United States General Accounting Office GAO Report to the Ranking Minority Member, Committee on Government Reform and Oversight, House of Representatives October 1997 MEDICAID MANAGED CARE Delays and Difficulties in Implementing California’s New Mandatory Program GAO/HEHS-98-2 United States GAO General Accounting Office Washington, D.C. 20548 Health, Education, and Human Services Division B-276078 October 1, 1997 The Honorable Henry A. Waxman Ranking Minority Member Committee on Government Reform and Oversight House of Representatives Dear Mr. Waxman: California’s Medicaid program, Medi-Cal, served 5.2 million beneficiaries—almost one-seventh of Medicaid beneficiaries nationwide—at a cost of nearly $18 billion in federal, state, and local Medicaid funds in fiscal year 1996. Over the past 2 decades, Medi-Cal has increasingly relied on managed care delivery systems with the aim of improving beneficiary access to quality care while reducing the rate of program cost growth. In 1992, California began planning a major expansion of its Medi-Cal managed care program—one that would eventually require more than 2.2 million beneficiaries in 12 counties to enroll in one of two managed care plans participating in each county. In a 1995 report, we expressed concern about California’s ability to successfully carry out such an expansion because of several weaknesses that we identified in the Medi-Cal managed care program, including the state’s potential inability to effectively monitor its contracts with managed care plans and to ensure that the services that plans were contracted to provide were actually provided.1 Now, nearly 5 years after planning began, the state has repeatedly delayed its completion date for full implementation of the expansion. In light of these delays and the magnitude of the state’s Medicaid program, you asked us to follow up on our earlier report and (1) determine the implementation status of California’s managed care expansion, including identifying the primary causes of delays; (2) assess the degree to which state efforts to educate beneficiaries about their managed care options and enroll them in managed care have encouraged beneficiaries to choose a plan; (3) evaluate the management of the state’s education and enrollment process for the new program, including state and federal oversight of enrollment brokers that the state contracted with to carry out these functions; and (4) make an initial assessment of the impact of the managed 1 Medicaid Managed Care: More Competition and Oversight Would Improve California’s Expansion Plan (GAO/HEHS-95-87, Apr. 28, 1995). Page 1 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 care expansion on current safety-net providers, such as community health centers, that serve low-income beneficiaries. To conduct our work, we interviewed officials from California’s Department of Health Services (DHS); DHS’ former and current enrollment brokers; selected managed care plans and advocacy groups; and the Department of Health and Human Services’ Health Care Financing Administration (HCFA), which oversees the Medicaid program. We also reviewed relevant state statutes and regulations and DHS policies and procedures on the education and enrollment process, as well as the enrollment broker contracts. For more detailed information on our scope and methodology, see the appendix. Despite California’s extensive planning and managed care experience, Results in Brief implementation of its 12-county expansion program is more than 2 years behind its initial schedule and is still incomplete. California originally had planned to implement the program simultaneously in all affected counties by March 1995. However, as a number of unforeseen difficulties arose, such as in contracting with and developing managed care plans, the state began to stagger implementation as it became clear that some counties would be ready before others. Still, as of July 1997, the program had been fully implemented in only seven counties. The most recent schedule estimated complete implementation in all 12 counties by December 1997, at the earliest. The state’s efforts to encourage beneficiaries to choose a health plan have been undermined by problems in the process for educating and enrolling beneficiaries. According to HCFA, beneficiary and provider advocates, and managed care plans, a number of problems contributed to confusion for many beneficiaries, including incorrect or unclear information about the mandatory Medi-Cal program and participating plans as well as erroneous assignments of beneficiaries to plans. Officials from one plan said that beneficiaries did not understand the changes in their health care coverage, and some beneficiaries thought that they were losing Medi-Cal benefits altogether. Available data show that, on average, almost half of affected beneficiaries have not actively chosen their own plan but instead have been automatically assigned to one by the state. Other problems were evident in DHS’ management of the program, such as insufficient performance standards for the enrollment brokers that DHS had contracted with to provide information to beneficiaries about their Page 2 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 managed care options and enroll them in the Medi-Cal program. The enrollment brokers also believed that difficult operating conditions—such as continual changes in state program and policy directives—contributed to the implementation problems. Poor internal communication and weak ties with advocacy and community-based organizations further exacerbated the difficulties DHS encountered in implementing its mandatory managed care program. California has taken a number of actions to improve the implementation and administration of its mandatory expansion program. For example, DHS has begun translating into a number of different languages and redesigning the enrollment materials to make them more comprehensible and has instituted on-site monitoring of the enrollment broker’s processes for enrolling beneficiaries. DHS also has taken steps to work more closely with community-based organizations to improve outreach efforts. However, these actions were taken too late to benefit the many beneficiaries who have already enrolled in the seven counties where full program implementation has been completed. And problems persist—some serious enough to have prompted HCFA to delay full implementation of the program in several counties earlier this year. HCFA is in the process of developing federal guidelines on designing and implementing an education and enrollment program. But these guidelines are not expected to be available before October 1997—too late to help influence design and early implementation issues for California’s program. Despite the fact that the state’s 12-county expansion program was designed to help ensure that federally qualified health centers, community and rural health centers, and other safety-net providers participate in the provider networks, some safety-net providers have reported difficulty maintaining their patient base. Though the new mandatory program provides some assurances that health plans assign beneficiaries to safety-net providers, it does not guarantee these providers any specified level of enrollments. Many beneficiaries who have chosen a primary care physician have opted to select a provider other than a participating safety-net provider. Page 3 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 Medi-Cal was implemented in 1965, the year the Medicaid statute was Background enacted.2 Administered by the California DHS,3 in fiscal year 1996, Medi-Cal provided a wide range of services to approximately 5.2 million low-income individuals at an estimated cost of about $17.7 billion—about 11 percent of national Medicaid expenditures. Medi-Cal managed care, which is composed of several programs, including the 12-county expansion program, is expected to serve over 3 million Medi-Cal beneficiaries once fully implemented. Since 1968, the state has contracted with prepaid health plans (PHP)—California’s equivalent of the federal definition of “health maintenance organizations”—to provide, on a capitated basis, preventive and acute-care Medicaid services, as well as case management. In the 1980s, the state established three additional managed care programs: Primary Care Case Management (PCCM), County Organized Health System (COHS), and Geographic Managed Care (GMC).4 In early 1993, the state completed conceptual development of its most ambitious program to date: the “two-plan model,” which requires more than 2.2 million Medi-Cal beneficiaries to enroll with one of two health plans participating in each of 12 counties.5 2 Established under title XIX of the Social Security Act, Medicaid finances health care for about 37 million low-income families, and aged, blind, and disabled individuals nationwide. Jointly funded by the federal government and the states, Medicaid is administered by states within broad federal guidance. 3 DHS determines policy, establishes fiscal and management controls, contracts with managed care plans, and reviews program activities. 4 PCCMs, operated primarily by physicians and physician groups, contract with the state to provide certain outpatient health care services for a capitated fee. Services not capitated are available to beneficiaries on a fee-for-service basis. COHSs—which operate in San Mateo, Santa Barbara, Solano, Orange, and Santa Cruz counties—are local entities that contract with DHS to administer a capitated, comprehensive, case-managed health care delivery system. Under the GMC model—currently operating in Sacramento County and planned for San Diego County—the state contracts directly with several managed care plans to provide covered services to beneficiaries on a capitated basis. PCCM enrollment is voluntary; COHS enrollment is mandatory for all Medicaid-eligible populations; and GMC enrollment is mandatory for Aid to Families With Dependent Children (AFDC) and AFDC-related beneficiaries. As of April 1997, about 1.2 million beneficiaries were enrolled in the PHP, PCCM, COHS, and GMC programs. 5 In January 1996, HCFA approved California’s request under section 1915(b) of the Social Security Act to waive three sections of the act. Section 1902(a), which requires a Medicaid program to be available throughout the state, was waived, enabling the state to implement the two-plan model in selected counties only. Section 1902(a)(10)(B), comparability of services, was waived, enabling the state to offer additional benefits not available to Medi-Cal beneficiaries not enrolled in the two-plan model. And section 1902(a)(23), freedom of choice, was waived, enabling the state to restrict beneficiary choice of providers under the two-plan model and to require certain beneficiaries to enroll. The 2-year, renewable waiver expires January 22, 1998. Page 4 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 The Two-Plan Model California’s managed care expansion program—often referred to as the two-plan model—was designed to ensure that each of the two managed care plans operating in each county could achieve an enrollment level sufficient to spread risk and that beneficiaries could obtain care from health plans that also served privately insured individuals. In addition, the model was developed to make the most of limited state resources by restricting the number of plans the state would need to monitor. Selection of the 12 counties to use the two-plan model was made on the basis of two criteria.6 First, the counties must have had a minimum of 45,000 Medicaid beneficiaries eligible to participate in managed care,7 and, second, the counties must have had an interest in the program or a significant managed care presence already established in the county. (See table 1 for the number of eligibles and current enrollees by county and plan.) 6 The 12 counties are Alameda, Contra Costa, Fresno, Kern, Los Angeles, Riverside, San Bernardino, San Francisco, San Joaquin, Santa Clara, Stanislaus, and Tulare. 7 Former AFDC and AFDC-related beneficiaries are required to enroll in the two-plan model. Supplemental Security Income (SSI) and SSI-related beneficiaries may enroll in managed care plans on a voluntary basis. California will continue to use this eligibility criteria until the Governor and state legislature agree on an approach to determining eligibility under the new Temporary Assistance for Needy Families (TANF) program, which replaced AFDC. Page 5 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 Table 1: Medi-Cal Eligibles and Enrollees by County and Plan Health plans Eligibles as of April 1997 Enrollees as Dates of initial Two-plan counties Mandatory Nonmandatory of July 1997 Names of plans operation Alameda 116,934 61,105 73,535 Alameda Alliance for Health 1/1/96 25,440 Blue Cross of California 7/1/96 Contra Costa 55,431 27,734 42,706 Contra Costa Health Plan 2/1/97 3,392 Foundation Health Systems 3/1/97 Fresno 145,558 44,352 105,015 Blue Cross of California 11/1/96 17,429 Foundation Health Systems 1/1/97 Kern 102,639 30,606 53,212 Kern Family Health Care 7/1/96 23,195 Blue Cross of California 9/1/96 Los Angeles 1,119,120 435,208 191,964 LA Care 4/1/97 256,812 Foundation Health Systems 7/1/97 Riverside and San 368,588 106,249 130,624 Inland Empire Health Plan 9/1/96 Bernardino N/A Molina Medical Centers Unknown San Francisco 44,155 58,408 23,079 San Francisco Health Plan 1/1/97 15,585 Blue Cross of California 7/1/96 San Joaquin 84,383 29,427 59,199 Health Plan of San Joaquin 2/1/96 11,329 OMNI 2/1/97 Santa Clara 97,815 51,029 42,917 Santa Clara Health 2/1/97 34,466 Authority Blue Cross of California 10/1/96 Stanislaus 63,901 21,410 N/A Blue Cross of California (as local 10/1/97 initiative) 9,145 OMNI 2/1/97 Tulare 71,608 19,945 N/A MediCo 4th quarter 1997 N/A Foundation Health Systems 4th quarter 1997 Total 2,270,132 885,473 1,119,044 Note: N/A = not applicable. In each county, beneficiaries are required to enroll in either the “local initiative”—a publicly sponsored health plan cooperatively developed by local government, clinics, hospitals, and other providers—or the commercial plan, under contract in a beneficiary’s county of residence.8 The local initiative concept was developed to support health care safety-nets—those providers, such as community health centers and federally qualified health centers, that provide health care services to the indigent. Minimum enrollment levels were set for both the commercial and local initiative plans to ensure their financial viability. A maximum enrollment level was also set for each commercial plan to further protect local initiatives and their subcontracted safety-net providers. The state 8 Fresno County did not develop a local initiative, so Fresno has two commercial plans. Page 6 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 contracted with the local initiatives on a sole-source basis, while the commercial plan contracts were awarded on a competitive basis. The situation in Los Angeles County, however, is unique. While California contracted with a local initiative and a commercial plan in Los Angeles County, the county has, in essence, 10 plans because the local initiative plan subcontracted with 7 plans, and the commercial plan subcontracted with 2 plans.9 Beneficiaries can choose a primary care physician from any one of the 10 plans. Health Care Options Medi-Cal beneficiaries required to enroll in the two-plan expansion Program Educates and program are informed about managed care and their choices of health care Enrolls Beneficiaries plans through DHS’ Health Care Options (HCO) program. HCO also enrolls and disenrolls beneficiaries in managed care plans.10 The state contracts with an enrollment broker to conduct HCO program activities. Beneficiaries are informed about the mandatory expansion program and their available choices primarily through an enrollment packet that they receive through the mail. The enrollment packet includes information on managed care, how to join a health plan, available plans and participating providers, phone numbers to call for assistance, and an enrollment form. The packet also includes the first of three standard notices that inform beneficiaries of the 30-day time frame in which they have to choose a plan and the plan to which they will be automatically assigned if they do not return an enrollment form.11 Beneficiaries also can learn about the two-plan model and their plan options at HCO presentations, which are often held daily at county social service offices. At these face-to-face presentations, HCO counselors provide information on managed care, plans available in the county, how to fill out 9 The plan partners in the local initiative—LA Care—are Blue Cross, Care 1st, LA Community Health Plan, Maxicare, United Health Plan, Tower, and Kaiser Foundation Health Plan. The plan partners in the commercial plan—Foundation Health Systems—are Universal Care and Molina Medical Centers. Unlike Foundation Health Systems, LA Care does not directly provide health care services. 10 DHS’ Medi-Cal Managed Care and Payment Systems divisions share responsibility for the HCO program. The Medi-Cal Managed Care Division makes all policy decisions regarding the program, while the Payment Systems Division implements and manages the HCO program and monitors HCO activities, which are contracted to an enrollment broker. The Payment Systems Division assumed this responsibility from the Medi-Cal Managed Care Division in March 1997. 11 The state assigns beneficiaries according to an established methodology, which generally stipulates that once the local initiative reaches a minimum number of enrollments, the state would assign every other beneficiary who did not choose a plan to the commercial plan. Beneficiaries who were already enrolled in one of the plans operating under the two-plan model are not re-assigned by the state. Beneficiaries have the option to change plans at any time. Page 7 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 the enrollment form, beneficiary rights and responsibilities, how to resolve problems with plans, and who to contact for more information. Enrollment materials are available at the presentations. Beneficiaries also can contact HCO’s toll-free call center to obtain enrollment packets and to have enrollment-related questions or concerns addressed. Since 1984, DHS has contracted with an enrollment broker to provide certain education and enrollment services.12 Initially, enrollment broker responsibilities consisted primarily of conducting HCO presentations in selected counties and helping beneficiaries complete enrollment forms. With the expansion of Medi-Cal’s mandatory program, broker responsibilities increased. In addition to distributing enrollment packets and providing HCO presentations, the broker was tasked with processing beneficiary enrollments and disenrollments in 18 counties with managed care and operating a call center to assist beneficiaries. Full implementation of Medi-Cal’s mandatory expansion program is more Implementation of than 2 years behind its initial implementation schedule. Originally, local Expansion Program Is initiatives and commercial plans in each of the 12 affected counties were More Than 2 Years to become simultaneously operational in March 1995. However, repeated delays in the awarding of contracts and the development of plans made it Behind Initial clear that some counties would be ready for implementation before others. Schedule Implementation therefore took place county by county. As of July 1997, plans in 7 of the 12 affected counties had been fully implemented, and full implementation in all counties was scheduled for the end of 1997 at the earliest. Figure 1 shows the 12 counties and their stages of implementation. As of July 1997, over 1.1 million beneficiaries were enrolled in the 12-county expansion program. 12 Between October 1991 and December 1996, Medi-Cal contracted with an Oregon-based enrollment broker, Benova, formerly HealthChoice, Inc. In 1995, the enrollment broker contract was re-bid and awarded to Virginia-based Maximus, which began operations January 1, 1997. Page 8 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 Figure 1: 12 Counties Participating in the Expansion Program and Their Stages of Implementation as of July 1997 Overly optimistic time frames and unanticipated difficulties resulted in a number of delays throughout the state’s planning and awarding of managed care contracts. Developing a Request for Applications for commercial plans and a Detailed Design Application for local initiatives took several months longer than expected. Once applications were submitted, the state did not at first meet its 90-day turnaround goal for approving submissions. Some plans protested the contract awards, further Page 9 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 delaying the contracting process 6 to 8 months. In addition, the state unexpectedly had to obtain—at the request of the developers of the local initiatives—additional state legislative authority, such as exemptions from regulations on public meetings that would enable the local initiatives to hold closed-door sessions to negotiate rates with providers. There also were delays in establishing local initiatives and commercial plans. Some local initiatives took 3 years to develop, instead of the expected 2 years. Unlike commercial plans, local initiatives had to develop health care plans from scratch and, as public entities, they had to interact with community stakeholders. In Fresno County, consensus on whether or not to develop a local initiative could not be reached. As a result, no local initiative was developed, and the state awarded a second commercial contract. The local government in Stanislaus County also had difficulty establishing a local initiative. Consequently, the local initiative contract was awarded to a commercial plan, which will operate in informal partnership with the county. It also took longer than expected for some commercial plans to begin operating under the two-plan model. In addition to obtaining approval of material modifications to their operating licenses, commercial plans had to develop provider networks in counties where the plans were not already operating. Even after implementation of the expansion program began—with Alameda County in January 1996—the state and HCFA took actions that further delayed implementation. For example, DHS delayed full implementation of the program in Fresno, Contra Costa, San Joaquin, and Santa Clara counties to allow the new enrollment broker to fully test its automated systems and capacity to handle all of the enrollment and disenrollment functions. Because of concerns about the education and enrollment process in Santa Clara, San Joaquin, and Los Angeles counties, HCFA temporarily prohibited the automatic assignment of beneficiaries who did not choose a plan and required DHS instead to maintain them in the fee-for-service system. As a result, the pace of enrollment was slowed in these counties, even though plans were allowed to receive voluntary enrollments. As of July 1997, the expansion program had been fully implemented in seven counties—Alameda, Kern, Fresno, San Francisco, Santa Clara, San Joaquin, and Contra Costa—with beneficiaries required to enroll in either the local initiative or the commercial plan. In four of the remaining counties—San Bernardino, Riverside, Stanislaus, and Los Angeles—the program was partially implemented, with only one plan operating in San Page 10 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 Bernardino, Riverside, and Stanislaus counties. Although Los Angeles County had both plans operating, the program was in effect only partially implemented because HCFA had delayed automatic assignment and the state had prohibited additional enrollment in the commercial plan until some remaining contract issues were resolved. In Tulare County, neither plan was operating. The December 1997 target date for full implementation may not be met since some of the plans in counties where the program has yet to be fully implemented have had difficulty developing and complying with regulations. For example, although both plans in Tulare County were tentatively scheduled to become operational by the end of the year, the plans were having difficulty organizing provider networks; implementation target dates have already been moved from spring 1997 to the end of the year. In San Bernardino and Riverside counties, the local initiative began operating in September 1996, but the commercial plan’s operation was delayed because it had not complied with the federal Medicaid requirement that effectively prohibited plan enrollment of Medicaid beneficiaries from reaching 75 percent.13 This requirement was repealed in August 1997; however, because of concerns the state has with other aspects of the plan’s operations, it is still not clear when this plan will begin operating under the two-plan model. Despite California’s efforts to encourage beneficiaries to choose a health Education and plan, many beneficiaries have been assigned to a plan by the state. Enrollment Problems Long-standing problems with California’s HCO program, which provides Contributed to Low beneficiaries with information about their managed care options and enrolls them in a plan, may have contributed to this and to widespread Beneficiary Choice confusion among beneficiaries. While many agree that the HCO program is Rate and Confusion running smoother now than in the past, deficiencies persist—some serious enough to have prompted HCFA to delay full implementation in several counties earlier this year. 13 Specifically, the commercial plan was in violation of Medicaid’s “75/25” restriction, which provides that a plan’s Medicaid (and Medicare, if any) enrollment must be less than 75 percent of its total enrollment. Under its current PHP contract, the commercial plan that serves both San Bernardino and Riverside counties had not complied with the requirement. The Balanced Budget Act of 1997, section 4703, repealed the requirement. Page 11 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 State’s Education Process To encourage Medi-Cal beneficiaries to choose their own managed care Has Not Resulted in plan, California’s HCO program provides them information on managed Beneficiary Selection of care and their available health plan options. Plans, advocates, and researchers agree that beneficiaries who are well informed about managed Plan care—and how it differs from fee-for-service—are more likely to choose a health plan, and those who choose a health plan are more likely to stay with that plan. Experts also believe that well-informed beneficiaries are more likely to use health services appropriately, such as relying more on a primary care physician and less on inappropriate use of emergency room services. Despite its efforts, the state estimated in January 1997 that the majority of enrollments had been the result of automatic assignments by the state. The automatic assignment rate for Alameda County at the beginning of implementation was estimated as high as 80 percent. Although automatic assignment rates have declined—the automatic assignment rate for two-plan counties averaged 45 percent from March to June 1997—the rates ranged widely from county to county. For example, the automatic assignment rate in Contra Costa County in April 1997 was 72 percent, while in Santa Clara County it was 32 percent.14 Unlike other states, California has not established a numeric goal for automatic assignments. Regardless, California’s automatic assignment rates have varied enough across counties to indicate potential problems with HCO’s program. HCFA, advocates, and managed care plans have expressed concerns about the adequacy of the state’s efforts to inform beneficiaries about their Medi-Cal managed care options. According to these groups, information in the enrollment packet was complex, lengthy, and written at too high a grade level.15 In some cases, the information was incorrect. For example, enrollment packets sent to some beneficiaries in San Bernardino and Riverside counties stated that automatic assignments would be made to Molina Medical Centers—a plan not contracted to serve beneficiaries in the expanded program in these counties at that time. Information in the enrollment packets could also be confusing. In anticipation of the Los Angeles County local initiative’s beginning operations in April 1997, thousands of beneficiaries in Los Angeles County received packets with cover letters dated January 8, 1997, that instructed them to respond by January 18, 1997—which did not allow beneficiaries the required 30 days 14 DHS believes that the default rate is high because many beneficiaries do not prefer one plan over the other or because they agree with the assignment that the state intends to make. 15 HCFA first identified problems with the content of the enrollment materials with the implementation of the GMC program in Sacramento in 1994. Page 12 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 to respond. DHS remailed the letters and provided additional time for beneficiaries to respond. And it has only been recently—more than a year after full implementation of the mandatory program in the first county—that many of the enrollment materials have begun to be translated into all of the state’s “threshold” languages.16 Although DHS has established a work group to address problems associated with the enrollment packet, all planned changes are not expected until November 1997, at which time many beneficiaries will have already been enrolled.17 Initially, there also were a number of problems with the toll-free call center, which was set up to provide beneficiaries access to additional information about how health plans operate and how to use them. The call center, however, often was a source of frustration and confusion because callers could not get through, messages went unanswered, voicemail boxes were full, or counselors provided incorrect information. However, a review of HCO’s recently instituted “problem log” revealed that the problems have largely disappeared under the current enrollment broker, Maximus, which expanded the call-center operation. There also have been problems with the HCO presentations. Through county-by-county preimplementation reviews, HCFA often found that the presentations were confusing, not conducted in the appropriate language, not accurate or performed as scripted or scheduled, or not sufficiently informative. In addition, beneficiary attendance has been low. State officials recognize that the limited number of presentation sites may make it difficult for beneficiaries to attend. For example, in June 1997, Los Angeles County—which comprises 88 cities and 136 unincorporated areas and covers over 4,000 square miles—had 35 presentation sites. Officials from one managed care plan we contacted believed that poor attendance at the HCO presentations was due in part to limitations in the state’s outreach to beneficiaries. The officials believed that by working closely with community-based organizations that beneficiaries know and trust, such as churches and legal aid services, more beneficiaries could be reached; in addition, these organizations could provide outreach services 16 The state requires the enrollment broker to provide linguistically appropriate services to a population group of mandatory Medi-Cal eligibles residing in a proposed service area whose primary language is not English if these eligibles meet a specific numeric “threshold” in a proposed service area. For example, in Alameda County, the first county in which the program was implemented, the number of eligibles whose primary language was not English exceeded the threshold for Spanish, Cantonese, Vietnamese, and Farsi. The state’s threshold languages are Cambodian, Cantonese, Farsi, Hmong, Lao, Russian, Spanish, Vietnamese, and Armenian. 17 A number of changes have already been completed, such as translation of some of the enrollment materials, including the enrollment exemption form and the list of important telephone numbers. Page 13 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 and thereby supplement HCO presentations. HCFA, advocates, and managed care plans have long called for increased outreach efforts—not only to beneficiaries, who can be difficult to reach, but to providers and others in the community as well. Some plans and advocates have, at their own expense, conducted outreach activities to fill the perceived gap in the state’s efforts. Yet even with high automatic assignment rates and poor attendance at the HCO presentations, it was not until October 1996 that DHS began development of an outreach campaign that was implemented in selected counties in March 1997. The campaign consisted of bus billboards and posters sent to HCO presentation sites, managed care plans, and community-based organizations. Brochures, a video, and radio announcements were also recently added. DHS has recently begun to explore additional ways to improve outreach and involve community-based organizations in HCO activities, such as participating in DHS-sponsored work groups. DHS asked community-based organizations to identify additional HCO presentation sites in Los Angeles County and plans to require Maximus to contract with a number of community-based organizations to provide HCO presentations to their clients. Recognizing that provider education could also be improved, DHS has begun to better disseminate information to participating providers on managed care programs, such as DHS provider bulletins that give HCO program updates. In addition, DHS created the HCO Education and Outreach Unit in June 1997 to develop and implement strategies to ensure beneficiaries, providers, legislators, advocates, and other interested parties are well informed and educated about the expansion program. Enrollment Processing Some of the problems with enrolling beneficiaries persisted throughout Improved, but Problems the state’s first year of implementation of its new mandatory program and Still Persist were exacerbated by the timing of the changeover between enrollment brokers. While many agree that enrollment processing is functioning much smoother now, there was enough lingering concern to have prompted HCFA to slow the pace of enrollment in several counties earlier this year. During the first year of implementation, the volume of enrollments may have overwhelmed Benova, the former enrollment broker. Enrollment materials were not always sent on time, and, in one county, it could not be determined whether they were sent at all. Enrollment data were not accurately or completely entered into the enrollment information system, Page 14 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 and some beneficiaries were enrolled in a plan other than the one they chose or were assigned to a plan that was not an option for them. State assignments of beneficiaries who did not choose a plan were not always timely, which meant that plans lost capitation revenue. The situation worsened when Benova lost its bid for the enrollment broker contract and began losing significant numbers of staff. HCFA and managed care plans agree that Medi-Cal’s enrollment process has begun to function more smoothly. Maximus has more resources to process and track enrollments, and the state has begun to implement long-needed fixes, such as improved monitoring of the enrollment broker. However, problems have continued to occur. For example, in April 1997, thousands of beneficiaries in Riverside County were sent letters with dates that implied beneficiaries had already been assigned to a plan. The state remailed the letters with corrected dates. Because of continuing concerns, HCFA slowed enrollment in several counties earlier this year. According to HCFA, it would not approve the February 1997 full implementation in Santa Clara and San Joaquin counties because it had found, during its preimplementation reviews, deficiencies in the education process that “grossly violated” the HCO process and the conditions of California’s waiver. For example, enrollment packets sent to beneficiaries were incomplete, and the state could not verify whether a subsequent mailing was sent. At the end of March 1997, HCFA decided to slow enrollment in Los Angeles County, prior to full implementation. HCFA took this action, in part, because the enrollment broker had not yet demonstrated an ability to send timely or accurate mailings to beneficiaries or to properly train HCO counselors to make accurate and informative presentations to beneficiaries. Adequately educating beneficiaries in Los Angeles about their plan options is especially difficult, since there are multiple plans from which beneficiaries can choose. Furthermore, with over 1 million beneficiaries who will be mandatorily enrolled, and another 400,000 voluntarily eligible, the consequences of enrollment errors in Los Angeles County could be significant. Potential Impact of Based on anecdotal evidence from HCFA, advocates, and managed care Education and Enrollment plans, the problems with the education and enrollment processes Problems on Beneficiaries throughout the implementation of the two-plan model have affected beneficiaries and plans alike. Officials from one plan said that and Plans Page 15 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 beneficiaries were not only confused but concerned because they did not understand what was happening to their health care coverage—some beneficiaries thought they were losing Medi-Cal benefits altogether. According to some plans, enrollment problems have resulted in significant financial loss due to lost capitation revenue and unanticipated operating and administrative costs. For example, if enrollment was delayed, some plans not only lost revenue but may have unnecessarily expended funds for staffing, facilities, and advertising. Officials at one local initiative claimed gross revenue losses of almost $2 million due to a 25-day delay in the mailing of enrollment materials. The lost capitation revenue required the plan to draw upon an existing line of credit—with interest—from the county. Because of long-standing problems and concerns over the implementation of the two-plan model, some groups wanted implementation either stopped or further delayed. Yet, some plans urged the state and HCFA not to delay implementation and enrollment further because of the financial repercussions. HCFA officials agreed that long delays in implementation could present financial hardship for some plans. Over the past several years, California has been criticized for a number of Weaknesses in State weaknesses in the management of its Medi-Cal managed care program. In Management of the a 1993 report, HCFA questioned whether DHS, with its existing staffing and HCO Program processes, could effectively monitor the state’s contracts with Medi-Cal managed care plans.18 Two years later, we echoed similar concerns. In Contributed to 1994, HCFA also cited a number of weaknesses in the implementation of Implementation Sacramento’s GMC program, including the need for early and ongoing local input into the planning process and deficiencies in the education and Difficulties enrollment process.19 More recently, Mathematica Policy Research, Inc., in its 1996 report on Medi-Cal managed care, cited limited time and resources 18 HCFA region IX, “Review of California’s Administration of Its Managed Care Program” (internal document, fiscal year 1993). 19 According to HCFA, there were a number of lessons learned from the GMC implementation experience, including the critical need for a well-informed provider and beneficiary population and the importance of an effective monitoring system, such as key performance indicators like disenrollment rates. Page 16 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 as the cause of initial enrollment problems experienced by beneficiaries in Sacramento’s GMC program.20 These and other management weaknesses—such as insufficient contract performance requirements for enrollment brokers, inadequate monitoring of the HCO program, and poor communication with and involvement of outside groups—contributed to the problems the state encountered in implementing its two-plan model.21 Benova and Maximus also cited reasons that made it difficult for them to perform as efficiently as possible. The state has taken a number of long-needed actions aimed at improving various aspects of the HCO program. However, the effect of some of these actions remains to be seen. Federal guidance on designing and implementing a mandatory managed care program, especially when education and enrollment functions are contracted to an enrollment broker, may have assisted the state in improving its program implementation in its earlier stages. Although HCFA is currently developing such guidance, HCFA’s oversight of California’s program has consisted primarily of approving the waiver application and conducting preimplementation reviews of each county prior to full implementation.22 Contracts Insufficient to DHS’ contract with Benova, the former enrollment broker, contained no Hold Enrollment Brokers specific performance standards. Performance standards should make clear Accountable the level of service expected of the broker and enable a state to gauge the sufficiency of the broker’s operations. When tied to payment, performance 20 Specifically, the Mathematica report said the initial enrollment process was “chaotic,” partially due to enrollment materials that were incomplete, confusing, and sometimes misleading and a call center that was overwhelmed with the volume of calls. Because of the magnitude and frequency of problems, the state provided beneficiaries additional time in which to choose. Mathematica Policy Research, Inc., Managed Care and Low-Income Populations: A Case Study of Managed Care in California (Washington, D.C.: Mathematica Policy Research, Inc., May 1996). 21 According to DHS, understaffing has also plagued the program. For fiscal year 1996, the Medi-Cal Managed Care Division requested an additional 126 staff to operate its managed care program. However, the state legislature approved somewhat less than two-thirds of these positions. The Payment Systems Division is currently seeking an additional 26 staff positions for the HCO program. Officials say that they need to request more staffing but have been unable to devote the resources needed to prepare the justification. 22 HCFA’s approval of California’s waiver was contingent upon several factors, including agreement that full implementation of the two-plan model would not commence in a county until HCFA had conducted a satisfactory on-site, preimplementation review that focused on policies and procedures regarding enrollment, beneficiary access, quality of care, and plans’ financial solvency. The waiver also included a requirement that the state demonstrate that it had allocated sufficient and appropriate staff to all areas of responsibility, particularly with regard to setting up and monitoring such a large and complex program. Page 17 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 standards can provide incentives for the enrollment broker to provide the services required and penalties for nonperformance. DHS’ contract with Maximus, the current enrollment broker, contained several performance standards; however, few were tied to payment. For example, although call-center staff were required to answer phones within three rings and process enrollment forms within 2 days, there was no penalty for noncompliance. More importantly, no performance standards that were tied to payment related to potential quality indicators, such as the rate of automatic assignment, beneficiary satisfaction with the education and enrollment process, or the rate of beneficiary disenrollment.23 California is planning to amend Maximus’ contract to include additional performance standards and to increase the number of standards that are tied to payment, which should help strengthen the contract and make it more enforceable. HCO Program Poorly According to HCFA, many of the problems with the state’s process for Monitored educating and enrolling beneficiaries were the result of inadequate monitoring of the HCO program. Until recently, DHS did not conduct on-site monitoring of enrollment broker activities nor did it have staff with the expertise to monitor the broker’s automated systems. In addition, HCO’s management information and reports were not adequate to effectively monitor the program. According to DHS, regular, on-site monitoring of Benova was difficult since Benova’s operations were about 80 miles from DHS headquarters in Sacramento. Without on-site monitoring, however, DHS could not guarantee that critical broker responsibilities, such as the mailing of enrollment packets, were carried out. For example, it was not until enrollment broker operations were transitioning to Maximus that DHS found that thousands of beneficiary enrollment packets had not been sent from a Benova mail facility. To help ensure this does not recur, as a condition of its contract, Maximus operations are located in or near Sacramento. DHS also has dedicated five full-time Payment Systems Division staff, four of whom have automated systems expertise, to conduct on-site monitoring at Maximus’ various locations. To help ensure Maximus complies with the terms of its contract, DHS staff observe the broker’s operations and test the automated systems. Staff also observe mail facility 23 DHS does not believe that it would be fair to tie performance standards on these indicators to payment because the enrollment broker contract does not provide the broker with much flexibility in how to conduct the HCO program. Page 18 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 operations to ensure the timeliness, completeness, and accuracy of the enrollment materials mailed to beneficiaries. Until recently, HCO program staff did not have the expertise to evaluate automated systems operations and ensure that their outputs were valid. Without such expertise, the state could not determine if beneficiaries had been assigned to plans as intended. Moving day-to-day HCO program operations from the Medi-Cal Managed Care Division to the Payment Systems Division provided the program with the expertise required to make such determinations. In addition, in March 1997, DHS contracted with a systems consultant, Logicon, to test Maximus’ automated systems and validate its output by July 1997. According to a DHS official, the testing and validation process will allow DHS to better understand the enrollment broker’s system and thus have greater confidence in its output. Validating system output will likely enhance the reliability of the information that the system generates, such as enrollment and disenrollment data. As of the end of August 1997, however, Logicon had yet to complete its contract. As a result, according to HCFA, there remains no external verification that the enrollment broker can effectively handle the increased volumes of enrollment that will result when plans in the remaining counties, like Los Angeles, become fully implemented. Management information and reporting also were not sufficient to effectively monitor the HCO program. According to one DHS official, HCO reports were not managerially useful. For example, while data were provided on the number of beneficiaries who chose a plan, the number who were automatically assigned to a plan,24 and the number who disenrolled from a plan, the reports did not include trend analyses. And while an automatic assignment rate was calculated, a disenrollment rate was not, which can serve as an important indicator of beneficiary satisfaction with plans.25 In addition, certain key terms, such as “disenrollment,” have yet to be defined, and the data have yet to be verified, which provides little confidence in its meaning or accuracy.26 As part of its contract, Logicon is required to ensure that numbers across 24 Prior to January 1997, DHS did not publish data on automatic assignments. 25 While DHS requires plans to conduct annual enrollee satisfaction surveys, there is no requirement to distinguish between beneficiaries who chose the plan and those who were automatically assigned. 26 For example, disenrollment can be involuntary due to loss of Medicaid eligibility. Voluntary disenrollments can be due to moving outside the plan service area or dissatisfaction with the plan or provider services. Analyzing reasons for disenrollment can provide valuable information about a plan’s performance. Page 19 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 reports are consistent and reconcilable and to identify reports that are needed for the state to effectively monitor enrollment broker activities. Finally, DHS initially had no system to determine whether problems reported to DHS were recorded or addressed. Although DHS began keeping an HCO “problem log” in January 1997 to capture and track the status of problems and complaints reported to either DHS, the enrollment broker, or the Medi-Cal managed care ombudsman,27 DHS had not summarized or systematically analyzed the information collected at the time of our review. Insufficient HCFA, managed care plans, and advocates have long expressed concern Communication and over a lack of effective state internal communication and timely Involvement of Outside communication with and involvement of outside groups in planning and decision-making. We found, for example, that until recently, HCO policy Groups decisions often were not officially documented or disseminated to the appropriate state staff. DHS has taken some steps to improve its internal communications, such as requiring HCO’s policy unit to provide written documentation of all HCO policy decisions to the chief of the Headquarters Management Branch, Payment Systems Division, for review and systematic dissemination. DHS has also increased its communication efforts with outside groups. To provide a forum to discuss and address issues and concerns, the state has convened or participates in several work groups. For example, the Policy Workgroup was formed in January 1997 to improve the education and enrollment process, such as by redesigning and translating the enrollment materials. The group includes representatives from DHS, HCFA, health plans, advocacy groups, and Maximus. The state also convened in June 1997 a Stakeholder Advisory Group to provide policy advice on and oversight of program implementation in Los Angeles County. The group is composed of advocates, provider representatives, DHS, Maximus, and the Los Angeles commercial plan and local initiative. It plans to meet monthly. 27 The Office of the Ombudsman began operating July 1996. Its purpose, in part, is to investigate and resolve complaints about Medi-Cal managed care and to provide information to and assist Medi-Cal beneficiaries by mediating on their behalf and verifying the resolution of complaints. Page 20 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 Enrollment Brokers Cite Benova and Maximus, the two enrollment brokers DHS has contracted Operating Conditions That with, also cited a number of factors that they believed adversely affected Affected Their their performance. According to these brokers, DHS made frequent policy and program changes and often provided little lead time to appropriately Performance implement these changes. According to Maximus, during the first 7 weeks of its contract period—which began January 1997—DHS made about 300 policy changes, sometimes giving Maximus little time to implement them. To comply with DHS’ time frames, Maximus believed it necessary to sometimes bypass quality assurance measures that it had established to ensure that such system changes did not have unintended consequences. In one instance, changes made to the mailing dates in one county caused Maximus to inadvertently halt mailings to another county. Benova believed that its performance as Medi-Cal’s enrollment broker suffered because of DHS’ often-changing directions and its lack of responsiveness. For example, DHS denied Benova’s request to transfer calls during peak times to call centers in other states—an arrangement Benova believed would have improved service. According to Benova, DHS also denied its request for cost-reimbursement for additional equipment needed to handle increasing volumes of enrollment. Benova and Maximus officials also stated that, relative to their experience with other states, California limited their contact with plans, advocacy groups, and community-based organizations. DHS was concerned about remaining informed about program operations and not burdening limited contractor staff with additional responsibilities. DHS recently has relaxed its policy and begun to allow the enrollment broker to participate in community meetings. Limited Federal Guidance HCFA’s oversight of California’s education and enrollment functions has on Education and consisted primarily of reviewing and approving the state’s waiver Enrollment Functions application to implement its mandatory managed care program and conducting preimplementation reviews in each county. As of August 1997, few federal guidelines existed for states to use for their process of educating Medicaid beneficiaries and enrolling them in mandatory managed care programs—two relatively new functions for states.28 In addition, guidelines did not exist for contracting out these functions. With 28 HCFA has issued guidelines to assist states in developing Medicaid managed care marketing standards, which could be applied broadly to the education process. Specifically, federal regulations require that states’ contracts with health plans specify the methods by which the plans will ensure that marketing plans, procedures, and materials are accurate and do not mislead, confuse, or defraud beneficiaries or the state. Page 21 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 such guidance, some of the problems that California experienced in expanding its Medi-Cal managed care program might have been avoided. HCFA is in the process of developing guidelines to assist states with designing and implementing an effective education and enrollment program, including contracting with enrollment brokers—an increasing trend. Earliest issuance of these guidelines was projected for October 1997.29 An expressed objective of the two-plan model was to protect existing Some Safety-Net health care safety nets in the new competitive environment of managed Providers Are care. Safety-net providers—such as federally qualified health centers, and Encountering community and rural health centers—provide health care services to the medically indigent. However, while the two-plan model provides some Difficulties assurances that plans will assign beneficiaries to safety-net providers, it does not guarantee that these providers will receive a specified level of enrollment, nor can it guarantee that they will maintain their enrollments. Some providers have reported that they are having difficulty operating under the two-plan model, especially in maintaining their former patient base. The two-plan model has several provisions and incentives aimed at protecting safety-net providers. The model’s local initiative arrangement enables counties to develop a plan that reflects local needs and priorities and includes county-operated health facilities. Once developed, the local initiative must contract with any safety-net provider that complies with the local initiative’s specific requirements and standards and accepts the rates offered. Although commercial plans are not required to contract with safety-net providers, they were awarded extra points during the evaluation process for the extent to which their networks included safety-net providers. The model also requires that automatic assignments be made to the local initiative until preestablished minimum enrollment levels are reached. In addition, the local initiatives and commercial plans are required to ensure—to the maximum extent possible—that existing patient-physician relationships are maintained. Furthermore, the local initiative must develop a process that “equitably assigns” to safety-net providers those beneficiaries who do not choose a primary care provider; similarly, the commercial plan must develop a process that 29 HCFA developed the guidelines with input from selected states; an expert researcher; a review of reports from the National Academy for State Health Policy and GAO; review of states’ requests for proposals and contracts; and information from advocacy groups, trade organizations, and the managed care industry. Page 22 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 “proportionately” assigns such beneficiaries.30 According to DHS, it did not require plans to assign a specific number of beneficiaries to safety-net providers because federal law requires states to ensure that beneficiaries have a choice of providers. Despite these protections, an initial assessment of the two-plan model’s impact on safety-net providers suggests that some are experiencing difficulties, especially in maintaining their levels of enrollment. According to the state and HCFA, a couple of factors have affected safety-net providers’ enrollment bases. Beneficiaries in managed care are required to designate only one provider as their primary care physician, although they may have visited more than one provider in fee-for-service care. Consequently, some safety-net providers say that they have seen fewer beneficiaries under the two-plan model. However, many beneficiaries who choose a provider are not choosing safety-net providers, and many who are assigned to these providers disenroll. HCFA has reported that in Los Angeles County, 12,600 beneficiaries—or 70 percent—who had been assigned to a safety-net provider chose to disenroll within 5 days. The two-plan model does not prescribe, other than in general terms, how plans are to assign beneficiaries to individual providers. However, a number of plans favor safety-net providers in their assignment methodology. One plan had designed a four-tier assignment methodology that gives priority to contracted safety-net providers and other providers that have at least a 50-percent Medi-Cal enrollment base. Another plan seeks to maintain a 60/40 assignment ratio, with approximately 60 percent of beneficiaries assigned to private providers and the remaining 40 percent assigned to county and community clinics.31 State Assessing Safety-Net The state has begun to assess measures that could be taken to assist Issue and Taking Some safety-net providers and has taken action in one county. To reduce the Steps to Assist Providers number of beneficiaries assigned by plans away from their safety-net providers, the state planned to provide information on beneficiaries’ last provider of record to plans beginning August 1997. With this information, plans could assign the beneficiary to that provider if the provider was part of the plan’s network. 30 “Proportionately” means that the number of enrollees assigned should approximate the proportion that the providers represent in the network. For example, if the safety-net providers represent 20 percent of a plan’s network, they should receive approximately 20 percent of beneficiaries. 31 According to a plan representative, the actual assignment ratio is closer to 70/30, since an average of 80 percent choose their provider, leaving few beneficiaries for the plan to assign—only 1,500 over the last 5 months. Page 23 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 Safety-net providers in Fresno County were particularly concerned about their viability since the county’s two-plan model did not include a local initiative. An agreement was reached between the state, providers, and the two commercial plans that addressed some of the short- and long-term concerns of these safety-net providers. For example, the two plans agreed to assign all state-assigned beneficiaries who had not designated a primary care physician to a safety-net provider. Over the longer term, a special team composed of state, plan, and provider representatives will be established to oversee the implementation of managed care in Fresno County. California’s expansion of its Medi-Cal managed care program is currently Conclusions the largest effort of its kind in the nation in terms of the number of beneficiaries involved. Although California invested nearly 5 years in both conceptual and implementation planning of its two-plan mandatory program, implementation has not been smooth. Many of the circumstances that contributed to implementation problems were within the state’s control, while others were not. For example, the timing of the transition from one enrollment broker to another undoubtedly contributed to the implementation delays and difficulties. Had the transition not occurred in the midst of the two-plan implementation in several counties, some problems might have been less severe. Many of the problems that occurred in implementing the new mandatory program were foreshadowed by the state’s earlier efforts to implement managed care. These earlier problems—documented in prior evaluations by other organizations—should have convinced the state that many of its policies and procedures needed retooling. The state is now taking certain actions to improve the program, but many are too late to benefit those beneficiaries already enrolled in the seven counties where implementation has been completed. HCFA’s preimplementation reviews enabled HCFA to identify problem areas in California’s implementation of its two-plan model; the reviews did not, however, always result in immediate improvements. At the same time that DHS was attempting to address these problems, managed care plans were exerting pressure to push ahead with program implementation since their large investments—and financial viability—were dependent on receiving enrollments and associated revenues according to set time frames. As a result, while HCFA identified the need for significant improvements, it did not halt program implementation to effect such changes. HCFA also did not Page 24 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 have sufficient written guidance in place to assist the state in developing and implementing its program. Despite these delays and difficulties, California’s experience can be instructive for other states as they develop, expand, or adapt their mandatory Medicaid managed care programs. Specifically, California’s experience points to several potential lessons learned: • Incremental implementation allows for adjustments and improvement. Simultaneous or quick-succession implementation in multiple areas does not give sufficient time for program modifications when unforeseen problems arise. • Sufficient staff—including individuals who have expertise in managed care program design and implementation—are needed to conduct program activities. Of particular importance are systems analysts and contract specialists. • Stakeholder and community input and involvement, sought early and often, can contribute significantly to effective education and enrollment processes and problem resolution. • Effective monitoring systems, including adequate management information and reporting, can ensure accountability for program operations—especially if there is heavy reliance on a contractor for integral parts of the program. Including performance standards for key areas of operation in enrollment broker contracts and tying these standards directly to broker payment might help to ensure maximum contractor performance. To help states design and implement Medicaid managed care programs Recommendation that ensure beneficiaries who enroll—especially those who are mandated to do so—are able to make an informed choice in selecting a plan, we recommend that the Secretary of Health and Human Services direct HCFA to promptly finalize guidelines for developing and operating an education and enrollment program. To help ensure accountability, these guidelines should include considerations regarding appropriate performance standards and measures and monitoring mechanisms, especially when a state contracts out these functions to an enrollment broker. We provided a draft of this report to the Administrator, HCFA; Director, Agency Comments California DHS; and officials of Benova and Maximus, the former and and Our Evaluation Page 25 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 current enrollment brokers. Each entity provided technical or clarifying comments, which we incorporated as appropriate. HCFA concurred with our recommendation and stated it is working to finalize its education and enrollment guidelines. For example, it sponsored a joint industry and Medicaid managed care meeting in September to discuss the draft guidelines. HCFA did not, however, indicate a target date for finalizing the guidelines. HCFA’s Administrator stated that, because the guidelines are not requirements, it is important to take the necessary time to reach consensus on them in order to obtain necessary buy-in and endorsement from those affected in order to give the guidelines credibility and acceptability. DHS agreed with our conclusions and recommendation, saying that the state has already adopted or is working toward implementing the lessons learned that were outlined in the conclusions. It acknowledged that there have been problems associated with California’s transition to managed care for its Medi-Cal population and emphasized its efforts to address these problems in partnership with HCFA, plan partners, medical providers, and advocacy groups; however, the state was concerned that the report did not sufficiently acknowledge its efforts in this regard. DHS provided to us additional information on its efforts to be responsive to identified problems, which we incorporated where appropriate. In terms of the evidence and findings presented in the report, DHS questioned the objectivity of information obtained from some sources, such as some contracted health plans and the former enrollment broker, with whom the state is involved in formal contract disputes or litigation. Being aware of these ongoing disputes and litigation during the course of our work, we were sensitive to the use of information obtained from all affected parties. In this regard, we either corroborated the testimonial evidence we obtained with independent sources or clearly attributed the information to its source in the report. Both Benova and Maximus generally concurred with our findings. Benova provided additional information on several findings in order to more fully explain its relationship with the state and the resulting impact on Benova’s performance. For example, Benova contends that its contract was not adequately funded to fulfill the enrollment contract functions. We chose, however, not to include these additional details because of ongoing litigation between the two parties. Maximus generally agreed with our assessment of the program and implementation issues. Despite the difficulties cited in the report, Maximus believed that it has gained sound Page 26 GAO/HEHS-98-2 California’s Managed Care Expansion B-276078 administrative control of the basic enrollment processes, such as the call center operations, the enrollment process, and the computer system operations. While Maximus endorsed holding all program participants accountable, it emphasized that establishing standards for functions that are not entirely within its control can be problematic—especially when these functions are tied to payment. Maximus added that the California experience has served as an important learning opportunity in its role as enrollment broker in other states. As arranged with your office, unless you announce its contents earlier, we plan no further distribution of this report until 30 days after its issuance date. At that time, we will send copies to the Secretary of Health and Human Services; the Administrator, HCFA; the Director, California DHS; and interested congressional committees. Copies of this report will also be made available to others upon request. If you or your staff have any questions about the information in this report, please call me or Kathryn G. Allen, Acting Associate Director, at (202) 512-7114. Other contributors were Aleta Hancock, Carla Brown, and Karen Sloan. Sincerely yours, William J. Scanlon Director, Health Financing and Systems Issues Page 27 GAO/HEHS-98-2 California’s Managed Care Expansion Contents Letter 1 Appendix 30 Scope and Methodology Table Table 1: Medi-Cal Eligibles and Enrollees by County and Plan 6 Figure Figure 1: 12 Counties Participating in the Expansion Program and 9 Their Stages of Implementation as of July 1997 Abbreviations AFDC Aid to Families With Dependent Children COHS County Organized Health System DHS Department of Health Services GMC Geographic Managed Care HCFA Health Care Financing Administration HCO Health Care Options PCCM Primary Care Case Management PHP prepaid health plan SSI Supplemental Security Income TANF Temporary Assistance for Needy Families Page 28 GAO/HEHS-98-2 California’s Managed Care Expansion Page 29 GAO/HEHS-98-2 California’s Managed Care Expansion Appendix Scope and Methodology To determine the status of California’s expansion of its Medi-Cal managed care program and identify potential reasons for delays in implementing the two-plan model, we interviewed officials from the California Department of Health Services (DHS) and reviewed their implementation schedules—the initial schedule and subsequent updates—for the two-plan model. We also interviewed Medicaid officials in HCFA’s region IX office in San Francisco and examined their preimplementation reviews, which are conducted in each affected county to determine the state’s readiness to implement the two-plan model in that county. To identify the state’s efforts to educate Medi-Cal beneficiaries about managed care and enroll them into one of the state-contracted plans, and to evaluate its management of the education and enrollment process, we interviewed DHS and HCFA region IX officials and obtained and reviewed relevant state law, regulations, policies, and procedures; the state’s strategic plan for expanding its Medi-Cal managed care program; the state’s two-plan model waiver application submitted to HCFA; Health Care Options (HCO) program documents, including enrollment materials; minutes from DHS’ Policy and Transition Workgroup meetings; HCO’s problem log; enrollment broker contracts and the 1995 Request for Proposal; HCO management reports, including monthly enrollment summaries; and HCFA’s preimplementation reviews. We also interviewed officials from two commercial and four local-initiative health plans that served 11 of the 12 two-plan counties; Benova, Medi-Cal’s previous enrollment broker, and Maximus, its current enrollment broker; and advocacy and consumer groups. We reviewed documents obtained from these officials, including minutes from the California Alliance of Local Health Plan Enrollment Workgroup meetings and written testimony of some stakeholders on the implementation of the two-plan model provided in February 1997 before the California state legislature. We also reviewed reports by Mathematica Policy Research, Inc., and the Medi-Cal Community Assistance Project that discussed issues and concerns about DHS’ expanded program. To evaluate the state and federal oversight of California’s enrollment broker, we obtained and analyzed California’s past and current enrollment brokers’ contracts and amendments. To obtain detailed information on specific DHS activities to monitor enrollment broker performance, we interviewed DHS and HCFA region IX officials. We also visited Maximus’ administrative office, which houses its systems operations and call center, and one of the subcontracted mail facilities to observe broker operations. At these facilities, we met with DHS and Maximus officials to discuss Page 30 GAO/HEHS-98-2 California’s Managed Care Expansion Appendix Scope and Methodology oversight activities and broker operations. We also reviewed program information generated by Maximus. To identify federal monitoring of contracted enrollment broker functions and guidance for states to use in monitoring contracted enrollment broker activities, we met with officials in HCFA’s Baltimore Office of Managed Care and region IX Medicaid officials. In addition to reviewing HCFA’s guidelines for state compliance with federal regulations on Medicaid managed care marketing, we obtained and reviewed HCFA’s “Managed Care Pre-Implementation Review Guide” and its draft guidelines to states for enrolling beneficiaries in managed care programs. To make an initial assessment of the two-plan model’s impact on safety-net providers, we interviewed officials from DHS, HCFA, and two commercial and two local initiative plans. We also reviewed the state’s strategic plan, which discusses how safety-net providers would be included under the two-plan model; state requirements for assigning beneficiaries to plans; and selected plan assignment methodologies. In addition, we reviewed reports by the Medi-Cal Community Assistance Project and Mathematica, which examined the experiences of some safety-net providers. We performed our work between January and August 1997 in accordance with generally accepted government auditing standards. (101533) Page 31 GAO/HEHS-98-2 California’s Managed Care Expansion Ordering Information The first copy of each GAO report and testimony is free. Additional copies are $2 each. Orders should be sent to the following address, accompanied by a check or money order made out to the Superintendent of Documents, when necessary. VISA and MasterCard credit cards are accepted, also. Orders for 100 or more copies to be mailed to a single address are discounted 25 percent. Orders by mail: U.S. General Accounting Office P.O. Box 37050 Washington, DC 20013 or visit: Room 1100 700 4th St. NW (corner of 4th and G Sts. NW) U.S. General Accounting Office Washington, DC Orders may also be placed by calling (202) 512-6000 or by using fax number (202) 512-6061, or TDD (202) 512-2537. Each day, GAO issues a list of newly available reports and testimony. To receive facsimile copies of the daily list or any list from the past 30 days, please call (202) 512-6000 using a touchtone phone. A recorded menu will provide information on how to obtain these lists. For information on how to access GAO reports on the INTERNET, send an e-mail message with "info" in the body to: firstname.lastname@example.org or visit GAO’s World Wide Web Home Page at: http://www.gao.gov PRINTED ON RECYCLED PAPER United States Bulk Rate General Accounting Office Postage & Fees Paid Washington, D.C. 20548-0001 GAO Permit No. G100 Official Business Penalty for Private Use $300 Address Correction Requested
Medicaid Managed Care: Delays and Difficulties in Implementing California's New Mandatory Program
Published by the Government Accountability Office on 1997-10-01.
Below is a raw (and likely hideous) rendition of the original report. (PDF)