Payment Rates and Methodology

The BBA substantially restructured the system for setting the rates that Medicare pays health plans. In general, Republican lawmakers wanted to increase the number of health plan options available to beneficiaries. The new payment system was intended to address many of the problems associated with the previous adjusted average per capita cost (AAPCC) payment system, such as inequity across counties and artificial ties to the Medicare fee-for-service system. Since its inception, critics of the AAPCC system have complained that HMO payment rates are uneven and complicated. Because the AAPCC rate paid 95 percent of estimated fee-for-service costs, critics argued that the payment method rewarded inefficient markets. Thus, many lawmakers wanted to sever the link between local fee-for-service costs and payment updates to plans.

In addition, rural counties typically received a low payment rate, which inhibited health plan participation. Several lawmakers pushed to establish a “minimum floor” to bolster the payment rates in these low penetration areas.

On the other hand, some lawmakers wanted to ensure that HMOs were no longer “overpaid” by Medicare. Several studies submitted to Congress have indicated that HMOs were paid more than actual costs as a result of both the payment formula and favorable risk selection. The former Physician Payment Review Commission (PPRC), in its 1997 Annual Report to Congress, estimated that Medicare was paying up to $2 billion in excess payments to managed care plans each year. The Congressional Budget Office has said HMO costs average roughly 87 percent of Medicare fee-for-service costs—less than 95 percent of costs that the AAPCC formula reimbursed. And the DHHS Office of Inspector General has said that Medicare HMOs overstated their anticipated administrative costs in 1994, 1995, and 1996 by as much as $2 billion. In response to these reports, some lawmakers sought to limit growth in payments to managed care plans.

The resulting payment system enacted under the BBA represents a political compromise. Under the new system, plans were to receive the greatest of three possible rates:

• a new minimum or “floor” payment ($379.84 in 1999),
• a minimum 2 percent increase over the previous year’s rate, or
• a blend of the county rate and a national rate.

Further complicating the equation, the BBA mandates that the new payment rates must achieve budget neutrality so that total payments under the new system are no greater than total payments would have been if calculated on a basis similar to the AAPCC. The BBA stipulates that, if aggregate payments are too high, adjustments must be made only to blended rates in order to protect the floor payment and minimum 2 percent increase from reductions.

As a result of these multiple moving parts and Medicare’s recent low growth rates, the increases in payment rates were not as high as many plans, HCFA, or Congress had anticipated. Application of the budget neutrality provision resulted in rates for all counties that were either at the floor or the minimum 2 percent increase. Thus, no county rates for 1998 and 1999 were based on the blend. The blended rates were expected to increase payments and encourage Medicare+Choice offerings in rural and other low payment areas.⁹

Several other changes to the payment rates will be occurring simultaneously over a five-year implementation period. Graduate medical education (GME) costs, which had been included in HMO payments under the old system, have been carved out of county rates over the five-year implementation period and will be paid directly to teaching hospitals. The blended local/national rate must reach a 50/50 balance over the same time period. The national rate, local rates, and the minimum payment amount will be annually updated based on per capita Medicare growth. Moreover, the BBA directs the secretary of health and human services to implement risk adjustments to payments by January 1, 2000.

As stated earlier, rural counties represented a large share of those affected by the plan withdrawals. Plans argue that the payment floor of $380 per member per month does not adequately cover the added costs of assembling networks and operating HMOs in rural areas in which they have low enrollments, little cost control or ability to negotiate with providers, and no real prospects for payment increases above the floor payment.

Nineteen ninety-eight was the first year in which Medicare payments in all but the rural counties were subjected to the 2 percent minimum increase. Plans report that health care costs are rising faster than Medicare reimbursements. In some cases, HMOs saw their medical-loss ratios—the percentage of premium dollars they pay for medical costs as opposed to administrative and other costs—exceed 100 percent. The blended payments, if they had been implemented, would have cushioned the shock of going from 8 to 10 percent annual rate increases to the minimum 2 percent increase for some plans. On the other hand, HCFA notes that the 2 percent increase was meant to be a guarantee. As designed, the blended payment rate is intended to move some dollars from high payment areas to low payment areas, which means some plans will receive less than in the past. The 2 percent increase insures that no plans receive less than the minimum update. Nevertheless, plans report that the prospect of long-term fixed growth at 2 percent caused some to withdraw from Medicare in certain markets because they were worried about their ability to achieve profitability over the long term, especially when health care inflation is expected to rise 6 percent annually over the next four years.¹⁰

The user fee associated with the beneficiary education information campaign has also been blamed for reducing the dollar payment update to health plans and causing plans to withdraw. Under the BBA, HCFA is authorized to withhold a fee from each Medicare+ Choice plan to finance a beneficiary education program to assist Medicare beneficiaries in making informed choices about their health plan options. Each plan is assessed a portion of the total $95 million outlay on the basis of its share of total Medicare+Choice payments (about $32 billion for 1998). The program is intended to provide beneficiaries with objective information about their health plan options. Plans argue that the costs should be spread across the entire Medicare market, not just the Medicare+Choice plans.

Some analysts have argued that the plans’ retreat from Medicare simply demonstrates that so-called easy money is no longer available. In testimony before Congress, Judith Feder, professor of public policy at Georgetown University, argued that the

relatively high historical payments to managed care plans provide them with a strong cushion to absorb lower rates of increase in the future and still earn healthy returns and provide broad benefits to Medicare enrollees, consistent with quality standards equivalent to those applied in the private sector.¹¹

Other Medicare experts say these changes in the marketplace should not be unexpected. The BBA was intended to slow growth in payments to private plans and to make payments more equitable across counties. “By their very nature, these policies will change payment rates in ways that lead plans to retrench in some areas and to take advantage of opportunities newly available to them in other areas,” noted Gail Wilensky, chair of the Medicare Payment Advisory Commission (MedPAC), in recent congressional testimony.¹²

In summary, health plans argue that payment rates are below costs in some areas and the 2 percent annual increase will not keep up with inflation. The opposing view is that plans have received overly generous payments for the past several years and the withdrawals represent a normal sorting out process associated with market competition and appropriate payment restructuring.

Increased Regulatory Requirements

HCFA issued interim final rules for the Medicare+ Choice program on June 26, 1998. Known as the “mega-reg,” the 800-page regulations set forth detailed requirements for plans that participate in the Medicare+Choice program, including:

• Quality assurance and performance improvement requirements. Medicare+Choice organizations will be required to achieve compliance through the use of HCFA’s Quality Improvement System for Managed Care (QISMC).¹³
• Specified enrollment information and encounter data collection requirements (including a requirement for 100 percent “accuracy, completeness, and truthfulness”).
• Rules governing access to health care such as direct access to specialists and emergency care.
• New appeal and grievance requirements.
• Marketing material revisions.
• New provider contracting requirements.

HMOs have said that the burden imposed by the Medicare+Choice regulation has caused some HMOs to stop serving Medicare beneficiaries. Because of the fluctuating market conditions in which they operate, plans contend that the regulations should provide flexibility rather than rigid, bureaucratic requirements. The plans argue that the Medicare+Choice regulation will require extensive new compliance activities that must be initiated even before, in some cases, HCFA has issued detailed guidance to supplement the regulations.

In addition, the rules require tight implementation schedules that plans have criticized as unrealistic. In testimony before Congress, Willis Gradison, then president of the Health Insurance Association of America, said “Even if fiscally possible, the procurement, installation, training, and validation time requirements would take years to fully implement.”¹⁴

It is important to note that most of these requirements and time lines were mandated by Congress as part of the BBA. These provisions were intended to provide more protections to beneficiaries and to move HCFA in the direction of operating more like a private purchaser, making performance demands on the plans with which they contract. Proponents of the new data requirements argue that they are very similar to what is currently required by large private purchasers. Nevertheless, plans say, the regulatory requirements contributed to their decisions to withdraw from the Medicare market and argue that HCFA’s approach has been “needlessly ambitious and complex.”¹⁵ In response to plans’ concerns, HCFA has announced some refinements since the regulations were released, especially relating to the quality improvement requirements. It has also announced that plans will have an additional year to implement compliance plans and complete contracts that incorporate new requirements with their current providers.

ACR Filings

As part of the Medicare+Choice structure, health plans had to submit their annual “adjusted community rate” (ACR) to HCFA in May 1998 instead of in November, the customary month for compliance. The deadline was moved up to give HCFA more time to compile information for distribution during the November Medicare beneficiary information campaign.

ACRs detail benefits beyond basic coverage that the plan will offer as well as the cost-sharing requirements for Medicare beneficiaries. According to the American Association of Health Plans (AAHP), a significant factor in the decisions of plans to reduce their service areas or withdraw from the Medicare program was their inability to revise ACR filings submitted last May, before the mega-reg had been issued. Indeed, AAHP embarked on an intensive lobbying campaign for HCFA to allow plans to amend their ACRs. The plans maintain that they could not anticipate the cost of complying with the Medicare+Choice interim final regulations or providing unexpectedly high cost trend items, such as prescription drugs.¹⁶

HCFA refused the plan’s request to resubmit their ACRs. HCFA’s position is that it would not be in the best interest of beneficiaries to allow HMOs to increase premiums and cost-sharing and reduce benefits to their Medicare enrollees. Several senior groups supported HCFA’s decision, saying that allowing plans to resubmit their ACRs would have “undermined the integrity of the whole competitive process of Medicare+Choice.”¹⁷

Provider Network Problems and Marginal Market Share

While low reimbursement rates are among the most frequently cited reasons for HMO withdrawals, other market conditions clearly impacted plans’ decisions. Some analysts have suggested that the pullouts are simply a “market correction,” stabilizing the rapid expansion of HMOs into the Medicare market over the last few years. In some cases, the termination of a contract was the result of a merger between two plans in which only one corporate entity will continue to contract with Medicare. Some plans have also terminated their contracts with the Federal Employees Health Benefits Program and Medicaid, suggesting that plans’ decision were based on reasons beyond Medicare payment issues.

For some plans, the withdrawal decision seems to be based more on their enrollment growth potential and the ability to be profitable in the long term. For example, Aetna pulled out of Massachusetts, where payment rates in Boston are as high as $650 per member per month, and Tolland County in Connecticut, where payments average $500 per beneficiary, the second highest in the state.¹⁸ Many of these plans had a limited capacity to reduce costs and minimal prospects for revenue growth. Most of the plans terminating Medicare contracts in New England, for example, had been losing money in these markets.¹⁹

A number of plans never gained sufficient market share to be sustainable. Prior to enactment of the BBA, many plans had crowded into high payment areas. In several cases, plans that did not renew had only a few thousand enrollees in a given market.²⁰

In some cases, health plans had negotiated provider contracts that were favorable to the providers to gain entry to the market and build networks. These contracts had left plans with few tools to manage or reduce costs. In California, for example, PacifiCare pulled out of some markets because it was unable to reach agreements with provider groups; this was also true for Allina Health System in Minnesota, where the plan was unable to contract with clinics, forcing it to leave certain markets. Some health plans have attributed provider network problems to the disparity in payments to managed care plans relative to fee-for-service. In testimony before Congress, AAHP President Karen Ignagni said, “Health plans are experiencing difficulties in maintaining adequate provider networks because providers are unwilling to accept lower payments.”²¹

Some analysts have suggested that plans viewed 1998 as a unique window in which they could terminate their old Medicare contracts and preserve their right to apply in the next few years as Medicare+Choice contractors. Once Medicare+Choice is fully in effect, plans that withdraw will face a five-year lock-out from participating in that service area again. Moreover, some have suggested that plans terminated more contracts in 1998 in order to “turn up the heat” on Congress to revisit the payment issue in 1999.

Risk Adjustment Uncertainty

The BBA requires HCFA to implement a risk adjustment method to set payment rates based on the “expected relative health status of each enrollee.” Risk adjustment is intended to ensure that health plans are not penalized for enrolling the chronically ill and do not benefit from favorable risk selection. In its 1996 Annual Report to Congress, PPRC estimated that HMOs received overpayments of about 5 to 6 percent per beneficiary because the populations they enrolled were healthier than the general Medicare population. A General Accounting Office study also found that HMOs tend to attract the least costly enrollees within each health status group.²² Health plans argue that their populations are just as sick but are better managed than those of their fee-for-service counterparts. In theory, risk adjustment would make beneficiaries equally attractive to enroll—regardless of health status—if the methodology works as intended. The BBA requires the system to be in place no later than January 1, 2000.

HCFA released a notice in September 1998 describing the risk adjustment method it intends to implement—a system that would group each Medicare beneficiary into one of several diagnostic cost groups based on demographic status and prior rates of hospitalization, using inpatient hospital encounter data. Payments to Medicare+Choice plans would be adjusted for each Medicare beneficiary, based on whether the individual’s “risk factor” is higher or lower than that of an average beneficiary.²³

This approach has been met with considerable criticism. Health plans argue that it would penalize plans that have effectively reduced inpatient hospital use and focused on providing more care on an outpatient basis. In testimony before Congress last October, Gradison said that

incentives created by risk adjustment methodology based exclusively on inpatient hospital data, even if applied only for a few years, would result in increased hospital use, increased avoidable costs, and a setback in the effort to realize greater efficiency in the health care system.²⁴

Several organizations have urged Congress to delay the implementation of risk adjustment methodology until information from a broader range of health care settings can be gathered and used. MedPAC has recommended that this method of risk adjustment be phased in to temper the effect of risk adjustment on payment rates. Furthermore, MedPAC urged HCFA to move as quickly as feasible to develop the capability to use diagnosis data from all sites of care for risk adjustment.²⁵ HCFA anticipates that the risk adjustment method will be refined as new and additional sources of encounter data become available. The interim final regulations require that all Medicare+Choice organizations provide physician, outpatient hospital, skilled nursing facility, and home health data beginning as early as October 1, 1999. The health insurance industry maintains that these time frames are too short and are unreasonable.

In recent testimony before Congress, HCFA officials said they expect to implement the system on schedule in 2000; recently, however, they have been signaling that they may agree to a phased-in implementation. In the meantime, HCFA has said it will provide each plan with an estimate of risk adjustment’s impact on their payments early in 1999. HCFA estimates that plans on average will see reductions in payments of 7 to 9 percent due to risk adjustment. Plans say the uncertainty surrounding risk adjustment also contributed to their decisions to pull out of certain markets.

Speakers

Robert Berenson, M.D., director of the Center for Health Plans and Providers, will summarize the latest data available on the number, location, and types of plans that have withdrawn or reduced their service areas, as well as information on the status of beneficiary coverage. In addition, he will outline HCFA’s strategy for stabilizing the Medicare+Choice program for the future (including legislative proposals to be included in the president’s budget to be submitted to Congress on February 1). Before joining HCFA, Dr. Berenson served as vice president of the Lewin Group and for ten years as a founder and medical director of the National Capital Preferred Provider Organization. He also practiced medicine for 12 years in a Washington, D.C., group practice.

Kathy Claunch, director of the Senior Health Insurance Program (SHIP) of the Illinois Department of Insurance, will review the reactions of Medicare beneficiaries to the pullouts and to the impact on their Medicare coverage. SHIP trains volunteers to assist Medicare beneficiaries and their caregivers with their questions and problems concerning Medicare, Medicare supplemental policies, long-term care policies, Medicare HMOs and other health insurance issues. Ms. Claunch also serves on the Illinois Medicare Beneficiary Advisory Committee, the Illinois Partners for Medicare Consumers Committee, and the Medicare+Choice work group sponsored by the Chicago Health Care Financing Administration Regional Office.

Sheila Meehan, director of Health Benefits and Life Services for the Bell Atlantic Corporation, will discuss the impact of the withdrawals on her company’s retirees and Bell Atlantic’s outlook on the future of Medicare managed care for corporate retirees. Ms. Meehan is responsible for the strategic planning and development for all of Bell Atlantic’s health and welfare benefits. Additionally, she oversees the fitness and wellness programs and occupational health services.

A panel of senior executives from three health plans from different regions of the country will continue the discussion. These speakers will discuss the market conditions in which their plans operate, the reasons their companies did or did not pull out of certain markets, and suggested policy options to prevent further withdrawals. Tom Anderson is vice president of Medicare Programs for United HealthCare, a large national health plan. Mr. Anderson oversees the company’s efforts in Medicare, which has focused on rapid growth in existing health plans as well as new geographic expansion.

Ellen Offner is vice president for Medicare Programs at Harvard Pilgrim Health Care, a nonprofit health plan located in Massachusetts. Prior to assuming her current position, she was vice president, product development and management, at Harvard Community Health Plan from 1992 to 1995 and held a variety of positions in planning and finance there between 1982 and 1992.

Stephen J. deMontmollin is vice president and general counsel of AvMed Health Plan, Florida’s oldest and largest nonprofit HMO, where he is responsible for corporate legal affairs, regulatory compliance, corporate risk management, public policy, and is the corporate compliance officer.