Milestones in Medicare Managed Care.

Medicare managed care has a long history, dating back to the beginning of the Medicare program. The role and prominence of managed care in Medicare have both changed over the years; though plan participation has waxed and waned, enrollment has grown steadily The greatest growth in Medicare managed care enrollment occurred in the middle to late 1990s, coinciding with the "managed care revolution." Enrollment growth has slowed in recent years, plan participation is declining, and the future of the program is not easy to predict.

Introduction

The President, in his message to Congress, decries the runaway inflation in health care costs, the inequities in access to health care, and the variation in the quality of health care across the Nation and across income classes. He champions a novel approach to national health care reform that would rely on market forces to bring discipline to the health care system. Congress balks and does not give the President what he wants, but in the years that follow, reform is achieved, after a fashion.

A familiar story? The President is, of course, Richard Nixon, conveying a message to Congress in 1971 and pointing out how much the Federal programs contributed to “this growing investment in health” as a portion of national expenditures (National Health Insurance Proposals, 1972). The novel approach he advocates is the “health maintenance strategy.” Unlike a later President's proposal—which specifically excluded Medicare from the novel managed competition approach—the Nixon Administration's health maintenance strategy would have begun with the then-relatively-new public programs, Medicare and Medicaid. Having started with the Federal Government programs, “the government's actions would catalyze similar restructuring in the private, largely employer-financed segment of the health economy that also was having difficulty coping with medical inflation” (Ellwood and Lundberg, 1996).

Medicare Managed Care in 1965

This historical tidbit illustrates the close ties that exist, or some hoped would exist, between Medicare and managed care. Although in 1965 the term “health maintenance organization” (HMO) had yet to be coined, what came to be known as HMOs, or their precursors (such as group practice prepayment plans), have been a part of the Medicare program since its inception in 1965. To be more precise, the Medicare program recognized prepaid health care plans as a different kind of entity for which a different kind of payment method was necessary. In 1965, prepaid plans were accommodated by permitting them to be paid on a reasonable cost basis for services (such as physician services) that the program would otherwise be paying on a reasonable charge basis.

This approach to payment made sense in that, if the HMO-like organization used salaried physicians, there would be no service-by-service billing by the physicians and, therefore, no “charge” other than the aggregate “charge”—a salary—for any and all services rendered by the physician. In choosing between the two available payment options that Medicare used in 1965—reasonable charge payments and reasonable cost payments—this seemed like a suitable approach, but not necessarily a perfect fit. Thirty-five years later, we are still in search of a “perfect fit” in Medicare's approach to payment of HMOs.

1972 Amendments

The history of managed care and Medicare can be described through milestones that generally coincide with legislative history. After 1965, as previously alluded to, the next major milestone in Medicare HMO provisions was historic—in a symbolic sense if not in a practical sense. Although Congress may not have given the then-President everything he asked for, one result of the Nixon Administration's push towards health care reform was passage of the HMO Act of 1973. Before the HMO Act, however, the first Federal legislation in which the term HMO was defined was the Medicare provisions of the 1972 amendments to the Social Security Act.

The 1972 amendments introduced Medicare HMO enrollment and contracting, as opposed to merely providing for a mechanism to secure reimbursement for services rendered by such organizations. HMOs had to meet certain standards, had to provide the full range of available Medicare services, and had to have open enrollment for all Medicare beneficiaries in the service area. However, the new payment methodology eventually agreed upon for Medicare HMOs proved to be not very popular with HMOs. The original version of what became the 1972 amendments (H.R. 1 of 1971) included, for Medicare HMOs, a prepaid, capitated payment methodology that was more consistent with the usual method of prepayment (and quite similar, in fact, to the methodology of the Tax Equity and Fiscal Responsibility Act of 1982 [TEFRA]), described later, incorporating a requirement that excess revenues be used for extra benefits). The bill that was passed had Medicare HMOs being paid on a “risk-sharing” basis, with a non-risk cost reimbursement contracting option also available. Group practices, union and employment-based plans, and HMOs could also continue to be paid under the pre-existing cost reimbursement method (and such organizations would not have to comply with the open enrollment requirements applicable to contracting HMOs).

Under a risk-sharing contract, interim payments would be made, and the costs incurred each year by a contracting HMO would be compared with the adjusted average per capita cost (AAPCC)—an estimate of program expenses that otherwise would have been incurred for the Medicare beneficiaries enrolled in the organization. If the organization achieved “savings” in relation to the AAPCC, up to 20 percent of the savings would be split equally between the Federal Government and the HMO. Savings in excess of 20 percent would go to the Federal Government. Losses were the responsibility of the HMO, but could be carried over into subsequent years and offset against savings. Very few HMOs took advantage of this option. In 1979, there were 32 group practice prepayment plans (the pre-existing cost reimbursement option), 32 HMO cost contractors, and only 1 risk-sharing HMO (Langwell and Hadley, 1989).

Demonstration Projects

Again, the perfect fit—the right payment methodology for Medicare HMO payments—had not been achieved. However, the 1972 amendments also introduced additional authority, beyond that introduced in 1967, for Medicare demonstration projects of such things as prospective payment methods and payment for comprehensive services. Continuing the quest for the perfect payment methodology, in 1980 the Medicare capitation demonstrations began, to be followed by the Medicare competition demonstrations and, eventually, the next legislative milestone, the TEFRA risk program.

TEFRA

Other than in demonstration projects, a contracting option for Medicare HMOs operating on a full risk basis would not be available until the changes made by TEFRA were to become effective, which, according to the legislation itself, would be 13 months after enactment or, if later, after “the Secretary…notifies… [Congress]… that the methodology to make appropriate adjustments…has been developed and can be implemented to assure actuarial equivalence in the estimation of the adjusted average per capita costs.” The TEFRA risk contracting program authorized in 1982 would begin in mid-1985, after publication of the final regulations in January 1985.

Under TEFRA, contracting HMOs or competitive medical plans (which were essentially HMOs that did not have a Federal qualification designation under the HMO Act of 1973) would be paid 95 percent of the AAPCC on a full risk basis. The 5 percent differential recognized the presumed greater efficiency of HMOs and their ability to reduce program expenditures. Any additional savings, determined through a prospective comparison of projected costs with projected AAPCC payments, had to be (a) returned to beneficiaries in the form of extra benefits or reduced cost sharing; or (b) used to fund future additional benefits; or (c) be returned to the Federal Government. HMOs were allowed the normal level of profit, or retained earnings, that they customarily received in the private sector.

There were certain changes to the earlier law. What had been a requirement that no more than 50 percent of an organization's membership could be over the age of 65 became the new 50/50 rule, limiting Medicare and Medicaid enrollment to no more than 50 percent of total enrollment—a provision that could be waived by the Secretary of Health and Human Services. Although the open enrollment requirement was continued, Medicare beneficiaries (of any age) with end stage renal disease (ESRD) were prohibited from enrolling in TEFRA HMOs, unless they were already members of the HMO (as pre-existing Medicare enrollees, or as non-Medicare enrollees continuing in the HMO). The cost contracting option continued to be available to HMOs, and HMOs (and other organizations) could continue to be paid under the health care prepayment plan (group practice prepayment plan) option.

AAPCCs were computed for each county of the United States, with separate rates for the disabled and elderly (and statewide rates for ESRD enrollees), and certain adjustment factors were applied to better approximate fee-for-service (FFS) costs: age, sex, institutional status, and Medicaid status. However, there was no direct health status adjuster. As would become evident, the lack of a health status adjuster meant that there was still not a perfect fit in the payment methodology.

The TEFRA program enjoyed a certain level of success in its earliest years. In 1985 there were 480 operating HMOs in the U.S. and 87 Medicare risk contractors. By the end of 1987, the number of risk contractors rose to 161 (out of 662 operating HMOs [Group Health Association of America, 1993])—the “high water mark” for the first 10 years of the program. Risk enrollment rose in every year, but not very dramatically, from 1.1 million at the end of 1985 to nearly 2 million by the end of 1990.

As measured by beneficiary interest in the program, Medicare managed care, in areas where it was offered and included additional benefits, was a highly successful program. In south Florida, for example, Medicare HMOs offered drug coverage and low out-of-pocket costs through “zero premium” plans (plans in which enrolled members did not have to pay an additional premium beyond the Medicare part B premium). The ability to provide additional benefits in certain areas was partly a function of the payment methodology, which recognized the extreme variation in Medicare FFS expenditures in different counties: some counties had per capita costs that were two to three times more than other counties. These payment differences were very visible to Medicare beneficiaries, with residents of Minnesota and Massachusetts, for example, less likely to have the kind of added benefits that were available in the Miami and Los Angeles areas.

Milestone of Another Sort

During the early TEFRA years, the largest Medicare contractor was International Medical Centers (IMC) of Florida, which began as a demonstration project and continued as a TEFRA contractor. It consisted almost exclusively Medicare enrollees, operating under the authority of a waiver of the 50/50 rule. The demise of IMC in 1986 was the low point in the history of Medicare managed care. The organization had enrolled over 100,000 Medicare beneficiaries. The following indicates the drama associated with this particular episode of Medicare managed care history. This is the text of the FBI international crime alert regarding the head of IMC:

“In 1986, a federal government task force was established to investigate charges of corruption and fraud on the part of Miguel Recarey, Jr. Recarey was then head of International Medical Centers, America's largest health maintenance organization. During its peak years, International Medical Centers received three-hundred sixty million dollars a year in U.S.- government Medicare funds. In April 1987, the first indictment was returned in Miami, Florida, against Recarey and three co-defendants for conspiracy, bribery, obstruction of justice, and illegal wiretapping.”

The IMC crisis for Medicare and for the Medicare enrollees of the organization was alleviated when Humana took over operation of the plan, which continues operating to this day. However, the image of Medicare HMOs was tarnished by the IMC experience for many years afterwards.

Fits and Starts in the Late 1980s

In the late 1980s and early 1990s, the number of Medicare-contracting HMOs began to decline (as did the number of operating HMOs in the U.S.). From the 1987 high of 161, the number of contracting plans declined to 93 by December 1991. In 1989, for example, Prudential Insurance, which had applied to have 30 contracts across the U.S., scaled back its Medicare contracting to only a few plans. Enrollment continued its rise, however, reaching 1.4 million beneficiaries in risk plans by the end of 1991, even though the number of contracts was at its post-1985 low point in 1991.

During this period, relatively large numbers of Medicare beneficiaries were affected by contract terminations or service area reductions (under a new policy of the mid-1980s which allowed HMOs to choose which counties they wanted to include in their Medicare contracts among those counties where they otherwise operated). For the years 1987-1989, terminations affected an average of nearly 7 percent of enrollees each year (not including those affected by service area reductions).

Not All Milestones Are Legislative

With the managed care revolution of the mid-1990s, HMOs burgeoned in the private sector as well as the public sector. Medicare HMO enrollment doubled between 1993 and 1996 (to 4.1 million enrollees), just as enrollment overall in HMOs doubled from January 1993 to a January 1999 level of 81 million (Interstudy, 1999). Between December 1994 and December 1998, Medicare risk HMO enrollment nearly tripled, rising to 6.1 million beneficiaries, or over 15 percent of the Medicare population. Within areas in which Medicare HMOs were available, one in five beneficiaries had elected to enroll in a plan, while in the private employer market, about one-third of individuals were covered by an HMO (Buckley and D'Amaro, 1998). While in 1993 only about one-half of Medicare beneficiaries resided in a county in which a risk plan was available, the interest in Medicare contracting expanded to such an extent that by 1998, 74 percent of beneficiaries had at least one Medicare plan available in their area.

Medicare+Choice

The most recent major legislative milestone was the Balanced Budget Act of 1997 (BBA). It was heralded as the most significant change in private plan contracting in the history of Medicare. The BBA introduced major revisions in the types of private plans that could have Medicare contracts, the contracting standards to be applied, beneficiary enrollment rules, and payment rules. The BBA also finally introduced a Part C of Medicare, “Medicare+Choice (M+C),” a quarter of a century after the Nixon Administration's proposal to add Part C. Under Part C, new types of organizations included provider-sponsored organizations, preferred provider organizations, medical savings account plans (on a demonstration basis), private FFS plans (the first “defined contribution” option, because there is no statutory limit on its Medicare premium), and religious fraternal benefit organizations.

Continuing to look for that perfect fit in the approach to payments, the BBA made a number of major changes in the method of computing Medicare capitation payments to health plans. The BBA introduced national/local blended rates, a payment floor for the lowest-paid counties, and a minimum update payment. Under the minimum update provision, all counties are guaranteed a payment increase of 2 percent over the preceding year's base rates. Annual payment increases after 1997 would be based on an update factor that is the rate of increase in projected Medicare expenditures each year, less a statutorily specified reduction (as opposed to the AAPCC methodology, under which each county's rate of increase would be based on a projection of the actual incurred Medicare expenditures in the county for the year in question).

In general, historically lower-paid counties (which are less likely to have had Medicare managed care plans) would receive higher payment increases as a result of the BBA's payment floor and the phased-in national/local blended payment rate—attempting to address the Miami versus Minnesota issue of M+C benefits varying as a reflection of Medicare FFS payment rates. Many counties that historically had higher payment rates had their rate increases reduced by the BBA.

The BBA also reduced M+C capitation rates by phasing in the removal of direct and indirect medical education payments from M+C rates beginning in 1998, providing instead for phasing in direct payment of these “carved out” amounts to the hospitals providing care to M+C enrollees.

The BBA applies a budget neutrality adjustment to the blended rates. The effect of this adjustment in 1998 and 1999, as well as in 2001, was to have payments at the “floor” level or at the minimum 2 percent update level for all counties. For the year 2000, blended payments were made for the first time. (The BBA Refinement Act [BBRA] modified some of the payment provisions: the update factor reduction for 2002 is changed to 0.3 percent rather than 0.5 percent; and bonus payments for 2000 and 2001 are provided to the first organization entering an area that has not had a M+C plan since 1997.)

The BBA also requires that health status be used to adjust payments to M+C plans, in light of the evidence over the years of favorable selection in HMOs—i.e., enrollees tend to be healthier than average Medicare beneficiaries but plans are paid based on costs for an average population (U.S. General Accounting Office, 2000a). HCFA chose to introduce the BBA risk adjustment methodology on a phased-in schedule, and the BBRA modified the phase-in schedule to more gradually phase in the share of payments that would be computed on a risk-adjusted basis.

Some of the BBA changes appear, to date, to have been more symbolic than practical, to repeat a phrase used about the 1972 amendments. Only one provider-sponsored organization contracted with Medicare after the BBA, and that organization will terminate its contract at the end of 2000. One private FFS plan is operating, and there are preferred provider organization applications pending. The BBAs repeal of the 50/50 rule has not resulted in any HMOs going into new areas as Medicare-only HMOs, and the payment floor and blended rates (when possible under the budget neutrality rules) have not resulted in increased access in rural areas.

Terminations in 1998-2000

For the 2001 contract year, over 900,000 Medicare beneficiaries—about 15 percent of all enrollees—will be affected by a plan termination or service area reduction. About 150,000 will not have access to another M+C plan (other than the private FFS plan in some areas). Overall, only 63 percent of Medicare beneficiaries will have access to an M+C coordinated care plan in 2001. In the preceding year, 327,000 enrollees were affected by terminations and service area reductions (5 percent of enrollees), with 79,000 left without an M+C plan available. In the preceding year, at the end of 1998, 407,000 enrollees were affected. For the first time since the beginning of the program, overall enrollment declined from one year to the next (December 1999 to January 2000, from 6.35 million to 6.19 million).

These changes may give one pause as to whether the BBA did the opposite of what it was intended to do (in terms of expanding private health plan choices for beneficiaries while also controlling Medicare expenditures). However, just as one could note that the increase in Medicare enrollment that coincided with the managed care revolution cannot be traced to any particular change in Medicare, one might also argue that other factors besides the BBA changes explain the post-BBA downturn in enrollment trends among Medicare beneficiaries and the leveling off of interest in Medicare on the part of HMOs and other private plans. The U.S. General Accounting Office, for example, argues as much (U.S. General Accounting Office, 1999; 2000).

Murky Future

What the future holds is difficult to say, other than perhaps to say that these things—the insurance cycle, actions by Congress to control Medicare expenditures—are cyclical (a redundancy in the case of the insurance cycle, except that some had claimed the insurance cycle had disappeared). It is probably wisest not to opine on this issue. Health policy analysts are notoriously bad at making predictions. When the HMO Act was enacted in 1973, the Nixon Administration announced a strategy calling for the development of 1,800 HMOs—a projection somewhat off the mark. To cite another example, in 1990, very few people would have predicted the managed care revolution, and the consequent pre-eminent role to be played by managed care plans and the virtual disappearance of FFS indemnity plans. Later, in the midst of the managed care revolution, perhaps few people would have thought that the future of managed care was not secure—its success at controlling costs would ensure that it would be the model for health care forever after, and its success in attracting Medicare beneficiaries would continue indefinitely.

Loose Ends

A different way to approach the question of predicting the future is merely to recite what the past has left undone. The perfect fit in payment has not been achieved (but may be forever elusive—would it be through some competitive pricing approach?). Extending managed care to rural areas remains problematic, and matching the needs of certain populations with managed care plans is an issue (e.g., those with ESRD, the disabled). Perhaps, with time, other viable models of managed care or private health plans will be developed for Medicare as a result of the BBA Thirty-five years just seems not to have been enough time to sort all this out, but as the history shows, people are at least aware of some of the issues worth pondering in Medicare managed care.