Committee on Ways and Means, Subcommittee on Health, 3-18-99 Testimony

Statement of James G. Newport, Vice President, Regulatory Affairs
PacifiCare Health Health Systems, on behalf of the American Association of Health Plans

Testimony Before the Subcommittee on Health
of the House Committee on Ways and Means

Hearing on the Medicare+Choice Program

March 18, 1999

I. Introduction

Mr. Chairman and members of the Subcommittee, thank you very much for the opportunity to comment on issues related to implementation of the Medicare+Choice program. I am Janet Newport, Vice President of Regulatory Affairs⁽¹⁾ for PacifiCare Health Systems, based in Santa Ana, California. PacifiCare provides health care coverage for more than 3.7 million enrollees in ten states - Arizona, California, Colorado, Kentucky, Nevada, Ohio, Oklahoma, Oregon, Texas and Washington - and the territory of Guam. Through Secure Horizons, our Medicare plan, we enroll nearly one million Medicare beneficiaries, the largest enrollment nationwide.

I am testifying today on behalf of the American Association of Health Plans (AAHP), which represents more than 1,000 HMOs, PPOs, and similar network health plans. Together, AAHP member plans provide care for more than 140 million Americans nationwide. AAHP appreciates this opportunity to comment on the Health Care Financing Administration's (HCFA) implementation of the Medicare+Choice provisions of the Balanced Budget Act of 1997 (BBA).

When passed by Congress, the goal of the BBA was to expand on the successful Medicare HMO program and to provide for additional private sector options for beneficiaries while allowing the program to expand into other parts of the country. AAHP and its member plans have long supported these efforts to modernize Medicare and give beneficiaries more choice. Today, more than 16 percent - or 6 million beneficiaries are enrolled in health plans, up from 6.2 percent five years ago.⁽²⁾

The transition from Medicare's risk contracting program to the Medicare+Choice program has not been without difficulties. As the Medicare+Choice program has been implemented, the result has been unintended negative consequences for beneficiaries and the health plans that serve them. Last year, health plans holding nearly 100 Medicare contracts reluctantly reduced their service areas or withdrew from the Medicare program largely because of the combination of reduced payments and the significant administrative burden of the many new regulatory requirements under the Medicare+Choice program. These decisions resulted in disruptions in care and a loss of benefits and increased out-of-pocket costs for more than 440,000 Medicare beneficiaries. Of these beneficiaries, 50,000 were left with no choice but to return to the traditional fee-for-service program. HCFA's decision to proceed with phased-in deep cuts under its new risk adjustment methodology will further contribute to the program's instability and penalize beneficiaries who have chosen to join Medicare+Choice plans.

AAHP has consistently supported the goal of ensuring that Medicare payments to health plans are accurate and that they fairly reflect the health care service needs of the Medicare beneficiaries who enroll. However, it is critical to place the program on a stable and predictable footing so that beneficiaries enrolled in the Medicare+Choice program can continue to receive the high quality services, benefits, and lower cost sharing that they have come to rely on. This statement examines three areas of Medicare+Choice implementation: 1) the proposed new risk adjustment method; 2) dissemination of health plan information to seniors; and 3) new plan requirements for quality measurement.

II. Risk Adjustment

The risk adjuster proposed by HCFA threatens the stability of the Medicare+Choice program in two ways: 1) by exacerbating the cumulative impact of payment reductions to Medicare+Choice plans; and 2) by creating unworkable and burdensome administrative processes that increase plan costs and raise the likelihood of inaccurate payment. Taken together, these problems will widen the growing disparity between payment to Medicare+Choice plans and reimbursement under fee-for-service. In addition, these problems will make it difficult for Medicare+Choice plans to operate in certain markets and to maintain the level of benefits and services to which beneficiaries have become accustomed. It is unrealistic for HCFA or Congress to assume that a disparity of this magnitude will have no adverse impact on providers, delivery of services, or health care options for seniors.

On January 15th, HCFA announced its new methodology for adjusting Medicare+Choice payment rates to reflect the health status of individual Medicare beneficiaries. AAHP has serious concerns about the impact of HCFA's new risk adjustment methodology on Medicare beneficiaries, as well as on Medicare+Choice participating organizations and their providers. Under HCFA's new methodology, between 2000 and 2003, Medicare+Choice payments will be based on a blend of rates adjusted for demographic characteristics only and rates adjusted for health status as well as demographic factors. During this period, the portion of the payment rate that reflects health status adjustment will be calculated under a model that uses only inpatient hospital data. The phase-in will be completed in 2004 when Medicare+Choice payments will be risk-adjusted using full encounter data - not just inpatient hospital data.

According to HCFA, fully implementing the PIP-DCG risk adjustment methodology in 2000 - without using any transition period - would have resulted in $15.7 billion in cuts over a five-year period. These reductions would have accounted for 71 percent of the $22 billion savings that Congress anticipated from the Medicare+Choice program enacted in 1997. With the transition period proposed by HCFA, aggregate payments to Medicare+Choice organizations will be cut $11.2 billion over a five-year period. This is an administratively imposed 50 percent increase in the $22 billion savings Congress anticipated from the BBA payment methodology. Congress did not anticipate this level of savings from the risk adjuster when the BBA was enacted. In fact, the Congressional Budget Office did not score the risk adjustment provision in the BBA. As discussed in our recommendations below, AAHP strongly believes the new risk adjustment methodology should be budget neutral to ensure that it does not reduce aggregate funding of the Medicare+Choice program. HCFA projects that, when fully implemented, its proposed risk adjuster will cut payments, on average, by a further 7.0 percent. HCFA's own data show that only 5 to 6 plans⁽³⁾ will receive increased payments as a result of the new risk adjustment methodology.

The BBA required that HCFA have its risk adjustment method evaluated by an outside, independent actuary. In response to this mandate, the American Academy of Actuaries Risk Adjustor Work Group conducted an analysis of HCFA's new health status risk adjustment methodology.⁽⁴⁾ The Academy issued only a "qualified review" of HCFA's methodology, which is significant because it means that the Academy was unable to analyze HCFA's methodology fully due to incomplete available data and information (p.4). In other words, the Academy was unable to replicate HCFA's approach and expressed "serious concerns" about the method's "implementation, operation and impact"(p.3). AAHP believes, and the Academy's report confirms, that the design of HCFA's risk adjustment methodology results in a bias against managed care due to the system's exclusion of one-day hospital stays and its reliance on inpatient-only hospital utilization data. Additionally, according to the Academy report, health plans could be underpaid if claims are "denied due to edits, or get caught in a data processing bottleneck" (p.27).

At PacifiCare, we have experienced problems with HCFA's ability to process our claims in an accurate and timely manner. While PacifiCare, like all plans, has faced challenges, particularly since the system relies on the reporting of retroactive data, HCFA's internal systems issues have been a more significant factor in slowing the establishment of the database necessary for risk adjustment. In addition, HCFA has admitted to errors in their estimates of our plans' average risk adjustment factors and has urged us to rely on our own estimates to determine next year's benefits. In order to prepare our adjusted community rate proposals, through which we determine the benefits we will offer and the premiums we will charge to beneficiaries, we must have accurate estimates of our plans' average risk adjustment factors from HCFA. Without such information, our own estimates may not be accurate and have the potential to translate into decreased benefits for the many beneficiaries that PacifiCare serves. Furthermore, as a requirement of the BBA, we must attest to the accuracy of our adjusted community rate proposal. This will be difficult to do without the assurance from HCFA that the data provided to us and which serves as the foundation for the ACR are accurate.

The initial use of a risk adjuster based solely on hospital inpatient data penalizes health plans that use disease management or other care management programs designed to reduce hospitalizations for chronically ill and other patients who would have otherwise been treated in inpatient settings. These programs are designed to provide superior quality care while preventing costly hospitalizations by treating patients in alternative settings. The Academy of Actuaries identifies "several drawbacks" to a risk adjustment system that uses only inpatient data, noting that "health plans which use outpatient alternatives to hospitalization would be financially penalized" (p.10, p.19). The Academy also notes "many observers recognize that using only inpatient data in the PIP-DCG risk adjustment model may result in a bias toward the FFS system" (p.30). Since the PIP-DCG method is restricted to inpatient claims, beneficiaries enrolled in health plans may appear healthier than they actually are because of the limits of what HCFA's new risk adjustment method can measure.

HCFA's decision not to count diagnoses related to one-day hospital stays in the risk adjustment formula reduces payment for health plan members. We believe that Medicare+Choice organizations have a higher proportion of one-day hospital stays than occurs in the fee-for-service Medicare program and that more of these stays involve diagnoses for which hospitalization is the common course of treatment. In its report, the Academy of Actuaries warns "this limitation could penalize efficient plans enough to make them leave the Medicare market" (p.19). The Academy also notes that excluding one-day stays reduces the predictive power of risk adjustment. The Academy concludes, "the disadvantages of excluding one-day hospitalizations may outweigh any possible gains" (p.13).

The Academy's report questions the administrative feasibility of implementing the new risk adjustment system given the timing and data collection challenges involved, some of which we have identified above. The Academy notes that other programs have had difficulty implementing the PIP-DCG method due to data collection problems and administrative challenges. Many of these programs conducted extensive simulations prior to implementation of risk adjustment including the Health Insurance Plan of California (HIPC), which conducted a one year test prior to the start of the program.

AAHP recommends the following with regard to implementation of the new risk adjustment methodology:

The new risk adjustment methodology should be budget neutral to ensure that it does not reduce aggregate funding of the Medicare+Choice program.
Payments to Medicare+Choice organizations should not be based on a new risk adjustment mechanism until, at the earliest, January 1, 2001.
In the meantime, HCFA should redesign the new risk adjustment method to eliminate serious inaccuracies with the data and the methodology itself.
HCFA should also move forward with the calculations and data collection initiatives that are necessary to support implementation of the model and allow simulations of its impact.

This additional time will provide an opportunity for Medicare+Choice organizations and HCFA to gauge more reliably the potential impact of the PIP-DCG model on health plan payments and the benefits and premiums offered to Medicare beneficiaries. Postponing implementation of the PIP-DCG model will also provide an opportunity to resolve numerous significant outstanding issues with the model and complete the data submission and processing that are necessary to support it.

III. Beneficiary Information and Education

The BBA instructs the Secretary to educate Medicare beneficiaries about their choices using a variety of approaches, including a handbook, toll-free number, an internet website, and community outreach. To finance these activities, the BBA authorizes HCFA to charge each Medicare+Choice organization and Medicare risk contractor a fee equal to the organization's pro rata share of HCFA's estimated costs of enrollment and information dissemination activities. While AAHP supports efforts to enhance informed beneficiary choice, we have significant concerns about the funding, costs and design of the program developed by HCFA.

We urge HCFA to revisit its plans for the 1999 beneficiary education campaign and ensure that it provides beneficiaries with information that will educate, not confuse. HCFA's 1998 beneficiary information and education campaign experienced numerous problems that confused beneficiaries and hindered access to the new Medicare+Choice program.

In Omaha, Nebraska, Baltimore, Maryland and West Virginia, the Spanish language brochures were sent to areas with no Spanish-speaking population.
In Eastern Washington and parts of Florida, the brochures were mailed with a statement that the information presented is incorrect and that the beneficiaries should call a toll-free number if they have any questions.
The toll-free call centers were expected to receive 15,000 calls per week per center. However, during the month of November 1998, the total number of calls received by all centers was only 9,400. Most of the calls regarded HCFA's mistake in sending Spanish language brochures, and requests for additional brochures.

While it is reasonable for health plans and their enrollees to contribute to funding HCFA's enrollment and information dissemination initiatives, their contribution should be in proportion to their participation in the Medicare program. Last year, Medicare risk HMOs and their enrollees represented 14.3 percent of the program but shouldered 100 percent of the cost of the information campaign.⁽⁵⁾ Requiring health plans and their members to bear 100 percent of the burden of this fee directly affects the premiums and benefits that plans can offer to their members. While AAHP supports disseminating information to all beneficiaries to enhance informed choice, we believe that an equitable funding mechanism is critical to the success of this effort. The goal of expanded choice is not served if the costs of underwriting the information campaign reduce the level of benefits that Congress sought to make available to more beneficiaries.

We also are concerned about the costs of the education campaign that HCFA intends to implement. Congress appropriated $95 million for these activities in FY 1998, less than one half of the $200 million allowed by the BBA. In FY 1999, HCFA requested the full $150 million allowed by the BBA, and Congress again approved only $95 million. The President's proposed FY2000 budget requests that Congress appropriate $150 million, $50 million more than the amount allowed by the BBA. In addition, the President's budget calls for a series of new user fees to generate an additional $194.5 million. Of these proposed new user fees, which also would apply to fee-for-service providers, $36.7 million would be levied on Medicare+Choice plans to support HCFA's application and renewal processing activities. At a time of growing instability in the Medicare+Choice program, we are concerned that these user fees set a dangerous precedent and translate into reduced choices for beneficiaries.

AAHP and its member plans will continue to work with HCFA, beneficiary groups and others to develop an education campaign that provides accurate, timely and meaningful information to beneficiaries without compromising the services to which they have become accustomed. The central goal of this initiative, to provide more and better information to beneficiaries about all of the options available to them, is critical to permitting beneficiaries to take advantage of the expanded range of choices envisioned under the new Medicare+Choice program.

IV. Quality Improvement System for Managed Care

One area of significant concern to our member plans is HCFA's Quality Improvement System for Managed Care (QISMC). QISMC is designed to establish a consistent set of quality oversight standards for health plans for use by HCFA and state Medicaid agencies under the Medicare and Medicaid programs, respectively. AAHP has long advocated coordination of quality standards for health plans in order to maximize the value of plan resources dedicated to quality improvement. While we believe that QISMC holds the promise of contributing to this important goal, we have a number of serious concerns regarding implementation. We urge HCFA to engage in intensive dialogue with health plans contracting under the Medicare and Medicaid programs to permit full consideration of their outstanding concerns about the QISMC standards and guidelines. Furthermore, we are also concerned that the Medicare program is not providing equal attention to the overall quality of care furnished under the fee-for-service program.

One of our primary concerns is that QISMC lacks clear coordination with existing public and private sector accreditation and reporting standards. Health plans currently meet voluntary private accreditation standards, such as those developed by the National Committee for Quality Assurance, in order to satisfy requirements of private sector purchasers and some states. Rather than coordinate with these existing standards, QISMC appears to establish an entirely new system of requirements. This adds to administrative cost while actually detracting from health care quality improvement. Additional concerns expressed by AAHP members include:

QISMC fails to establish realistic goals for health plan activities and performance that take into consideration available resources and health plan responsibilities for the delivery of quality care to beneficiaries. The significant additional resources that would be required to meet the new QISMC standards and the prescriptive nature of the QISMC standards would seriously hinder health plan pursuit of quality improvement projects focused on the needs of their enrolled Medicare and Medicaid beneficiaries.
Far more consideration should be given to standards that would be applicable to all types of Medicare+Choice plans. While QISMC requirements are modeled after a health plan using the primary care gatekeeper model, organizations participating in the Medicare+Choice program will have many different structures.
In establishing goals for health plan performance, QISMC lacks recognition of differing characteristics of Medicare and Medicaid beneficiaries and the programs and policies under which they receive health care.
We have urged HCFA to engage in an intensive dialogue with health plans contracting under the Medicare and Medicaid programs to permit full consideration of their serious outstanding concerns about the draft QISMC standards and guidelines. We are eager to work collaboratively with HCFA to develop an approach to future health plan quality oversight activities under Medicare and Medicaid that is responsive to the interests of beneficiaries, health plans, and federal and state responsibilities under the Medicare and Medicaid programs.

As the Committee considers reforms to the Medicare+Choice program, we encourage you to examine the experience of our member plans in implementing quality initiatives such as QISMC. As numerous studies demonstrate, health plans provide high quality care. In recent years, private sector quality initiatives have provided the leadership in this area. We are concerned that government programs are not adequately recognizing these initiatives and coordinating better with them.

V. Conclusion

Health plans have valuable experience to share with Congress and HCFA on implementation of many of the provisions of the Balanced Budget Act of 1997. AAHP appreciates this opportunity to comment on the Medicare+Choice program and its implementation to date. We urge you to undertake mid-course corrections to implementation of the Medicare+Choice program to ensure the program's future stability. We look forward to continuing to work with members of the Subcommittee, other members of Congress, and HCFA to ensure the successful implementation of the Medicare+Choice program.

1. Ms. Newport also serves as a commissioner on the Medicare Payment Advisory Commission.

2. Includes enrollees in risk, cost, and HCPP contractors.

3. Out of 195 plans that submitted substantially complete data for the base period July 1, 1997 to June 30, 1998.

4. American Academy of Actuaries Risk Adjustor Work Group, Actuarial Review of the Health Status Risk Adjustor Methodology, January 14, 1999.

5. Includes enrollees in risk contracts only.