Summary

Department of Health Services: Despite Shortcomings in the Department’s Monitoring Efforts, Limited Data Suggest Its Two-Plan Model Does Not Adversely Affect Quality of and Access to Health Care July 1999 99102 The first copy of each California State Auditor report is free. Additional copies are $3 each. You can obtain reports by contacting the Bureau of State Audits at the following address: California State Auditor Bureau of State Audits 555 Capitol Mall, Suite 300 Sacramento, California 95814 (916) 445-0255 or TDD (916) 445-0255 x 248 OR This report may also be available on the World Wide Web http://www.bsa.ca.gov/bsa/ Permission is granted to reproduce reports. C S A ALIFORNIA TATE UDITOR KURT R. SJOBERG MARIANNE P. EVASHENK STATE AUDITOR CHIEF DEPUTY STATE AUDITOR July 14, 1999 99102 The Governor of California President pro Tempore of the Senate Speaker of the Assembly State Capitol Sacramento, California 95814 Dear Governor and Legislative Leaders: As requested by the Joint Legislative Audit Committee, the Bureau of State Audits presents its audit report concerning the Medi-Cal Managed Care Two-Plan Model (two-plan model) and its impact on quality of and access to health care. The two-plan model is a method of operating managed care in which the Department of Health Services (department) contracts with two health plans in each county to provide care. This report concludes that based on limited available data, the department’s implementation of the two-plan model does not appear to have an adverse impact on the quality of and access to medical services provided to the Medi-Cal population. To ensure that health plans are providing high quality and easily accessible medical services, the department monitors the health plans. However, the department’s monitoring efforts thus far have been incomplete because its approach for monitoring health plans is not well organized. Respectfully submitted, KURT R. SJOBERG State Auditor BUREAU OF STATE AUDITS 660 J Street, Suite 300, Sacramento, California 95814 Telephone: (916) 445-0255 Fax: (916) 327-0019 CONTENTS Summary 1 Introduction 5 Chapter 1 Data Currently Available Suggest That the Two-Plan Model Does Not Adversely Affect Quality of and Access to Care for Medi-Cal Beneficiaries 11 Recommendations 27 Chapter 2 The Department Needs More Effective Monitoring of Its Health Plans to Maintain Quality Health Care and Access to Services 29 Recommendations 43 Appendix The Department of Health Services’ Methodology for Developing Capitation Rates 47 Response to the Audit Department of Health Services R-1 SUMMARY RESULTS IN BRIEF T he Department of Health Services (department) has shown deficiencies in its monitoring of health plans that contract with the California Medical Assistance Program (Medi-Cal), and these shortcomings could potentially under- Audit Highlights . . . mine the State’s delivery of medical services to the financially needy. Nonetheless, limited data indicate that the department’s Limited available data recently developed model for managed care gives Medi-Cal suggest that the Medi-Cal beneficiaries adequate access to quality health care. Under the Managed Care Two-Plan Model does not adversely Medi-Cal Managed Care Two-Plan Model (two-plan model), each affect quality of and access to participating county offers beneficiaries a choice between a health care for Medi-Cal health plan operated by a local entity and one operated by a beneficiaries. However, the commercial health maintenance organization. Both types of Department of Health Services’ (department) efforts plans pay for and manage all medically necessary services for to monitor plans’ services thus their beneficiaries while the department compensates the plans far have been incomplete and according to a predetermined fixed rate. not well organized. Specifically, the department: Current statistics from one health plan, as well as our own (cid:254) Does not sufficiently observations of five health plans that administer Medi-Cal monitor the number of managed care, suggest that managed care in general offers physicians and specialists available to serve benefits that fee-for-service systems do not necessarily provide, beneficiaries. especially in the area of preventive care. However, data the (cid:254) health plans submitted to the department are insufficient for us Has not met its goal for performing regular site to evaluate the overall quality of health care furnished to benefi- visits of the health plans. ciaries in counties with the two-plan model. The data, which (cid:254) include information on services, providers, and beneficiaries, are Fails to promptly review problematic because the department has not validated them. the corrective actions that health plans have The department has also faced difficulties in obtaining the data proposed to address because providers lack incentives for supplying detailed informa- deficiencies identified tion to the health plans when the providers receive a fixed fee in audits. for their services regardless of what services they actually (cid:254) May be less effective in provided. Further, the department inadvertently discourages its efforts because it does health plans from supplying some information because it not coordinate between requires them to use two separate forms to report services fur- managed care and audit staff. nished under two different programs that cover children’s medical care. To facilitate its collection of data on medical services, the department recently began to withhold a portion of the health plans’ monthly payments until the plans meet reporting goals. The department is also taking other steps to measure the plans’ quality of care. C A L I F O R N I A S T A T E A U D I T O R 1 In addition to its difficulties in acquiring necessary data, the department’s efforts to monitor health plans have been incomplete and poorly organized. Even though it designed a comprehensive system for overseeing health plans, the depart- ment does not supply staff members with specific guidelines to direct their monitoring activities, track the status of documents used for monitoring, or summarize in a formal document the results of its efforts to evaluate plans’ compliance with Medi-Cal requirements. These gaps in the department’s procedures have contributed to its failures to analyze adequately whether health plans have enough primary care physicians and specialists who can serve beneficiaries, meet its goal for visiting health plan sites regularly, review promptly health plans’ proposed corrective actions to address weaknesses the department identified in its audits, and inform its monitoring staff about trends in com- plaints against the health plans. Finally, the department’s audits may be less effective because the department did not coordinate efforts among its staff. Such inefficiencies within the department could eventually delay or even prevent the delivery of quality medical services to those who most need the State’s assistance for health care. RECOMMENDATIONS To obtain complete, reliable data for measuring the success of the Medi-Cal Managed Care Two-Plan Model, the Department of Health Services should do the following: • Validate the accuracy of data received from the health plans that provide care in counties that participate in the two-plan model. • Periodically assess the effectiveness of its withholding provi- sion and whether this provision encourages an increase in reporting of data by health plans. If necessary, the department should modify the provision or impose sanctions to further encourage the prompt submission of reliable data on services, providers, and beneficiaries. • Address the inefficiencies caused by its existing practice of requiring health plans to complete two different forms that use different coding systems for Children’s Health and Dis- ability Prevention program services and for Medi-Cal services. 2 C A L I F O R N I A S T A T E A U D I T O R In addition, the department should continue to promote quality improvement among the health plans through its various reviews and develop new approaches to address emerging health care issues. To ensure that it monitors adequately health plans that provide care under Medi-Cal managed care, the department should take these steps: • Implement formal guidelines for monitoring that describe the department’s expectations for various tasks, such as evalua- tions of the existing provider networks for health plans, site reviews of health plans, and the communication of trends pertaining to grievances. • Develop a tracking tool that will better enable its contract managers to assess whether the health plans have submitted all reports required by the department and whether the department’s staff has promptly reviewed the reports. • Require its contract managers to prepare written documenta- tion describing their monitoring efforts. • Maintain an ongoing record for each health plan that encapsulates the results of the department’s overall monitor- ing efforts and also the corrective actions not yet taken by the plan. • Coordinate efforts between its managed-care division and its audits and investigations program to ensure consensus on roles in performing audits of the health plans. At the same time, both sides should continue efforts to resolve differences in their perspectives on the audits to ensure that these reviews directly address the expectations of the managed-care division. AGENCY COMMENTS The Department of Health Services agrees with our audit findings and recommendations and has committed to specific improvements of its monitoring of the health plans. n C A L I F O R N I A S T A T E A U D I T O R 3 Blank page inserted for reproduction purposes only. 4 C A L I F O R N I A S T A T E A U D I T O R INTRODUCTION BACKGROUND T he California Medical Assistance Program (Medi-Cal), which the Department of Health Services (department) administers, enables the financially needy to have access to inpatient and outpatient medical services; nursing-home care; laboratory and X-ray services; home health care; and early and periodic screening, diagnosis, and treatment services. Medi-Cal receives one-half of its funding through Social Security Act Title XIX appropriations from the federal Department of Health and Human Services, while various state funds, including the State’s General Fund, supply the balance of Medi-Cal’s funding. Chapter 95, California Statutes of 1991, directed the Medi-Cal program to increase its efforts to use managed-care health plans similar to those available to the general public to provide the Medi-Cal population with equivalent quality of services and access to care. Additionally, the department’s philosophy in using managed care is that health plans will focus on preventive care to keep members healthy, thus reducing the need for more expensive care, such as hospital stays and emergency room services. Under managed care, the plans receive payment for Medi-Cal services through a process called “capitation” in which the department multiplies the number of beneficiaries the plan serves by a predetermined fixed rate. The plan agrees to provide all medically necessary services and assume the risk that capita- tion payments may not cover all the services it renders. As illustrated in Figure 1, the department uses specific models of managed care in 20 of the State’s 58 counties. Each of the 20 counties operates under one of the following three models: Two-Plan Model—The department contracts with two health plans in each county to provide care. A local entity, such as a county government or an independent health commission, must operate one of the plans, and a commercial health maintenance organization (HMO) must operate the other. The department refers to the plans as the “local initiative” and the “commercial plan”, respectively. In those counties in which a C A L I F O R N I A S T A T E A U D I T O R 5 local initiative is not available, the department contracts with two commercial plans. Each Medi-Cal beneficiary may choose a plan and has the flexibility to switch plans. Geographic Managed Care—Under this model, the department contracts with several commercial HMOs in the county. The beneficiary may select a health plan from any of these HMOs. County Organized Health System—A county using this model has a single, locally organized health plan that serves the Medi- Cal beneficiaries in the county, and the plan is similar to the local initiative in the two-plan model. The county’s board of supervisors often organizes this type of plan. California’s 38 other counties operate under a fee-for-service system in which the department pays each doctor or other medical provider a fixed amount per service performed. 6 C A L I F O R N I A S T A T E A U D I T O R FIGURE 1 Medi-Cal Managed Care in California Two-Plan Model Geographic Managed Care County Organized Health System NAPA MENTO SOLANOSACRA Kern L o F c a a m l il I y n H it e ia al t t i h v e Care SAN CONTRA JOAQUIN COSTA Commercial Plan SAN FRANCISCO SAN MATEO ALAMEDA STANISLAUS Blue Cross SANTA CLARA SANTA CRUZ FRESNO TULARE Local Initiative Alameda Alliance for Health KERN Commercial Plan Blue Cross SAN BERNARDINO SANTA BARBARA LOS ANGELES RIVERSIDE ORANGE Local Initiative SAN DIEGO L.A. Care Commercial Plan Health Net C A L I F O R N I A S T A T E A U D I T O R 7 Figure 2 illustrates the relationships typical in managed care and the corresponding flow of activity. Most health plans do not provide medical care directly but instead contract with health care providers. In some cases, health plans contract directly with individual physicians. In other cases, the plans contract with independent practice associations and medical groups who, in turn, contract with individual physicians. A health plan may also contract with another health plan to serve a segment of its members. As a result, it is possible to have several entities in- volved in the delivery of services. However, in all cases, the health plans that contract with the department assume the financial risk of ensuring that their members receive services. Furthermore, the department has the ultimate responsibility to ensure that health plans adhere to the contractual requirements regarding quality of care and access to services. FIGURE 2 Flow of Medical Services, Payments, and Data for Medi-Cal Managed Care Data Data Services Payments Health Payments Plan* Health Care Provider† Department of Health Services Medi-Cal Beneficiary * Some health plans contract with other health plans, who in turn contract with health care providers. For example, L.A. Care contracts with seven other health plans. Therefore, multiple health plans may be involved in the flow of payments and data. † Health care providers include medical groups, independent practice associations (IPAs), and individual physicians. Because medical groups and IPAs contract with physicians to provide medical services, more than one level of providers may be involved in the flow of payments and data. 8 C A L I F O R N I A S T A T E A U D I T O R SCOPE AND METHODOLOGY The Joint Legislative Audit Committee (JLAC) asked the Bureau of State Audits to determine whether the implementation of the two-plan model has affected the access and range of medical services provided to eligible beneficiaries. In addition, JLAC was interested in the department’s ability to monitor effectively and efficiently the quality and cost of services delivered under the model. We reviewed state laws and regulations governing the Medi-Cal program and the two-plan model. To obtain an understanding of the department’s monitoring responsibilities, we interviewed staff members in the Medi-Cal Managed Care Division about their roles in monitoring health plans. Further, we reviewed key monitoring documents, including those submitted by the plans as well as reports generated internally by the department. We also identified the extent to which the department’s staff members documented their monitoring efforts. To understand the health plans’ roles and responsibilities, we reviewed their contracts with the department and identified the requirements regarding the delivery of services. We visited the five health plans that constitute the two-plan model in Alameda, Kern, and Los Angeles counties. (Blue Cross partici- pates in the two-plan model in both Alameda and Kern coun- ties.) We inquired about the plans’ operations and methods for providing access to care and maintaining quality of services, and we also asked about the plans’ interactions with the department and health care providers. To determine the extent to which the department monitors the cost of managed care, we evaluated the department’s procedures for setting capitation payment rates for the health plans. We reviewed for reasonableness and consideration of relevant factors the department’s methodology for developing rates, but we did not evaluate the source or validity of the data and actu- arial assumptions used to develop the rates. The Appendix describes the department’s methodology for developing these rates. Further, we examined a sample of the department’s calcu- lations of capitation payments made to the five plans. To assess whether the two-plan model has affected Medi-Cal beneficiaries’ quality of or access to care, we identified and analyzed different methods that the department and others use to measure quality and accessibility, including the department’s C A L I F O R N I A S T A T E A U D I T O R 9 accumulation and study of data on beneficiaries’ medical services. In addition to reviewing this data, we evaluated the department’s system for collecting the data. Moreover, we tried to review the data for four quality-of-care indicators: well-child care, physician visits per member, prenatal care, and initial health assessments of new members within 120 days of their joining a health plan. We also attempted to compare the data to statistics maintained by the federal Health Care Financing Administration and other sources. Finally, we inquired about other efforts that the department and health plans make to evaluate quality of health care provided to Medi-Cal beneficiaries. n 10 C A L I F O R N I A S T A T E A U D I T O R CHAPTER 1 Data Currently Available Suggest That the Two-Plan Model Does Not Adversely Affect Quality of and Access to Care for Medi-Cal Beneficiaries CHAPTER SUMMARY A ccess to and range of services are critical components in the effective delivery of health care. These elements are particularly important when a change in health care services occurs, as it did when the State implemented in some counties the Managed Care Two-Plan Model (two-plan model) of the California Medical Assistance Program (Medi-Cal). Nothing we observed suggests that the two-plan model has negatively affected the access to and range of medical services for Medi-Cal beneficiaries. Although patient contact and treatment statistics are limited in availability, the overall design of managed-care systems promotes access to medical services for its members. Also, the health plans we reviewed that administer medical care to Medi-Cal beneficiaries educate their members on health matters and focus on ensuring that medical professionals give quality care to the plans’ members. Furthermore, the Department of Health Services (department) has recently begun to assess the performance of the health plans by measuring the volume of their encounter data, which is information about health services provided and which the health plans must submit to the department. However, as yet there are not sufficient data for the department to fully assess the two-plan model on its overall quality of and access to care. The department is also developing quality-improvement studies. These efforts appear consistent with the department’s objective of ensuring quality of care and access to services for all Medi-Cal beneficiaries enrolled in managed care. C A L I F O R N I A S T A T E A U D I T O R 11 A MANAGED-CARE SYSTEM EMPHASIZES MEMBERS’ ACCESS TO CARE Managed care encourages health care professionals to focus on maintaining a healthy population through preventive care, which minimizes such expensive treatments as emergency room visits and hospital stays. Generally, all managed-care members select a health plan that contracts with a network of health care providers, including general practitioners and specialists, and all members then select a primary care physician. As the central coordinator of health care, the primary care physician ensures that members receive all medically necessary services, including additional care from specialists. In contrast, under the Medi-Cal fee-for-service system, individuals seeking care are responsible for locating physicians and specialists that accept Medi-Cal beneficiaries, although the individuals may have difficulty finding providers that accept new patients. As part of managed care, health plans offer services intended to Health plans offer assist members in developing healthy lifestyles and obtaining additional services to preventive care. These services include regular health examina- develop healthy lifestyles tions and recommended immunizations. Furthermore, as this and provide preventive report’s Introduction explains, the two-plan model is unique in care. that it gives Medi-Cal beneficiaries the choice of enrolling in either a locally operated health plan (local initiative) or a commercial plan. Some beneficiaries may choose the local initiative because of its familiarity with and close ties to the county it serves. Others opt for the commercial plan because of its favorable reputation for serving the general population. The Managed-Care Plans Try to Focus on Quality of Care and Access to Services We visited the five health plans that comprise the two-plan model in Alameda, Kern, and Los Angeles counties. One of these plans, Blue Cross, is the commercial plan in both Alameda and Kern counties. All five appeared to have effective member services programs and policies to ensure members’ access to medical treatment. For example, each plan has a 24-hour information line available for members to ask medical questions and obtain referrals for covered services. In addition, the health plans have member complaint and grievance processes, and each offers and promotes health education services. Further, although the medical review reports from the department’s audits and investigations program identify areas in which the health plans need to improve, the reports nevertheless support 12 C A L I F O R N I A S T A T E A U D I T O R our conclusion that the plans are generally meeting their obligations to ensure Medi-Cal beneficiaries can readily access quality medical care. In addition to supplying required services, some health plans make efforts above and beyond contract provisions in order to serve Medi-Cal beneficiaries. For example, the local initiative in Alameda County, Alameda Alliance for Health, pays for taxi and public transportation services to routine medical appointments to increase its members’ access to care. Another local initiative, Kern Family Health Care, offers more than 100 health education classes for its members and will also arrange transportation for medical services if necessary. In addition, Blue Cross has an asthma program that identifies members who suffer from asthma and offers them a variety of services, including phar- macy education sessions, asthma management equipment, and home health visits. The Department Concluded That the Managed-Care Plans Adequately Protect Members’ Rights and Ensure Availability of Services In its annual medical reviews, the department’s audits and Each year, the investigations program gave the five health plans we visited department audits plans high scores, indicating that the plans substantially or to assess compliance with significantly complied with contract requirements for protecting contract requirements members’ rights. Each year, the audits and investigations and laws. program visits the plans’ headquarters and provider facilities to assess their compliance with contract requirements and with state and federal laws governing members’ rights and availability and accessibility of services. A team of reviewers determines whether the health plans are meeting specific requirements for services, including providing 24-hour access to interpreter services, maintaining a member complaint and grievance system, and providing urgent care within 48 hours. The team also monitors the time patients spend in waiting rooms and whether patients receive sufficient medication at emergency rooms until they can fill their prescriptions. Limited Data Suggest That Well-Child Care Under the Two-Plan Model Is Comparable to Care Provided Under Fee-for-Service Arrangements As mentioned earlier, our audit focused on four quality-of-care indicators. Well-child care, or preventive care for children, is one indicator for which the department has developed a method for C A L I F O R N I A S T A T E A U D I T O R 13 assessing how well health plans are performing. Routine health examinations for children and immunizations against childhood diseases are two important components of well-child care. Under Medi-Cal managed care, health plans must follow the most recent standards for pediatric care and immunizations that the American Academy of Pediatrics recommends. In assessing how well the health plans are performing well-child care, the department compares the number of actual preventive services to the number of services recommended by the academy. For example, the academy recommends that children between the ages of one and two have a physical examination by a physician three times per year. To assess whether each plan provides a reasonable level of services to these members, the department calculates the actual number of visits during the year for the children from this age group and compares this number to the academy’s recommended goal. The department makes such comparisons for children who receive services under the fee-for-service system as well as for those children in a managed-care system. As Table 1 shows, Blue Cross’s data suggest that in all categories Children enrolled in the but one, children covered under its plan receive more medical Blue Cross health plan in services recommended for well-child care than do children in two counties are receiving the fee-for-service system. Blue Cross is similar to the other more preventive care managed-care health plans because it receives capitation services than children in payments from the State; however, Blue Cross reports more data the fee-for-service system. on its services than the other plans we reviewed because it contracts with most of its providers on a fee-for-service basis and requires these providers to submit detailed data on claims for payment. Because Blue Cross collects more data from its providers, it reports more data to the department than do the other health plans. The other managed-care plans we reviewed, including the local initiatives in Alameda and Kern counties and the two health plans in Los Angeles County, have had less success collecting such data from their providers primarily because many providers have a capitated payment arrangement and lack an incentive to provide the data. Consequently, the department is unable to determine whether the providers simply do not supply many services or whether the providers do not report the services they supply. In addition, a more complete picture of the services rendered by health plans other than Blue Cross is difficult to obtain. Unlike Blue Cross, which has been operating for many 14 C A L I F O R N I A S T A T E A U D I T O R years and has previous experience collecting encounter data for services it provides to its members who are not Medi-Cal benefi- ciaries, the local initiatives were just recently established under the two-plan model in January 1996. Therefore, the data for fiscal year 1997-98 is from the first or second year of operations for most of these plans. TABLE 1 Children Received More Medical Services Under Managed Care (Blue Cross) Than Under the Fee-for-Service (FFS) System in Alameda and Kern Counties Fiscal Year 1997-98 Well Child Well Child Well Child Well Child Well Child Childhood Visit (<1 year Visit (1 year Visit (2-4 Visit (5-10 Visit (11-21 Immunization Health Plan old) old) years old) years old) years old) (Age <2) Managed care with claims Blue Cross Alameda 88.5% 93.9% 111.0% 58.7% 37.8% 77.0% Blue Cross Kern 94.4 83.0 139.9 78.1 50.0 76.1 FFS with claims Alameda County FFS 53.8 60.6 86.2 53.9 43.5 49.0 Kern County FFS 57.4 52.2 83.6 49.0 30.2 49.1 Source: Quality Folder in Panorama View, Department of Health Services Management Information System/Decision Support System *The Panorama View report on which this table is based includes an annual physical examination for seven and nine year olds; however, the most recent guidelines of the American Academy of Pediatrics do not include physical examinations for these ages. As the data for Blue Cross indicate, Kern County children in nearly all age groups received more of the recommended physi- cal examinations under the Blue Cross plan than did children in the fee-for-service system. For example, children under one year old received 94 percent of the recommended number of physical examinations, while the corresponding group under the fee-for- service system received only 57 percent of the recommended number. Similarly, children under age two enrolled in Blue Cross also received a higher percentage of the recommended immuni- zations for their age group than did those in the fee-for-service system. Although the comparison shown in Table 1 is limited to one of the health plans that operates in two counties, the avail- able data suggest that Medi-Cal managed care is working as the department intended. C A L I F O R N I A S T A T E A U D I T O R 15 In Two Counties, More Beneficiaries Receive Physician Services Under Managed Care Than Under Fee-for-Service In addition to evaluating pediatric preventive care, we focused on the number of physician visits per member. Because the department’s new system reports on this particular measure in terms of the number of physician services per member, we compared the percentage of Medi-Cal beneficiaries receiving physician services under Blue Cross in Alameda and Kern coun- ties to the percentage of those beneficiaries obtaining services in the fee-for-service system. As Table 2 shows, the data reported by Blue Cross in Kern and Alameda counties suggest that under managed care, more beneficiaries are receiving at least one physician visit than are beneficiaries in the fee-for-service system. Again, because Blue Cross contracts with most of its providers on a fee-for- service basis, it reports more data on physician services than do the other plans in the three counties we reviewed. TABLE 2 Comparison of the Number of Medi-Cal Beneficiaries Who Received Physician Services Under Managed Care to the Number Who Received Similar Services in the Fee-for-Service (FFS) System Fiscal Year 1997-98 Percentage Medi-Cal of Medi-Cal Beneficiaries Beneficiaries Who Received Who Received Medi-Cal Physician Physician Health Plan Beneficiaries Services Services Blue Cross Alameda 22,337 18,480 82.7% Blue Cross Kern 21,935 18,484 84.3 Alameda County FFS 21,102 14,773 70.0 Kern County FFS 23,744 15,259 64.3 Source: “Physicians Trend by Paid Date,” Utilization by Category of Service report from Panorama View in the department’s management information system/decision support system. This represents encounters with physicians and not pharmacy visits, lab tests, or other services. 16 C A L I F O R N I A S T A T E A U D I T O R ENCOUNTER DATA IS KEY TO ASSESSING HEALTH PLANS’ PERFORMANCE In addition to the department, many other interested groups, such as taxpayers, lawmakers, and advocacy groups, are also concerned about the quality, accessibility, and cost of care for Medi-Cal patients. These stakeholders rely on the department to measure the performance of health plans in the State’s Medi-Cal managed-care program to ensure the goals of managed care are met. In addition to using other methods, the department in- tends to assess the performance and participation of health plans and their providers by analyzing patterns of care for specific diagnoses or procedures and by exploring areas of concern in the delivery of services. The department’s system is in the final stages of development, and its success is premised upon reliable, complete data from the health plans and their providers. To assess the health plans adequately, the department is relying, By gathering data on in large part, on its ability to collect complete, accurate encoun- services provided, the ter data from health plans. Encounter data is information that department can assess health plans maintain in their computer databases about the whether members have services their members receive from providers. Like the informa- ready access to medical tion included on a claim for payment that a provider submits care. under the fee-for-service system, encounter data includes infor- mation about the beneficiaries in addition to information about all services the beneficiaries obtained as well as about the pro- viders of those services. Encounter data can be used to track the services provided to members and enables the department to determine whether members can readily access medical care. An analysis of encounter data can identify service utilization patterns for a specific health plan. By comparing these patterns among health plans, the department can identify plans that underuse a specific service, such as a laboratory test, and also identify the providers who may have denied services. Similarly, the department and the plans can more effectively control costs by identifying members who overuse costly services, such as hospital emergency care, and then educating the members about appropriate uses of those services. The department can also use encounter data to identify patterns in the delivery of services and compare these patterns across plans and individual providers. Examining these patterns could help the department to identify providers who may not be following accepted protocols for care. For example, providers C A L I F O R N I A S T A T E A U D I T O R 17 generally schedule a follow-up laboratory procedure for a patient Analyzing service recently released from a hospital after an illness to identify and patterns can help the treat recurring problems before these problems necessitate a department identify second hospital stay. Encounter data could allow the department providers who may not be to determine whether providers actually schedule their patients following accepted for the follow-up procedures. protocols for care. To enable it to report on the quality of care patients receive under Medi-Cal managed care, the department is currently developing a system to accumulate and report encounter data for its Medi-Cal population. This system will use encounter data to report on the level of services beneficiaries are receiving, service utilization patterns, and trends for a specific category of service, such as inpatient hospital or laboratory services. The department is requiring the health plans to collect and submit encounter data from their providers. As of June 1999, the department’s system is in the fourth of five phases of develop- ment and is scheduled to be fully operational in April 2000. Meanwhile, individual applications of the system are opera- tional and can analyze the data reported thus far. Gathering complete, accurate encounter data will also enable the department to develop competitive capitation rates, which are the monthly rates the health plans are paid to supply ser- vices to members. Currently, the department develops its rates by comparing costs from the fee-for-service system. However, the department’s ability to rely on fee-for-service costs is eroding as fee-for-service arrangements are quickly being replaced by managed care, so the department must prepare to develop rates based on managed-care plans alone. AVAILABLE DATA ARE NOT SUFFICIENT FOR ASSESSING OVERALL ACCESS TO AND QUALITY OF CARE Although we were able to draw preliminary conclusions from the available data on well-child care and physician services, the department has not collected enough data to assess the overall access to and quality of care Medi-Cal beneficiaries are receiving under the two-plan model. The department’s contracts with managed-care health plans require the plans to submit data on all services for which they are paid on a capitated payment basis. The plans in turn rely on providers to report this information, but many providers are not submitting sufficient data. For example, we could not draw conclusions from data for prenatal care and initial health assessments. Moreover, until recently, the 18 C A L I F O R N I A S T A T E A U D I T O R department had not identified the segment of services, such as baseline physical examinations and health histories, that pertain to the initial health assessment. The department is currently working on finalizing a method for plans to use to identify these services and then to use encounter data to report on the initial health assessment. Nevertheless, until the plans and providers report sufficient information, meaningful conclusions cannot be drawn. In addition, data collection efforts are further complicated because the encounter data requirements are new for the Medi-Cal program in general, while the local initiatives, most of which have been operating for fewer than three years, have had no prior experience with collecting encounter data. As a result, the data that health plans have forwarded to the department are not adequate for it to draw reasonable conclusions. Providers Lack Incentive to Report on Services Provided Through Capitation The department realizes that under managed care, some provid- Providers are less likely ers, including primary care physicians who probably supply the to submit service majority of medical services to members, no longer have an information when it is not incentive to furnish data on their services because these provid- required for payment. ers receive capitated payments before they provide any services. Unlike a fee-for-service payment system, which requires all providers to document services provided before receiving payments, primary care physicians and other providers under a capitated payment agreement have no similar incentive to provide such information, and thus are less likely to report on the services they supplied. Further, because most Medi-Cal managed-care providers have previous experience with the fee-for-service system, the absence of such a monetary incentive in a managed-care system makes collecting complete, timely encounter data from providers a difficult task for health plans. The Process for Reporting Services Provided to Children Is Cumbersome Additionally, reporting on medical services provided to children is an added burden for managed-care providers as well as for the department. If providers and health plans must spend an excessive amount of time processing forms, they may be reluctant to report needed information on their services. Under the current reporting system, providers fill out similar forms for both Medi-Cal and the Children’s Health and Disability C A L I F O R N I A S T A T E A U D I T O R 19 Prevention (CHDP) program, and the health plans process these forms. Additionally, the department receives the encounter data from plans that have converted the data into standardized codes using the plans’ own methodologies, and this conversion may render it unreliable. Because many children eligible for Medi-Cal are also eligible for CHDP services, these reporting problems are somewhat widespread. The department’s Children’s Medical Services Branch administers the State CHDP program, which promotes preventive health screenings, such as vision and hearing tests, for children from low-income families to ensure that the children’s health problems are detected early and that the children receive immunizations for childhood diseases as recommended by the American Academy of Pediatrics. Like Medi-Cal, the CHDP program also requires physicians to report To capture all of the data on services provided; therefore, some health plans require they need, some plans providers paid under a fee-for-service arrangement and who require their providers to render services for both programs to submit a standard Medi-Cal submit two forms for the billing form and a CHDP form for the same services. The health same services. plan uses information from the Medi-Cal form as its source for encounter data and for identifying the amount to pay the provider. Additionally, as required by its managed-care contract, the plan submits the CHDP form directly to the Children’s Medical Services Branch. Because they must use standard medical codes for their encounter data, the plans convert the CHDP codes, which are not standardized, into the standardized codes. In fact, some plans require their providers to convert the CHDP codes before submitting their forms. On the other hand, providers working under a capitation arrangement need only to submit CHDP forms to the health plans. However, if one of these providers performs a physical examination, which is a CHDP service, and also treats the patient for an illness during the same visit, the provider submits to the health plan both a CHDP form for the physical examina- tion and a Medi-Cal form for the additional services. The burden of using these two forms to report the same information creates inefficiencies for both providers and health plans. Collecting the CHDP data also presents a burden for the depart- ment. The health plans must use their encounter data for their internal quality-of-care studies as well as for the annual study an external review organization conducts. However, the depart- ment cannot rely on the CHDP data because the plans use their own methodologies to convert the CHDP codes before reporting 20 C A L I F O R N I A S T A T E A U D I T O R the data. Although some plans and providers convert the CHDP codes to standard medical codes, the descriptions of many CHDP services do not mirror clearly the descriptions of services listed in the standard medical codes. Consequently, health plans have developed their own conversion methodologies. Because it cannot be sure that the plans’ translation methods are consis- tent with each other, every month the department’s information technology services division discards all CHDP data the plans submit and replaces them with data the Children’s Medical Services Branch compiles from the forms it receives. Before the information technology services division uses the replacement data, it translates the CHDP service codes into standard Medi-Cal codes. The redundancy and inefficiency of this system creates extra work for the department and may discourage health plans and providers from reporting their encounter data. The Department Does Not Validate Encounter Data The department has yet to develop a method to ensure that the encounter data it has collected thus far are accurate. In its April 1998 report on encounter data, the department’s audits and investigations program concluded that nearly half of the sample data from six plans it reviewed had discrepancies when compared to actual medical and billing records. As previously mentioned, the department plans to use the encounter data to develop reports for better assessing the adequacy of and access to services for beneficiaries of Medi-Cal managed care. However, because the reliability of the encounter data currently is questionable, the department and other interested parties will not be able to rely on the data to draw valid conclusions. According to the chief of the department’s Office of Clinical Although the department Standards and Quality, reviewing medical records to validate concluded that encounter encounter data is very labor-intensive and costly because it data was not reliable, it requires medical expertise. Also, according to the chief, within has not yet developed a the managed-care industry, various efforts are under way to method to assure validate encounter data, but there is no agreement on an appro- accuracy. priate and cost-effective method. As a result, the department’s efforts to check the data are limited to its annual external qual- ity review process, which involves both an audit of specific encounter data used to develop quality-of-care indicators and a review of the systems and processes that plans have for produc- ing these indicators. C A L I F O R N I A S T A T E A U D I T O R 21 The Department Has Begun to Withhold Payments From Plans That Do Not Meet Minimum Reporting Requirements By identifying unexplained gaps in encounter data in service areas that Medi-Cal beneficiaries typically use, the department has determined that health plans are not providing all encounter data. The department recognizes that the health plans have not complied with their responsibility to report encounter data. As previously mentioned, the health plans have had difficulty getting the physicians they contract with to report on the services they provide to Medi-Cal beneficiaries. In response to our inquiries about this situation, the department told us that getting the health plans to meet their responsibilities to provide the encounter data was going to take time. The department learned from similar “data-reporting efforts” in other parts of The department learned the country that it will need at least three years to get the from similar efforts providers to collect and report the data. Furthermore, the nationwide that it will department stated that during these first few years of the take at least three years transition to managed care in the two-plan model counties, it to get providers to collect had focused on other priorities, such as ensuring access to care, and report needed data. and making certain that the plans provide complete medical coverage to their members and operate in a fiscally prudent manner. However, during the second year of managed care in many of the two-plan model counties, the plans had not progressed in obtaining encounter data from their providers and reporting the data to the department. For this reason, the department determined that it needed to develop an arrangement that would encourage improved performance by the health plans. As a result, the department recently began withholding a portion of the monthly capitation amount from the plans until they meet contractual requirements for reporting encounter data each quarter. The requirements focus on the reporting of encounter data in two major areas: CHDP program services for children within a specified age range and also outpatient and emergency department services. Beginning with the April 1999 capitation payments, the depart- ment is withholding either 1 percent of each health plan’s monthly capitation payment or $100,000 per month, whichever is less, until the plan meets quarterly reporting goals. Once a health plan meets the quarterly goal, the department will release the withheld amount. For example, for the CHDP services under health plans in Alameda County, the department has established a cumulative goal of 32 percent. In other words, the plan would have to submit enough encounter data to indicate that at least 22 C A L I F O R N I A S T A T E A U D I T O R 32 percent of the children within a specified age range received at least one preventive service during the previous 12 months. The cumulative goals for other health plans under the two-plan model range from 23 percent to 31 percent. According to the department, it based the cumulative goals on its analysis of fee- for-service data from the CHDP program in fiscal year 1992-93, which indicated that 32 percent of the children in Alameda County had received a preventive service. Moreover, although 32 percent seems low, the department has an additional purpose in making this percentage a reporting goal: The department hopes to demonstrate that managed care is providing at least the same volume of services as under the fee-for-service system. Because the department only recently imposed the withholding provision on health plans, it is too soon to determine the provision’s effectiveness. In response to the department’s withholding provision, the In response to the health plans have developed their own mechanisms to withhold provision, encourage their providers to submit encounter data regularly. health plans have For example, in Los Angeles County, Health Net offers a developed mechanisms to financial incentive to some providers to encourage regular encourage providers to reporting. Another plan, Alameda Alliance for Health, regularly submit established a reporting incentive to encourage primary care encounter data. physicians to submit encounter data. Other plans, such as Blue Cross and Kern Family Health Care, get more data on the services their providers render by paying them for some or all of the services they provide under their plans on a fee-for-service basis. As previously explained, under a fee-for-service arrangement, providers must document their services on a detailed billing form. Developing Meaningful Data Is a Nationwide Concern The challenge of collecting enough encounter data to draw meaningful conclusions about the provision of health care is not limited to Medi-Cal in California. In May 1997, the federal General Accounting Office (GAO) reviewed the managed-care programs in four states: Arizona, Pennsylvania, Wisconsin, and Tennessee. The GAO reported that three of the four states it reviewed have invested significant resources to gather and develop reliable, useful encounter data for their Medicaid programs. One of the states, Tennessee, has begun to use its encounter data to analyze service utilization patterns, even though in March 1999, the Tennessee Division of State Audits reported that the state had significant problems with validating and underreporting of encounter data. In contrast, according to C A L I F O R N I A S T A T E A U D I T O R 23 a representative of its Medicaid program, Arizona has experienced fewer problems getting providers to report The GAO reported that encounter data than other states have experienced. According other states are to the representative, this situation exists probably because experiencing similar Arizona’s Medicaid program was first established as a managed- problems with collecting care program and Arizona has required providers to report this complete, reliable information since its Medicaid program was established. encounter data. Arizona’s program is unlike managed-care programs in California and other states, which are moving their Medicaid populations from a fee-for-service system to managed care. As a result of its success collecting encounter data, Arizona has begun to use its encounter data to develop standards for measuring the quality of care. The GAO also reported that the four states it reviewed have contracts with health plans that include data-collection requirements designed to hold plans accountable for providing appropriate care to their members. Nevertheless, few benchmarks for defining appropriate levels of care have been established, so the states have few standards for determining whether the plans are indeed providing adequate care. For example, using Medicaid encounter data and billing data from fee-for-service claims from states, the federal Health Care Financing Administration (HCFA) produces statistics on the number of inpatient days in hospitals and the number of days of long-term care patients receive; however, most of its other statistics are not useful for comparing the quality of care under managed-care plans. Finally, benchmarks derived from providers and patients in the fee-for-service system may not offer appropriate comparisons to managed care since service utilization patterns should be different under managed care. In the absence of sufficient encounter data and available benchmarks to develop meaningful analyses and performance measures, states are using other accountability measures and processes, such as focus studies and member surveys, to assess the quality of care that Medicaid beneficiaries receive. THE DEPARTMENT HAS TAKEN OTHER STEPS TO MEASURE THE QUALITY OF CARE THAT BENEFICIARIES RECEIVE To ensure that the Medi-Cal population receives quality care and adequate access to services under the two-plan model, the department has conducted quality-of-care studies and member satisfaction surveys. Although the quality-of-care studies use 24 C A L I F O R N I A S T A T E A U D I T O R encounter data that the department has not validated, the department has contracted with an external organization to oversee the studies conducted by the health plans, to validate a sample of the encounter data used in the studies, and to report on the studies’ findings. The department contracts with an external quality review organization (EQRO) to conduct annual, independent quality studies of Medi-Cal managed-care plans by using statistical sampling methods. The first set of studies conducted by the EQRO was at Alameda Alliance for Health (Alameda Alliance) because this plan was the first to operate under the two-plan model. However, the EQRO has since completed reviews for the remaining plans. The results of these initial studies will provide a baseline from which the department can measure improvement. The department required the EQRO to perform comprehensive studies in two areas: prenatal care and pediatric preventive care. In addition, as part of their quality-improvement programs, the plans were required to perform their own studies on nine additional topics of their choice. Topics studied by Alameda Alliance included cervical cancer screening, frequency of selected procedures, use of the emergency room, and customer satisfaction. As the basis for the studies, the EQRO used quality indicators from the Health Employer Plan Data Information Set (HEDIS) developed by the National Committee on Quality Assurance (national committee), a nonprofit organization whose primary objective is to develop strategies and systems to establish accountability in the managed-care industry. The EQRO also used other quality indicators developed by the department. The results of all 11 initial studies for Alameda Alliance were The results of the first poor. For example, for the prenatal care study, the medical quality of care study records for 60 percent of the pregnant women enrolled in the completed were poor, in plan lacked documentation of prenatal care. Although the part because the health results were due to a variety of factors, a primary factor was that plan was new and the period covered in the studies was the first year of the plan’s providers did not submit operations. Additionally, many of Alameda Alliance’s providers all of the necessary did not produce copies of the medical records necessary for the medical records. study. Generally, the national committee allows plans to use encounter data or a combination of actual medical records and encounter data as the source of information for HEDIS quality indicators. However, according to the chief of the department’s Office of Clinical Standards and Quality, the department C A L I F O R N I A S T A T E A U D I T O R 25 required the EQRO to use only medical records as the source of service data for its baseline studies because most of the health plans were new and did not have enough time to collect sufficient, valid encounter data. Moreover, some of the HEDIS indicators require beneficiaries’ continuous enrollment in the plan, which is a difficult requirement for the health plans to meet with a traditionally mobile Medi-Cal population. According to the department, the EQRO is in the process of completing its final reports for its baseline studies of the remaining health plans. The second round of EQRO studies began in January 1999 for all Results of the second of the health plans. To simplify the reviews, the department round of quality reviews reduced the number of quality indicators to the following seven are scheduled for HEDIS measures: childhood and adolescent immunization completion in status; prenatal care in the first trimester; checkups after October 1999. delivery; and preventive care for babies, older children, and adolescents. These reviews are scheduled for completion in October 1999. Unlike the first EQRO studies, which required the plans to collect all of their study data from medical records, the second studies will allow the health plans to use a combination of encounter data and medical records, thereby reducing the need to rely solely on medical records. The Health Plans Are Performing Internal Focus Studies In addition to the external reviews conducted on behalf of the State, the department also requires health plans to perform similar internal focus studies as part of their quality improve- ment programs. Currently, plans must select four topics for concurrent study—two clinical and two nonclinical—with all topics approved by the department. The department will oversee the studies using HCFA guidelines. Once a plan demon- strates improvement, it may end the study after obtaining approval from the department. The department has requested that the plans submit their selected topics by October 1, 1999. Once the department receives and approves a plan’s topics, the first series of studies will begin. Anticipated topics include the improvement of childhood immunization rates, the improvement of adolescent well-care visit rates, and the reduction in emergency room visits. In the future, the department will require the plans to add two additional projects—one in 2001 and one in 2002—for a total of six projects per year. 26 C A L I F O R N I A S T A T E A U D I T O R Member Surveys Are in Progress In addition to the quality-of-care studies the health plans are conducting, the department is in the process of administering a Consumer Assessment of Health Plans Study (surveys) through its EQRO contractor. The national committee and the Agency for Health Care Policy and Research developed the surveys to gather standardized information about members’ overall satisfaction with the quality and accessibility of their health care. The results of the surveys will allow Medi-Cal beneficiaries enrolled in managed care to compare plans. According to the department, once the EQRO has compiled the results of the surveys and the health plans have had sufficient time to implement interven- tions designed to improve identified areas, the department plans to repeat the surveys to determine if the plans have improved measurably in those areas. RECOMMENDATIONS In order to strengthen its ability to use encounter data effectively, the Department of Health Services should do the following: • Determine the accuracy of encounter data by validating the data received from managed-care health plans. • Periodically assess the effectiveness of its withholding provi- sion and whether this provision has resulted in an increase in data reporting. If necessary, the department should modify the provision or impose sanctions to further encourage the prompt, reliable submission of encounter data. • Address the inefficiencies caused by its existing practice of requiring health plans to use one type of form with one coding system for Children’s Health and Disability Prevention services and another form with another coding system for Medi-Cal services. In addition, the department should continue to promote quality improvement among the health plans by performing its various reviews and also develop new approaches to address emerging health care issues. n C A L I F O R N I A S T A T E A U D I T O R 27 Blank page inserted for reproduction purposes only. 28 C A L I F O R N I A S T A T E A U D I T O R CHAPTER 2 The Department Needs More Effective Monitoring of Its Health Plans to Maintain Quality Health Care and Access to Services CHAPTER SUMMARY T he Department of Health Services (department) monitors health plans, including those in the 12 counties under the Medi-Cal Managed Care Two-Plan Model (two-plan model), to ensure that these plans are providing high-quality, accessible health care to California Medical Assistance Program (Medi-Cal) beneficiaries. However, even though it has designed a comprehensive system for overseeing health plans, the department’s monitoring efforts thus far have been incomplete and poorly organized. Specifically, the department does not regularly identify whether health plans have sufficient networks of primary care physicians and specialists available to serve the needs of Medi-Cal beneficiaries. The department also has not met its goal for conducting regular site visits of the health plans, nor has the department promptly reviewed corrective action that health plans propose to address deficiencies that its audits have noted. Moreover, these audits may be less than effective because the department did not coordinate efforts between its managed-care staff and the staff that performed the audits. Finally, the department’s system for addressing grievances is incomplete because it does not have a formal process for inform- ing its monitoring units about trends in beneficiaries’ com- plaints about the health plans. Such inadequate monitoring of health plans by the department could lead to future concerns regarding the quality and delivery of medical care to Medi-Cal beneficiaries. Several factors have contributed to the gaps in the department’s monitoring efforts. First, the department has not adopted formal guidelines directing the monitoring activities of its staff. In addition, the department does not have an effective process for tracking the status of documents that it uses for monitoring the health plans. Furthermore, some staff do not consistently C A L I F O R N I A S T A T E A U D I T O R 29 document their monitoring efforts, and the department does not summarize its collective efforts to address the overall compliance of health plans. MULTIPLE DEPARTMENT UNITS COLLECTIVELY MONITOR THE HEALTH PLANS The department’s Medi-Cal Managed Care Division (managed- care division) is responsible for administering managed care in the State. Initially, this division oversaw the implementation of managed care for Medi-Cal patients. In particular, the division focused on expanding managed care in a short time through intensive development and implementation of contracts with health plans. As Medi-Cal managed care evolved, the depart- ment shifted its focus to monitoring the health plans. As a result, in 1998, the department reorganized the managed-care division to address the change in its responsibilities. The department’s monitoring covers a wide spectrum of activities, including performing audits and site visits of the health plans, advising the plans of their responsibilities to their members and to the department, and reviewing documents and reports that the plans are required to submit. A significant A significant portion of portion of the department’s monitoring efforts are split between the department’s two branches of the managed-care division. The division’s Plan monitoring efforts are Management Branch (plan management) consists of contract split between two managers, who serve as the liaisons between the department branches of the managed and the health plans. The contract managers perform certain care division. monitoring activities, such as reviewing health care provider networks and certain health plan policies. The contract managers may also assist other department staff in conducting site reviews or audits of the plans. The Plan Monitoring/Member Rights Branch (plan monitoring) is responsible for specific monitoring tasks, such as performing site visits, reviewing corrective action plans, and reviewing health plan policies. This branch also has the responsibility for calculating and processing capitation payments. Plan monitor- ing also includes the Office of the Ombudsman (ombudsman unit), which is responsible for addressing complaints and grievances from health plan members. 30 C A L I F O R N I A S T A T E A U D I T O R Finally, through a memorandum of understanding with the managed-care division, the department’s audits and investiga- tions program (Audits and Investigations) audits health plans. Within Audits and Investigations, the Financial Audits Branch conducts financial audits that focus on assessing contract compliance of selected health plans. These audits include reviewing internal controls, financial requirements, and management practices. In addition, the Medical Review Branch conducts annual medical reviews of all plans that target compli- ance with medical and clinical requirements in the contract. The plan management and plan monitoring branches share joint responsibility to review corrective action plans that the health plans submit to address recommendations in Audits and Investigations’ reports. In addition to the department’s monitoring activities, other In addition to the external parties oversee managed-care programs. The Depart- department, other ment of Corporations licenses health plans under the external parties oversee Knox-Keene Health Care Act and the federal Department of the managed-care Health and Human Services Health Care Financing Administra- programs. tion (HCFA) oversees Medi-Cal managed care. The department is required to submit a semiannual report to HCFA documenting its monitoring of the two-plan model. THE DEPARTMENT USES VARIOUS RESOURCES IN ITS MONITORING EFFORTS Monitoring of health plans is necessary to ensure that beneficiaries receive needed medical services. We observed the department’s use of several resources in its efforts to monitor plans. One such tool, a formal grievance system, provides beneficiaries opportunities to file directly with the department their complaints against health plans or providers. Investigators address the complaints by contacting the necessary parties and resolving the complaints. Another means of monitoring involves meetings between the department’s medical professionals and the health plans’ medical directors. During these meetings, the medical professionals discuss various issues regarding health plan members’ quality of care and access to services, examples of which include problems relating to the coordination of care for children with special needs, the creation of an immunization registry to track all Medi-Cal beneficiaries, C A L I F O R N I A S T A T E A U D I T O R 31 and requirements of plans to conduct access-to-care studies. Another monitoring resource available to the department is its contract managers, who communicate directly with the health plans. Contract managers provide guidance to the plans on Medi-Cal eligibility, contract requirements, and other policies. THE DEPARTMENT DOES NOT FULFILL CERTAIN IMPORTANT MONITORING RESPONSIBILITIES Despite having a comprehensive process with several compo- nents for monitoring the health plans, the department has not addressed certain key areas. Specifically, the department is not sufficiently monitoring to ensure that an adequate number of primary care physicians are available to serve beneficiaries in each of the health plans. In addition, the department has only conducted a limited number of site reviews of health plans since the managed-care division’s reorganization. Moreover, the department has not performed prompt reviews of corrective action proposed by the health plans to address deficiencies identified as a result of the department’s audits. Finally, although the department has a formal grievance process, this process is incomplete because it does not include the use of formal reports that the department can use to inform staff responsible for monitoring the plans about trends in beneficia- ries’ complaints and grievances against the health plans. The Department Does Not Perform Key Monitoring Tasks Designed to Ensure That Ongoing Accessibility to Providers Is Adequate Enhancing beneficiaries’ access to providers was a fundamental The department has not objective of the department’s conversion to managed care; adequately reviewed however, the department does not sufficiently monitor whether whether health plans the health plans are adequately ensuring that their members can have made available to readily obtain medical services because it does not review key its members a complete areas pertaining to ongoing accessibility of services. To ensure network of primary care that it achieves this objective, the department has included physicians and specialists. provisions in its contracts with the health plans requiring that they maintain an acceptable level of provider accessibility. When the department first contracted with the plans, it deter- mined members’ accessibility to the providers by performing various reviews, including calculating ratios of members and primary care physicians, determining geographical locations of primary care physicians, and reviewing the plans’ policies. 32 C A L I F O R N I A S T A T E A U D I T O R However, the five plans we reviewed have been under contract for two or more years, yet the department has not continued to monitor accessibility sufficiently. Procedures for ensuring that each health plan has a sufficient A combination of reports network of primary care physicians and specialists are just part on the changes to the of the system that the department has designed for ensuring provider network and the access to medical services. The department has various resources GeoAccess system can available to determine accessibility. Each quarter, the health provide needed data to plans are required to submit to the department for review assess the availability provider directories and reports describing the impact of of providers. changes to the provider network (impact reports). The impact reports furnish the department with useful information that enables the department to determine whether the health plan has provided sufficient access to medical services. The department can use the impact reports to validate thresholds for member-to-physician ratios and distances between members and primary care physicians. In addition, the department has a computer application, GeoAccess, which maps out the location of each provider to determine whether a health plan has a sufficient network of accessible primary care physicians by determining whether these physicians are located within acceptable distances of the members. When we reviewed whether the department monitored the plans during the first quarter of 1999 for a sufficient network of health care providers, we learned that in some cases, the depart- ment could not provide evidence that it reviewed updated provider directories. Therefore, the department could not ensure that members are relying on current, accurate information for contacting providers. In fact, for one of the plans it did review, the department found various discrepancies in the information listed in the provider directory. Additionally, the department did not ensure that each plan submitted required impact reports. Therefore, the department cannot verify that the plans’ provider networks sufficiently meet contract requirements, such as those governing specific ratios of members to physicians and time-and-distance standards. Finally, the department has not used GeoAccess in more than a year to monitor any of the five health plans. In fact, when we reviewed the department’s use of GeoAccess to determine provider accessibility when the plans first began operations, the department was unable to demonstrate the use of appropriate member and provider data in its analyses. The department’s analyses appear to have compared member and provider C A L I F O R N I A S T A T E A U D I T O R 33 locations from significantly different periods in time, thereby depicting relationships that may not have existed. Specifically, for four of the five health plans, the dates of each plan’s member and provider data files varied from 13 to 26 months apart. Therefore, because the department’s conclusions regarding initial accessibility were based on questionable comparisons, we believe that the need for monitoring ongoing accessibility of health care has become increasingly important. The Department Has Conducted Only a Limited Number of Site Reviews of the Health Plans Although the plan monitoring branch conducts focused site The department has reviews of health plans, it has performed only a limited number fallen far short of its goal of these reviews since the managed-care division’s reorganiza- to complete 24 site visits tion. The site reviews evaluate the plan’s performance pertaining per quarter. to specific contract requirements, such as initial health assess- ments and notices to members when services are denied, deferred, or modified, and the reviews can also evaluate other areas, such as the plan’s claims processing system. The plan monitoring branch’s goal is to conduct site visits of all plans every quarter, so this branch would target the 2 health plans in each of the 12 counties participating in the two-plan model and thus make 24 site visits per quarter. However, the branch has yet to meet this goal. In fact, since October 1998, the branch has only conducted 6 site visits and has not yet issued the review results to the plans for 2 of the 6 visits. In addition to these visits, the branch also conducted a special review of one plan’s comprehensive claims processing system. According to the acting assistant chief of the managed-care division, staffing vacancies limit the department’s ability to conduct all these reviews. 34 C A L I F O R N I A S T A T E A U D I T O R The Department Does Not Promptly Review Corrective Action Plans As mentioned earlier, Audits and Investigations performs medical and financial audits of health plans, and the department requires the plans to submit corrective action plans (CAPs) that address deficiencies identified during the audits. To ensure that the health plans correct these deficiencies, the department needs to review the CAPs promptly. However, the department has taken several months to evaluate some CAPs; in fact, it is still reviewing other CAPs received six or more months ago. As Figure 3 shows, the review and approval process for CAPs involves assessments from many department staff. FIGURE 3 Several Staff Members at the Department Review Corrective Action Plans (CAPs) Department of Health Services Plan Management Health Branch Plan CAP Plan CAP Review by Contract Monitoring Manager Branch 1. Analysis by Nurse Evaluator or analyst(s) Plan CAP 2. Review by Unit Chief Health Response 3. Review by Section Management Plan Chief Branch 4. Review by Branch AND Chief 1. Review by Contract Analysis Manager 2. Review by Unit Chief C A L I F O R N I A S T A T E A U D I T O R 35 The department has still not approved three of the five CAPs we reviewed even though the health plans submitted them to the department several months ago. For example, the department has yet to approve the CAP for the 1997-98 medical audit of the Contra Costa Health Plan even though the plan submitted the CAP in August 1998. In fact, the plan management branch, which initially receives CAPs, took three months to forward the CAP to the plan monitoring branch. Furthermore, even though the staff person in the plan monitoring branch reviewed the CAP and prepared a formal analysis, the respective management staff have yet to review the analysis. Similarly, the department has not approved the CAP for the fiscal year 1997-98 financial audit of Alameda Alliance for Health despite receiving the CAP in November 1998. Table 3 displays the review timetables for the five CAPs we examined. TABLE 3 The Department Has Not Reviewed and Approved Promptly Some Health Plans’ Corrective Action Plans (CAPs) Review Department CAP Received by CAP Sent From Returned to Communicated Department Plan Management Plan Results of Total (Plan Management Branch to Plan Management Review to Time Audit Branch) Monitoring Branch Branch Health Plan Elapsed* San Francisco June 1998 Dec 1998 May 1999 Requested 7 months Health Plan 1997 additional corrective medical audit action in January 1999 Approved CAP in March 1999 Alameda Alliance Sep 1998 Dec 1998 April 1999 Requested 8 months for Health 1997-98 additional corrective medical audit action in May 1999 Contra Costa Aug 1998 Nov 1998 April 1999 Pending as 9 months Health Plan 1997-98 of May 1999 medical audit Santa Clara Family Oct 1998 Dec 1998 March 1999 Pending as 7 months Health Plan 1997-98 of May 1999 medical audit Alameda Alliance Nov 1998 Dec 1998 Pending as Pending as 6 months for Health 1997-98 of May 1999 of May 1999 financial audit *We calculated the total review time from the month the department received the health plan’s CAP to the month the department first communicated review results to the plan. If the department had not communicated results to the health plan as of May 1999, we used that month as the end date for the calculation. 36 C A L I F O R N I A S T A T E A U D I T O R As indicated earlier, the fact that the review process involves many individuals appears to contribute to the delays in the department’s approving the CAPs. Additionally, some CAPs and analyses are held up in the review process for significant periods of time. For example, as Table 3 indicates, the plan management branch took from two to six months to forward four different CAPs to the plan monitoring branch. Similarly, the plan moni- toring branch took three to five months to analyze and review the CAPs before returning them to the plan management branch. According to the acting assistant chief of the managed- care division, staffing vacancies also limit the department’s ability to perform the CAP reviews regularly. He informed us that the department is in the process of recruiting new staff to assist in this function. The Department’s System for Addressing Grievances Is Missing a Significant Element The department’s system for addressing grievances is incomplete The system lacks a formal because it does not have a formal process for informing its process to inform monitoring units about trends in beneficiaries’ complaints and monitoring units of grievances against the health plans. Although the ombudsman trends in complaints and unit gathers data from the complaints it receives, the unit does grievances against health not prepare formal reports summarizing trends. The ombuds- plans. man unit previously prepared reports that compiled its data, but the unit discontinued this practice in October 1998 because of a failure in the computer system used to record the data. The chief of the department’s plan management branch stated that the earlier reports were not useful for developing monitoring trends because the reports presented raw data that did not consider the membership volume of the plans. The current chief of the ombudsman unit, who joined the unit in January 1999, informed us that the unit is in the process of reformatting the reports to address the needs of the plan man- agement branch. However, without formal trend reports, it is more difficult and time-consuming for the ombudsman unit to identify and communicate such trends as an unusually high rate of service denials by a particular plan. Thus, until it establishes a formal process for communicating trends in complaints and grievances, the department is missing a key element in its system for identifying potential unresolved problems within the plans. C A L I F O R N I A S T A T E A U D I T O R 37 SEVERAL FACTORS CONTRIBUTE TO INCOMPLETE MONITORING EFFORTS In our view, several factors have contributed to the department’s incomplete monitoring of health plans. First, the department never adopted a formal set of guidelines that its staff could follow in carrying out their monitoring responsibilities. Although the department initially prepared a monitoring manual, it never formally adopted this manual as the department’s approach for overseeing the health plans, nor did it update and implement the manual when it later reorganized the monitoring responsibilities among its various units. Further, some contract managers do not ensure that the department has received all required documents from the health plans. Unlike other monitoring staff, the department’s contract managers also do not consistently document their monitoring efforts. More- over, the department does not summarize its collective efforts to address the overall compliance of the health plans. In the absence of an organized approach, the department’s ability to manage its monitoring responsibilities properly is impaired. The Department Has Not Established Formal Expectations for Monitoring Activities That Staff Perform The department has not adopted formal guidelines directing the monitoring activities of its staff. When the division reorganized, the department split certain monitoring responsibilities between the plan management branch and the plan monitoring branch. For example, the contract managers within the plan manage- ment branch review documents submitted by the health plans, such as provider directories and corrective action plans. Likewise, the units in the plan monitoring branch review corrective action plans, and they also conduct site visits, review plan policies, and resolve member complaints through the Although the department ombudsman unit. However, the department has not imple- developed a draft mented procedure manuals describing the nature, extent, and monitoring manual in timing of monitoring activities for either branch. Although it May 1997, staff were developed a draft monitoring manual in May 1997 that never required to use it. described staff responsibilities, the department never required staff to use it. According to the chief of the plan management branch, even though the department has not adopted a formal procedure manual for contract managers to follow, he has implemented two procedures to assist his staff in their monitoring responsi- bilities. One procedure describes a filing system to organize the 38 C A L I F O R N I A S T A T E A U D I T O R department’s monitoring efforts. The other outlines steps for verifying the sufficiency of the provider network and the accu- racy of the provider directories. However, despite the fact that the chief has expressed his expectations that staff follow these procedures, we have seen no evidence that the contract manag- ers for the health plans we reviewed are using a uniform filing system. Furthermore, we observed significant differences in the We observed significant amount of monitoring performed by the contract managers in differences in the review their reviews of provider directories. For example, the contract of provider directories by managers varied in the number of providers they contacted to contract managers. verify the accuracy of information in the directories. Without communicating its expectations of an acceptable level of review, the department cannot be assured that its contract managers are sufficiently monitoring the health plans. The acting assistant chief of the managed-care division stated that the department initially created the draft monitoring manual under a different approach that emphasized monitoring health plans through a heavy reliance on site visits of providers. The current approach that he intends to use will focus on using data from a variety of sources to identify potential problems for specific review. These reviews will focus on each plan’s efforts to monitor its providers rather than on the site visits. Even so, the plan monitoring branch believes that certain sections of the manual are still applicable, and this branch is in the process of identifying and extracting such sections for its use. Contract Managers Do Not Consistently Ensure That Health Plans Submit All Required Documents The department’s contract managers do not always ensure that the health plans submit required documents even though the department needs these documents to verify that the plans comply with their contracts and meet the goals of managed care. Contract managers are responsible for ensuring that the health plans submit required documents, such as their policies, member materials, and corrective action plans. In addition, the contract manager forwards some of these documents to other units for review. At any given time, the department may require several documents from plans, and its various staff may be reviewing many other documents. Despite this level of activity, the contract managers for the five plans we reviewed did not formally track the status of the documents. Three of the five contract managers we spoke with informed us that they were aware of the status of certain required documents from their C A L I F O R N I A S T A T E A U D I T O R 39 respective health plans. However, the other two contract manag- ers were unaware that certain documents were overdue from their plans. Although the individual contract managers do not formally track documents, the plan management branch maintains a correspondence log tracking those documents that this branch has forwarded to the plan monitoring branch. In addition, one supervisor maintains a log that identifies specific tasks assigned to the contract managers. However, neither log identifies outstanding documents that have yet to be received from the plans. In order to ensure the integrity of contract compliance, each contract manager should maintain a tracking log specific to the health plan that lists the required documents and their current status in the department’s review process. In fact, the department developed a tracking log designed for use by the contract managers and included it in its draft monitoring manual. However, the department does not require its contract managers to use any type of log or other formal process to track key documents. Contract Managers Sometimes Fail to Document Their Monitoring Efforts Contract managers perform specific activities to ensure health plans’ compliance with contract requirements, such as reviewing plan policies to determine whether they conform to newly issued managed-care policy directives for Medi-Cal. However, some contract managers do not always prepare written summa- ries describing their efforts and the results of their analyses. The department documents in written reports the results of site visits and the outcomes of medical and financial audits, including any recommendations made during these reviews. However, the five contract managers we spoke with did not consistently prepare or maintain documentation of their efforts. For example, one contract manager maintained written notes on her review of the Contract managers health plan’s provider directory. However, another contract do not consistently manager informed us that he had reviewed another plan’s prepare or maintain provider directory but did not document his review. In the documentation of absence of written summaries, the department must rely on the their efforts. contract managers’ recollections of their monitoring efforts. In fact, two contract managers we spoke with informed us that they had taken over the management of their respective health 40 C A L I F O R N I A S T A T E A U D I T O R plans in October 1998 but had little knowledge of the activities performed by their predecessors. Therefore, the depart- ment is in a potentially difficult position of determining whether it is sufficiently overseeing certain areas of the health plans’ operations. The Department Does Not Summarize Its Collective Monitoring Efforts The department does not summarize the results of its collective Operational inefficiencies monitoring efforts of health plans to conclude on the overall result from a lack of a compliance of the health plans. The department initially devel- central record describing oped an outline of this type of summary and included it as part the results of the various of its proposed monitoring manual. The manual referred to this monitoring activities. summary as an important tool for evaluating vast amounts of information and making informed decisions on each plan’s overall compliance. However, the department’s current approach creates an incomplete picture of its monitoring efforts. Even though various monitoring activities are occurring throughout the year, the department does not maintain a central record summarizing the department’s findings of the health plans and its follow-up of the plans’ efforts to address corrective action. In the event that the department must defend its actions, it will likely spend additional time retracing its past activities in order to summarize the results of its collective efforts. By maintaining a central record, the department would have documented support for its judgments on whether plans are complying with requirements for Medi-Cal managed care. Also, the department could rely on this record when it prepares its semiannual monitoring status report to the Health Care Financing Administration. DIFFERING PERSPECTIVES CONTRIBUTE TO INEFFECTIVE MONITORING OF HEALTH PLANS Not only has the department failed to perform certain monitor- ing tasks, but its various divisions also do not agree on their respective roles in the monitoring process. Specifically, the department’s approach for monitoring health plans through medical reviews performed by its audits and investigations program (Audits and Investigations) is ineffective because the parties involved in this task have not reached consensus. Although the managed-care division and Audits and Investiga- tions communicate before and during audits, they are unable to C A L I F O R N I A S T A T E A U D I T O R 41 resolve differences of opinion regarding the role of Audits and Investigations. Audits and Investigations views itself as an independent evaluator of the managed-care plans’ contract compliance. Although the managed-care division believes that Audits and Investigations’ independence is appropriate and necessary for program integrity, it also believes that Audits and Investigations should consider input from managed care when developing and refining audit tools and scoring criteria. Further- more, the managed-care division believes that it, rather than Audits and Investigations, has the ultimate responsibility for interpreting contract requirements, policies, and regulations. These different perspectives seem to contribute to ineffective communication between the divisions regarding the audits. Specifically, Audits and Investigations believes its role is to report audit findings for those contract provisions to which the health plans do not strictly adhere. In contrast, the managed- care division believes that the findings should reflect a Audits and Investigations reasonable flexibility in evaluating health plans’ performance in and the managed-care meeting program standards and requirements. Because of this division disagree difference of opinion, the results of the medical reviews may not regarding the role reflect the goals of the managed-care division. We believe that of audits. this ineffective coordination between divisions has resulted in audit findings that add questionable value to the department’s overall monitoring efforts. Furthermore, the department spent additional resources generating these audit findings and reviewing the resulting corrective action plans. The lack of consensus between divisions is evident in certain findings and recommendations. For example, Audits and Inves- tigations reviews health plans to ensure that individual health education behavioral assessments are performed within 120 days of beneficiaries’ enrollment. In October 1996, the managed-care division wrote to the plans postponing implementation of the assessments until it developed a standardized assessment tool that all the plans could use. Although the managed-care division maintains that the postponement letter clearly communicates the desire to waive this contract provision, Audits and Investigations informed us that the divisions have a formal process for resolving differences in policy interpretation. Meanwhile, Audits and Investigations continues to audit the plans to determine whether behavioral assessments have been performed. 42 C A L I F O R N I A S T A T E A U D I T O R In another situation, Audits and Investigations concluded that Although the managed- several plans did not comply with the contract requirement to care division believes that provide initial health assessments to their new members even the department should though Audits and Investigations concluded that these plans acknowledge a plan’s followed adequate procedures to encourage the members to efforts to comply with obtain assessments. Audits and Investigations determines com- contract requirements, pliance with this provision by selecting a sample of member Audits and Investigations medical records and identifying those documenting initial continues to interpret the assessments. However, the evaluation does not account for a requirements more member who never scheduled an appointment. Without an narrowly. appointment, a plan is unable to provide the assessment. Although Audits and Investigations concluded that some health plans did not comply with the contract requirements for initial health assessments—despite factors outside of the plans’ control—the managed-care division believes the department should consider each plan’s outreach efforts when evaluating its compliance. The managed-care division and Audits and Investigations are working to improve their communication. They recently amended their memorandum of understanding to include a process to work out disagreements during the audit process. Additionally, a work group composed of staff from both divisions meets periodically to discuss issues, and the group is currently developing a format to streamline the process for clarifying policy interpretations. However, until the managed- care division and Audits and Investigations agree on their respective roles, communication during the audit process will continue to suffer and reduce the effectiveness of the department’s audits. RECOMMENDATIONS In order to ensure that it adequately monitors the health plans under Medi-Cal managed care, the Department of Health Services should take the following steps: • Implement formal standards for monitoring health plans that describe the department’s expectations for various tasks, such as evaluations of the existing provider networks for health plans, CAP reviews, site reviews of health plans, and the communication of trends pertaining to members’ grievances. These expectations should describe the nature, extent, and timing of the monitoring tasks. Staff should have the C A L I F O R N I A S T A T E A U D I T O R 43 flexibility to adjust its level of monitoring when justified, but staff should document its reasons for deviating from the department’s expectations. • Develop a tracking tool that will better enable its contract managers to assess whether the health plans have submitted all reports required by the department. The tracking tool can also assist the department in ensuring that its staff have promptly reviewed the reports. • Require its contract managers to prepare written documenta- tion describing their monitoring efforts. This documentation will assist the department in holding its contract managers accountable for their responsibilities. Also, the existence of written documentation facilitates the transfer of responsibili- ties when a contract manager takes over for another manager. • Maintain an ongoing record for each health plan that encap- sulates the results of the department’s overall monitoring efforts. This record would also identify any corrective actions not yet taken by the plans. • Coordinate efforts between the managed-care division and Audits and Investigations to ensure consensus on Audits and Investigations’ role in performing audits. At the same time, both divisions should continue efforts to resolve differences in perspectives to ensure that audits directly address the expectations of the managed-care division, including the department’s reporting only those deficiencies within the health plans’ control. 44 C A L I F O R N I A S T A T E A U D I T O R We conducted this review under the authority vested in the California State Auditor by Section 8543 et seq. of the California Government Code and according to generally accepted government auditing standards. We limited our review to those areas specified in the audit scope section of this report. Respectfully submitted, KURT R. SJOBERG State Auditor Date: July 14, 1999 Staff: Steven Hendrickson, Audit Principal Linus A. Li, CPA, CMA Debra Maus, CPA Vince J. Blackburn, Esq. Deborah M. Ciarla Theresa Gartner Alan Ma C A L I F O R N I A S T A T E A U D I T O R 45 Blank page inserted for reproduction purposes only. 46 C A L I F O R N I A S T A T E A U D I T O R APPENDIX The Department of Health Services’ Methodology for Developing Capitation Rates T he Department of Health Services (department) makes monthly capitation payments to health plans under managed care for the California Medical Assistance Program (Medi-Cal). Capitation consists of a predetermined fixed payment per health plan member per month for which the health plan agrees to supply a defined set of medical services. Although we did not evaluate the source or validity of the data used to develop the capitation rates, we reviewed the rate meth- odology and believe it considers relevant factors. For the Medi-Cal Managed Care Two-Plan Model (two-plan model), capitation rates are constrained by federal regulation, which limits the capitation payments to an upper limit that represents no more than the cost of medical care for the same services for a comparable population under a fee-for-service system. The department determines the limit using fee-for- service claims for services delivered to the State’s Medi-Cal population in all but 6 counties. From these claims, the department calculates a base cost for these services for groups of members, called aid code groupings, and adjusts the statewide base costs to reflect the two-plan model population in the 12 counties. Adjustments include factors such as age and sex demographics, eligibility, legislative mandates, trends in costs and utilization of services, and administrative cost savings to the State. The department multiplies these fee-for-service equivalents by the projected number of eligible members in the associated aid code grouping and then adds these individual amounts to arrive at the total amount available for capitation payments to the health plans. Once it determines the limit, the department calculates separate capitation rates for each local initiative and commercial plan. The department uses managed-care data from the Santa Barbara Health Authority, a county organized health system that the department has determined is a well-managed health plan with the best data available. The department adjusts the average costs determined from the data for each plan to account for C A L I F O R N I A S T A T E A U D I T O R 47 demographic differences in the population, varying service provider costs, and trends in costs and utilization of services. Next, it multiplies the resulting total adjusted cost per member by the projected number of eligible beneficiaries for each aid code grouping and adds these amounts to obtain the total capitation costs. The department then adjusts this sum downward, if necessary, to ensure that the capitation costs do not exceed the limit. For fiscal year 1998-99, it set the aggregate value of the rates at a level equal to 94 percent of the limit. In a January 1999 letter to the health plans in the two-plan model, the department communicated its decision to discount the fiscal year 1998-99 capitation rates, citing concerns for the applicability of fee-for-service base costs and the State’s budget- ary constraints. The letter also stated that the plans’ average rate increase for fiscal year 1998-99 is 4.6 percent over the previous year’s rates. We asked the department to clarify further the budget con- straints and adjustment factors referred to in its letter. The department told us that several factors led to the capitation rate calculation. The department is concerned about the long-term viability of the fee-for-service base costs as more of the Medi-Cal population moves into managed care, thus reducing the fee-for-service population base. Furthermore, preliminary rate calculations resulted in some plans potentially receiving rate increases that were considerably higher than other health care market increases and cost-of-living increases. The department therefore adjusted the capitation rates, taking into consideration certain significant cost items such as higher pharmaceutical costs, additional hospital inpatient days, and legislative increases for provider rates. 48 C A L I F O R N I A S T A T E A U D I T O R RR--11 R-2 RR--33 R-4