strategies, we analyzed data from nations with widely varying health care systems. Three of the nations— England, Scotland, and Wales—are within the United Kingdom. Each has a public National Health Service (NHS) funded by taxes, but the three systems vary in their hospital funding. Canada has a single-payer public insurance system in each province. France has a system akin to a single-payer social insurance model. However, payments are funneled through several nominally separate insurance funds. Germany and the Netherlands have compulsory, multipayer social insurance systems, but the Netherlands is transitioning to a market-based payment system. The United States has a largely private, multipayer health care system. For each nation we obtained official hospital cost accounting data that covered most or all hospitals. The data were for 2010 or 2011. Starting with the comprehensive Medicare Cost Reports submitted by US hospitals, we developed a classification scheme that apportioned costs between clinical and administrative functions, including information technology (IT).We distributed a few costs, such as employee benefits, between the clinical and administrative categories.We allocated capital costs to administrative and clinical cost centers based on each center’s share of total operating expenses. We excluded research and teaching costs. These methods emulate those employed in previous analyses of US and Canadian hospitals.9 The level of detail in the Medicare data allowed us to identify administrative costs incurred at any US hospital location—for example, costs for a ward secretary or a clinic receptionist. Some administrative arms of clinical functions, such as nursing administration, were categorized separately. In other cases, Medicare required hospitals to allocate administrative costs incurred in clinical units to administrative categories. Data for Canada, the Netherlands, England, Scotland, and Wales were sufficiently detailed to allow full replication of this analysis. However, in the German and French data, clerical work performed at clinic or ward locations was sometimes charged to a clinical cost center, as were some IT costs. Hence, for these two nations we could not fully apply the US-based classification scheme. Instead, we constructed an alternative, narrower measure for the German and French data, which we called central administration costs. This category excluded IT costs and administrative or clerical work on wards and at other clinical locations. Data to calculate this narrower measure were available for all but the UK nations. For each of the eight nations we reviewed detailed documentation describing hospital expense categories, and we mapped those categories to the US ones. In most cases, this mapping was straightforward, because the available documentation provided sufficiently detailed descriptions or lists of items subsumed under each category to resolve ambiguities.When uncertainties remained, we obtained additional specific descriptions of the items included in the category from national experts and officials. In some cases, we also consulted Medicare auditors to ascertain where such items would be classified in the US cost reporting scheme. The online Appendix summarizes the data sources and classification schemes employed for each nation.14 However, the voluminous documentation of the cost reporting schemes for several nations precluded listing all of the available details even in the Appendix. For instance, the instruction manual for Medicare Cost Reports is over 500 pages long. To generate per capita cost estimates, we assumed that the administration share of costs at hospitals for which we lacked data (for example, those in Quebec and private hospitals in England) was the same as the administration share at other hospitals in that nation. All figures were adjusted to US dollars using purchasing power parities for the appropriate year. Time trend data on administrative costs were available only for the United States and Canada. However, time trend data on administrative fulltime equivalents (FTEs) as a share of total FTEs (which likely tracks trends in the administration share of costs) in the hospital and community health sectors were available for the United Kingdom. This allowed us to assess precise time trends for administrative costs in the United States and Canada and approximate time trends in the United Kingdom. Limitations Several caveats apply to our findings. First, nations differ in many ways besides health care financing. The mix of services provided by hospitals, especially their role in ambulatory care, varies across nations. Many US hospitals operate outpatient clinics that provide both specialty and primary care. In contrast, hospitals in most other nations provide only specialty outpatient services. Similarly, our figures for US, Canadian, and Dutch hospitals excluded most physician compensation. In contrast, the hospital spending figures in the other nations included substantial physician compensation for care delivered on the premises. For instance, German hospitals employ large numbers of physicians whose average pay is relatively low. S ep t e m b e r 20 1 4 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. 3 3: 9 Sarah Thomson is an associate professor in the Department of Social Policy, London School of Economics, and a senior research associate at the European Observatory on Health Systems and Policies, in London, England. Marie-Amelie Vinet is a health economist at the Paris Health Services and Health Economics Research Unit at the Assistance Publique– Hôpitaux de Paris and also a member of the ECEVE team (UMR 1123) of the French National Institute of Medical Research. Steffie Woolhandler is an internist; a professor at the School of Public Health and Hunter College, CUNY; and a lecturer at Harvard Medical School. Health Affairs 1587 Hospital Costs Even the definition of hospital may vary somewhat both within and across nations. For instance, in some nations, hospital accounts include the costs of ambulance services. Some US hospitals’ Medicare Cost Reports include some services that are provided by affiliated home care agencies, while others’ reports cover only those activities carried out within the hospital’s walls—as is generally the case with financial figures for hospitals in some other nations. However, these differences across nations should not have greatly distorted our estimates. In all nations, the core inpatient services account for the bulk of budgets. Moreover, previous studies have found that at least for the United States and Canada, administrative costs associated with physician compensation (equivalent to 26.9 percent of physicians’ gross incomes in the United States versus 16.1 percent in Canada) were similar, in percentage terms, to hospital administrative costs.9 In contrast, Dutch hospital expenditures include some costs of administering reimbursements for physicians not employed by the hospitals, which would have led us to slightly overstate hospital administrative costs. A further limitation is that our data sources excluded some hospitals in most of the nations we studied (notably, eight university centers in the Netherlands) and a larger number of institutions (NHS Foundation Trust and private hospitals) in England. However, limited data from NHS Foundation Trusts’ audited year-end accounts for 2010–11 indicate that their administrative staffing levels are similar to those of the NHS hospitals in England that we studied. UK private hospitals’ administrative costs may be higher than those of NHS hospitals, but they account for a small proportion of expenditures. Furthermore, the omission of a few large Dutch university hospitals is unlikely to distort our estimates, since size was not related to administrative costs among the hospitals in the Netherlands for which we had data. For the United States, we lacked data on military hospitals and those in the Department of Veterans Affairs, which do not file Medicare Cost Reports. The exclusion of these federal hospitals with global budgets, which probably have low administrative costs, might have caused us to slightly overestimate US administrative costs. However, Medicare Cost Reports omit profits and most advertising, which cannot be billed to Medicare. This would have caused us to underestimate US overhead costs. Other limitations are that there is no international standard for hospital cost accounting, and that our alignment of categories was imperfect. Our analysis allocated some capital costs to ad1588 H e a lt h A f fai r s September 2014 The proportion of hospital costs devoted to administration was highest in the United States, at 25.3 percent. ministration, based on the administration share of operating expenses. Our analysis handled capital costs uniformly across the eight nations. However, it should be noted that Dutch hospitals’ capital costs are higher than those in the United States, and about double those of the other European nations. Our data do not address the question of which components of administrative spending drive international differences. However, fragmentary data from other sources suggest that a larger number of managers and clerical workers—not differences in wage levels, benefit costs, or nonwage costs—explains much or all of the higher administrative costs in US hospitals compared to hospitals in the other nations we studied.8,11,15,16 Finally, our study did not include the administrative costs of insurers and regulators who deal with hospital payments. Study Results Exhibit 1 presents an overview of the health systems and hospital funding mechanisms of the eight nations. For additional details on coverage and hospital payment in the eight nations, see Appendix Exhibit A1.14 Canada, Scotland, and Wales pay hospitals global operating budgets (similar to the way in which a US firehouse is funded), with separate grants for capital needs such as new buildings and expensive new equipment. France and Germany use tightly regulated all-payer diagnosisrelated group (DRG) payment systems, with separate public grants for most capital needs. England also uses all-payer DRGs, but hospitals negotiate contracts for some services with local agencies. The Netherlands combines elements of DRG-like payment with market-based pricing (for example, pricing based on bargaining between individual hospitals and individual insurers). In both England and the Netherlands, hospitals increasingly depend on operating surpluses or profits to meet their capital needs.17,18 Health care spending in 2010 ranged from 33:9 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. Exhibit 1 Principal Hospital Financing Characteristics Of Eight Nations, 2011 Nation Insurance coverage Funding for hospital operating budgets Primary source of capital funds US Multipayer; loosely regulated; substantial OOP; many people uninsured Per patient payments; mechanisms (such as DRGs, per diem, and FFS), regulations, and rates differ by payer Operating surpluses or profits Canada Single public payer in each province; universal coverage for hospital and physician care; minimal OOP; private coverage only for items not covered by public plan Global, lump-sum budgets Funds allocated directly by the provincial government France Universal social insurance; minimal OOP; optional private coverage reimburses patients’ cost sharing Tightly regulated, multipayer social insurance; minimal OOP; higher-income people may opt for private insurance with enhanced services and higher premiums Regulated, multipayer, private insurance; compulsory basic benefit package; optional supplementary coverage; minimal OOP DRGs, uniform for all patients Lump-sum payments for capital and other public missions DRGs, uniform for all patients Lump-sum payments from the states DBCs (DRG-like system): about 30,000 DBCs; rates uniform for 2/3 of DBCs, negotiated between hospital and insurer for 1/3 Operating surpluses and capital add-ons included in the uniform DBC rates, but not in negotiated DBCs rates England Universal NHS coverage; prominent market features; most services purchased at local level by groups of GPs; minimal OOP; private coverage for care outside the NHS 60% from DRGs with uniform rates; 40% from lump-sum contracts negotiated with local agencies Operating surpluses, with a central review of planned major investments Scotland Universal NHS coverage with few market features; virtually no OOP; private coverage for care outside the NHS Universal NHS coverage with decreasing market features since 1999; virtually no OOP; private coverage for care outside the NHS Global, lump-sum budgets Funds allocated directly by the government Global, lump-sum budgets Funds allocated directly by the government Germany Netherlands Wales SOURCE Authors’ analysis. NOTES OOP is out-of-pocket, or patients’ spending. DRG is diagnosis-related group. FFS is fee-for-service. DBC is diagnostic-treatmentcombination. NHS is National Health Service. GP is general or family practitioner. 9.6 percent of GDP in the United Kingdom to 17.6 percent in the United States (Exhibit 2). Germany had the largest supply of both hospital beds and physicians per 1,000 population, while the United States had the most specialists, measured as a percentage of all physicians. The US population had smaller percentages of elderly people and smokers, compared to the populations of other countries, but its percentage of obese people was second only to Scotland’s (Exhibit 2). Life expectancy was similar in the United States and Scotland, trailing that in the other nations by about two years. Hospitals’ Total Administrative Costs The proportion of hospital costs devoted to administration was highest in the United States, at 25.3 percent (Exhibit 3). This was more than twice the percentages for Canada and Scotland, which spent the least on administration. Hospitals’ administrative costs were notably higher in the Netherlands than in other European nations. Differences were more marked when expressed as a percentage of GDP or in dollars per capita. For example, hospital administration costs ranged from 1.43 percent of GDP in the United States ($667 per capita) to 0.41 percent of GDP ($158 per capita) in Canada (Exhibit 3). Among the UK nations, Scotland’s administrative costs were lowest, England’s were highest, and Wales’s were in between (Exhibit 3). This ranking correlates roughly with the role of market mechanisms in those nations’ health care systems. The NHS internal market reforms introduced throughout the United Kingdom during the 1990s separated the commissioning and provision of care, with price-based competition among hospitals. Scotland reversed these market-based reforms soon after devolution in 1999; Wales did so somewhat later, in 2009. In the United States, for-profit hospitals had higher administrative costs (27.2 percent) than did nonprofit (25.0 percent) or public (22.8 percent) institutions. Teaching hospitals, few of which are for-profit, had lower-than-average administrative costs (23.6 percent), as did rural facilities (24.7 percent, compared to 25.5 percent for urban hospitals). Administrative costs for hospitals in Maryland September 2014 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. 33:9 H ea lt h A f fai r s 1 589 Hospital Costs Exhibit 2 Demographic Characteristics And Health Expenditures, Resources, And Indicators For Eight Nations UK US Canada France Germany Netherlands England Scotland Wales Population older than 64 (%) GDP per capita (PPP-adjusted US $) Smokers (percent of population older than 14)c Obese people (percent of population older than 14)c People with insurance (percent of population) 13.1 46,747 15.1 28.1 81.3d 14.4 39,070 16.3 17.5 100.0d 17.3 34,136 23.3 12.9 99.9d 20.7 37,402 21.9d 17.3d 100.0d 15.6 42,166 20.9 11.6 98.8d 16.2 35,687 21.5d 26.1 100.0d 16.8 32,215a 24.0 28.2 100.0d 18.6 32,239b 23.0 22.0 100.0d Expenditures Health care spending Per capita (PPP-adjusted US $) Percent of GDP Health insurance overhead and government health administration per capita (PPP-adjusted US $) 8,233 17.6 4,445 11.4 3,974 11.6 4,338 11.6 5,056 12.0 3,433e 9.6e 587 147 274 233 183 —f —f —f Resources Physicians Number (per 1,000 population) Percent specialists Hospital beds (per 1,000 population) Average length of acute care hospital stay (days) 2.6 87.7 3.1d 5.4 2.4 53.0 3.2d 7.7 3.3 51.3 6.4 5.2 4.1 58.0 8.3 7.3 2.9d 57.7 4.7d 5.6 2.7 70.9e 3.0 6.6 2.3b 2.5 3.3 4.8 4.0 6.2 Health indicators Life expectancy (years) Females Males Infant mortality (per 1,000 live births) 81.1 76.2 6.1 83.1g 78.5g 5.1g 84.7 78.0 3.6 83.0 78.0 3.4 82.7 78.8 3.8 82.6 78.6 4.2 80.6d 76.0d 3.7 81.8 77.6 4.0 Demographic characteristics SOURCE Authors’ analysis of health data from the following sources: (1) Organization for Economic Cooperation and Development. OECD health statistics (see Note 1 in text). (2) Scottish Government. Health and community care [Internet]. Edinburgh: Scottish Government; [cited 2014 May 7]. Available from: http://www.scotland.gov.uk/ Topics/Statistics/Browse/Health. (3) Welsh Government. Health statistics Wales [Internet]. Cardiff: Welsh Government; 2012 [cited 2014 May 20]. Available from: http:// wales.gov.uk/docs/statistics/2012/120927hsw12en.pdf. NOTES Data are for 2010 except where otherwise indicated. PPP is purchasing power parity. aExcludes costs for care outside of Scotland. bData are for 2011. cOlder than fifteen for Scotland and Wales. dData are for 2009. eData are for England, Scotland, Wales, and Northern Ireland. f Not available. gData are for 2008. Exhibit 3 Total Hospital Administrative Costs And Spending In Eight Nations, 2010 UK US Canada France Germany Netherlands England Scotland Wales Total hospital expenditures Per capita, (PPP-adjusted US $) Share of GDP (%) 2,634 5.63 1,271 3.25 1,357 3.98 1,245 3.33 1,631 3.87 1,458a 4.09a 1,416 4.39 1,482 4.60 Central administrationb Share of hospital costs (%) 15.51 7.40 8.77 9.00 10.85 —c 25.32 1.43 667 12.42 0.41 158 —c —c —c —c —c —c 19.79 0.77 323 15.45 0.63a 225a —c —c Hospital administration Share of hospital costs (%) Share of GDP (%) Expenditures per capita (PPP-adjusted US $) 11.59 0.51 164 14.27 0.66 211 SOURCE Authors’ analysis of data from the following sources: (1) Organization for Economic Cooperation and Development. OECD health statistics 2014 (see Note 1 in text). (2) Information Services Division, NHS National Services Scotland. Net expenditure, by board of treatment, by care type [Internet]. Edinburgh: NHS National Services Scotland; 2012 [cited 2014 Jul 23]. Available from: http://www.isdscotland.org/Health-Topics/Finance/Publications/2011-11-29/Costs_R300s_2011.xls. (3) Welsh Government. Health statistics Wales. Cardiff: Welsh Government; 2012. (4) Form TFR3E, the Final Accounts NHS Trusts TFR (Treasury Financial Reports) for 2011. (5) Monitor—independent regulator of NHS foundation trusts. NHS foundation trusts: consolidated accounts 201/11 [Internet]. London: Stationery Office; 2011 Jul 14 [cited 2014 May 7]. Available from: http://www.monitor-nhsft.gov.uk/sites/default/files/NHS%20Foundation%20Trusts%20Consolidated%20Accounts%201011 %20website%20file.pdf. NOTES Data for the Netherlands are for 2011. Data for England, Scotland, and Wales are for April 1, 2010–March 31, 2011. Figures for Scotland and Wales are for National Health Service (NHS) hospitals only. PPP is purchasing power parity. GDP is gross domestic product. aIncludes NHS Trusts and Acute NHS Foundation Trusts. bCentral administration costs exclude costs of information technology and of administrative or clerical work on wards and at other clinical locations. cNot available. 1590 Health Affairs S ep t e m b e r 20 1 4 33 : 9 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. (the only state with all-payer hospital rate setting, the type of reform that some policy experts suggest might reduce administrative costs)19 were 25.2 percent of total hospital costs. This did not differ from the national average (p ¼ 0:94). Despite Maryland’s all-payer rate-setting system, copayments, deductibles, documentation requirements, clinical guidelines, and so forth differ across payers.20 Hospitals’ Central Administration Costs Hospitals’ central administration costs followed a pattern similar to that for total administrative costs. Central administration costs were highest in the United States, followed by the Netherlands (Exhibit 3). Time Trends US hospital administrative costs rose from 23.5 percent of total hospital costs ($97.816 billion) in 2000 to 25.3 percent ($215.369 billion) in 2011. In the same period, the hospital administration share of GDP rose from 0.98 percent to 1.43 percent (Exhibit 4). The proportion spent on administration by Canadian hospitals fell slightly from 1999 (12.9 percent)9 to 2011 (12.4 percent). The administration share of hospital FTEs in the United Kingdom rose from 13.8 percent in 1980 to 23.9 percent in 2009.21 This change reflects mostly trends in England, where 84 percent of the UK population lives, and coincided with market-oriented reforms. The UK time trends are shown in Appendix Exhibit A2.14 Discussion Hospitals’ administrative overhead varied more than twofold across the nations we studied as a share of total hospital costs and more than fourfold in absolute terms. These costs were far higher in the United States than elsewhere. What Lies Behind These Differences ? In all nations, hospital administrators must procure and coordinate the facilities, supplies, and personnel needed for good care. In nations where administrators have few responsibilities beyond these logistical matters, administration seems to require about 12 percent of hospital expenditures. Modes of hospital payment can increase the complexity and costs associated with two additional management tasks: garnering operating funds and securing capital funds for modernization and expansion. Garnering operating funds requires little administrative work in nations such as Canada, Scotland, and Wales, where hospitals receive global, lump-sum budgets. In contrast, per patient billing (for example, using DRGs) requires additional clerical and management personnel and special-purpose IT systems. This is true even in countries—such as France and Germany— where payment rates, documentation, and billing procedures are uniform. Billing is even more complex in nations where each hospital must bargain over payment rates with multiple payers, whose documentation requirements and billing procedures often vary, as is the case in the United States and the Netherlands. Differences in how hospitals obtain capital funds also appear to affect administrative costs. The combination of direct government grants for capital with separate global operating budgets— as in Scotland and Canada—was associated with the lowest administrative costs. (Wales has recently transitioned to such a system, reversing previous market reforms.) Hospitals in France Exhibit 4 US Hospital Administration Costs As A Percentage Of Gross Domestic Product (GDP), 2000–11 SOURCE Authors’ analysis of data from Medicare Hospital Cost Reports. September 2014 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. 33:9 H e a lt h A f fai r s 1591 Hospital Costs and Germany, where direct government grants account for a substantial share of hospital capital funding, have relatively low administrative costs despite per patient, DRG-based billing. Administration is costliest in nations where surpluses from day-to-day operations are the main source of hospital capital funds: the United States and, increasingly, the Netherlands and England. In such health care systems, the need to accumulate capital funds for modernization and expansion stimulates administrators to undertake the additional work that is needed to identify and pursue profit opportunities. This entrepreneurial incentive rewards hospitals that cut unnecessary operating costs and thereby improves efficiency. However, it can also reward hospitals for devoting resources to activities that decrease efficiency, such as advertising; upcoding bills—that is, exaggerating the severity of patients’ illnesses in order to bill for higher DRGs;22 and cherry-picking profitable patients, physicians and services while avoiding unprofitable ones. The performance of US for-profit hospitals— whose explicit goal is profitability and whose administrative costs are high—helps clarify whether, on balance, entrepreneurial incentives improve efficiency. Compared to other US hospitals, for-profit institutions spend less on clinical personnel such as nurses23 but provide costlier care.24,25 Similarly, in Germany for-profit hospitals don’t appear to be more efficient than other hospitals.26 The divergence between Scotland and England is also instructive. Administrative costs are low in Scotland, where hospitals don’t bill for individual patients and capital projects are funded by direct government grants—which leaves administrators little leeway for financial entrepreneurship. In contrast, the administration share of costs is higher (and apparently rising) in England, where per patient billing has largely replaced global hospital budgets and recent market-based reforms encourage entrepreneurialism. Hospital administrative costs appear to be driven by the complexity of the reimbursement system and the mode of capital funding. However, other factors could explain our findings. The greater intensity of care in US hospitals might explain why administrative costs are higher in that country than elsewhere. But the relatively low administrative costs of US teaching hospitals (which have high care intensity) argues against this explanation. A heavier regulatory burden in the United States and the Netherlands than elsewhere might also impose administrative costs on hospitals. Some of this burden—for example, regu1592 Health Affairs S ep t e m b e r 20 1 4 Hospital administrative costs appear to be driven by the complexity of the reimbursement system and the mode of capital funding. lations regarding privacy and translators in the United States—is unrelated to payment. Nonetheless, much of it reflects the tussle over reimbursement. Our findings could also reflect a shift of responsibility (and costs) for some planning and budgeting tasks out of hospital offices and into the offices of government agencies and insurers in nations that have more centrally directed hospital systems. Perhaps the use of global budgets, regulated DRG pricing, and centralized capital allocation increases out-of-hospital costs to administer hospital payments and to monitor hospitals’ activity and compliance. Our hospitalbased analysis would not capture such costs, but they must be modest: Other nations spend far less than the United States on administration by government and insurers (Exhibit 2). Do Higher Administrative Costs Yield Benefits ? If more administration eliminated clinical waste or enhanced patients’ choices and market competition, administration’s share might rise, but total costs would fall.27 However, we found the opposite pattern: Total hospital costs were highest in the nations that had the highest hospital administrative costs. Moreover, Americans enjoy the widest choice of insurers, but patients in several nations with low administrative costs are free to choose to receive care at any hospital. Nor do higher administrative costs appear to be associated with better care within the United States. A comprehensive meta-analysis of fifteen studies found that death rates at for-profit hospitals (adjusted for severity of illness, patients’ socioeconomic status, and hospitals’ teaching status) were 2 percent higher than those at nonprofit hospitals.28 For-profit hospitals also score lower on Medicare quality measures,29 and their patients perceive their care less favorably,30 compared to nonprofit institutions. 33 : 9 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. Reforming the US health care system so that it operated on a single-payer basis could result in large savings on administration. Overall, there is no evidence that the high administrative costs in the United States translate into superior care.31 Policy Implications Our data hold lessons for policy makers. Hospital payment strategies can shift vast sums from care to administration, and vice versa. In the United States, administration consumes an increasing share of hospital budgets—a share that is far higher than in nations with simpler and less market-oriented payment schemes. To put the differences in perspective, in 2011 rolling back US spending for hospital administration This research was supported by a grant from the Commonwealth Fund to the London School of Economics and Political Science. The funder did not play any role in the design and conduct of the study; the collection, management, analysis, and interpretation of the data; the preparation, review, or approval of the to the 2000 level (adjusted for inflation and population growth) would have saved $74.4 billion. Reducing US spending to Canada’s or Scotland’s level on a per capita basis would have saved $158 billion or $156 billion, respectively—equivalent to 1 percent of the US GDP. Reforming the US health care system so that it operated on a single-payer basis could result in large savings on administration. In contrast, current policy initiatives may boost administrative costs. Pay-for-performance schemes add new documentation requirements and incentives for data mining of patients’ records to ferret out exceptions (for example, finding the phrase “patient refused test” in free-text entries). Similarly, DRGs have long given hospitals incentives to find and document clinically insignificant comorbidities among inpatients, and the transition to accountable care organizations (ACOs) adds incentives to extend upcoding to outpatients. The ACO strategy also stimulates hospitals to develop bureaucratic structures to carry out tasks that resemble components of managed care, such as referral management, underwriting, and utilization review. In other nations, policy makers should take into account the added administrative costs of moving to activity-based funding (for example, DRGs) and market-based allocation of new capital investments for hospital modernization and expansion. The administrative burdens of promarket reforms should be weighed against their putative benefits. ▪ manuscript; or the decision to submit the manuscript for publication. David Himmelstein and Steffie Woolhandler founded and remain active in Physicians for a National Health Program, which advocates for single-payer health reform in the United States. They have received no financial compensation from that organization. The authors are indebted to Farhad Mehrtash for his assistance in obtaining and interpreting official data on Canadian hospital expenditures, to John Evans for his assistance in obtaining official data on Welsh hospital expenditures, and to Douglas Cameron for his assistance in obtaining and interpreting official data on Scottish hospital expenditures. September 2014 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. 33:9 H ea lt h A f fai r s 1 593 Hospital Costs NOTES 1 Organization for Economic Cooperation and Development. OECD health statistics 2014 [Internet]. Paris: OECD; [cited 2014 Jul 23]. Available from: http://www.oecd .org/els/health-systems/healthdata.htm 2 Squires DA. Explaining high health care spending in the United States: an international comparison of supply, utilization, prices, and quality [Internet]. New York (NY): Commonwealth Fund; 2012 May [cited 2014 May 7]. Available from: http://www.commonwealthfund .org/~/media/Files/Publications/ Issue%20Brief/2012/May/1595_ Squires_explaining_high_hlt_ care_spending_intl_brief.pdf 3 Starfield B, Shi L, Macinko J. Contribution of primary care to health systems and health. Milbank Q. 2005;83(3):457–502. 4 Patented Medicine Prices Review Board. Annual report 2010: comparison of Canadian prices to foreign prices [Internet]. Ottawa (ON): PMPRB; [last modified 2011 Jun 9; cited 2014 May 7]. Available from: http://www.pmprb-cepmb.gc.ca/ english/view.asp?x=1503&mid= 1331 5 Laugesen MJ, Glied SA. Higher fees paid to US physicians drive higher spending for physician services compared to other countries. Health Aff (Millwood). 2011;30(9):1647–56. 6 Office of Technology Assessment. International comparisons of administrative costs in health care [Internet]. Washington (DC): Government Printing Office; 1994 Sep [cited 2014 May 7]. (Pub. No. OTABP-H-135). Available from: http:// govinfo.library.unt.edu/ota/Ota_1/ DATA/1994/9417.PDF 7 Himmelstein DU, Woolhandler S. Cost without benefit. Administrative waste in U.S. health care. N Engl J Med. 1986;314(7):441–5. 8 Himmelstein DU, Lewontin JP, Woolhandler S. Who administers? Who cares? Medical administrative and clinical employment in the United States and Canada. Am J Public Health. 1996;86(2):172–8. 9 Woolhandler S, Campbell T, Himmelstein DU. Costs of health care administration in the United States and Canada. N Engl J Med. 2003;349(8):768–75. 1594 Health Affa irs S ep t e m b e r 20 1 4 10 Pozen A, Cutler DM. Medical spending differences in the United States and Canada: the role of prices, procedures, and administrative expenses. Inquiry. 2010;47(2):124–34. 11 Cutler DM, Ly DP. The (paper) work of medicine: understanding international medical costs. J Econ Perspect. 2011;25(2):3–25. 12 Casalino LP, Nicholson S, Gans DN, Hammons T, Morra D, Karrison T, et al. What does it cost physician practices to interact with health insurance plans? Health Aff (Millwood). 2009;28(4):w533–43. DOI: 10.1377/hlthaff.28.4.w533. 13 Kahn JG. Excess billing and insurance-related administrative costs. In: Yong PL, Saunders RS, Olsen LA, editors. The healthcare imperative: lowering costs and improving outcomes: workshop series summary. Washington (DC): National Academies Press; 2010. p. 142–51. 14 To access the Appendix, click on the Appendix link in the box to the right of the article online. 15 Worldsalaries.org. International average salary income database: office clerk salaries—international comparison [Internet]. [place unknown]: Worldsalaries.org; c2008 [cited 2014 May 7]. Available from: http://www.worldsalaries.org/ officeclerk.shtml 16 Bureau of Labor Statistics. International comparisons of hourly compensation costs in manufacturing, 2012 [Internet]. Washington (DC): Department of Labor; 2013 Aug 9 [cited 2014 May 7]. Available from: http://www.bls.gov/fls/ichcc.pdf 17 Boyle S. United Kingdom (England): health system review. Health Syst Transit. 2011;13(1):1–483. 18 Maarse H, Jeurissen P, Ruwaard D. Concerns over the financial sustainability of the Dutch healthcare system. CESifo DICE Report. 2013; 11(1):32–36. 19 Reinhardt UE. The many different prices paid to providers and the flawed theory of cost shifting: is it time for a more rational all-payer system? Health Aff (Millwood). 2011;30(11):2125–33. 20 Maryland Hospital Association. Achievement, access, and accountability: Maryland’s all-payor hospital payment system [Internet]. Elkridge 3 3: 9 Downloaded from HealthAffairs.org on February 15, 2019. Copyright Project HOPE—The People-to-People Health Foundation, Inc. For personal use only. All rights reserved. Reuse permissions at HealthAffairs.org. 21 22 23 24 25 26 27 28 29 30 31 (MD): The Association; [cited 2014 May 7]. Available from: http://www .hscrc.state.md.us/documents/ HSCRC_PolicyDocumentsReports/ GeneralInformation/AshbyReport 2007.pdf Hawe E, Yuen P, Baillie L. OHE guide to UK health and health care statistics. London: Office of Health Economics; 2011 Jul. Silverman E, Skinner J. Medicare upcoding and hospital ownership. J Health Econ. 2004;23(2):369–89. Woolhandler S, Himmelstein DU. Costs of care and administration at for-profit and other hospitals in the United States. N Engl J Med. 1997; 336(11):769–74. Silverman EM, Skinner JS, Fisher ES. The association between forprofit hospital ownership and increased Medicare spending. N Engl J Med. 1999;341(6):420–6. Devereaux PJ, Heels-Ansdell D, Lacchetti C, Haines T, Burns KE, Cook DJ, et al. Payments for care at private for-profit and private not-forprofit hospitals: a systematic review and meta-analysis. CMAJ. 2004; 170(12):1817–24. Tiemann O, Schreyögg J, Busse R. Hospital ownership and efficiency: a review of studies with particular focus on Germany. Health Policy. 2012;104(2):163–71. Danzon PM. Hidden overhead costs: is Canada’s system really less expensive? Health Aff (Millwood). 1992;11(1):21–43. Devereaux PJ, Choi PT, Lacchetti C, Weaver B, Schünemann HJ, Haines T, et al. A systematic review and meta-analysis of studies comparing mortality rates of private for-profit and private not-for-profit hospitals. CMAJ. 2002;166(11):1399–406. Jha AK, Orav EJ, Epstein AM. Lowquality, high-cost hospitals, mainly in the South, care for sharply higher shares of elderly black, Hispanic, and Medicaid patients. Health Aff (Millwood). 2011;30(10):1904–11. Jha AK, Orav EJ, Zheng J, Epstein AM. Patients’ perception of hospital care in the United States. N Engl J Med. 2008;359:1921–31. National Research Council. U.S. health in international perspective: shorter lives, poorer health. Washington (DC): National Academies Press; 2013. MEDICARE SPENDING AND FINANCING A Primer 2011 MEDICARE SPENDING AND FINANCING A Primer February 2011 Prepared by: Lisa Potetz Health Policy Alternatives, Inc. and Juliette Cubanski and Tricia Neuman The Henry J. Kaiser Family Foundation A PRIMER ON MEDICARE SPENDING AND FINANCING INTRODUCTION For 45 years, Medicare has successfully provided access to health care services for the elderly ages 65 and over and many nonelderly people with disabilities, and currently covers 47 million Americans. Persistently high rates of growth in national health expenditures combined with demographic trends, however, pose a serious challenge to the financing of Medicare in the 21st century. This paper provides a detailed overview of Medicare spending and financing, beginning with a review of the factors contributing to the growth in Medicare spending, including the effects of the 2010 health reform law. Next, it explains the structure of the Medicare program’s financing, reviews various measures of fiscal status, and discusses the expected effects of rising Medicare costs on beneficiaries. The paper concludes with a discussion of the program’s long-run financial challenges. With Medicare being the nation’s single largest health insurance program covering a large population for a broad range of health services, the program’s influence extends well beyond the assistance it provides to its beneficiaries. Medicare expenditures and the policies under which the program operates have a large impact on the nation’s health care system. One in five dollars used to purchase health services in 2008 came through the Medicare program, which finances nearly four in ten hospital stays nationally.1 TRENDS IN MEDICARE SPENDING Since its enactment in 1965, spending on Medicare has grown steadily, as measured in absolute dollars, as a share of the federal budget, and as a share of the gross domestic product (GDP), and these trends are expected to continue (Exhibits 1 and 2). In fiscal year 2010, Medicare’s $524 billion in total expenditures represented 15 percent of all federal outlays, exceeded only by Social Security benefits and defense spending, which each Exhibit 1 accounted for 20 percent (Exhibit 3).2 By Medicare Spending as a Share of 2020, Medicare is projected to reach 17 Federal Budget Outlays, 1970-2020 percent of budget outlays and 4 percent Projected Actual of the GDP. 17.4% Between 1985 and 2009, growth in Medicare spending averaged almost 9 percent annually, compared with 5 percent growth in both the GDP and medical care inflation during those years (Exhibit 4). The average annual growth in aggregate Medicare spending (9 percent) exceeds the average growth in Medicare per capita spending (7 percent) during this period because it includes costs attributable to the growth in the 15.1% 12.1% 8.5% 5.8% 3.5% Total Medicare spending in billions 1970 $7 1980 1990 2000 2010 2020 $35 $110 $219 $524 $949* SOURCE: Congressional Budget Office, Budget and Economic Outlook, January 2010 (for 1970 data) and January 2011 (for 1980-2020 data, except 2010 which comes from CBO August 2010 Baseline: Medicare). Historical total spending for 1970-2000 from 2010 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. NOTE: *Estimates for 1970-2010 represent total Medicare outlays, estimate for 2020 represents projection of mandatory Medicare outlays. CBO (August 2010) projects discretionary Medicare outlays will be $9 billion in 2020. 1 Medicare population, which increased by almost 2 percent annually during this period. The addition of the Part D prescription drug benefit in 2006 contributed to the rate of growth; excluding Part D, average annual Medicare growth in total spending was just under 8 percent overall and 6 percent per enrollee. Looking to the decade ahead, Medicare spending is projected to grow more slowly (about 6 percent annually), and on a per-beneficiary basis, will be closer to growth in GDP and general inflation.3 However, current projections of Medicare spending also assume large cuts in physician fees that will occur under current law due to the physician payment formula known as the Sustainable Growth Rate (SGR). If reductions in physician fees are avoided in the future, as they have been numerous times in recent years, Medicare spending will exceed the current projections. For example, under one alternative scenario, the Medicare actuaries estimate that Medicare spending could be about 9 percent higher in 2019 than it is projected to be under current law.4 Medicare spending growth generally reflects trends in national health spending, which for many years has outpaced growth in the economy, rising from 7 percent of GDP in 1970 to nearly 18 percent in 2009, and is projected to reach 20 percent by the end of the decade.5 Over the long run, average growth in Medicare spending per beneficiary has been slightly lower than per capita growth in private health spending for comparable benefits, although over some periods of time the opposite has been true (Exhibit 5). Exhibit 2 Medicare Spending as a Share of Gross Domestic Product (GDP), 1970-2020 Actual Projected 4.2% 3.6% 2.2% 1.9% 1.2% 0.7% 1970 1980 1990 2000 2010 2020 SOURCE: Congressional Budget Office, Budget and Economic Outlook, January 2010 (for 1970 data) and January 2011 (for 1980-2020 data). Exhibit 3 Medicare Spending as a Share of Total Federal Outlays, FY 2010 Social Security 20% Defense 20% Medicare1 15% Medicaid 8% All Other Combined2 31% Net Interest 6% FY 2010 Total Federal Outlays = $3.5 trillion SOURCE: Kaiser Family Foundation based from Congressional Budget Office, Historical Budget Data, January 2011. NOTE: FY is fiscal year. 1Amount for Medicare is mandatory spending and excludes offsetting premium receipts (premiums paid by beneficiaries, amounts paid to providers and later recovered, and state contribution (clawback) payments to Medicare Part D). 2”All Other Combined” category includes other mandatory outlays, offsetting receipts, and negative outlays for Troubled Asset Relief Program. Exhibit 4 Average Annual Growth in Medicare Spending, 1985-2009 Compared with Economic Benchmarks 8.5% 6.7% 5.2% 5.1% 4.1% 2.9% Medicare spending Medicare spending per enrollee GDP GDP per capita CPI CPI-Medical care SOURCE: Economic Report of the President, 2010, except Medicare data from the 2010 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. NOTE: GDP is gross domestic product. CPI is consumer price index. 2 The distribution of Medicare spending across beneficiaries has been and continues to be highly skewed, as is health spending for the broader population. In any given year, a relatively small share of the Medicare population, ten percent, accounts for a majority (58 percent) of all Medicare spending (Exhibit 6). Average Medicare spending for beneficiaries who are in the top ten percent of the Medicare population ($48,210) was nearly six times greater than the average across all beneficiaries in 2006 ($8,344). Exhibit 5 Annual Change in Medicare and Private Health Insurance Spending, 1970-2009 Medicare (Average Annual Growth, 1970-2009 = 8.3%) Private Health Insurance (Average Annual Growth, 1970-2009 = 9.3%) 22% 20% Medicare 18% 16% Private Health Insurance 14% 12% 10% 8% 6% 4% 2% 0% 1970 1975 1980 1985 1990 1995 2000 2005 SOURCE: Centers for Medicare & Medicaid Services, Office of the Actuary, National Health Statistics Group, 2011. Exhibit 6 Distribution of Medicare Fee-For-Service Enrollment and Spending, 2006 Average per capita Medicare FFS spending: $8,344 10% 58% Average per capita Medicare FFS spending among top 10%: $48,210 90% 42% Total Number of FFS Beneficiaries: 35.9 million Average per capita Medicare FFS spending among bottom 90%: $3,910 Total Medicare FFS Spending: $299 billion SOURCE: Kaiser Family Foundation analysis of the CMS Medicare Current Beneficiary Survey Cost and Use file, 2006. NOTE: Analysis excludes Medicare Advantage enrollees. FFS is fee-for-service. WHAT FACTORS CONTRIBUTE TO GROWTH IN MEDICARE SPENDING? The Medicare actuaries have identified a number of factors that increase health care costs, including those paid for by Medicare, Medicaid, and private health insurance. Specifically, these are increases in the prices paid per service and increases in the volume and complexity of services provided per beneficiary. Medicare costs are also affected by growing program enrollment and an aging population, along with other factors that are unique to the Medicare program. 3 Prices, Volume, and Complexity of Services The rates Medicare pays for specific services are generally indexed to reflect inflation in the prices of goods and services used to produce those services. For example, hospital payment rates are tied to changes in the price of a “market basket” of goods and services that hospitals must purchase, including wages and benefits paid to nurses and other employees, drugs, food, and medical equipment and instruments. In addition to prices, medical advances and other changes in the practice of medicine over time have increased the average volume and intensity of the services provided to Medicare beneficiaries. Together, these effects have increased program expenditures over time and are expected to continue to do so in the future. Increased Enrollment and an Aging Population Often discussed as a driver of Medicare spending is the accelerating growth in program enrollment that will occur with the retirement of the post-World War II “baby boom” generation, who began to turn 65 in 2011. Between 1995 and 2009, as the cohort of individuals born during the Great Depression and World War II became eligible for benefits, Medicare enrollment grew by an average of 623,000 beneficiaries annually. Looking to the future, net Medicare enrollment growth is expected to average more than 1.6 million beneficiaries annually between 2010 and 2030, and the program will reach a total of 80 million enrollees in 2030 – double the number of enrollees in 2000.6 The contribution of increased enrollment and an aging population to growing Medicare spending, however, is modest relative to the effects of rising health care costs. CBO projects that increased program enrollment along with the aging of the Medicare population will increase Medicare spending from 4 percent of GDP in 2020 to 5 percent of GDP by 2035; however, when all factors affecting Medicare and rising health care costs are also included, Medicare spending is projected to rise to 7 percent of GDP that year.7 As one would expect, per capita Medicare spending increases as beneficiaries age. For example, in 2006, per capita Medicare spending for beneficiaries in the traditional fee-for-service program who were age 85 or older totaled $12,059, more than double the $5,887 average for beneficiaries ages 65 to 74.8 As the baby boom generation ages into Medicare, however, the age mix of the program’s beneficiaries will initially be younger than it is today. Only after 2030, when the bulk of baby boomer beneficiaries reach an older age level, is age mix expected to contribute more to program spending than it does now.9 Unique Medicare Policy Issues Certain policy issues unique to the Medicare program have significantly affected program spending and projections. For example, addition of the Medicare Part D prescription drug benefit in 2006 increased program outlays considerably, accounting for two-thirds of the $72 billion increase from 2005 to 2006. Additionally, in recent years rising enrollment of Medicare beneficiaries into private Medicare Advantage (MA) plans increased program expenditures because the average per enrollee payments made to these private plans has exceeded the cost of the traditional fee-for-service program. These payments in excess of the cost of the Medicare fee-for-service program averaged 13 percent in 2010.10 As a result of recent policy changes that reduced payments to MA plans, MA plan enrollment is projected to decline in the future.11 4 Administrative Costs The costs of administering the Medicare program have remained low over the years – less than 2 percent of program expenditures. As such, program administration is not a contributing factor to Medicare’s expenditure growth. Administrative costs include all expenses by government agencies in administering the program (HHS, Treasury, the Social Security Administration, and the Medicare Payment Advisory Commission). Also included are the cost of claims contractors and other costs incurred in the payment of benefits, collection of Medicare taxes, fraud and abuse control activities, various demonstration projects, and building costs associated with program administration. HOW WILL HEALTH REFORM AFFECT MEDICARE SPENDING? Implementation of the Patient Protection and Affordable Care Act (ACA) of 2010 will have a major effect on Medicare spending and policy.12 Most of the Medicare provisions in the ACA will reduce program spending, but some will increase it, and on net, Medicare spending will be reduced by an estimated $424 billion for the 10-year period from fiscal year 2010 through fiscal year 2019, a 6 percent reduction from spending that had been projected Exhibit 7 for that period (Exhibit 7).13 While this Net Effect of Major Legislation on Medicare Spending amount is not insignificant, it is also not Net Spending/Savings as a Share of Projected Medicare Spending over 10 Years unprecedented, and represents a smaller DRA PPACA BBA BBRA BIPA MMA MIPPA share of projected 10-year baseline (2005) (2010) (1997) (1999) (2000) (2003) (2008) 15% Medicare spending than the spending 12% 10% reductions included in the Balanced 9% Budget Act (BBA) of 1997. At the time it 6% 3% was enacted, the BBA was projected to Net 3% 1% spending 0% result in a 12 percent reduction in Net -0.02% -0.4% savings -3% projected baseline Medicare spending.14 The ACA includes a number of provisions that are expected to reduce Medicare spending. (See Appendix A for the cost estimate for the major Medicare provisions in the ACA.) -6% -9% -12% -15% 10-yr Medicare baseline amounts (in $ trillions): -6% -12% $3.4 $3.2 $3.2 $3.8 $5.6 $6.3 $7.1 SOURCE: Kaiser Family Foundation analysis of Congressional Budget Office (CBO) estimates. NOTE: Shares are rounded to the nearest whole number. Net spending as a percent of baseline for MIPPA is rounded up from -0.02%; estimate for DRA is rounded from -0.47%. Baseline amounts are based on CBO projections of 10-year Medicare baseline spending prior to enactment of legislation. Reduced Payments to Providers and Medicare Advantage Plans More than half the Medicare savings comes from two provisions: one that institutes a productivity adjustment which will reduce annual fee-for-service provider payment updates and one that makes changes to payments for Medicare Advantage plans. Additional savings come from reducing payments for preventable hospital readmissions and home health services. The annual increase in payment rates for various Medicare services, generally adjusted to reflect inflation, will be reduced by a measure of economy-wide productivity improvement. The compounding effect of this change will significantly reduce growth in program spending in perpetuity. The Medicare actuaries have questioned whether the savings will be sustainable for the long run, while others believe the baseline sets a new achievable target for promoting provider efficiency.15 For Medicare Advantage plans, the ACA phases out payments made in excess of fee-for-service costs, consistent with recommendations of the Medicare Payment Advisory Commission. In addition, the 5 productivity adjustment and other Medicare fee-for-service payment changes generate additional savings from MA plans because plan payments are linked to fee-for-service spending levels. Independent Payment Advisory Board (IPAB) The newly-created IPAB is a 15-member panel charged with recommending a set of Medicare program changes, within certain constraints, if program spending growth exceeds specified targets, beginning in 2015. These targets are estimated to require IPAB to identify nearly $16 billion in savings for the years 2015 to 2019. The IPAB recommendations will be sent to the Congress under special procedures; disapproval will require a supermajority vote. They may not include changes to increase revenues, beneficiary premiums or cost-sharing; restrict benefits or modify eligibility criteria. Prior to 2019, certain providers are exempt from reductions. Beneficiary Premiums The ACA modified a provision in current law that requires higher-income Medicare Part B enrollees to pay a higher monthly Part B premium. The law temporarily eliminates the indexing of the income thresholds at which the higher premium must be paid. As a result, between 2010 and 2019, the share of Medicare beneficiaries paying an income-related Part B premium will rise from 5 percent to 14 percent.16 In addition, the law established a new income-related premium for Medicare beneficiaries enrolled in Part D prescription drug plans, reducing government contributions for Part D coverage by increasing premiums for higher-income beneficiaries who are also subject to the income-related Part B premium. The thresholds for the Part B and Part D income-related premium are fixed at $85,000 for an individual beneficiary and $170,000 for couples through 2019. Delivery System Reforms Other provisions of the ACA are aimed at changing the health care delivery system in ways that seek to produce long-run efficiencies and program savings. A general theme of these provisions is to shift provider incentives away from increasing the volume of services toward incentives for improving quality and care coordination. Examples include establishment of a Medicare Shared Savings program for Accountable Care Organizations (ACOs) and bundling payments around a hospital stay. A new Center for Medicare and Medicaid Innovation will test additional innovative payment and service delivery models, a number of which are expected to focus on Medicare and Medicaid dual eligibles and other high-need populations. Other Provisions That Increase Medicare Spending Some provisions will increase program spending and partially offset the savings generated from the provisions discussed above. Most notably, the law gradually phases in coverage to close the Part D coverage gap (or “doughnut hole”) by 2020. The law also includes an annual wellness visit and other improvements in Medicare coverage of recommended preventive services. Revenue Provisions Finally, although not affecting program spending, the law included two provisions that will provide new revenue streams dedicated to financing Medicare benefits. These are an increase in the Medicare payroll tax on certain high earners (earnings more than $200,000 for an individual or $250,000 for a couple) and a new fee on drug manufacturers. 6 HOW IS MEDICARE FINANCED? In financing Medicare, the government draws from several sources of revenue: a dedicated Medicare payroll tax, general revenue (primarily federal income taxes), premiums collected from beneficiaries, a tax on Social Security benefits, and, since 2006, payments from states for the Medicare drug benefit, which shifted some state Medicaid program expenditures to Medicare (Exhibit 8). Exhibit 8 Sources of Medicare Funding, 2009 1% 22% 38% Medicare payroll tax Interest on Trust Funds 4% 13% 85% 2% Beneficiary premiums 74% Social Security tax General revenue 42% Operationally, Medicare financing is Transfers from States/other 7% 1% conducted through two trust fund 6% 2% 3% 1% accounts. The Hospital Insurance (HI) Total HI Trust Fund SMI Trust Fund Trust Fund finances inpatient hospital $508 billion $225 billion $283 billion SOURCE: Kaiser Family Foundation based on 2010 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal care and other services covered under Supplementary Medical Insurance Trust Funds, Table II.B1. Medicare Part A, and the Supplementary Medical Insurance (SMI) Trust Fund finances physician and other services covered under Medicare Part B along with the Part D prescription drug benefit. Both trust funds are used to pay MA plans for providing benefits to their enrollees under Parts A and B and, where applicable, Part D. (See Appendix B for detail on the sources and uses of Medicare Trust Fund revenue.) HI Trust Fund Medicare payroll taxes and certain other dedicated revenue are credited to the HI Trust Fund. The Medicare payroll tax, requiring contributions of 1.45 percent of wages each from the employer and employee, is a primary source of HI Trust Fund revenue. In 2009, the payroll tax provided 85 percent of all the revenue contributed to the HI Trust Fund and 38 percent of Medicare revenue overall. In any given year, the revenue dedicated to the HI Trust Fund may be greater or less than the expenditures from the fund. When income exceeds expenditures, the excess HI Trust Fund revenue amounts are loaned to the federal government and used to pay for other federal obligations. Interest on the loans is credited to the Trust Fund as income. Interest payments are not actually transferred out of general revenue unless these amounts are needed to pay Medicare claims. As a result, the amounts collected in Medicare payroll taxes and other dedicated revenue but loaned out of the HI Trust Fund, along with the associated interest payments, represent a claim on future general revenue funds. The HI Trust Fund balance, which totaled $279 billion at the end of fiscal year 2010, is a measure of these future claims that have accumulated to date, to be drawn upon when payroll taxes and other dedicated revenue are insufficient to cover program obligations. Supplementary Medical Insurance (SMI) Trust Fund The SMI Trust Fund is financed primarily through the monthly Part B and Part D premiums paid by beneficiaries and by general revenue. Beginning in 2011, revenue from the new fee on drug manufacturers established by the ACA will be credited to this trust fund. In 2009, general revenue accounted for 74 percent of the SMI Trust Fund revenue and 42 percent of all Medicare revenue, while 7 total beneficiary premiums made up 22 percent of the SMI Trust Fund revenue and 13 percent of Medicare revenue overall. Unlike the HI Trust Fund, SMI Trust Fund financing is not structured in a way that will produce yearly excess revenue or shortfalls. In this case, beneficiary premiums and general revenue contributions are adjusted each year in order to cover trust fund obligations. HOW IS MEDICARE’S FISCAL STATUS MEASURED? Serious concerns have been raised about the long-term financial health of the Medicare program. The program’s financial status is often measured in terms of the HI Trust Fund solvency or Medicare spending as a share of the federal budget and of the overall national economy. New benchmarks were established under the ACA for purposes of determining whether IPAB will need to recommend changes to the Medicare program in order to achieve specified levels of savings. Each measure addresses a different perspective on the program’s financing and points toward different potential solutions to Medicare’s long-term financing challenges. Because economic forecasts and estimates of Medicare and total federal spending are always being adjusted due to changes in forecasts and better information, it is difficult to isolate the effects of the ACA from other factors affecting Medicare’s fiscal status. For the measures discussed here, where possible, using the information available when the law was passed, a sense of the order of magnitude of the effects of the ACA is included. Trust Fund Solvency Solvency of the HI Trust Fund is the measure of Medicare’s financial health that typically receives the most attention (Exhibit 9). A report on the financial status of the HI Trust Fund is released annually, as required by law, including short-run and long-run financial forecasts prepared by the Medicare actuaries. The report is issued by the Medicare Trustees, an oversight panel comprised of the Secretaries of Health and Human Services (HHS), Labor, and Treasury; the Commissioner of Social Security; and two public trustees appointed by the President. The most recent annual report underscores the precarious financial health of the HI Trust Fund.17 Reflecting the recession’s effect on payroll tax contributions, in each year since 2008, total payments from the HI Trust Fund exceeded total income to the Fund. When such a shortfall occurs, the Trust Fund reserves are drawn upon through general revenue transfers to make up the difference. The shortfall was $17 billion in 2009 and is projected to continue to accumulate for several more years, further drawing down the HI Trust Fund reserves. 8 Exhibit 9 Medicare Part A Trust Fund Balance, 2000-2019 Under High Cost, Low Cost, and Intermediate Assumptions Fund balance as % of annual expenditures: 250% 225% Actual Projected 200% 175% Low cost 150% 125% Intermediate 100% 75% 50% High cost 25% 0% 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 SOURCE: Kaiser Family Foundation based on 2010 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. NOTE: The Medicare Trustees recommend that the HI Trust Fund assets should be maintained at a level of at least 100% of annual expenditures. Reversing the trend of annual shortfalls, annual surpluses are now expected to occur from 2014 until after 2020 when shortfalls will begin anew and the Trust Fund balances are expected to be exhausted in 2029. The balances are exhausted at the point when even if all the payroll tax amounts that were previously loaned to the rest of the federal government are repaid with interest, the Trust Fund will not have sufficient funds to cover the entire cost of inpatient hospital care and other Medicare Part A services. The 2029 date is 12 years later than projected prior to enactment of the ACA. The Medicare actuaries and the CBO have each noted, however, that because the ACA included other provisions that will increase federal spending obligations in the future, the fiscal position of the federal government will still be challenged to find the future funding needed to meet program obligations.18 A range around the 2029 insolvency date is bounded by the actuaries’ more pessimistic and optimistic assumptions about future economic and demographic factors and health-care costs (shown in Exhibit 9 as “high cost” and “low cost” assumptions). That is, assuming slower economic growth or more rapidly growing health care costs would move up the insolvency date to 2017, while assumptions of faster economic growth and slower growth in use of health services would push back the insolvency beyond the end of the 75-year projection period in 2085. The projection of HI Trust Fund exhaustion in 2029 does not mean that the Medicare program will be “bankrupt”, or that there will be no funds available to pay for Medicare Part A benefits that year, since revenue will continue to flow to the HI Trust Fund. Rather, it means that there will be insufficient funds to meet all the Trust Fund obligations. What makes the problem especially serious, however, is that it is not temporary—the shortfalls in HI benefit financing will continue to accumulate each year unless something changes either to increase the revenue coming into the Trust Fund or to decrease total Trust Fund expenditures. No process currently exists for addressing insufficiencies in the HI Trust Fund; new legislation would be required to make up the difference. Medicare’s per capita spending rate is not the only factor affecting Trust Fund solvency. Demographic factors also are important. Not only will Medicare need to provide for more beneficiaries, but there will also be fewer workers per beneficiary making payroll contributions to help cover the costs (Exhibit 10). In 2010, 3.4 workers were contributing Exhibit 10 taxes for each beneficiary; by 2030 that Change in the Medicare Population and the Number of figure is projected to fall to 2.3 and Workers per Beneficiary continue to decline to 2.1 workers per Number of beneficiaries (millions) Millions beneficiary by 2080.19 As a result, even at Number of workers per beneficiary 90 4.5 a healthy rate of economic growth, 80 80 4 Medicare payroll taxes would not keep 4.0 70 3.5 64 pace with program growth. This worker3.4 60 3 to-retiree ratio problem is not unique to 47 50 2.5 2.8 the United States. In fact, the 40 40 2 2.3 proportional decline in workers is 30 1.5 expected to be much worse in Japan and 20 1 many European countries.20 10 While technically, the SMI Trust Fund cannot become insolvent, financing the 0.5 0 0 2000 2010 2020 2030 SOURCE: Kaiser Family Foundation based on the 2010 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds. 9 projected growth in spending for Part B and Part D services would require increasing general revenue and beneficiary premium contributions. The increase in general revenue contributions has important implications for the federal budget, which offers another way to measure Medicare financing. Medicare Spending as a Share of Gross Domestic Product (GDP) One common way of evaluating the burden of financing a rapidly growing Medicare program is to consider Medicare spending in relation to the overall U.S. economy. While no particular amount of the GDP is the “correct” amount for Medicare spending, the implication of having more economic output devoted to Medicare is that fewer resources are available to meet other needs. As referenced earlier, Medicare spending grew from 2.2 percent of the GDP in 2000 to 3.6 percent of the GDP in 2010, a period covering the addition of the Part D prescription drug benefit. During the coming decade, Medicare spending as a share of GDP is projected to grow at a somewhat slower rate, to about 4 percent by 2020, as the savings provisions in the ACA are implemented.21 By 2035, under CBO’s long-term alternative Medicare baseline trend (which differs from current law), Medicare spending will reach 7 percent of GDP.22 This baseline begins with growth in per enrollee spending equal to the 1985-2008 trend of GDP plus 1.7 percentage points, but gradually slows under the assumption that as health care eventually begins to crowd out consumption of other necessary goods and services to a degree that is Exhibit 11 unsustainable, slower growth in health Projections of Medicare Spending as a Share of GDP Based on Various Growth Rate Scenarios care costs will result even in the absence Total Medicare Outlays as of changes in federal law. Substantial Medicare Spending as % of Gross Domestic Product (GDP) Share of: savings would result if program 8% GDP+2% expenditures were to grow more slowly. 7% GDP+1.7% (CBO For example, if beginning in 2021, projection)* 6% GDP+1% growth in Medicare spending per 5% beneficiary were equal to growth in the GDP+0% (enrollment 4% trends) GDP plus 1.0 percentage point, by 2035, 3% Medicare spending as a share of GDP 2% would be reduced from about 7 percent 1% to about 6 percent (Exhibit 11). (See 0% Text Box, “Putting Medicare Spending 2010 2015 2020 2025 2030 2035 Growth Rate Differentials in Context.”) IPAB Benchmarks SOURCE: Congressional Budget Office. Data combining all federal health spending can be found in The Long Term Budget Outlook, June 2010, Figure 2-5. NOTE: *CBO’s “Alternative Fiscal Scenario” which is not current law baseline. It assumes a gradual increase in physician payments through 2020; for 2021 and beyond it assumes a long-term rate of growth in Medicare spending that begins at GDP+1.7% and slows to GDP+1% by 2084. The Medicare spending targets under which IPAB will operate represent another measure of Medicare’s fiscal status. The targets are established using five-year averages, and prior to 2018 are based on an average of the increase in the Consumer Price Index (CPI) and the medical care expenditure category of the CPI. Beginning in 2018, the target for Medicare spending per beneficiary is equal to growth in the GDP plus 1.0 percentage point. If spending is determined to exceed the applicable target, IPAB is directed to recommend reductions in program spending up to a limit that begins at 0.5 percent of program expenditures in 2015 and rises to 1.5 percent of program expenditures by 2018. So, although the target provides a limit that triggers action by IPAB, the required savings are not set to constrain total program spending to meet the growth target. The CBO projects Medicare spending will exceed targets in 2015 though 2019. 10 Putting Medicare Spending Growth Rate Differentials in Context The difference between an annual growth rate target for per beneficiary Medicare spending of GDP plus 1.0 percentage point referenced by some Medicare savings proposals and the GDP plus 1.7 percentage point trend that CBO assumes as the basis of its long-term projections for the period beginning in 2021 might not seem significant, but it is. For example, in 2020 CBO projects that Medicare outlays will total about $950 billion. The one-year difference between a 1.7 percent increase from 2020 ($16.2 billion) and a 1.0 percent increase ($9.5 billion), or 0.7 percent, is $6.7 billion. The impact of a reduction in Medicare spending based on a lower growth rate compounds over time. For example, if the annual growth in Medicare spending from 2011-2020 were 0.7 percentage points lower than CBO currently projects, the total 10year spending would be reduced by about $280 billion, roughly a 4 percent reduction. The impact gets larger as the time horizon extends. Another perspective is that the GDP is large—about $24 trillion in 2020 under the CBO projections—so any measurable difference in the size of the Medicare program as a proportion of the GDP will be a large dollar amount. For example, in 2009 the Medicare actuaries projected that by 2020 Medicare would comprise 4.53 percent of GDP. In the 2010 report, after enactment of the ACA, the actuaries’ 2020 current law Medicare forecast fell to 3.91 percent of GDP—a decline of 0.62 percent of GDP that translates into a reduction in projected Medicare spending in 2020 of about $149 billion. Medicare Spending as a Share of the Federal Budget As noted earlier, Medicare is one of the largest and fastest growing federal programs. Budget experts have expressed concern about the long-run fiscal implications of the federal obligation for spending on Medicare and other health programs. For example, the National Commission on Fiscal Responsibility and Reform estimated that by 2025, federal revenue will be sufficient only to pay for Medicare, Medicaid, Social Security and interest on the national debt.23 As a result, recent proposals by this group and others for improving the nation’s fiscal status would, among other actions, reduce Medicare spending. The Medicare Solvency “Trigger” Another measure of Medicare’s claim on the federal budget is commonly referred to as the “45 percent trigger.” Implemented as part of the Medicare Modernization Act of 2003, the trigger provision requires the Medicare Trustees to estimate, using a particular formula, a ratio measuring the extent to which program expenditures exceed dedicated revenue. If, for two consecutive years, the actuaries project that the ratio is expected to exceed 45 percent within seven years, a “Medicare funding warning” is issued by the Trustees. The trigger is intended to draw attention to Medicare’s financial situation and to prompt the President and Congress to develop a response, but no automatic spending reductions or other changes in the program are set to occur as a result of the warning. The Medicare Trustees have issued a Medicare funding warning each year since 2007. Most recently, in the 2010 annual report, they estimated that the ratio would reach 45 percent in 2010. This is much earlier than previous reports, a change they attribute to the effects of the poor economy on Medicare payroll tax receipts. To date, the Congress has not voted on any legislation offered in direct response to the warning, and this measure has been criticized as arbitrary, and promoting certain policy solutions over others.24 11 Medicare’s Long-Term “Unfunded Obligation” As part of their long-range analysis of the Hospital Insurance Trust Fund, the Medicare actuaries calculate the 75-year “unfunded obligation”, as the present value of future Trust Fund expenditures less future income, decreased by the trust fund balance on hand at the beginning of the projection period. In the most recent estimates from 2010, this totals $2.4 trillion, an amount dramatically lower than the $13.4 trillion estimated in 2009, a reduction which is due to the provisions of the ACA. The reduction is largely attributed to the ongoing savings associated with the annual productivity adjustment to the provider payment rate updates. Uncertainties in Projections Given the complexity of both the U.S. economy and health care system, Medicare financing projections are always uncertain. They rely on a variety of predictions about the economy, demographics, and health care spending trends. Economic factors affect both spending and revenue projections. For example, future payroll taxes are tied to growth in wages, while annual increases in payments to hospitals and other providers are linked to measures of price inflation. The longer the time horizon under consideration, the greater the uncertainty introduced into the forecasts. Current projections are perhaps more uncertain than ever, however, because of difficulties in predicting a path for the current economy, uncertainties about the effects of health reform, and indecision about the long-term treatment of physician payments in the Medicare program. The Medicare actuaries, for example, have developed an alternative scenario for Medicare spending that assumes the physician payment cuts required under current law will not take effect, with physician fees instead updated annually by the Medicare Economic Index.25 It also assumes that the productivity adjustment will prove unsustainable and will be phased out Exhibit 12 over 15 years beginning in 2020. Under Medicare Spending as a Share of GDP under this alternative scenario, the HI Trust Different Projections Fund would be exhausted in 2028, only 2009 projection (pre-ACA) 2010 projection (current law, includes ACA) 2010 illustrative scenario one year earlier than projected, but the 12% long-run implications for HI spending 11.2% 10.7% 10.5% obligations would be much greater when 9.9% 9.6% 10% 9.0% 8.7% measured as a percent of taxable payroll 8.2% 8.0% 26 8% 7.3% or share of GDP. Long-run total 6.4% 6.4% 6.3% 6.1% 6.0% 5.9% Medicare spending as a share of GDP 5.8% 6% 5.1% 4.5% would remain lower than predicted in the 4.3% 3.9% 4% 3.5% 3.6% 3.6% 2009 Trustees Report, but would be much higher than the current law forecasts. For 2% example, in 2030, Medicare would 0% represent 6 percent of GDP, compared 2010 2020 2030 2040 2050 2060 2070 2080 with 5 percent projected under current SOURCE: Office of the Actuary, Centers for Medicare and Medicaid Services, Projected Medicare Expenditures Under an Illustrative Scenario with Alternative Payment Updates to Medicare Providers, August 5, 2010. law in the 2010 Trustees report (Exhibit NOTE: ACA is Affordable Care Act. 12). 12 HOW DOES THE RISING COST OF MEDICARE AFFECT BENEFICIARIES? In addition to monthly premiums, Medicare beneficiaries contribute to the cost of their care through cost sharing at the point of service—for example, Medicare deductibles and coinsurance—which are not reflected in data on Medicare spending or financing. For 2011, the standard monthly Part B premium is $115.40, with higher-income beneficiaries paying as much as $369.10 per month.27 Part D premiums vary depending on the private plan chosen, but average about $30 a month.28 In addition, beneficiaries face a $1,132 deductible for Part A inpatient hospital services, a $162 deductible for services covered under Part B, a 20 percent coinsurance for many services covered by Part B, and, in some cases, an additional amount for physician services, known as “balance billing” amounts. Medicare beneficiaries also pay for health care items and services not covered by Medicare, including most vision, dental, and hearing services and long-term care. Overall, Medicare paid 48 percent of beneficiaries’ total medical and long-term care costs in 2006; beneficiaries paid 15 percent of the total for Medicare-covered and other services and another 10 percent for premiums for Part B and supplemental insurance; and third-party payers (Medicaid, private supplemental “Medigap” plans, and employersponsored health plans) paid 26 percent of the total on behalf of beneficiaries (Exhibit 13). Because of premium and cost-sharing requirements, the growing cost of Medicare creates a financial burden on beneficiaries as well as the federal government. The Medicare Trustees project that over time beneficiaries will pay an increasing share of their Social Security income for their Medicare coverage. In 2010, premiums and costsharing for Part B and Part D together accounted for 27 percent of the average Social Security benefit (premiums accounted for 13 percent and average cost sharing absorbed another 14 percent). By 2030, Medicare premiums and cost sharing for Parts B and D are estimated to grow to 36 percent of average Social Security benefits (Exhibit 14).29 Additionally, beneficiaries will face rising premiums for private Medicare supplemental coverage. The impact on Exhibit 13 Sources of Payment for Health Care Services to Medicare Beneficiaries, 2006 Total beneficiary out-of-pocket spending $4,403 Beneficiary spending on services $2,651 Beneficiary spending on premiums $1,752 Medicare $8,344 48% 15% 10% 26% Third party payments $4,485 Total Per Capita Spending, 2006 = $17,232 SOURCE: Kaiser Family Foundation analysis of the CMS Medicare Current Beneficiary Survey Cost and Use file, 2006. NOTE: Excludes Medicare Advantage enrollees. Exhibit 14 Total Part B and Part D (SMI) Out-of-Pocket Spending as a Share of the Average Social Security Benefit, 1967-2084 Average Social Security benefit payment, 2010: $1,093/month; $13,116/year Average out-of-pocket spending on Part B and Part D, 2010: $291/month; $3,492/year 60% Average out-of-pocket spending on SMI premiums Average out-of-pocket spending on SMI cost sharing 50% 40% 30% 27% 20% 10% 0% 6% 7% 7% 3% 3% 4% 4% 4% 3% 9% 5% 4% 12% 7% 5% 15% 8% 7% 14% 8% 6% 17% 14% 8% 9% 29% 31% 16% 17% 33% 19% 36% 21% 38% 22% 40% 24% 42% 44% 46% 27% 25% 26% 47% 48% 49% 50% 50% 29% 30% 30% 28% 29% 18% 19% 20% 20% 20% 20% 16% 17% 17% 18% 13% 13% 13% 14% 15% 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 2055 2060 2065 2070 2075 2080 SOURCE: Kaiser Family Foundation analysis of data from 2010 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary Medical Insurance Trust Funds, Figure III.C1. NOTE: SMI is Supplemental Medical Insurance. Out-of-pocket spending includes SMI premiums and out-of-pocket cost-sharing expenses for SMI covered services. Estimates do not include the income-related premium under which higher income beneficiaries pay an amount greater than the standard monthly premium. Sources of beneficiary income other than Social Security benefits are also excluded. Estimates may not sum to totals due to rounding. 13 individual beneficiaries will vary—those who use fewer health services may be less affected by costsharing requirements and those with higher incomes may be better able to afford to pay more for their Medicare benefits. Assistance is provided to the lowest-income Medicare beneficiaries through federal subsidies for Part D premiums and through the Medicare Savings Programs which provide Medicaid subsidies for Part B premiums and in some instances, Medicare cost sharing.30 However, many low-income Medicare beneficiaries do not have Medicaid coverage: more than one-third of Medicare beneficiaries with incomes below 100 percent of the federal poverty level do not have Medicaid coverage and two-thirds of those with incomes between 100 and 150 percent of poverty do not have Medicaid coverage.31 In addition, some beneficiaries may be shielded from premium increases due to a provision in law known as the “hold harmless,” which caps the Part B premium increase to prevent monthly Social Security income from falling as the Part B premium increases. The hold-harmless provision does not apply to Part D premiums. Exhibit 15 Most beneficiaries are not protected against increases in Medicare’s costsharing amounts, however. Coinsurance amounts rise annually along with provider payment rates. With health costs rising faster than income for Medicare beneficiaries, median out-ofpocket health spending as a share of beneficiaries’ income increased from 11.9 percent in 1997 to 16.2 percent in 2006; among the top quartile of beneficiaries, out-of-pocket health care costs as a share of income rose to 30 percent or more by 2006 (Exhibit 15). Out-of-Pocket Health Care Spending As a Percent of Income Among Medicare Beneficiaries, By Spending Percentile, 1997–2006 35% 30% 25% 23.9% 23.9% 24.9% 26.2% 20% 15% 11.9% 11.8% 12.0% 12.8% 27.4% 29.2% 29.9% 30.1% 29.9% 30.1% 75th percentile 15.5% 15.6% 15.6% 16.2% 14.0% 14.9% 50th percentile (median) 10% 5% 0% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 SOURCE: Kaiser Family Foundation analysis of CMS Medicare Current Beneficiary Survey Cost and Use File, 1997-2006. WHAT IS THE OUTLOOK FOR THE FUTURE? Policymakers are always challenged to balance the interests of Medicare beneficiaries, taxpayers, health care providers and manufacturers, but national economic and fiscal constraints in the near term will make the task more difficult than ever. Rising costs of health care pose similar challenges to all payers— employers and individuals as well as Medicare and other government programs. The long-run fate of Medicare depends on solving the larger problem of rising health care costs. The Affordable Care Act of 2010 takes initial steps to use Medicare’s leverage to carve a path that slows cost growth and improves quality, but it will take a sustained effort to play out all the changes required to rein in rising health care costs. There appears to be a consensus among policymakers and stakeholders that Medicare price reductions alone will not work; some set of delivery system reforms will also need to take hold. Further, to address the long-term financing challenges of the HI Trust Fund, and to meet the needs of an aging population, additional revenues will be needed to forestall major Medicare benefit reductions or relatively drastic reductions in provider payments. 14 Because of Medicare’s size, policymakers must also consider the broader effects of changes made to the program. Major reductions in Medicare payments to providers could put upward pressure on the prices they charge to private payers or negatively impact beneficiary access to services. Additional payments to teaching hospitals and those located in rural areas and serving low-income urban populations are explicitly made to address social needs beyond the care of Medicare patients, and substantially reducing or eliminating these payments would disadvantage the communities that rely on these facilities. Raising costs for Medicare beneficiaries would alleviate fiscal pressure for the government, but shift the burden to beneficiaries—many of whom have modest incomes—and reduce their access to needed health services. At least for the foreseeable future, Medicare policy will be shaped as much by concerns about the size of the federal budget deficit and national debt as by concerns about the program’s financial sustainability. The budget deficit is expected to reach nearly $1.5 trillion in 2011, or 10 percent of GDP, and the publicly-held national debt of $10 trillion will continue to grow as large deficits are projected for many years to come, even as the health of the economy improves.32 Because Medicare represents a large share of federal spending and is growing faster than other parts of the federal budget, changes to Medicare will inevitably be a focus of deficit reduction debates. Many of the Medicare proposals put on the table as part of broader deficit reduction recommendations are variations on recurring themes. These include introducing caps on Medicare spending growth, increasing beneficiary contributions, reducing provider payments, delaying the age of Medicare eligibility, raising dedicated revenue, expanding the scope of the IPAB, and completely restructuring the Medicare entitlement—each of which could have significant implications for beneficiaries and providers. The nature of the proposals being offered underscores the scale of changes that may be in store for the Medicare program in the future. The potential effects of these changes on Medicare beneficiaries and providers of care mean that debate over these changes will be contentious. Policymakers will face the difficult challenge of finding cost-reducing strategies that sustain or improve quality of care, and possibly new sources of revenue as well, while balancing the needs of beneficiaries, taxpayers, and health care providers. 15 APPENDIX A: MEDICARE SAVINGS AND SPENDING IN THE PATIENT PROTECTION AND AFFORDABLE CARE ACT (P.L. 111-148), AS AMENDED BY THE HEALTH CARE AND EDUCATION RECONCILIATION ACT OF 2010 (P.L. 111-152) MEDICARE SAVINGS PROVISIONS Annual provider payment updates Medicare Advantage payment reforms Other Medicare Advantage plan savings (interactions with fee-for-service savings) Home health payments Part B premiums for higher-income enrollees Disproportionate Share Hospital (DSH) payments Medicare Improvement Fund Independent Payment Advisory Board Part D premiums for higher-income enrollees Fraud, waste, and abuse Reducing hospital readmissions Part D enrollment and other consumer protections Delivery system pilot programs Other provisions TOTAL 10-YEAR GROSS MEDICARE SAVINGS COST ESTIMATE (in $ billions) $157 $136 $70 $40 $25 $22 $21 $16 $11 $7 $7 $6 $5 $7 $528 MEDICARE SPENDING PROVISIONS Part D coverage gap discount program and new federal subsidies Premium reductions (interactions with fee-for-service savings) Physician payment reforms Preventive services Other provider payments Medicare Savings Programs and Part D low-income subsidies Disproportionate Share Hospital (DSH) payments Part D enrollment and other consumer protections Medicare Advantage reforms Other provisions Interactions* TOTAL 10-YEAR GROSS MEDICARE SPENDING NET 10-YEAR MEDICARE SAVINGS OTHER RELATED REVENUE PROVISIONS Raise Medicare payroll tax on high earnings (Deposited in HI Trust Fund) Fee on drug manufacturers (Deposited in SMI trust fund) Eliminate Part D employer deduction $43 $38 $7 $5 $1 $1 $1 $1 $1 $4 $2 $104 4 $424 $87 $27 $5 NOTE: *Spending interactions include implementation of Medicare changes, Part D interactions with Medicare Advantage provisions, and Part B interactions with Part D provisions. SOURCE: Kaiser Family Foundation analysis of Congressional Budget Office (CBO) cost estimates as provided on March 20, 2010; Revenue estimates based on Joint Committee on Taxation estimates as provided on March 20, 2010. 16 APPENDIX B: MEDICARE’S TRUST FUNDS Sources of Funds Hospital Insurance (HI) Trust Fund Supplementary Medical Insurance (SMI) Trust Fund The HI Trust Fund is the repository for the Medicare payroll tax contributions (1.45 percent each for employee and employer), which constituted 85 percent of Trust Fund revenue in 2009. Premiums paid by beneficiaries constituted 22 percent of SMI Trust Fund revenue in 2009. General revenue contributed 74 percent of the total; transfers from states to offset state savings from implementation of the Medicare drug benefit accounted for 3 percent; interest on the Trust Fund balance was 1 percent. Other sources of funding include some of the income taxes paid on Social Security benefits by those exceeding certain income thresholds (6 percent of revenue); interest earned on trust fund balances (7 percent), and enrollee premiums (1 percent). Use of Funds Medicare Part A benefits are financed out of the HI Trust Fund. Individuals become eligible for Medicare Part A when they turn age 65 if they have made sufficient payroll tax contributions or choose to pay a premium to enroll; disabled individuals and those with end-stage renal disease may qualify at a younger age. Part A benefits include inpatient hospital care (55 percent of net HI expenditures for health services in 2009); limited skilled nursing facility Beneficiary premiums include the standard monthly premium paid for Medicare Part B ($115.40 in 2011); premiums paid by beneficiaries electing to enroll in Medicare Part D for their prescription drug coverage, which vary based on the plan they choose; and beginning in 2007, an income-related Part B premium paid by higher income beneficiaries. In 2011, the thresholds are $85,000 individual/$170,000 couple; these amounts are fixed through 2019. The total premium paid by these beneficiaries ranges from 40 percent to 220 percent higher than the standard premium, depending on income. The SMI Trust Fund is used to pay for benefits under Medicare Part B and to pay premiums to private prescription drug plans under Medicare Part D. Unlike Part A, eligible individuals must elect to enroll in Medicare Parts B and D and pay a monthly premium. Part D benefits in 2009 accounted for 23 percent of all SMI expenditures. Part B benefits include physician care (23 percent of SMI expenditures in 2009); outpatient hospital services (11 percent); and home health care (4 17 Hospital Insurance (HI) Trust Fund care (11 percent), home health (3 percent) and hospice (5 percent). Some 24 percent of payments from the HI Trust Fund are made to cover the costs of services to beneficiaries enrolled in private Medicare Advantage plans. The remaining 1 percent of expenditures pays for Medicare program administration, including government costs incurred in the payment of benefits, collection of taxes, fraud and abuse control activities, and various demonstration projects. Financial Status The financial status of the HI Trust Fund depends on the extent to which the Medicare payroll tax and other revenue that is dedicated to the Trust Fund covers the Part A expenditures that are obligated to be financed by the fund. At the end of calendar year 2009, the HI Trust Fund had a balance of $304 billion. Over the next decade the Trust Fund is projected to have sufficient income to cover expenditures each year; over the longer term, however, the Medicare actuaries project that the trust fund balances will be exhausted and therefore there will be insufficient funds to pay all obligations beginning in 2029, under the Medicare actuaries’ intermediate (most likely) assumptions. Supplementary Medical Insurance (SMI) Trust Fund percent). About 20 percent of payments from the SMI Trust Fund are made to cover the costs of services to beneficiaries enrolled in private Medicare Advantage plans. When combined, other benefits, including durable medical equipment, laboratory and ambulance services, clinic care and other services, account for about 17 percent of SMI expenditures. The remaining 1 percent of expenditures pays for Medicare program administration, including government costs incurred in the payment of benefits, collection …
Purchase answer to see full attachment