Can Social Insurance For Long-Term Care Work? The Experience Of Germany

Transcription

1 L o n g - T e r m C a r e S y s t e m s Can Social Insurance For Long-Term Care Work? The Experience Of Germany Germany may be the only country in which most of the beneficiaries and the money are in community-based long-term care settings. by Alison Evans Cuellar and Joshua M. Wiener 8 GERMANY PROLOGUE: Americans have often looked with envy at the German health care system, where citizens enjoy universal access to a comprehensive set of health benefits, all for about half of what Americans pay, per capita. As if that weren t enough, outcomes and satisfaction in Germany are at least as good as (if not better than) those in the United States. Until recently, however, Germans held no apparent advantage in the long-term care arena. Before 1994 Germany s system of public support for long-term care bore a resemblance to the U.S. system in that both were means-tested and state-administered. Growing financial pressure on states and discomfort with the disparity between policy and the German ideal of social solidarity led to reforms that have now put long-term care financing on a par with acute care in Germany. In this paper Alison Evans Cuellar and Joshua Wiener review Germany s experience with these reforms and draw lessons for other nations. Cuellar has worked extensively on Medicaid and Medicare policy issues for a decade, much of it at the Urban Institute. She holds a master of business administration degree from the University of Texas and is a doctoral student in health services and policy analysis at the University of California, Berkeley. Wiener, a renowned expert in long-term care, is a principal research associate at the Urban Institute. He holds a doctorate in sociology from Harvard University. H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r Project HOPE ThePeople-to-PeopleHealth Foundation, Downloaded Inc. from HealthAffairs.org on January 29, 2018.

2 G E R M A N Y ABSTRACT: In 1994 Germany enacted a universal-coverage social insurance program for long-term care to largely replace its means-tested system. The program has achieved many of its stated policy goals: shifting the financial burden of long-term care off the states and municipalities; expanding home and community-based services; lessening dependence on means-tested welfare; and increasing support of informal caregivers. Many of these goals were reached without exploding caseloads or uncontrolled expenditures. We examine the German long-term insurance program, focusing on issues of financing, eligibility and assessment, benefits, availability of services, and quality assurance. Germany enacted a universal-coverage social insurance program for long-term care in 1994 to largely replace its means-tested system. This major expansion of social insurance and long-term care spending was proposed by the conservative government of Chancellor Helmut Kohl at a time when Germany faced the dual burdens of reuniting East and West Germany and reducing high unemployment. The program provides extensive coverage of both home care and nursing home services. It was implemented quickly with surprisingly few difficulties and, after five years, enjoys broad popular and political support. If anything, the labor government of Chancellor Gerhard Schroeder seeks to expand the program slightly. The program has achieved many of its stated policy goals: shifting the financial burden of long-term care off the Länder (states) and municipalities; expanding home and community-based services; increasing the supply of long-term care services; lessening dependence on means-tested welfare; and increasing support of informal caregivers. 1 To the surprise of some outside observers, the new program has met many of its policy goals without exploding caseloads or uncontrolled expenditures. Policy debate in Germany has shifted from basic implementation challenges to questions of quality of care and removing remaining perceived inequities. In this paper we examine the German long-term care insurance program, focusing on issues of financing, eligibility and assessment, benefits, availability of services, and quality assurance. Background For twenty years West German policymakers criticized the wide gap between the financing of acute care and of long-term care. 2 Acute conditions were covered by universal health insurance, while long-term care was covered only for the most needy through a Landbased, means-tested program. 3 Policymakers criticized the lack of parallel coverage on the grounds of social equity, viewing social assistance and the required impoverishment needed to become eligible as degrading. 4 Social solidarity is a constant theme in German LONG-TERM 9 CARE SYSTEMS H E A L T H A F F A I R S ~ M a y / J u n e

3 L o n g - T e r m C a r e S y s t e m s 10 GERMANY social policy debates. There is a very strong belief that collective arrangements that protect everyone, not just the poor, against the exigencies of life (such as sickness, retirement, and disability) are desirable. Acute care in Germany is financed primarily through sickness funds, which are quasi-public, quasi-private insurers heavily regulated by the national government. Benefits and funding rules are set at the national level. Financing is through mandatory, incomerelated premiums that vary by sickness fund. 5 Reimbursement rates are negotiated at the Land level by committees representing associations of providers and sickness funds. These sickness funds historically covered little in the way of long-term care services, distinguishing illness from custodial care. In contrast to acute care, financing for long-term care prior to the 1994 reforms was largely through means-tested welfare, and as such, the responsibility of Länder and local governments, with no direct federal contribution. 6 Long-term care was a growing financial burden on the subnational governments. The number of persons receiving social assistance for long-term care in West Germany grew from 260,000 in 1970 to 546,280 in Approximately 80 percent of nursing home residents in the former West Germany and 100 percent of nursing home residents in the former East Germany had their care financed by means-tested social assistance; the remainder were self-pay. 8 In addition, 25 percent of the relatively few persons receiving paid home care received social assistance. Länder and local governments had few options to lower social assistance outlays for long-term care because federal law determined most policies regarding minimum eligibility and coverage, while locally influential (often church-related) nursing homes resisted payment rate cuts. In the early 1990s Länder and local governments became increasingly vocal in their demand for financial relief. Financing The German legislature enacted a mandatory, universal social insurance program for long-term care Soziale Pflegeversicherung in 1994 that provides for extensive nursing home and home-care benefits for people of all ages without regard to financial status. Home care benefits became available 1 April 1995, and nursing home benefits, 1 July Although administered by the sickness funds, long-term care is fiscally separate from acute care. 9 The long-term care insurance premium is uniform and fixed by law at 1.7 percent of salary, which is shared equally by employers and employees. Retirees pay half of the premium, and pension funds pay the other half. The employment-based premium was chosen despite objections from H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

4 G E R M A N Y employers about high labor costs. To appease employers, one mandatory paid holiday was eliminated. 10 Because means-tested programs were viewed as demeaning, all serious debate focused on how universal coverage should be achieved. Public coverage funded through income-based taxes was rejected by conservatives and became untenable in the wake of high taxes resulting from reunification. Private insurers themselves rejected subsidized private coverage because their target market historically had been high-income families, and their products had always been sold at premiums independent of income. By comparison, the use of the sickness fund structure had several advantages, including the use of familiar, existing administrative bodies. Contributions to social insurance programs via sickness funds are viewed as distinct from taxes paid to the government, despite the fact that social insurance contributions are mandatory. 11 Consistently, contributions have far greater public support than taxes have. Furthermore, the expenditures of sickness fund programs do not appear on government budgets. Not only is the mandatory premium not perceived as a tax, but at 1.7 percent of salary, it is thought to be a relatively modest premium when compared with the average of 39.6 percent of income that is already paid for health, pension, and unemployment benefits. 12 Thus, it seemed a relatively small price to pay to solve a major social problem. To ensure that spending did not mushroom uncontrollably, the long-term care program has several cost-control mechanisms that distinguish it from acute care. If spending exceeds agreed-upon levels, deliberate political choices by government authorities are needed to balance funds; no automatic mechanisms have been built in. Both revenues and benefits are capped to some degree. Revenues are limited by the fixed contribution rate. On the benefit side, maximum monthly benefits per eligible person are fixed by disability level and setting. Finally, benefits do not automatically increase with inflation; they must be legislatively raised. While the choice of sickness funds as administrative bodies had many advantages, the relationship between the acute care and longterm care programs has not been completely smooth. Because the line between acute and long-term care is difficult to draw, charges of cost shifting between the two programs are common. For this reason, one of the explicit goals of the long-term care legislation to favor rehabilitation services over long-term care has not come to fruition. The difficulty is that more rehabilitation services would be funded by the health program, whereas any savings would accrue to the long-term care program. Part of the goal of the new program was to spend more money on LONG-TERM 11 CARE SYSTEMS H E A L T H A F F A I R S ~ M a y / J u n e

5 L o n g - T e r m C a r e S y s t e m s 12 GERMANY long-term care, and the program has clearly succeeded in doing so. Spending for the new program totaled DM 31 billion ($15.5 billion) in 1998 (Exhibit 1). 13 As a result of the new program, Land and municipal budgets for social assistance for long-term care declined by 58 percent, from DM 18 billion ($9 billion) in 1994 to DM 7.6 ($3.8 billion) in 1997, providing substantial fiscal relief. 14 The program also largely achieved its goal of expanding coverage to noninstitutional settings. As of 1998 approximately half of expenditures under the new program are for noninstitutional settings, and almost threequarters of beneficiaries are in noninstitutional settings (Exhibit 2). The decline in the number of persons receiving social assistance for long-term care has been less than the reduction in expenditures, and a substantial number of people remain dependent on welfare, which is a disappointment to some. According to national figures, the proportion of nursing home residents dependent on social assistance has fallen from 80 percent in the former West Germany and 100 percent in the former East Germany to about 50 percent overall. 15 Some observers, however, argue that this is primarily a transition issue that will decline as the institutionalized population that depleted its assets prior to the new program dies off. The proportion requiring social assistance for home care is negligible. 16 Despite fears of substantial cost overruns, spending for the German long-term care insurance program initially ran well below EXHIBIT 1 German Social Long-Term Care Insurance Program Income And Expenditures, Billions Of Dollars, Income $8.20 $11.76 $15.59 $15.65 Expenditures Cash benefit Home-based service benefit Partial day care and short-term care Respite care and pension contribution for caregivers Assistive devices Institutional care Institutional care, developmentally disabled Administrative costs Other a a b b b Total expenditures Excess of income over expenditures Surplus SOURCE: Ministry for Labor and Social Affairs, (27 March 2000). NOTES: Deutsche marks converted to dollars using a factor of 0.5. Does not include private insurance. a Not available, because coverage began in July b Less than $0.01 billion. H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

6 G E R M A N Y EXHIBIT 2 Levels Of Disability In Germany s Social Long-Term Care Insurance Program Level I: substantial Level II: severe Level III: very severe Two or more ADL limitations and need for help with IADL Two or more ADL limitations and need for help with IADL Two or more ADL limitations and need for help with IADL ADL: at least once per day IADL: several times per week ADL: at least three times daily IADL: several times per week ADL: day and night IADL: several times per week SOURCE: German Sozialgesetzbuch XI, section 15. NOTES: ADL is activities of daily living. IADL is instrumental activities of daily living. Min. 90 minutes for combined ADL and IADL, with at least 45 minutes for ADL Min. 3 hours for combined ADL and IADL, with at least 2 hours for ADL Min. 5 hours for combined ADL and IADL, with at least 4 hours for ADL original estimates and resulted in significant surpluses (Exhibit 1). The surplus was related to several factors: program contributions began four months before entitlement to benefits; the contribution rate has not increased, but the taxable income base has; and more beneficiaries than expected chose lower-cost cash benefits rather than services (explained below). For 1999 the government estimates that expenditures exceeded revenues by $30 million in Higher costs over time are attributable to a gradually rising number of beneficiaries, minor benefit modifications, the modest shift from lower-cost cash to higher-cost service benefits, and more people in higher eligibility categories. The surpluses are projected to decline until 2005 and then stabilize for several years at the minimum required level (DM 4 billion, or $2 billion). 18 There is no current discussion about either increasing contributions or changing benefits significantly. LONG-TERM 13 CARE SYSTEMS Eligibility, Assessment, And Case Management Eligibility for benefits in the German program depends solely on functional status, making how disability is measured a key concern. Assessments performed by the medical offices of the sickness funds are critically important because they determine eligibility and are the sole mechanisms by which resources are allocated. When the program began, the medical offices encountered several implementation difficulties, involving the enormous volume of applications, staff training, and assessment procedures, but these have largely been resolved. In fact, public confidence in the medical offices competency is considered a key aspect of the program s popularity. The eligibility criteria, which were developed to fit the estimated funds available, are federally established and written into law. Under the social insurance program, persons are eligible if they have a mental or physical condition that results in a need for assistance with activities of daily living (ADLs) and is expected to last at least six months. The minimum threshold for obtaining benefits is limita- H E A L T H A F F A I R S ~ M a y / J u n e

7 L o n g - T e r m C a r e S y s t e m s 14 GERMANY tions in two ADLs and need for help in some instrumental activities of daily living (IADLs). 19 There are three eligibility categories, which vary mostly by the time required for care and the frequency assistance is needed rather than by the number of problems with ADLs or IADLs (Exhibit 2). 20 The eligibility criteria for each category are the same for institutional and home care and apply to those insured through the sickness funds or through private insurance. In June 1998 the actual number of beneficiaries 1,784,805 was quite close to the number estimated to be eligible based on extensive household sample surveys in At the beginning of 1998, 2.3 percent of the total population participated in the program, including 29.6 percent of the population age eighty and older. 22 As expected, persons in institutions are more disabled than are persons electing home care or cash (Exhibit 3). Eligible persons at any level of disability can choose institutional or noninstitutional care. Technically, the medical office of the long-term care funds can decline a nursing home placement if the person can be cared for at home, but in practice this is unheard of, since family cannot be forced to provide care. 23 In fact, the number of persons in nursing homes at the lowest disability level may seem high, but it reflects historical patterns of institutional use. 24 When the program began, nearly 24 percent of nursing home residents did not meet the new eligibility criteria. In contrast, 12 percent of noninstitutionalized beneficiaries are in the most disabled category (which requires at least five hours of care, including during the night), which raises concerns about adequacy and quality of care and family burnout. While the eligibility criteria have not changed since the beginning of the program, there are active debates about expanding eligibility for the cognitively impaired. Critics contend that existing criteria do not adequately address the needs of persons with cognitive impairments, such as Alzheimer s disease. Persons with cognitive impairments need time-consuming general supervision, which EXHIBIT 3 German Social Long-Term Care Insurance Beneficiaries, By Level Of Disability, 1998 All beneficiaries 1,206, ,492 Level I: substantial Level II: severe Level III: very severe 50.3% % SOURCE: Ministry for Labor and Social Affairs, (27 March 2000). NOTE: Does not include private insurance. H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

8 G E R M A N Y Germany consciously rejects case management and resource allocation methods that target services based on need. is not factored into eligibility assessments. Many policymakers were concerned, however, that counting supervision hours in the assessment process would place virtually all cognitively impaired persons in the most costly, most impaired category. Nonetheless, the Schroeder government has made changing the eligibility criteria a high priority, although it is likely that any proposal to accommodate more persons with dementia will involve tightening eligibility for others, to make any change budget-neutral. 25 Although there is functional assessment and it is extremely important, there is almost no case management or social assessment. The level of services and cash payments in the new program is completely independent of family caregiving availability. To do otherwise is thought to be inconsistent with the insurance principles of equal treatment of similarly disabled persons. Thus, Germany consciously rejects case management and resource allocation methods that target services based on individually determined need. Case management at the individual client level, to advocate for clients or to assist them in their choice of services, also is absent. Persons and their families must make their decisions about type of benefits and choice of providers on their own or with the help of a provider. Increasingly, there are discussions about how to assist consumers more effectively. Some observers view the need for consumer assistance to be much broader than long-term care and to include the need for information on housing, health care, and other services. Benefits And Utilization Germany s social insurance program covers extensive institutional and home-care services. Maximum expenditures per person are capped at levels that vary by disability level and institutional status (Exhibit 4). For people receiving care outside of an institution, the program allows the choice of cash rather than services that are paid for by sickness funds up to a set amount. The service benefit may be thought of as a voucher for approved services, the cash benefit as an income supplement. Persons electing cash receive less than half the value of the service benefits, but the use of cash is unrestricted. The benefit structure has been stable for nearly five years, although a few relatively minor modifications were made in June Benefits have not been updated with inflation. Because LONG-TERM 15 CARE SYSTEMS H E A L T H A F F A I R S ~ M a y / J u n e

9 L o n g - T e r m C a r e S y s t e m s EXHIBIT 4 Monthly Benefits Paid For Home And Institutional Care In Germany, 1999 Level I: substantial Level II: severe Level III: very severe Hardship cases $ a $ ,400 1,875 $ ,400 a $1,000 1,250 1,400 1,650 SOURCE: German Sozialgesetzbuch XI, sections 36-37, 41, 42. NOTE: Deutsche marks converted to dollars using a factor of 0.5. a Not applicable. 16 GERMANY monthly costs per beneficiary are capped, program outlays do not depend on the amount of services used per person or provider payment levels, but instead, on whether a person is eligible, what disability level they are categorized as having, and whether beneficiaries choose cash or services. The overwhelming majority of beneficiaries 74 percent in 1998 receive care outside of nursing facilities (Exhibit 5). Among the community-based population, a substantial majority (76 percent) chose cash rather than services, although there has been a slight increase over the past few years in the proportion of beneficiaries choosing services or a combination of services and cash. The proportion varies by level of disability but is still quite high among the most severely disabled. Sixty-five percent of the most disabled group in community settings chose cash only (not in combination with services), compared with 80 percent of the least disabled. The reason for the overwhelming popularity of cash benefits is not clear and might be due to available services, high unemployment, or lack of information. Studies of who chooses cash over serv- EXHIBIT 5 Average Number Of German Social Long-Term Care Insurance Beneficiaries And Percentage Distribution, By Benefit Type And Year, Service Cash Combination service/cash Partial day or night care Short-term and respite care Institutional 82, ,403 82,293 1,777 14,432 a 7.7% a 105, , ,305 3,639 12, , % , , ,543 5,065 9, , % , , ,764 6,774 10, ,750 Total 1,068,695 1,562,088 1,727,414 1,794, % SOURCE: Ministry for Labor and Social Affairs, (27 March 2000). NOTE: Does not include private insurance. a Not available, because institutional benefits began in July H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

10 G E R M A N Y ices have not been undertaken. There is very little concern among government officials over how the cash benefit is used or misused. Persons electing cash benefits do not have to account for how the funds are spent, and there is no requirement that funds be used to buy long-term care services. Dementia is not a reason to deny cash and require services, and a large proportion of beneficiaries have cognitive or psychological impairments. However, the medical office of the sickness funds must ascertain that care is adequate before cash is granted. Furthermore, beneficiaries electing cash are subject to periodic visits every four to six months, depending on disability level, to ensure that adequate care is being provided to the beneficiary. Although systematic data are not available, the common assumption is that most disabled elderly give at least some of the cash to their informal caregivers. If the cash is given to close relatives, it is not taxed as income. From this perspective, the cash benefit was designed to encourage family caregiving. There is clearly a preference for family-provided care if at all possible. To further the family caregiving goal, respite care is provided for informal caregivers (up to four weeks), and pension credit is awarded to persons providing high levels of unpaid services. In 1998 approximately 550,000 persons received pension contributions as caregivers; more than 93 percent of these were women, and 55 percent were between fifty and sixty-five years old. 27 Not all groups benefited from the cash option. Most observers believed that younger persons receiving personal assistance had actually been harmed by the introduction of cash benefits under social insurance, relative to the former social assistance program. Under that program, persons wishing to live independently but requiring personal assistance could arrange for their own caretakers and have the services reimbursed (referred to as the employer model ). Under the new program, however, only participating providers of formal home care can be reimbursed for services. Thus, the beneficiary must elect the cash option, which provides much lower benefit levels. By special exemption, 500 to 800 persons already receiving personal assistance may continue to do so under the new program. 28 Persons electing home and community-based services choose among providers who have contracts with sickness funds. Assistance with personal hygiene, eating, mobility, and household chores is available. However, it is not clear how much is spent out of pocket to obtain these services. Sickness fund data in 1996 from Baden- Wuerttemberg found that the new long-term care program did not fully cover the cost of services for 43 percent of recipients. Average costs not paid by the sickness funds were DM 654 ($327) per month. 29 Nine percent of recipients paid more than DM 2,000 LONG-TERM 17 CARE SYSTEMS H E A L T H A F F A I R S ~ M a y / J u n e

11 L o n g - T e r m C a r e S y s t e m s 18 GERMANY ($1,000) monthly out of pocket. In contrast, a more recent study of partial day care services in Nordrhein-Westphalen found that beneficiaries typically spent up to the benefit limit and no more. 30 Certainly, the common perception is that people do not have large out-of-pocket costs for home care because they do not purchase additional services beyond the value of the benefit. Nursing home coverage includes basic care, medical care, and therapeutic social activities, but not room and board or capital costs. Residents are responsible for no less than 25 percent of the costs of nursing home care. The sickness funds pay flat monthly amounts for institutional care, depending on the level of disability. Consequently, if the costs of care exceed the flat payment, residents must pay the remainder. The proportion of costs that a beneficiary must cover rises with disability level, because program payment does not increase as sharply as nursing home costs do. Consequently, persons with the highest levels of disability may have the greatest disincentive to live in institutions. For the approximately 140,000 younger disabled persons who live in specialized institutions, which provide mostly rehabilitation and special education services (and are roughly equivalent to facilities for persons with developmental disabilities in the United States), benefits are much lower than for nursing homes. The program designers had not intended to cover this group, whose care is predominantly covered by social assistance. Public outcries, however, led to changes in the law so that the program makes a modest contribution (DM 500, or $250, per month). Länder responded, perhaps predictably, by redesignating many institutions as nursing homes to receive higher payments. No coverage is provided for services received in semi-institutional settings, such as sheltered workshops. Availability Of Services One goal of the long-term care program was to increase the availability of nursing home and home-care services and to foster competition across providers. Overall supply appears adequate to meet need. An oversupply of home-care providers is thought to exist in some areas, partly as a result of the unanticipated number of persons electing cash benefits rather than services. According to several experts, waiting lists for nursing home placement have largely disappeared. Only short-term and partial day care facilities were not considered to be adequately available throughout the country. One measure of success is that the number of home-care providers in Germany has risen dramatically. The number of agencies rose from 4,300 in 1992 to 11,800 in 1999, with most of the increase occurring in the first two years of the program. 31 An additional H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

12 G E R M A N Y 60,000 workers are now employed in the home-care industry. 32 Most agencies are small operations with only ten to fourteen employees. In the past home-care providers were primarily connected with charitable welfare organizations, but most of the growth has been among the private, for-profit agencies. Today, 5 percent are publicly owned; 49 percent are owned by charitable, nonprofit organizations; and 46 percent are privately owned. 33 The supply of nursing homes has grown as well from approximately 4,300 in 1992 to approximately 8,100 in 1999, with most of the growth occurring in the first two years. 34 Although some new institutions have been built, particularly in the East, most of the growth in nursing homes came from recategorizing residential facilities as nursing homes. Historically, providers sponsored by charitable welfare organizations received special treatment by social assistance programs and were the dominant suppliers of long-term care services. Under the new insurance program, sickness funds are required to treat nonprofit and private agencies equally and to favor them over public agencies. This has caused tensions as the nonprofit organizations argue that their higher costs result from better-trained, higher-quality staff than exist in for-profit organizations. The new program was intended to foster price competition, but this has not been very successful. As in the acute care sector, sickness funds must contract with all providers that meet minimum standards and cannot selectively contract with providers based on price. Beneficiaries have complete freedom of choice within this list. Rather than establishing competitive bidding, sickness funds use traditional, collective rate negotiations with provider associations, as in the acute care realm. One change, however, is that Länder and communities may not preclude new providers from entering markets, for example, through certificate-of-need. Quality Of Care The new insurance program has focused increased attention on quality-of-care problems. A government study of nursing homes found that workers underestimated residents abilities to act independently, did not motivate self-care, treated quality documentation as tedious and unnecessary, and had difficulties during shortterm staff shortages. 35 In audits of long-term care providers, including both home-care agencies and nursing homes, the medical offices of the long-term care funds found 10 percent of providers to have high-quality, sound quality assurance (QA) systems; 40 percent had begun their QA systems and were seeing positive results; 45 percent were introduced to the need for QA by the medical office LONG-TERM 19 CARE SYSTEMS H E A L T H A F F A I R S ~ M a y / J u n e

13 L o n g - T e r m C a r e S y s t e m s 20 GERMANY and were responding positively; and about 5 percent provided irreparably poor care. 36 As a result of media attention to poor quality, politicians were increasingly interested in improving quality standards legislatively, especially for nursing home care. 37 The existing QA system is an amalgam of earlier federal standards and new initiatives by the sickness funds. A federal nursing home law, Heimgesetz, which predates the long-term care insurance program, is still in effect. Although it specifies such things as building and staffing requirements and involves routine inspections, the standards often were not enforced. 38 There is no similar law for home-care providers, although one is now being debated. Sickness funds, provider associations, and social assistance representatives developed a national quality framework in 1995, which specifies standards and review procedures for institutions and home-care providers participating in the long-term care program. Under existing national guidelines, providers are supposed to put in place internal quality assessment and assurance systems including for example, staff quality committees and consumer surveys. 39 Increasingly, providers are pursuing total quality management (TQM) systems or procedures called for by the International Standards Organization. However, the new standards are minimal, are vaguely worded, and stress mostly structural issues, not outcomes. 40 External reviews are performed by the medical office of the sickness funds, when complaints are made against providers. Recently, two Länder started requiring random audits as well. The medical office has attempted to make its audits a consultative, rather than punitive, process. By 1998 the medical office had audited only 6 percent of all participating providers. In response to public concern over quality, the sickness funds are preparing to release additional quality guidelines, including random provider audits in all Länder. Many observers, however, lamented the poor state of research on quality mechanisms in both institutional and home care. This has hampered the sickness funds ability to negotiate stricter standards with providers and to use quality measures when negotiating rates. Invariably, calls for improved quality were answered by providers demands for higher across-the-board reimbursement, whereas the sickness funds wanted to be able to reward only the higher-quality providers. While sickness funds are preparing new guidelines, lawmakers are drafting legislation to provide sickness funds with greater leverage over quality issues. The only enforcement action now available to sickness funds is to terminate the participation contract; expanding the range of sanctions that could be applied against poorly performing providers is under discussion. There also is interest in hav- H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

14 G E R M A N Y The prevailing view is that care among family members is a personal issue, largely outside the regulatory realm. ing annual, rather than one-time, quality agreements with providers. Incorporating greater consumer input and perspectives into the QA process is just now being debated, and the new emphasis on consumer involvement may reflect the shift from social assistance to social insurance. At the policy level, however, there have been no structures to permit direct consumer input. Instead, coalitions of sickness funds negotiate with coalitions of providers without the involvement of consumers. However, at the provider level, consumer input is starting through residents and families satisfaction ratings, which are part of the internal and external review processes. Although the quality of formal care is being addressed at a number of levels, the quality of informal care is another matter entirely. The prevailing view is that care among family members is a personal issue, largely outside the regulatory realm. For persons receiving cash, the initial medical office assessment and the agency control visits are the primary quality mechanisms for informal and family care. As part of the more widespread quality initiatives, the content of the control visits and the provider responsibilities were being more closely examined. Some experts viewed the cash benefit skeptically, arguing that the insurance program had done little to improve services and quality of care for the elderly. To them, a high proportion of beneficiaries receiving cash means a high percentage of severely disabled persons receiving untrained assistance, which could be especially problematic for persons with severe disabilities. While family members might better understand a person s preferences, a trained aide may take more time to identify physical and mental changes. Some effort is being put into training family members, but several observers have called for more activity in this area. From the government s and sickness funds perspective, no additional quality measures are needed in the family and informal care realm. 41 They argue that while neglect exists, it is not caused or exacerbated by the long-term care program or its cash benefits. Family members should not be treated with distrust, since, after all, those who take on caregiving are taking on a burdensome task, which requires considerable personal commitment. LONG-TERM 21 CARE SYSTEMS H E A L T H A F F A I R S ~ M a y / J u n e

15 L o n g - T e r m C a r e S y s t e m s 22 GERMANY Conclusions And Policy Implications Nearly everyone interviewed for this study regarded the new longterm care insurance program as an immense political and social accomplishment. It had achieved or made substantial progress toward several important goals, including giving security and support to informal caregivers, shifting the balance of long-term care from institutional to home-care services, increasing attention to quality of care, providing Länder with fiscal relief, reducing dependence of social assistance, increasing the supply of providers, and increasing consumer choice. The program was implemented quickly and remarkably smoothly, although there were enormous challenges in establishing a large program quickly. Importantly, despite the possibility of an explosion in the number of beneficiaries, expenditures began well below projections and now, although rising, are well within a politically acceptable range. Germany s changing demographics will place strains on the program. The percentage of the population age eighty and older the age group most likely to need long-term care services will increase from 3.6 percent to 6.3 percent from 2000 to Because of its relatively small size, the program faces challenges that pale in comparison to the health and pension areas. Nonetheless, policy discussions about how to maintain the pay-as-you-go funding through changes in contributions or benefits are being put off. n Lessons from Germany. As other countries examine the experience of Germany, what lessons can they draw to inform their own reform efforts? There are several. First, it is not a law of nature that new social programs, especially non-means-tested ones for longterm care, must cost far more than originally estimated. The German long-term care insurance program has an enrollment fairly close to what was originally projected, and spending has been lower than anticipated. Second, the political success of the program can be tied to some of its design characteristics. Most importantly, the German reform plan provided substantial fiscal relief to the regional and municipal governmental units that funded long-term care. Since there was no maintenance-of-effort requirement, the Länder were enthusiastic proponents of the new long-term care program. In addition, the German insurance program established an entitlement to a specific set of benefits that people could understand and believed they would receive if they met the eligibility criteria. Thus, the population was able to see, in a very concrete way, what benefits they would receive for their new contributions. The program offset the limited range of services that could be covered as an entitlement by H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

16 G E R M A N Y providing a cash alternative, which can be used for any purpose and thus has maximum flexibility. Third, in the face of large increases in provider capacity, assuring quality of care has been a challenge. The program must balance the need for quality assurance against the costs of extensive quality monitoring, both on the part of agencies and on the part of providers who must meet the requirements and collect the data. In addition, efforts to improve quality will most likely lead to higher prices. This may push more people to the cash option, where quality is barely monitored at all. Fourth, Germany s strong insistence on a uniform national program has meant that there is very little variation in services across geographic areas or across individuals in the same disability category. This also has resulted in a fairly rigid program that does not flexibly tailor benefit levels to individual needs or take into account local needs and desires. Fifth, establishing boundaries between acute and long-term care is difficult because individuals have a combination of acute and long-term care needs. Maintaining a separate funding stream for long-term care protects those funds against being swallowed up by the much larger and more powerful acute care system but creates problems of cost shifting and coordination. Sixth, Germany also illustrates the classic conflict between equity and efficiency. Germany s cash payments can be justified on an equity basis in that they make family caregivers better off. On moral grounds, policymakers want to reward informal caregivers for their sacrifices. But from an efficiency perspective, long-term care funds are spending a great deal of money to accomplish relatively little behavioral change. Extensive informal care is being provided, now as before. For people receiving the cash benefit, it is not clear that much has changed in the way they receive care, although some observers think that it is too early to tell. LONG-TERM 23 CARE SYSTEMS The dominant long- term care policy issue in developed countries is the balance between institutional and noninstitutional services. Most countries believe that they are not providing enough home and community-based services. Rather than reallocating funds, Germany s strategy of creating a more balanced delivery system was simple but costly they set out to spend more for noninstitutional services and have done so. Germany may be the only country where both a majority of the beneficiaries and a majority of the expenditures are in community-based rather than institutional benefits. H E A L T H A F F A I R S ~ M a y / J u n e

17 L o n g - T e r m C a r e S y s t e m s Information for this paper was obtained largely from in-person interviews with government officials, representatives of provider and consumer organizations, and researchers in Germany during 1997 and 1999, supplemented by publicly available papers and government documents. The authors thank the German respondents who gave so generously of their time and insights. This paper was presented at the Commonwealth Fund s 1999 International Symposium on Health Care Policy, entitled Financing, Delivering, and Ensuring Quality of Health and Long-Term Care for an Aging Population, in Washington, D.C., October GERMANY NOTES 1. German Parliament, 13th Voting Period, First Report on Developments in the Long-Term Care Insurance Program, vol. 13/9528 (Bonn: 19 December 1997), 8 9 (in German). 2. This section refers to the historical situation in West rather than East Germany. 3. J. Alber, The Debate about Long-Term Care Reform in Germany, in Caring for Frail Elderly People: Policies in Evolution, ed. P. Hennessy (Paris: Organization for Economic Cooperation and Development, 1996), ; U. Goetting et al., The Long Road to Long-Term Care Insurance in Germany, Journal of Public Policy 14, no. 3 (1994): ; and U.S. General Accounting Office, Long-Term Care: Other Countries Tighten Budgets while Seeking Better Access, Pub. no. GAO/ HEHS (Washington: GAO, 1994). 4. Kuratorium Deutsche Altershilfe, Review of Institutional Care of Sickness in Old Age and the Assumption of Costs by thesickness Funds (Cologne: KDA, 1974) (in German). 5. Higher-income Germans can opt to buy private insurance if they have incomes over DM 6,375 ($3,188) per month, but it is difficult for individuals to return to the sickness funds once they have left. 6. However, Länder receive the equivalent of a federal general revenue sharing block grants, which provide much of their funding and reduce revenue differences across geographic areas. 7. Report of the Ministry for Labor and Social Affairs to Parliament (Bonn: Ministry for Labor and Social Affairs, 7 March 1997) (in German). 8. Ibid. 9. High-income persons who opt out of sickness funds for acute care must receive long-term care through their private insurers. Their monthly premium may not exceed 1.7 percent of the opt-out income threshold. In 1998, of the 8.5 million privately insured, 89,000 were receiving long-term care benefits, compared with 1.8 million out of 71.4 million persons insured through sickness funds ( March 1999). 10. Paid holidays are mandatory for most businesses in Germany. One of the Länder, Sachsen, voted not to eliminate a holiday and to have workers pay the entire premium. 11. In Germany contributions are mandatory only for persons below set income thresholds. Persons above these thresholds can purchase alternative private insurance and are not subject to the tax. 12. Social Expenditures Increase Despite Contribution Reductions, Süddeutsche Zeitung, 4 March 1999 (in German). 13. All dollar estimates are based on 1998 purchasing power parities from the Organization for Economic Cooperation and Development, Main Economic Indicators (Paris: OECD, January 1999). 14. German Ministry of Health, unpublished data; and German Parliament, 13th Voting Period, First Report on Developments. 15. German Ministry of Health, unpublished data. H E A L T H A F F A I R S ~ V o l u m e 1 9, N u m b e r 3

18 G E R M A N Y 16. German Parliament, 13th Voting Period, First Report on Developments. 17. Die Finanzentwicklung der sozialen Pflegeversicherung, www. bmgesundheit.de/themen/pflege/zahlen/tabs/ent.html (27 March 2000). 18. German Ministry of Health, unpublished data. 19. ADLs include personal hygiene (bathing, using the toilet, shaving, brushing teeth), eating (including food preparation), and mobility (transferring in and out of bed, dressing, walking, standing, climbing stairs, leaving and returning home). IADLs are household activities (shopping, cooking, cleaning, washing clothes, washing dishes, heating the home). 20. The medical office is allowed to designate certain persons already at Level III to be hardship cases, if they need at least two simultaneous caregivers at night or if they need at least seven hours of ADL assistance per day with at least two hours occurring at night. See German Parliament, 13th Voting Period, First Report on Developments. 21. Ministry for Labor and Social Affairs, Social Long-Term Care Insurance, Bundesarbeitsblatt (March 1999): 127 (in German). 22. Medical Office of the Association of Sickness Funds, unpublished data. 23. J. Wilbers, Long-Term Care in Germany (Unpublished paper, Trier, Germany, 1998) (in German). 24. Ibid. 25. Ministry for Labor and Social Affairs, Germany s Way into the Twenty-first Century: Excerpts from the Coalition Agreement, Stabilizing Long-Term Care (Bonn: Ministry for Labor and Social Affairs, 1999) (in German). 26. German Parliament, memorandum 14/1203 (22 June 1999). 27. Ministry for Labor and Social Affairs (March 1999). 28. Report of the Ministry for Labor and Social Affairs to Parliament. 29. Ibid. 30. Kuratorium Deutsche Altershilfe, Pilot Study on Partial Day Care in Nordrhein- Westphalen (Cologne: KDA, 1997) (in German). 31. Ministry for Labor and Social Affairs, First Report of the Ministry for Labor and Social Affairs to Parliament in Accordance with S. 10 subpara. 4, Sozialgesetzbuch XI (Bonn: Ministry for Labor and Social Affairs, 12 September 1997); and J. Forster, Sharper Controls in Nursing Homes, Süddeutsche Zeitung, 10 March 1999 (in German). 32. J. Wilbers, Long-Term Care in Germany. 33. M. Aberle, One Toothbrushing, 3.30DM, Frankfurter Allgemeine, 10 November Ministry for Labor and Social Affairs, First Report; and Forster, Sharper Controls in Nursing Homes. 35. G. Niehoerster et al., Identifying the Potential for Independent Living, vol (Bonn: Ministry for Family, Seniors, Women, and Youth, 1998). 36. P. Pick, Quality and Quality Assurance in Long-Term Care, Die Ersatzkasse (October 1998): D. Mittler, Caritas Takes On the Sickness Funds, Süddeutsche Zeitung, 8 April 1999 (in German). 38. Wilbers, Long-Term Care in Germany. 39. Medical Office of the Association of Sickness Funds, MDK Concept Paper on Quality Assurance in Long-Term Care according to SGB XI (Essen: MDK, October 1998) (in German). 40. Pick, Quality and Quality Assurance in Long-Term Care. 41. German Parliament, 13th Voting Period, First Report on Developments. 42. G. Anderson, Health and Population Aging: A Multinational Comparison (New York: Commonwealth Fund, October 1999). LONG-TERM 25 CARE SYSTEMS H E A L T H A F F A I R S ~ M a y / J u n e