Executive Summary
High rates of growth in the Medicare home health benefit have made ita focal point in efforts to reduce the federal budget deficit and slow the rate of growth in Medicare spending. The home health benefit covers skilled nursing, home health aide services, medical social services, and three therapies (physical, speech, and occupational) for beneficiaries who are homebound, under the care of a physician, and have a need for skilledcare (i.e., who need intermittent or part-time skilled nursing care, physical therapy, or continuing occupational therapy). From 1988 to 1996, Medicare home health spending increased eightfold and now accounts for 9 percent of all Medicare spending (Congressional Budget Office 1997).
BACKGROUND
The history of Medicare home health care is one of periods of rapid growth followed by the imposition of restrictions to limit that growth. The growth spurt of the 1980s was promoted partially by changes in program rules: in 1972, cost sharing for home health under Part B of the program was discontinued, and further liberalization in 1980 eliminated the 100-visit limit and Part A's prior-hospitalization requirement. Also in 1980, proprietary agencies were permitted to operate without licensing requirements, leading to a rapid expansion in the supply of home health services.
The federal government began serious cost containment efforts in the early 1980s, when it subjected home health care to a period of extremely tight regulatory control. These efforts resulted in a virtual "Catch 22," where eligible individuals had to be so incapacitated as to be unable to operate outside the home, but well enough to need only part-time care. This regulatory strategy proved to be very effective, resulting in declines in per capita expenditure growth for several years. This occurred even though, at about the same time, the hospital reimbursement system was changed in ways that would have been expected to increase demand for such services.
In 1989, reinterpretation of Medicare rules again broadened coverage of home health services and contributed substantially to the recent growth in expenditures. This growth may also be due to two other factors: the absence of copayments and the use of a cost-based reimbursement system. Unlike most other parts of Medicare, home health services require no copayment, and with no out-of-pocket expenses, beneficiaries have little incentive to limit their service use. Further, Medicare reimburses home health agencies on the basis of their costs up to ceiling levels, thus providing no reward for efficient service provision. The home health program also has few controls on the amount of services provided—since the content and duration of visits are not monitored, agencies have incentives to make as many visits as possible and to keep them as short as possible.
While growth in home health services has occurred across the entire benefit, the greatest increases have come in the average number of visits provided per user. From 1988 to 1995, the share of enrollees receiving Medicare home health services increased from 5 percent to 9 percent and the average number of visits provided per user more than tripled, from 24 to 80. Moreover, much of this growth is attributable to users receiving 100 or more visits, indicating that the Medicare home health benefit is increasingly used for long-term care (Health Care Financing Administration 1997). High-volume users appear to be receiving proportionately more home health aide services and fewer skilled services and to be doing so over an extended period relative to the low-volume users (Lewin-VHI 1995).
Payments to home health agencies have also risen over time, but not by nearly as much as the use of services. Further, parallel with the steep growth in spending has been a large increase in Medicare-certified agencies: from 1988 to 1995, their numbers grew by 57 percent. Particularly striking has been the increase in proprietary agencies, which constituted 43 percent of all agencies in 1995. For reasons that are not well understood, proprietary agencies provide more services (an average of 40 percent more visits) to their Medicare clients relative to other agencies.
Another dimension of home health care receiving scrutiny is the substantial variation across regions and states in use of the benefit. While regional variation in Medicare home health use may be due partially to differences in need for services, most is due to other factors. Some of this variability seems to be related to proprietary agencies promoting use of home health services in some areas of the country (Schore 1994). In other areas, particularly the South, Medicaid programs have actively and effectively sought to obtain Medicare services for their beneficiaries (Kenney et al. 1996).
THE NEED FOR CHANGE
Slowing the rate of growth in home health spending substantially could generate billions of dollars in savings to Medicare. Regional variation in rates of use, potential abuse and even fraud by some agencies, and rapid rates of growth among high-volume users have set off alarm bells regarding home health. However, while overall use is likely too high and should be limited, this over-utilization does not apply across the board. Some of the expansion in services has been desirable, particularly as it reflects the changing health care system, in which delivery of services is increasingly moving out of institutions and into settings such as the home. This use of home health in the post-acute period likely saves Medicare money, and discouraging such expansions makes little sense.
Considerable evidence suggests that Medicare is becoming more of a long-term care benefit for chronically ill patients, which is a new area of service for the program. As such, the benefit fills a potentially important gap for many older and disabled persons, since long-term care is not well covered by other insurance. But while these benefits are highly valued by users, the program lacks standards for judging when care is appropriate, and payment levels may be overly generous. Anecdotal evidence suggests abuse of these services by some agencies and beneficiaries.
OPTIONS FOR SLOWING GROWTH IN SPENDING ON HOME HEALTH
In seeking limits, should the focus be on the supply of services and the behavior of providers, or on demand by beneficiaries? If problematic growth in home health stems more from the supply side of the program, cost sharing may not be the best solution, since it would affect all users of home health services and would not affect providers' incentives. Instead, changing the reimbursement system or improving oversight might be more appropriate. Medicaid's possible role in maximizing use of Medicare benefits is also a complicating factor. Given how rapidly home health is growing, seeking reforms that will affect both the supply and demand sides may be advisable.
Prospective Payment Options
Most of the focus on altering Medicare's reimbursement system for home health services has been on replacing the cost-based, per-visit reimbursement system with some variant on a prospective system. Moving the program away from a cost-based system would help to improve the incentives for providing care more efficiently. Further, this payment system could be established either on a per-visit or per-episode basis. The advantage of an episode-based approach is in moving the benefit further from a payment structure that encourages excess use.
Many problems remain in the development of such a system, including the lack of standards regarding what care is necessary and the lack of effective case mix adjustors to distinguish among types of patients. Case mix adjustors are essential for an episode-based system. Further, the problem of starting from a base that contains great variations in level of payments and use across different types of agencies and around the country makes establishing national standards difficult.
As a result of these problems, an interim adjustment makes sense, with more permanent reforms such as prospective payment adopted later. Moving immediately through an interim adjustment is important, since a freeze on payments that had helped to limit payment levels was lifted in 1996. Development of the most appropriate form of a new prospective system could be based on the establishment of the adjustments described above.
Asking Beneficiaries to Pay More
Payment reforms alone are unlikely to slow growth in the use of services sufficiently over time. One often-proposed additional option is the reintroduction of coinsurance on home health services, which would make beneficiaries more aware of the costs of services and thus help to reduce use. Because the typical coinsurance amount for Medicare is 20 percent, that figure is often mentioned as the appropriate amount. But such a level of coinsurance would add a substantial burden on users of home health services, raising their required cost sharing by more than $1,200 per year on average. This would be particularly burdensome for very old and frail beneficiaries, who tend to have low incomes and be disproportionate users of home health services. Thus, a 20 percent coinsurance requirement would likely prevent use of appropriate as well as inappropriate services.
A more sensible approach, which would make beneficiaries sensitive to use but not form a barrier to care, would be to limit copayments to small amounts per visit (such as $3 or $5). Additional limitations on beneficiary liability could exempt days after a hospital stay or be restricted to home health aide visits, which is the fastest growing part of the program and the one most likely to be used for long-term rather than acute care.
In conjunction with modest copayments, shifting home health to Part B and then increasing the premium that people are asked to pay would both raise revenues and provide additional relief for the Part A trust fund, since all home health costs would be shifted out of that part of the program. The rationale would be to not only move home health into the part of the program that covers other ambulatory services, but also to achieve further contributions that would be more evenly distributed among beneficiaries than coinsurance. Options for greater beneficiary contributions of any type, however, should be sensitive to the particular vulnerability of low income beneficiaries.
Other Ways to Reduce Service Use
In addition to payment reforms and cost sharing, other options that rely on some type of limit on use may also be appropriate. They could be in the form of aggregate limits by region that would reduce future payment levels if use of services rose too rapidly, or limits on a per capita basis that would simply restrict how many services are available to any beneficiary over a defined period. A less restrictive version of per capita limits would be greater case management and oversight of benefits, particularly for beneficiaries whose high use triggers review. Finally, greater efforts to reduce fraud and abuse, including such simple approaches as providing beneficiaries with a notice of benefits claimed on their ehalf, would be useful.
The search for reasonable ways to reduce growth in spending on Medicare's home health benefit while maintaining this service for beneficiaries in need will likely prove difficult. In addition to making short-term changes, it would be useful to step back, examine the benefit, and determine what makes sense in terms of its overall role in the Medicare program.
The full report is not available at this time.