This is the accessible text file for GAO report number GAO-08-359 entitled 'Medicare Advantage: Increased Spending Relative to Medicare Fee-for-Service May Not Always Reduce Beneficiary Out-of-Pocket Costs' which was released on February 28, 2008. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to Congressional Requesters: United States Government Accountability Office: GAO: February 2008: Medicare Advantage: Increased Spending Relative to Medicare Fee-for-Service May Not Always Reduce Beneficiary Out-of-Pocket Costs: GAO-08-359: GAO Highlights: Highlights of GAO-08-359, a report to congressional requesters. Why GAO Did This Study: In 2006, the federal government spent about $59 billion on Medicare Advantage (MA) plans, an alternative to the original Medicare fee-for- service (FFS) program. Although health plans were originally envisioned in the 1980s as a potential source of Medicare savings, such plans have generally increased program spending. Payments to MA plans have been estimated to be 12 percent greater than what Medicare would have spent in 2006 had MA beneficiaries been enrolled in Medicare FFS. Some policymakers are concerned about the cost of the MA program and its contribution to overall spending on the Medicare program, which already faces serious long-term financial challenges. MA plans receive a per member per month (PMPM) payment to provide services covered under Medicare FFS. Almost all MA plans receive an additional Medicare payment, known as a rebate. Plans use rebates and sometimes additional beneficiary premiums to fund benefits not covered under Medicare FFS, reduce premiums, or reduce beneficiary cost sharing. This report examines for 2007 (1) MA plans’ projected rebate allocations; (2) additional benefits MA plans commonly covered and their costs; (3) MA plans’ projected cost sharing; and (4) MA plans’ allocation of projected revenues and expenses. GAO analyzed data on MA plans’ projected revenues and covered benefits for the most common types of MA plans, accounting for 71 percent of all beneficiaries in MA plans. What GAO Found: In 2007, plans projected that relatively little of their rebates would be spent on additional benefits compared to cost-sharing and premium reductions. Of the average projected rebate amount of $87 PMPM, plans projected they would allocate about $10 PMPM (11 percent) to additional benefits, about $61 PMPM (69 percent) to reduced cost sharing, and about $17 PMPM (20 percent) to reduced premiums. Using funding from both rebates and additional premiums, plans covered a variety of additional benefits not covered by Medicare FFS in 2007, including dental and vision benefits. On the basis of plans’ projections, GAO estimated that rebates would pay for approximately 77 percent of additional benefits and additional beneficiary premiums would pay for the remaining 23 percent. MA plans projected that, on average, beneficiaries in their plans would have lower cost sharing than Medicare FFS cost-sharing estimates, although some MA plans projected that their beneficiaries would have higher cost sharing for certain service categories, such as home health care and inpatient services. Because cost sharing was projected to be higher for some categories of services, beneficiaries who frequently used these services could have had overall cost sharing that would be higher than under Medicare FFS. On average, MA plans projected that they would allocate about 87 percent of total revenue ($683 of $783 PMPM) to medical expenses; approximately 9 percent ($71 PMPM) to non-medical expenses, including administration, marketing, and sales; and approximately 4 percent ($30 PMPM) to the plans’ margin, sometimes called the plans’ profit. About 30 percent of beneficiaries were enrolled in plans that projected they would allocate less than 85 percent of their revenues to medical expenses. Whether the value that MA beneficiaries receive in the form of reduced cost sharing, lower premiums, and additional benefits is worth the additional cost is a decision for policymakers. However, if the policy objective is to subsidize health care costs of low-income Medicare beneficiaries, it may be more efficient to directly target subsidies to a defined low-income population than to subsidize premiums and cost sharing for all MA beneficiaries, including those who are well off. As Congress considers the design and cost of MA, it will be important for policymakers to balance the needs of beneficiaries and the necessity of addressing Medicare’s long-term financial health. In commenting on a draft of this report, the Centers for Medicare & Medicaid Services expressed concern that the report was not balanced because it did not sufficiently focus on the advantages of MA plans. GAO disagrees. This report provides information on how plans projected they would use rebates and identified instances in which MA beneficiaries could have out-of-pocket costs higher than they would have experienced under Medicare FFS. To view the full product, including the scope and methodology, click on [hyperlink, http://www.GAO-08-359]. For more information, contact James Cosgrove at (202) 512-7114 or cosgrovej@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: MA Plans Projected That They Would Allocate Relatively Little of Their Rebates to Additional Benefits and the Majority to Reduced Cost Sharing: MA Plans Used Rebates and Additional Premiums to Cover Additional Benefits Such as Dental, Hearing, and Vision: MA Plans Projected That MA Beneficiaries, on Average, Would Have Lower Cost Sharing Than if They Were in Medicare FFS, but Some MA Beneficiaries Could Pay More: Approximately 87 Percent of Total Revenue Projected to Be Allocated to Medical Expenses, but Projections Varied among Individual Plans: Concluding Observations: Agency and Other External Comments and Our Evaluation: Appendix I: Example of a Rebate Calculation: Appendix II: Scope and Methodology: Appendix III: Plan Variation in Rebate Amounts: Appendix IV: Plan Variation in the Out-of-Pocket Maximum: Appendix V: Comments from the Centers for Medicare & Medicaid Services: Appendix VI: GAO Contact and Staff Acknowledgments: Tables: Table 1: Rebate Amount PMPM Allocated to Additional Benefits, Premium Reductions, and Cost-Sharing Reductions by Plan Type, 2007: Table 2: Percentage of Beneficiaries in Plans That Charge an Additional Premium and Average Amount of Additional Premium by Plan Type, 2007: Table 3: Average Projected PMPM Costs of Additional Benefits by Service Category and Plan Type for Plans That Offered Benefits and Reported Costs, 2007: Table 4: Beneficiaries in MA Plans with Higher Projected Cost Sharing Than Medicare FFS for a Given Service Category by Plan Type, 2007: Table 5: MA Plans That Exclude Some Services under a Service Category from Their Out-of-Pocket Maximum: Table 6: The Calculation of the Rebate for Two Hypothetical MA Plans: Table 7: Rebate Amount Allocated to Additional Benefits, Premium Reductions, and Cost-Sharing Reductions by Plan Type, 2007: Table 8: Variation in Values of Out-of-Pocket Maximum by Plan Type, 2007: Figures: Figure 1: Projected Rebate Allocation to Additional Benefits, Premium Reductions, and Cost-Sharing Reductions by Plan Type, 2007: Figure 2: Percentage of Beneficiaries in Plans Covering Additional Benefits by Plan Type, 2007: Figure 3: Average Projected Cost Sharing for MA Beneficiaries Compared to Their Cost Sharing in Medicare FFS, by Plan Type, 2007: Figure 4: Example of an MA Plan with Inpatient Cost Sharing Different from the Medicare FFS Program: Figure 5: Beneficiaries in MA Plans by Out-of-Pocket Maximum Amount and Plan Type, 2007: Figure 6: Percentage of Beneficiaries in MA Plans That Project Allocating Less Than 85 Percent of Total Revenues to Medical Expenses, by Plan Type, 2007: Figure 7: MA Plans' Projected Marketing and Sales Expenses by Plan Type, 2007: Abbreviations: AHIP: America's Health Insurance Plans: CHAMP Act: Children's Health and Medicare Protection Act of 2007: CMS: Centers for Medicare & Medicaid Services: FFS: fee-for-service: HMO: Health Maintenance Organization: MA: Medicare Advantage: MedPAC: Medicare Payment Advisory Commission: MMA: Medicare Prescription Drug, Improvement, and Modernization Act of 2003: MSA: Medical Savings Account: PFFS: Private Fee-for-Service: PMPM: per member per month: PPO: Preferred Provider Organization: PSO: Provider-Sponsored Organization: SNP: Special Needs Plan: [End of section] United States Government Accountability Office: Washington, DC 20548: February 22, 2008: Congressional Requesters: In 2006, the federal government spent an estimated $59 billion on the Medicare Advantage (MA) program, an alternative to the original Medicare fee-for-service (FFS) program.[Footnote 1] The MA program provides health care coverage to Medicare beneficiaries through private health plans, referred to as MA plans. As of August 2007, 8.1 million people--about one out of every five Medicare beneficiaries--were enrolled in an MA plan. Although private health plans were originally envisioned in the 1980s as a potential source of Medicare savings, such plans have generally increased overall program spending. Medicare spending on private health plans has increased rapidly since the enactment of the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (MMA),[Footnote 2] rising 64 percent from 2004 to 2006, while enrollment has increased by more than 50 percent. The MMA increased payment rates for private health plans and allowed for larger annual rate increases, among other things.[Footnote 3] These payment increases enabled MA plans to spend more money on additional benefits relative to those available under Medicare FFS, such as vision and hearing coverage; reductions in cost sharing--the amount a beneficiary pays for covered services; and reductions in the premiums that many Medicare FFS beneficiaries pay for coverage of outpatient services and outpatient drugs. Beginning in 2006, MA plans were required to submit bids for providing Medicare-covered services. MA plans that submitted bids below predetermined benchmarks received additional payments, known as rebates, and were required to spend their rebates on additional benefits, reduced cost sharing, reduced premiums, or a combination of the three. As the MA program has grown, some policymakers and congressional advisors have raised concerns about the design and cost of the program as well as its effect on overall Medicare spending. The Medicare Payment Advisory Commission (MedPAC) found that payments to MA plans in 2006 exceeded by 12 percent what Medicare would have paid had MA beneficiaries received services through Medicare FFS.[Footnote 4] The Congressional Budget Office estimated that $54 billion in projected Medicare spending from 2009 through 2012 is the result of setting MA plan payments above Medicare FFS spending.[Footnote 5] MA plans' payments thus place an additional financial burden on the Medicare program, which the Comptroller General and others have noted already faces serious long-term financial challenges resulting from rising health care costs and the retirement of the baby boom generation.[Footnote 6] Proponents of the MA program assert that the current level of MA plan payments has allowed plans to offer valuable additional benefits and make health care services more affordable for beneficiaries, particularly in rural areas where private plan options had been very limited. Further, they note that the MA program provides beneficiaries with private plan choices and enables them to select plans that reflect their preferences for premiums and cost sharing. They also point out that individuals with low incomes who do not qualify for other government health care coverage may receive some financial relief by enrolling in an MA plan. Critics of the current MA program suggest that if the policy objective is to subsidize the health care of individuals with low incomes, it would be more efficient to directly target subsidies to a well-defined low-income population instead of subsidizing the health care costs of all MA beneficiaries. Program critics also assert that a large portion of the additional payments to MA plans goes to profit and administrative costs and that some MA beneficiaries face higher cost sharing than they would if they received coverage through Medicare FFS. Questions have also been raised that while the MA program provides beneficiaries with many health plan choices, it can be difficult for even a sophisticated buyer to understand the implications of different cost-sharing arrangements. In addition, some policymakers are concerned that because premiums paid by beneficiaries in Medicare FFS are tied to both Medicare FFS and MA program spending, the excess payments to MA plans result in higher premiums for all Medicare beneficiaries. Medicare pays MA plans a per member per month (PMPM) amount that is based on a plan's bid--its projection of the revenue it requires to provide a beneficiary with services that are covered under Medicare FFS, and a benchmark--the maximum amount Medicare will pay the plan to serve an average beneficiary. Benchmarks vary by county, and in 2007, every county in the United States had a benchmark that was at least as high as average Medicare FFS spending PMPM in that county. If the plan's bid is higher than the benchmark, Medicare pays the plan the amount of the benchmark, and the plan must charge beneficiaries a premium to collect the amount by which the bid exceeds the benchmark.[Footnote 7] If the plan's bid is lower than the benchmark, Medicare pays the plan the amount of the bid and makes an additional rebate payment to the plan equal to 75 percent of the difference between the benchmark and the bid. Plans use the rebate to provide their beneficiaries with additional benefits beyond those offered in Medicare FFS, reduce premiums, reduce cost sharing, or any combination of the three. In 2007, the total amount of rebates paid to MA plans was about $8.3 billion. (See app. I for more information about how rebates are calculated.) Regardless of whether a plan's bid is above or below the benchmark, a plan may charge its beneficiaries an additional premium to provide additional benefits or reductions in cost sharing that are not otherwise financed by rebates.[Footnote 8] Given the additional spending--including rebates--for the MA program, you asked that we undertake a study on MA plans' rebates, benefit packages, and revenues. This report examines for 2007 (1) how MA plans projected they would allocate the rebates they receive, (2) what additional benefits MA plans commonly covered with the rebates and additional premiums and the projected costs of these additional benefits, (3) how MA plans' projected beneficiary cost sharing overall and by type of service compared to Medicare FFS, and (4) how MA plans projected they would allocate their revenue to medical and other expenses. We used two primary data sources in our analyses, the 2007 Bid Pricing Tool data and the 2007 Plan Benefit Package data that MA plans submitted to the Centers for Medicare & Medicaid Services (CMS), the agency that administers Medicare. The bid pricing data contain MA plans' projections of their revenue requirements and revenue sources. Specifically, the bid pricing data contain information on the amount of rebates and additional premiums plans project they will require to fund additional benefits, reduced premiums, and reduced cost sharing. The bid pricing data also contain information about how plans' projected cost sharing compared to estimates of cost sharing in Medicare FFS and plans' projections of revenue requirements--spending on medical expenses, spending on non-medical expenses (such as marketing, sales, and administration) and their margins.[Footnote 9] The benefit package data contain detailed information on the benefits and cost-sharing arrangements of plans. We analyzed bid pricing data and benefit package data from four different plan types, which together account for 98 percent of MA enrollment--including Health Maintenance Organizations (HMO), Private Fee-for-Service (PFFS) Plans, Preferred Provider Organizations (PPO), and Provider-Sponsored Organizations (PSO).[Footnote 10] Because there were only 22 PSOs and enrollment in those plans was only 1 percent of total MA enrollment, we did not report results separately for PSOs, but included them in the aggregated results we reported for all MA plans. We excluded plans that have restrictions on enrollment--such as employer plans and Special Needs Plans (SNP)--and bids for plans that only cover certain Medicare FFS services.[Footnote 11] We also excluded plans with service areas that are exclusively outside the 50 states and the District of Columbia. After all exclusions, we had 2,055 plans in our study that accounted for 71 percent of all beneficiaries in MA plans. Unless otherwise noted, the analyses were based on these 2,055 plans and their beneficiaries. To address our study questions, we did the following: * To determine how plans projected they would allocate the rebates they receive, we used the bid pricing data. We applied the proportion of the combined rebate and additional premium allocated to additional benefits, reduced premiums, and reduced cost sharing to the projected total. We restricted this analysis to those plans that received a rebate--1,874 of the 2,055 plans. * To identify the additional benefits MA plans commonly covered with rebates and additional premiums, and the projected costs of these additional benefits, we analyzed both the benefit package and bid pricing data. We used the benefit package data to identify the additional benefits plans covered and used the bid pricing data to identify the projected cost of these additional benefits. When we analyzed the projected cost of additional benefits, we included both the rebate payments and additional premiums. We included rebates and additional premiums, rather than solely considering the effects of rebates, because rebates and premiums together fund the additional benefits that MA beneficiaries will receive. If we had estimated the cost of additional benefits funded only by the rebates, that amount would have been lower than the amount we report. * To compare projected beneficiary cost sharing in the MA and Medicare FFS programs, we used both the bid pricing and the benefit package data. We used the bid pricing data to quantify the projected cost- sharing reduction, using the plans projections of the average cost- sharing expenditure on a PMPM basis, and compared this to CMS estimates of what the average PMPM cost-sharing expenditure would be in Medicare FFS. To obtain details on the specific cost-sharing arrangements used by the plans, we used the benefit package data. As was the case for our analysis of additional benefits, the amounts we reported for average PMPM cost sharing and cost-sharing reductions were based on the amounts projected by the plans and included funding from both rebates and additional premiums. If we had estimated the amount of cost sharing funded only by the rebates, the PMPM cost-sharing amounts would have been higher and the cost-sharing reduction amounts would have been lower. * To identify how plans projected they would allocate their revenue to medical and other expenses, we used the bid pricing data. Throughout the report, dollar amounts are adjusted to reflect a beneficiary of average health status. Where noted, we used August 2007 MA plan enrollment numbers to weight our results. To determine the reliability of the bid pricing, benefits, and enrollment data, we spoke with CMS officials about the strengths and limitations of these data sets. We also conducted logic tests to ensure that the bid pricing data were reasonable and consistent, and compared the bid pricing and benefits data to ensure consistency, where applicable, across the data sets. In some cases, there were discrepancies between the two data sources. For example, some plans indicated that they had an additional benefit in the benefit package data, but did not price that additional benefit in the bid pricing data. CMS officials indicated that these discrepancies could be due, in part, to the different purposes of the benefit package and bid pricing data sets, and resulting different benefit categorizations. CMS officials said discrepancies may also be the result of some plans with low projected amounts for additional benefits categorizing those benefits as Medicare-covered services, or the bid pricing data may accurately reflect low projected prices that round to zero. In general, based on CMS's recommendations, we used the benefit package data as the most reliable data source for identifying specific benefits covered by plans, and used the bid pricing data to identify costs. We determined that the data used were sufficiently reliable for the purposes of this report. However, verifying that the projections presented in the bid pricing data actually reflect plan revenues and expenditures was beyond the scope of our work. See appendix II for more details on our scope and methodology. We conducted our work from April 2007 through February 2008 in accordance with generally accepted government auditing standards. Results in Brief: In 2007, MA plans that received rebates projected that relatively little of the rebates would be spent on additional benefits compared to cost-sharing and premium reductions. Of the average projected rebate amount of $87 PMPM, plans projected that they would allocate about $10 PMPM (11 percent) to additional benefits, about $61 PMPM (69 percent) to reduced cost sharing, and about $17 PMPM (20 percent) to reduced premiums. Using funding from rebates, additional premiums, or both, plans covered a variety of additional benefits in 2007, including dental, hearing, and vision benefits. The average projected PMPM costs of specific additional benefits across all MA plans ranged from $0.11 PMPM for international outpatient emergency services to $4 PMPM for dental care. On the basis of plans' projections, we estimated that rebates would pay for approximately 77 percent of these additional benefits, and additional beneficiary premiums would pay for the remaining 23 percent. MA plans projected that, on average, beneficiaries in their plans would pay less in cost sharing than what their cost sharing would be in the Medicare FFS program, although some MA plans projected that their beneficiaries would have higher cost sharing for certain service categories. For example, 19 percent of MA beneficiaries were in plans that projected higher cost sharing for home health services and 16 percent of beneficiaries were in plans that projected higher cost sharing for inpatient services. Because cost sharing was projected to be higher for some categories of services, beneficiaries who frequently used these services could have had overall cost sharing that would be higher than under Medicare FFS. Similar to payments for additional services, we estimated that rebates would pay for about 77 percent of the cost-sharing reduction and the remainder would be paid for with additional beneficiary premiums. Plans' total revenues in 2007 were $783 PMPM, on average, of which plans projected they would allocate approximately 87 percent ($683 PMPM) to medical expenses--referred to as a medical loss ratio of 0.87. In addition, they projected that they would allocate approximately 9 percent of total revenue ($71 PMPM) to non-medical expenses, and approximately 4 percent ($30 PMPM) to the plans' margin--sometimes called a profit. About 30 percent of beneficiaries were enrolled in plans with a medical loss ratio of less than 0.85. Medicare spends more per beneficiary in the MA program than it does for beneficiaries in Medicare FFS, at an estimated additional cost to Medicare of $54 billion from 2009 through 2012. MA beneficiaries generally, but not always, receive additional value in the form of reduced cost sharing, lower premiums, and extra benefits, compared to Medicare FFS beneficiaries. Whether the additional value that MA beneficiaries receive is worth the additional cost to Medicare FFS beneficiaries and other taxpayers is a decision for policymakers. If the policy objective is to subsidize health care costs of low-income Medicare beneficiaries, it may be more efficient to directly target subsidies to a defined low-income population than to subsidize premiums and cost sharing for all MA beneficiaries, including those who are well off. As Congress considers the design and cost of the MA program, it will be important for policymakers to balance the needs of MA beneficiaries and Medicare FFS beneficiaries with the necessity of addressing Medicare's long-term financial health. In commenting on a draft of this report, CMS stated that we did not consider that the majority of MA benefit packages in 2007 were better than Medicare FFS and expressed concern that the report was not balanced because it did not sufficiently focus on the advantages of MA plans. They also noted that while they did not disagree with our finding that some beneficiaries in MA plans could have higher out-of- pocket costs, we did not recognize certain factors that would have mitigated the impact of the finding. We disagree with CMS. Specifically, we recognized in the report that, on average, plans projected MA beneficiary cost sharing that was 42 percent of estimated cost sharing in Medicare FFS. Our report provides an assessment of how MA plans projected they would use their rebates in 2007, and identified important issues related to cost sharing. America's Health Insurance Plans (AHIP) indicated that they agreed with our methodology, but raised certain points that they thought the report should have made or emphasized. We added these points to the report as appropriate. Background: MA plans are required to cover benefits that are covered under the Medicare FFS program.[Footnote 12] Medicare FFS consists of Part A; hospital insurance--which covers inpatient stays, care in skilled nursing facilities, hospice care, and some home health care, and Part B, which covers certain physician, outpatient hospital, and laboratory services, among other services. Persons aged 65 and older who meet Medicare's work requirement, certain individuals with disabilities, and most individuals with end-stage renal disease receive coverage for Part A services and pay no premium.[Footnote 13] Individuals eligible for Part A can also enroll in Part B, although they are charged a Part B premium.[Footnote 14] For 2007, the monthly Part B premium was set at $93.50, although high-income beneficiaries paid more. Most Medicare beneficiaries who are eligible for Medicare FFS can choose to enroll in the MA program instead of Medicare FFS.[Footnote 15] MA plans operate under Medicare Part C. All Medicare beneficiaries, regardless of their source of coverage, can choose to receive prescription drug coverage through Medicare Part D. Medicare FFS beneficiaries can enroll in stand-alone prescription drug plans, which are operated by private plan sponsors, and they generally must pay a premium to receive Part D coverage. MA beneficiaries who opt for prescription drug coverage generally receive that coverage through their MA plans, which may or may not charge an additional premium for Part D coverage. Beneficiaries enrolled in a PFFS plan that does not offer Part D coverage are allowed to enroll in a stand-alone prescription drug plan. Beneficiaries in both Medicare FFS and MA face cost-sharing requirements for medical services. Cost sharing gives beneficiaries a financial incentive to be mindful of the costs associated with using services. Medicare FFS cost sharing takes different forms. It includes both a Part A and a Part B deductible, which is the amount a beneficiary pays for services before Medicare FFS begins to pay. For 2007, Medicare FFS required a deductible payment of $992 before it began paying for an inpatient stay, and $131 before it began paying for any Part B services. Cost sharing also includes coinsurance--a percentage payment for a given service that a beneficiary must pay, such as 20 percent of the total payment for physician visits, and copayments--a standard amount a beneficiary must pay for a medical service, such as $248 per day for days 61 through 90 of an inpatient stay in 2007. Medicare allows MA plans to have cost-sharing requirements that are different from Medicare FFS's cost-sharing requirements. Plans may require more or less cost sharing than Medicare FFS for a given service, although, on average, a plan cannot require overall cost sharing that exceeds what beneficiaries would be expected to pay under Medicare FFS. MA plans may establish dollar limits on the amount a beneficiary spends on cost sharing in a year of coverage. In contrast, Medicare FFS has no total cost-sharing limit.[Footnote 16] Plans can use both out-of-pocket maximums, limits that can apply to all services but can exclude certain service categories, and service-specific maximums, limits that apply to one service category. These limits help provide financial protection to beneficiaries who might otherwise have high cost-sharing expenses. CMS officials said that they evaluate the cost-sharing arrangements of MA plans to determine if cost sharing is too high for services likely to be used by a beneficiary with below average health status. According to CMS officials, in 2007, if an MA plan (1) had no out-of-pocket maximum, (2) had an out-of-pocket maximum above $3,100, or (3) had an out-of-pocket maximum of $3,100 or below and excluded certain categories of service from that maximum, CMS compared the plan's cost sharing for certain service categories to thresholds that CMS based on Medicare FFS cost-sharing levels.[Footnote 17] If a plan exceeded one or more thresholds, CMS may have sought to negotiate with the plan over its cost sharing. According to CMS officials, the decision to negotiate was based on various factors, including the extent to which the thresholds were exceeded, local market comparisons, and the extent to which high cost sharing in one category was balanced with low cost sharing in another.[Footnote 18] MA Plans Projected That They Would Allocate Relatively Little of Their Rebates to Additional Benefits and the Majority to Reduced Cost Sharing: MA plans that received rebates projected, on average, that their rebates would be $87 PMPM. The plans projected that they would allocate a relatively small amount to additional benefits, compared to cost- sharing and premium reductions. Plans projected that, on average, about 11 percent of their rebates would be allocated to additional benefits, 69 percent to reduced cost sharing, 17 percent to Part D premium reductions, and 3 percent to Part B premium reductions. The average projected rebate allocation to additional benefits and reduced premiums varied by plan type. For example, PPOs projected that they would allocate less to Part D premium reductions and more to additional benefits than other plan types. PFFS plans projected that they would allocate less to additional benefits than other plan types. (See fig. 1) Figure 1: Projected Rebate Allocation to Additional Benefits, Premium Reductions, and Cost-Sharing Reductions by Plan Type, 2007: [See PDF for image] This figure is a stacked vertical bar graph depicting the following information: Plan type: HMO [plans: 1,179; beneficiaries: 3,747,087]; Percentage of rebate, Cost-sharing reduction: 69%; Percentage of rebate, Part B premium reduction: 2%; Percentage of rebate, Part D premium reduction[B]: 18%; Percentage of rebate, Additional benefits: 12%; Total: 100%. Plan type: PFFS [plans: 367; beneficiaries: 1,361,668]; Percentage of rebate, Cost-sharing reduction: 73%; Percentage of rebate, Part B premium reduction: 5%; Percentage of rebate, Part D premium reduction[B]: 14%; Percentage of rebate, Additional benefits: 8%; Total: 100%. Plan type: PPO [plans: 306; beneficiaries: 268,460]; Percentage of rebate, Cost-sharing reduction: 73%; Percentage of rebate, Part B premium reduction: 2%; Percentage of rebate, Part D premium reduction[B]: 8%; Percentage of rebate, Additional benefits: 16%; Total: 100%. Plan type: All plans[A] [plans: 1,874; beneficiaries: 5,454,573]; Percentage of rebate, Cost-sharing reduction: 69%; Percentage of rebate, Part B premium reduction: 3%; Percentage of rebate, Part D premium reduction[B]: 17%; Percentage of rebate, Additional benefits: 11%; Total: 100%. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Percentages may not sum to 100 due to rounding. Percentages are weighted by August 2007 plan enrollment. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. This analysis includes only the 1,874 plans that received a rebate. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [B] Of the 1,874 plans that received a rebate, 1,423 offered Part D benefits to their beneficiaries. Of those that offered Part D, 1,037 reduced Part D premiums. [End of figure] In dollar terms, the average projected rebates varied by plan type, from $55 PMPM for PPOs to $93 PMPM for HMOs. The dollar portions of the rebates that plans allocated to cost sharing varied, reflecting the variation in the average amount of the rebate. For example, on average, both PFFS plans and PPOs projected that they would allocate 73 percent of their rebate to cost-sharing reductions, but PFFS plans projected this would average $51 PMPM while PPOs projected this would average $41 PMPM.[Footnote 19] (See table 1.) For more information on the variation in how plans allocated rebates and the rebate amounts, see appendix III. Table 1: Rebate Amount PMPM Allocated to Additional Benefits, Premium Reductions, and Cost-Sharing Reductions by Plan Type, 2007: Rebate average: HMO Plans = 1,179, Beneficiaries = 3,747,087: $93.29; PFFS Plans = 367, Beneficiaries = 1,361,668: $70.06; PPO Plans = 306, Beneficiaries = 268,460: $55.26; All plans[A] Plans = 1,874, Beneficiaries = 5,454,573: $87.44. Amount of rebate allocated to: Additional benefits[B]; HMO Plans = 1,179, Beneficiaries = 3,747,087: $11.36; PFFS Plans = 367, Beneficiaries = 1,361,668: $5.58; PPO Plans = 306, Beneficiaries = 268,460: $9.08; All plans[A] Plans = 1,874, Beneficiaries = 5,454,573: $9.95. Amount of rebate allocated to: Part D premium reduction[C]; HMO Plans = 1,179, Beneficiaries = 3,747,087: $16.35; PFFS Plans = 367, Beneficiaries = 1,361,668: $9.51; PPO Plans = 306, Beneficiaries = 268,460: $4.51; All plans[A] Plans = 1,874, Beneficiaries = 5,454,573: $14.70. Amount of rebate allocated to: Part B premium reduction; HMO Plans = 1,179, Beneficiaries = 3,747,087: $1.59; PFFS Plans = 367, Beneficiaries = 1,361,668: $3.62; PPO Plans = 306 Beneficiaries = 268,460: $1.06; All plans[A] Plans = 1,874, Beneficiaries = 5,454,573: $2.29. Amount of rebate allocated to: Cost-sharing reduction[B]; HMO Plans = 1,179, Beneficiaries = 3,747,087: $63.99; PFFS Plans = 367, Beneficiaries = 1,361,668: $51.34; PPO Plans = 306, Beneficiaries = 268,460: $40.61; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $60.51. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Values are weighted by August 2007 plan enrollment and are standardized to represent a beneficiary of average health status. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia are excluded from the analysis. This analysis included only the 1,874 plans that received a rebate. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [B] The rebate amounts allocated to cost sharing and additional benefits included some non-medical expenses, such as administrative costs and plans' margins. [C] Of the 1,874 plans that received a rebate, 1,423 offered Part D benefits to their beneficiaries. Of those that offered Part D, 1,037 reduced Part D premiums. [End of table] While nearly all MA enrollees were in plans that received rebates, some plans charged additional premiums either in addition to the rebate or as the sole funding source to pay for additional benefits, reduced cost sharing, or a combination of the two. In 2007, approximately 41 percent of beneficiaries (about 2.3 million people) were enrolled in an MA plan that charged an additional premium. There were differences in the extent to which plans charged additional premiums by plan type. For example, 31 percent of beneficiaries enrolled in PFFS plans were charged an additional premium, compared to 83 percent of beneficiaries enrolled in PPOs. Of plans that charged an additional premium, the average additional premium was $58 PMPM.[Footnote 20] (See table 2.) Plans that received rebates and charged additional premiums had lower rebates ($54 PMPM on average), than plans that received rebates and did not charge an additional premium ($107 PMPM on average), and these plans allocated less of their rebates to premium reductions and more to additional benefits and cost-sharing reductions.[Footnote 21] Table 2: Percentage of Beneficiaries in Plans That Charge an Additional Premium and Average Amount of Additional Premium by Plan Type, 2007: Percentage of beneficiaries in plans that charge an additional premium and do not receive a rebate: HMO Plans = 1,209, Beneficiaries = 3,977,161: 6; PFFS Plans = 479, Beneficiaries = 1,408,103: 3; PPO Plans = 345, Beneficiaries = 301,746: 11; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: 5. Percentage of beneficiaries in plans that charge an additional premium and receive a rebate: HMO Plans = 1,209, Beneficiaries = 3,977,161: 36; PFFS Plans = 479, Beneficiaries = 1,408,103: 28; PPO Plans = 345, Beneficiaries = 301,746: 72; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: 35. Average amount of additional premium (PMPM): HMO Plans = 1,209, Beneficiaries = 3,977,161: $61.87; PFFS Plans = 479, Beneficiaries = 1,408,103: $42.09; PPO Plans = 345, Beneficiaries = 301,746: $60.47; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: $58.00. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Values are weighted by August 2007 plan enrollment and are standardized to represent a beneficiary of average health status. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [End of table] MA Plans Used Rebates and Additional Premiums to Cover Additional Benefits Such as Dental, Hearing, and Vision: MA plans covered several common additional benefits with the rebates, additional premiums, or both. These benefits included: * dental benefits, which may include oral exams, teeth cleanings, fluoride treatments, dental X-rays, or emergency dental services; * health education benefits, which may include nutritional training, smoking cessation, health club memberships, or nursing hotlines; * hearing benefits, which may include coverage for hearing tests, hearing aid fittings, and hearing aid evaluations; * inpatient facility stays, which may include additional inpatient facility days beyond those covered under Medicare FFS; * international coverage for outpatient emergency services; * skilled nursing facility stays, which include days in a skilled nursing facility beyond those covered under Medicare FFS; and: * vision benefits, which may include coverage for routine eye exams, contacts, or eyeglasses (lenses and frames). Almost all plans covered international outpatient emergency services and additional days in a skilled nursing facility and inpatient facility beyond what Medicare FFS covers. The percentage of plans covering dental, vision, or hearing services varied by plan type. For example, PFFS plans were more likely to cover hearing and less likely to cover dental and vision services than HMOs and PPOs. (See fig. 2.) Figure 2: Percentage of Beneficiaries in Plans Covering Additional Benefits by Plan Type, 2007: [See PDF for image] This figure is a multiple vertical bar graph, depicting the Percentage of Beneficiaries in Plans Covering Additional Benefits by Plan Type, 2007, as follows: Service category: Dental[A]; Plan type and percentage of beneficiaries: HMO (plans = 1,209; beneficiaries = 3,977,161): 33; Plan type and percentage of beneficiaries: PFFS (plans = 479; beneficiaries = 1,408,103): 24; Plan type and percentage of beneficiaries: PPO (plans = 345; beneficiaries = 301,746): 45. Service category: Health Education[B]; Plan type and percentage of beneficiaries: HMO (plans = 1,209; beneficiaries = 3,977,161): 83; Plan type and percentage of beneficiaries: PFFS (plans = 479; beneficiaries = 1,408,103): 80; Plan type and percentage of beneficiaries: PPO (plans = 345; beneficiaries = 301,746): 89. Service category: Hearing[C]; Plan type and percentage of beneficiaries: HMO (plans = 1,209; beneficiaries = 3,977,161): 61; Plan type and percentage of beneficiaries: PFFS (plans = 479; beneficiaries = 1,408,103): 89; Plan type and percentage of beneficiaries: PPO (plans = 345; beneficiaries = 301,746): 52. Service category: Inpatient stays[D]; Plan type and percentage of beneficiaries: HMO (plans = 1,209; beneficiaries = 3,977,161): 96; Plan type and percentage of beneficiaries: PFFS (plans = 479; beneficiaries = 1,408,103): 88; Plan type and percentage of beneficiaries: PPO (plans = 345; beneficiaries = 301,746): 98. Service category: International outpatient emergency; Plan type and percentage of beneficiaries: HMO (plans = 1,209; beneficiaries = 3,977,161): 95; Plan type and percentage of beneficiaries: PFFS (plans = 479; beneficiaries = 1,408,103): 95; Plan type and percentage of beneficiaries: PPO (plans = 345; beneficiaries = 301,746): 94. Service category: Skilled nursing facility stays[D]; Plan type and percentage of beneficiaries: HMO (plans = 1,209; beneficiaries = 3,977,161): 95; Plan type and percentage of beneficiaries: PFFS (plans = 479; beneficiaries = 1,408,103): 89; Plan type and percentage of beneficiaries: PPO (plans = 345; beneficiaries = 301,746): 94. Service category: Vision[E]; Plan type and percentage of beneficiaries: HMO (plans = 1,209; beneficiaries = 3,977,161): 89; Plan type and percentage of beneficiaries: PFFS (plans = 479; beneficiaries = 1,408,103): 49; Plan type and percentage of beneficiaries: PPO (plans = 345; beneficiaries = 301,746): 82. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: The percentages of beneficiaries in plans that have additional benefits are as of August 2007. This analysis included additional benefits funded by both rebates and additional premiums. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. [A] Dental benefits may include oral exams, teeth cleanings, fluoride treatments, dental X-rays, or emergency dental services. [B] Health education benefits may include nutritional training, smoking cessation, health club memberships, or nursing hotlines. [C] Hearing benefits may include coverage for hearing tests, hearing aid fittings, and hearing aid evaluations. [D] Inpatient stays and skilled nursing facility stays may include additional days beyond what Medicare FFS covers. [E] Vision benefits may include coverage for routine eye exams, contacts, or eyeglasses (lenses and frames). [End of figure] The average projected dollar amount of the common additional benefits across all MA plans ranged from $0.11 PMPM for international outpatient emergency services to $4 PMPM for dental care. These estimates were based on the subset of plans that provided cost projections in the categories associated with the benefits. The number of plans included in the averages varies from the number of plans offering the benefits in part because some plans did not consistently include the same additional services in the same benefit categories. For example, some plans categorized all or part of the costs associated with additional vision benefits in other categories, such as professional services.[Footnote 22] These estimates are also based on plans' reported funding for additional benefits from both rebates and additional premiums. Had we limited our analysis to additional benefits funded only from rebates, the estimated amounts of the additional benefits would have been lower. On the basis of plan projections, we estimated that rebates would pay for most of the additional benefits plans provided (77 percent), while additional premiums would pay for the remainder (23 percent). Table 3 provides a summary of the projected costs of additional benefits. Table 3: Average Projected PMPM Costs of Additional Benefits by Service Category and Plan Type for Plans That Offered Benefits and Reported Costs, 2007: Dental[B]: HMO: Number of plans: 435; HMO: Average cost (PMPM): $3.72; PFFS: Number of plans: 29; PFFS: Average cost (PMPM): $4.34; PPO: Number of plans: 80; PPO: Average cost (PMPM): $5.79; All plans[A]: Number of plans: 555; All plans[A]: Average cost (PMPM): $4.00. Health education[C]: HMO: Number of plans: 641; HMO: Average cost (PMPM): $2.01; PFFS: Number of plans: 97; PFFS: Average cost (PMPM): $1.12; PPO: Number of plans: 165; PPO: Average cost (PMPM): $1.95; All plans[A]: Number of plans: 920; All plans[A]: Average cost (PMPM): $1.88. Hearing[D]: HMO: Number of plans: 865; HMO: Average cost (PMPM): $0.86; PFFS: Number of plans: 185; PFFS: Average cost (PMPM): $0.97; PPO: Number of plans: 235; PPO: Average cost (PMPM): $1.51; All plans[A]: Number of plans: 1301; All plans[A]: Average cost (PMPM): $0.92. Inpatient stays[E]: HMO: Number of plans: 966; HMO: Average cost (PMPM): $1.74; PFFS: Number of plans: 255; PFFS: Average cost (PMPM): $1.31; PPO: Number of plans: 240; PPO: Average cost (PMPM): $1.75; All plans[A]: Number of plans: 1482; All plans[A]: Average cost (PMPM): $1.69. International outpatient emergency: HMO: Number of plans: 698; HMO: Average cost (PMPM): $0.13; PFFS: Number of plans: 165; PFFS: Average cost (PMPM): $0.05; PPO: Number of plans: 204; PPO: Average cost (PMPM): $0.06; All plans[A]: Number of plans: 1083; All plans[A]: Average cost (PMPM): $0.11. Skilled nursing facility stays[E]: HMO: Number of plans: 576; HMO: Average cost (PMPM): $1.33; PFFS: Number of plans: 119; PFFS: Average cost (PMPM): $0.38; PPO: Number of plans: 94; PPO: Average cost (PMPM): $1.55; All plans[A]: Number of plans: 801; All plans[A]: Average cost (PMPM): $1.14. Vision[F]: HMO: Number of plans: 1,076; HMO: Average cost (PMPM): $3.41; PFFS: Number of plans: 182; PFFS: Average cost (PMPM): $2.37; PPO: Number of plans: 280; PPO: Average cost (PMPM): $5.76; All plans[A]: Number of plans: 1559; All plans[A]: Average cost (PMPM): $3.37. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Dollar amounts are weighted by August 2007 plan enrollment and are standardized to represent a beneficiary of average health status. We considered an MA plan to have covered an additional benefit if it projected that it would allocate at least $.01 PMPM of revenue to the additional benefit. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there are only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [B] Dental benefits may include oral exams, teeth cleanings, fluoride treatments, dental X-rays, or emergency dental services. [C] Health education benefits may include nutritional training, smoking cessation, health club memberships, or nursing hotlines. [D] Hearing benefits may include coverage for hearing tests, hearing aid fittings, and hearing aid evaluations. [E] Inpatient stays and skilled nursing facility stays may include additional days beyond what Medicare FFS covers. [F] Vision benefits may include coverage for routine eye exams, contacts, or eyeglasses (lenses and frames). [End of table] MA Plans Projected That MA Beneficiaries, on Average, Would Have Lower Cost Sharing Than if They Were in Medicare FFS, but Some MA Beneficiaries Could Pay More: For 2007, MA plans projected that MA beneficiary cost sharing would be 42 percent of estimated cost sharing in Medicare FFS. (See fig 3.) Plans projected that their beneficiaries, on average, would pay $49 PMPM in cost sharing, and they estimated that Medicare FFS equivalent cost sharing for their beneficiaries was $116 PMPM. On the basis of plans' projections, we estimated that about 77 percent of the reduction in beneficiary cost sharing was funded by rebates with the remainder being funded by additional beneficiary premiums. Figure 3: Average Projected Cost Sharing for MA Beneficiaries Compared to Their Cost Sharing in Medicare FFS, by Plan Type, 2007: [See PDF for image] This figure is a vertical bar graph depicting the following information: Plan type: HMO; Percentage of Medicare FFS cost-sharing estimate: 40%. Plan type: PFFS; Percentage of Medicare FFS cost-sharing estimate: 51%. Plan type: PPO; Percentage of Medicare FFS cost-sharing estimate: 45%. Plan type: All plans[A]; Percentage of Medicare FFS cost-sharing estimate: 42%. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. Numbers are weighted by August 2007 plan enrollment. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [End of figure] Although plans projected that beneficiaries' overall cost sharing was lower, on average, than Medicare FFS cost-sharing estimates, some MA plans projected that cost sharing for certain categories of services was higher than Medicare FFS cost-sharing estimates. For example, 19 percent of MA beneficiaries were enrolled in plans that projected higher cost sharing for home health services, on average, than Medicare FFS, which has no cost sharing for this service at all, and 16 percent of beneficiaries were enrolled in plans that projected higher cost sharing for inpatient services compared to Medicare FFS estimates.[Footnote 23] (See table 4.) Because cost sharing is higher for some categories of services, some beneficiaries who frequently use these services can have overall cost sharing that is higher than what they would pay under Medicare FFS. Table 4: Beneficiaries in MA Plans with Higher Projected Cost Sharing Than Medicare FFS for a Given Service Category by Plan Type, 2007: Home health services[B]: HMO Plans = 1,209, Beneficiaries = 3,977,161: Number: 422,078; HMO Plans = 1,209, Beneficiaries = 3,977,161: Percentage: 11; PFFS Plans = 479, Beneficiaries = 1,408,103: Number: 393,523; PFFS Plans = 479, Beneficiaries = 1,408,103: Percentage: 28; PPO Plans = 345, Beneficiaries = 301,746: Number: 253,242; PPO Plans = 345, Beneficiaries = 301,746: Percentage: 84; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Number: 1,069,023; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Percentage: 19. Inpatient services[C]: HMO Plans = 1,209, Beneficiaries = 3,977,161: Number: 699,763; HMO Plans = 1,209, Beneficiaries = 3,977,161: Percentage: 18; PFFS Plans = 479, Beneficiaries = 1,408,103: Number: 170,737; PFFS Plans = 479, Beneficiaries = 1,408,103: Percentage: 12; PPO Plans = 345, Beneficiaries = 301,746: Number: 66,746; PPO Plans = 345, Beneficiaries = 301,746: Percentage: 22; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Number: 937,246; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Percentage: 16. Skilled nursing facility services: HMO Plans = 1,209, Beneficiaries = 3,977,161: Number: 384,960; HMO Plans = 1,209, Beneficiaries = 3,977,161: Percentage: 10; PFFS Plans = 479, Beneficiaries = 1,408,103: Number: 67,017; PFFS Plans = 479, Beneficiaries = 1,408,103: Percentage: 5; PPO Plans = 345, Beneficiaries = 301,746: Number: 47,094; PPO Plans = 345, Beneficiaries = 301,746: Percentage: 16; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Number: 499,071; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Percentage: 9. Durable medical equipment, prosthetics, and supplies: HMO Plans = 1,209, Beneficiaries = 3,977,161: Number: 92,070; HMO Plans = 1,209, Beneficiaries = 3,977,161: Percentage: 2; PFFS Plans = 479, Beneficiaries = 1,408,103: Number: 110,147; PFFS Plans = 479, Beneficiaries = 1,408,103: Percentage: 8; PPO Plans = 345, Beneficiaries = 301,746: Number: 13,324; PPO Plans = 345, Beneficiaries = 301,746: Percentage: 4; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Number: 215,541; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Percentage: 4. Part B drugs[D]: HMO Plans = 1,209, Beneficiaries = 3,977,161: Number: 88,458; HMO Plans = 1,209, Beneficiaries = 3,977,161: Percentage: 2; PFFS Plans = 479, Beneficiaries = 1,408,103: Number: 7,975; PFFS Plans = 479, Beneficiaries = 1,408,103: Percentage: 1; PPO Plans = 345, Beneficiaries = 301,746: Number: 4,806; PPO Plans = 345, Beneficiaries = 301,746: Percentage: 2; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Number: 101,416; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Percentage: 2. Outpatient facility services[E]: HMO Plans = 1,209, Beneficiaries = 3,977,161: Number: 31,359; HMO Plans = 1,209, Beneficiaries = 3,977,161: Percentage: 1; PFFS Plans = 479, Beneficiaries = 1,408,103: Number: 0; PFFS Plans = 479, Beneficiaries = 1,408,103: Percentage: 0; PPO Plans = 345, Beneficiaries = 301,746: Number: 138; PPO Plans = 345, Beneficiaries = 301,746: Percentage: 0; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Number: 31,497; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Percentage: 1. Professional services[C]: HMO Plans = 1,209, Beneficiaries = 3,977,161: Number: 14,641; HMO Plans = 1,209, Beneficiaries = 3,977,161: Percentage: 0; PFFS Plans = 479, Beneficiaries = 1,408,103: Number: 5,781; PFFS Plans = 479, Beneficiaries = 1,408,103: Percentage: 0; PPO Plans = 345, Beneficiaries = 301,746: Number: 26,611; PPO Plans = 345, Beneficiaries = 301,746: Percentage: 9; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Number: 47,033; All plans[A] Plans = 2,055, Beneficiaries = 5,764,368: Percentage: 1. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [B] Home health services include skilled nursing services, home health aides, and certain therapy services, all provided in the home setting. [C] Many MA plans include cost sharing for professional services, such as physician visits received during a hospital stay, in their inpatient cost-sharing amount. As a result, the cost sharing for professional services may be understated for MA plans, while the inpatient cost sharing may be overstated for MA plans. Professional services include physician visits, therapy, and radiology, among other services. [D] Part B drugs are drugs that are covered under Medicare Part B, and they include drugs that are typically administered by a physician. Many plans excluded Part B drugs from the out-of-pocket maximum if they were obtained from a pharmacy, but according to CMS, did not exclude Part B drugs administered by a physician. [E] Outpatient facility services include surgery, emergency, and other services provided in an outpatient facility. [End of table] Cost sharing for particular categories of services varied substantially among MA plans. For example, we found significant variation in cost sharing for inpatient services. Some MA beneficiaries were in plans with no cost sharing for inpatient services. More than half a million MA beneficiaries, representing 9 percent of MA beneficiaries, were in 193 plans with no deductibles, copayments, or coinsurance requirements for inpatient services as of August 2007. Beneficiaries in these plans with long or frequent hospital stays could have saved thousands compared to what their cost sharing would have been if they were enrolled in Medicare FFS, which typically included a $992 deductible, a $248 daily copayment for hospital stays lasting between 61 and 90 days, and additional coinsurance payments for professional services provided in the hospital.[Footnote 24] Other MA beneficiaries, however, could have paid substantially more than Medicare FFS beneficiaries for inpatient care. We found 80 MA plans that charged a daily copayment of $200 or more for the first 10 days of a hospital admission and placed high or no limits on out-of- pocket costs for inpatient services.[Footnote 25] These 80 MA plans also had more than half a million beneficiaries. Beneficiary cost sharing in these 80 plans could have been $2,000 or more for a 10-day hospital stay, and $3,000 or more for three average-length hospital stays.[Footnote 26] Figure 4 provides an illustrative example of an MA plan that could have exposed a beneficiary to higher inpatient costs than under Medicare FFS. While the plan in this illustrative example had lower cost sharing than Medicare FFS for initial hospital stays of 4 days or less as well as initial hospital stays of 30 days or more, for stays of other lengths the MA plan could have cost beneficiaries more than $1,000 above out-of-pocket costs under Medicare FFS. The disparity between out-of-pocket costs under the MA plan and costs under Medicare FFS was largest when comparing additional hospital visits in the same benefit period, since Medicare FFS does not charge a deductible if an admission occurs within 60 days of a previous admission. Figure 4: Example of an MA Plan with Inpatient Cost Sharing Different from the Medicare FFS Program: [See PDF for image] This figure is a multiple line graph depicting an example of an MA Plan with inpatient cost sharing different from the Medicare FFS Program. The vertical axis of the graph represents total cost to beneficiaries from 0 to $3,500. The horizontal axis of the graph represents length of stay in days from 1 to 35. Three lines are depicted, representing the following information: MA plan cost sharing consisting of a copayment for days 1-10 of a hospital stay: This line shows a cost of about $275 on day one, rising steadily to about $2,750 by day ten, and remaining constant for the remaining number of days. There is a $275 per day copayment under the MA plan for days 1-10 and no per day copayment under the MA plan for days 11-90. Medicare FFS estimated cost sharing for an initial hospital stay consisting of coinsurance for physician services received in the hospital and a deductible: This line shows a cost of about $1,100 on day one, rising steadily to about $3,000 by day 35. Medicare FFS estimated cost sharing for a subsequent hospital stay consisting of coinsurance for physician services received in the hospital (no deductible): This line shows a cost of 0 on day one, rising steadily to about $2,000 by day 35. Source: GAO analysis of 2007 CMS Plan Benefit Package data and CMS actuarial data. Notes: In this example, the MA plan charged a $275 daily copayment for the first 10 days of the hospital stay, and charged no additional copayment for days 11 through 90. The plan had a $4,000 out-of-pocket maximum. In contrast, in 2007 Medicare FFS charged a $992 deductible for an initial hospital stay in a benefit period and $248 per day for days 61 through 90 of a hospital stay. Medicare FFS beneficiaries paid no deductible for a subsequent hospital stay if it occurred within 60 days of the previous stay in an inpatient facility. In addition, Medicare FFS beneficiaries must pay coinsurance for physician services received while in the hospital. The charges associated with these physician services averaged $73 per day for the first 4 days of the hospital stay, and $58 per day for the remaining days of a hospital stay through 90 days. This example assumes that the beneficiary was charged the average coinsurance. The actual amount of coinsurance a beneficiary pays varies based on the amount of services a beneficiary receives, and charges can be above or below the average. [A] Nearly 88 percent of hospital stays under Medicare were 10 days or less in 2004 according to CMS data. About 3 percent of hospital stays were 20 days or longer, and 1 percent of stays were longer than 30 days. [End of figure] As of August 2007, about 48 percent of MA beneficiaries were enrolled in plans that had an out-of-pocket maximum, which helps protect beneficiaries against high spending on cost sharing.[Footnote 27] (See fig. 5.) Of the three most common MA plan types, beneficiaries in PFFS plans were the most likely to be in a plan with an out-of-pocket maximum, but PFFS plans also had the highest average out-of-pocket maximum. For MA plans that had an out-of-pocket maximum, the average amount was $3,463. See appendix IV for further details on out-of-pocket maximums. Figure 5: Beneficiaries in MA Plans by Out-of-Pocket Maximum Amount and Plan Type, 2007: [See PDF for image] This figure is a stacked vertical bar graph depicting the following information: Plan type: HMO, Plans = 1,209, beneficiaries = 3,977,161: Percentage of beneficiaries, Overall out-of-pocket maximum of $3,100 or below: 20; Percentage of beneficiaries, Overall out-of-pocket maximum of greater than $3,100: 16; Percentage of beneficiaries, No overall out-of-pocket maximum: 64. Plan type: PFFS, Plans = 479, beneficiaries = 1,408,103: Percentage of beneficiaries, Overall out-of-pocket maximum of $3,100 or below: 28; Percentage of beneficiaries, Overall out-of-pocket maximum of greater than $3,100: 50; Percentage of beneficiaries, No overall out-of-pocket maximum: 23. Plan type: PPO, Plans = 345, beneficiaries = 301,746: Percentage of beneficiaries, Overall out-of-pocket maximum of $3,100 or below: 57; Percentage of beneficiaries, Overall out-of-pocket maximum of greater than $3,100: 11; Percentage of beneficiaries, No overall out-of-pocket maximum: 32. Plan type: All plans[A], Plans = 2,055, beneficiaries = 5,764,368: Percentage of beneficiaries, Overall out-of-pocket maximum of $3,100 or below: 24; Percentage of beneficiaries, Overall out-of-pocket maximum of greater than $3,100: 24; Percentage of beneficiaries, No overall out-of-pocket maximum: 52. Notes: For 2007, CMS generally applied less stringent criteria in evaluating a plan's cost-sharing requirements if the plan had an out- of-pocket maximum of $3,100 or below. If a plan had two out-of-pocket maximums--one for in-network services and one for combined in-and out- of-network services, then we used the lower value for this analysis. Some plans without an out-of-pocket maximum did have a service-specific maximum. Twenty-one percent of plans with no out-of-pocket maximum had a service-specific maximum for inpatient acute services, and 16 percent of plans with no out-of-pocket maximum had a service-specific maximum for inpatient psychiatric services. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. Some numbers do not add up to 100 due to rounding. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [End of figure] An out-of-pocket maximum does not always cover all categories of services. Some MA plans excluded some services from the out-of-pocket maximum. Beneficiaries who use these excluded services may pay more in total cost sharing than is indicated by the plan's out-of-pocket maximum. Part B drugs, which include drugs that are typically physician-administered drugs, were most often excluded from the out-of- pocket maximum--29 percent of MA plans with an out-of-pocket maximum excluded some Part B drugs from that maximum.[Footnote 28] (See table 5.) Plans that excluded a certain service category from the out-of- pocket maximum did not necessarily exclude all services from that category. For example, many plans excluded Part B drugs from the out- of-pocket maximum if they were obtained from a pharmacy, but according to CMS, did not exclude Part B drugs administered by a physician. Table 5: MA Plans That Exclude Some Services under a Service Category from Their Out-of-Pocket Maximum: Part B drugs[A]: Number of plans, Plans = 1,016: 296; Percentage of plans, Plans = 1,016: 29; Number of beneficiaries, Beneficiaries = 2,738,531: 1,107,876; Percentage of beneficiaries, Beneficiaries = 2,738,531: 40. Outpatient substance abuse: Number of plans, Plans = 1,016: 233; Percentage of plans, Plans = 1,016: 23; Number of beneficiaries, Beneficiaries = 2,738,531: 645,997; Percentage of beneficiaries, Beneficiaries = 2,738,531: 24. Physician specialist, excluding psychiatric: Number of plans, Plans = 1,016: 230; Percentage of plans, Plans = 1,016: 23; Number of beneficiaries, Beneficiaries = 2,738,531: 641,270; Percentage of beneficiaries, Beneficiaries = 2,738,531: 23. Mental health, non-physician: Number of plans, Plans = 1,016: 230; Percentage of plans, Plans = 1,016: 23; Number of beneficiaries, Beneficiaries = 2,738,531: 630,504; Percentage of beneficiaries, Beneficiaries = 2,738,531: 23. Psychiatric: Number of plans, Plans = 1,016: 218; Percentage of plans, Plans = 1,016: 21; Number of beneficiaries, Beneficiaries = 2,738,531: 602,500; Percentage of beneficiaries, Beneficiaries = 2,738,531: 22. Home health services: Number of plans, Plans = 1,016: 211; Percentage of plans, Plans = 1,016: 21; Number of beneficiaries, Beneficiaries = 2,738,531: 569,618; Percentage of beneficiaries, Beneficiaries = 2,738,531: 21. Prosthetics and medical supplies: Number of plans, Plans = 1,016: 128; Percentage of plans, Plans = 1,016: 13; Number of beneficiaries, Beneficiaries = 2,738,531: 603,952; Percentage of beneficiaries, Beneficiaries = 2,738,531: 22. Durable medical equipment: Number of plans, Plans = 1,016: 116; Percentage of plans, Plans = 1,016: 11; Number of beneficiaries, Beneficiaries = 2,738,531: 565,413; Percentage of beneficiaries, Beneficiaries = 2,738,531: 21. Outpatient hospital: Number of plans, Plans = 1,016: 72; Percentage of plans, Plans = 1,016: 7; Number of beneficiaries, Beneficiaries = 2,738,531: 192,182; Percentage of beneficiaries, Beneficiaries = 2,738,531: 7. Inpatient hospital, psychiatric: Number of plans, Plans = 1,016: 37; Percentage of plans, Plans = 1,016: 4; Number of beneficiaries, Beneficiaries = 2,738,531: 149,105; Percentage of beneficiaries, Beneficiaries = 2,738,531: 5. Skilled nursing facility: Number of plans, Plans = 1,016: 34; Percentage of plans, Plans = 1,016: 3; Number of beneficiaries, Beneficiaries = 2,738,531: 100,700; Percentage of beneficiaries, Beneficiaries = 2,738,531: 4. Inpatient hospital, acute: Number of plans, Plans = 1,016: 19; Percentage of plans, Plans = 1,016: 2; Number of beneficiaries, Beneficiaries = 2,738,531: 29,937; Percentage of beneficiaries, Beneficiaries = 2,738,531: 1. Source: GAO analysis of 2007 CMS Plan Benefit Package data. Notes: We considered an MA plan to have an out-of-pocket maximum if the plan had either an in-network out-of-pocket maximum or an out-of-pocket maximum for both in-network and out-of-network services. A plan was considered to have excluded a service category from the out-of-pocket maximum if the out-of-pocket maximum did not cover that service category and if the plan had no service-specific maximum for that category. Plans that excluded a certain service category from the out- of-pocket maximum did not necessarily exclude all services from that category. HMOs, PFFS plans, PPOs, and PSOs were included in the analysis. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that were exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. Only plans with an out-of-pocket maximum were included in this analysis. [A] Many plans excluded Part B drugs from the out-of-pocket maximum if they were obtained from a pharmacy, but according to CMS, did not exclude Part B drugs administered by a physician. [End of table] Approximately 87 Percent of Total Revenue Projected to Be Allocated to Medical Expenses, but Projections Varied among Individual Plans: For 2007, MA plans projected that of their total revenues ($783 PMPM), they would allocate approximately 87 percent ($683 PMPM) to medical expenses, resulting in an average medical loss ratio of approximately 0.87. MA plans projected that they would allocate approximately 9 percent of total revenue ($71 PMPM) to non-medical expenses, and approximately 4 percent ($30 PMPM) to the plans' margin, on average.[Footnote 29] While there was little variation in the average projected medical loss ratio by plan type, there was variation among individual plans. For example, we found that about 30 percent of beneficiaries--about 1.7 million--were enrolled in plans with a medical loss ratio of less than 0.85--the threshold included in the Children's Health and Medicare Protection Act of 2007 (CHAMP Act).[Footnote 30] (See fig. 6.) A CMS official we spoke to stated that the medical loss ratio may vary for reasons other than utilization and the cost of providing care. For example, some MA plans may categorize the costs of delivering care management services as a medical expense, while other plans may include this as a non-medical expense. Figure 6: Percentage of Beneficiaries in MA Plans That Project Allocating Less Than 85 Percent of Total Revenues to Medical Expenses, by Plan Type, 2007: [See PDF for image] This figure is a vertical bar graph, depicting the following information: Plan type: HMO, Plans = 1,209; beneficiaries = 3,977,161; Percentage of beneficiaries: 33%. Plan type: PFFS, Plans = 479; beneficiaries = 1,408,103; Percentage of beneficiaries: 24%. Plan type: PPO, Plans = 345; beneficiaries = 301,746; Percentage of beneficiaries: 7%. Plan type: All plans[A], Plans = 2,055; beneficiaries = 5,764,368; Percentage of beneficiaries: 30%. Source: GAAO analysis of 2007 CMS Bid Pricing Tool data. Notes: A CMS official indicated that the percentage of revenues allocated to medical expenses (the medical loss ratio) may vary across plans for reasons other than utilization and the cost of providing care. For example, some MA plans may categorize the costs of delivering care management services as a medical expense, while other plans may include this as a non-medical expense. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [End of figure] MA plans projected expenses separately for four distinct non-medical expense categories--marketing and sales, direct administration, indirect administration, and the net cost of private reinsurance.[Footnote 31] On average, MA plans projected allocating total revenue to non-medical expenses approximately as follows: * 2.4 percent to marketing and sales; * 2.9 percent to direct administration, such as customer service and medical management; * 3.7 percent to indirect administration, such as accounting operations and human resources; and: * 0.1 percent to the net cost of private reinsurance. Of these four non-medical expense categories, the largest difference between plan types' allocation of revenue to non-medical expenses was in the category of marketing and sales. On average, as a percentage of total revenue, projected marketing and sales expenses were 2 percent ($16 PMPM) for HMOs, 3.6 percent ($27 PMPM) for PFFS plans, and 2 percent ($17 PMPM) for PPOs. (See fig. 7.) Figure 7: MA Plans' Projected Marketing and Sales Expenses by Plan Type, 2007: [See PDF for image] This figure is a vertical bar graph, depicting the following information: Plan type: HMO, Plans = 1,209; beneficiaries = 3,977,161; Percentage of total revenue: 2.0%. Plan type: PFFS, Plans = 479; beneficiaries = 1,408,103; Percentage of total revenue: 3.6%. Plan type: PPO, Plans = 345; beneficiaries = 301,746; Percentage of total revenue: 2.0%. Plan type: All plans[A], Plans = 2,055; beneficiaries = 5,764,368; Percentage of total revenue: 2.4%. Source: GAAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Percentages are weighted by August 2007 enrollment. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [End of figure] Concluding Observations: Medicare spends more per beneficiary in MA than it does for beneficiaries in Medicare FFS, at an estimated additional cost to Medicare of $54 billion from 2009 through 2012. Under the current payment system, the average MA plan receives a Medicare rebate equal to approximately $87 PMPM, on average. In 2007, MA plans projected that they would use the vast majority of their rebates--approximately 89 percent--to reduce enrollees' premiums and to lower their out-of-pocket costs for Medicare-covered services. Plans projected that they would use a relatively small portion of their rebates--approximately 11 percent--to provide benefits that are not covered under Medicare FFS. Although the rebates generally have helped to make health care more affordable for many beneficiaries enrolled in MA plans, some beneficiaries may face higher expenses than they would in Medicare FFS. Further, because premiums paid by beneficiaries in Medicare FFS are tied to both Medicare FFS and MA costs, the additional payments to MA plans have increased the premiums paid by beneficiaries in Medicare FFS as well as contributed to the substantial long-term financial challenge that Medicare faces. Whether the value that MA beneficiaries receive in the form of reduced cost sharing, lower premiums, and extra benefits is worth the increased cost borne by beneficiaries in Medicare FFS and other taxpayers is a decision for policymakers. However, if the policy objective is to subsidize health care costs of low-income Medicare beneficiaries, it may be more efficient to directly target subsidies to a defined low-income population than to subsidize premiums and cost sharing for all MA beneficiaries, including those who are well off. As Congress considers the design and cost of the MA program, it will be important for policymakers to balance the needs of beneficiaries-- including those in MA plans and those in Medicare FFS--with the necessity of addressing Medicare's long-term financial health. Agency and Other External Comments and Our Evaluation: CMS provided us with written comments on a draft of this report which are reprinted in appendix V, and AHIP, a national association that represents companies providing health insurance coverage, provided us with oral comments. CMS Comments: In general, CMS commented that the report did not recognize that the majority of MA benefit packages in 2007 were better and provided more protection for out-of-pocket costs than Medicare FFS. It stated that the report failed to acknowledge that MA plans provide beneficiaries with the ability to choose a plan that best meets individual medical and financial needs. CMS also expressed concern that the report was not balanced because it did not sufficiently focus on the advantages of MA plans. We disagree with CMS that we did not consider that most MA plans offered better cost sharing than Medicare FFS. We noted in the first paragraph of our cost sharing finding that, overall, plans projected MA beneficiary cost sharing that was 42 percent of estimated cost sharing in Medicare FFS. Regarding the absence of information about MA plans providing beneficiaries with choices, this was not the focus of our research. However, we agree the issue provides important context and therefore we noted in the report's introduction the additional choice MA plans provide Medicare beneficiaries. We disagree that the report is not balanced. We provided a fact-based assessment of how rebates were projected to be used in 2007, and identified important issues related to cost sharing. Even though cost sharing would be less, on average, in MA plans than in Medicare FFS, an important finding of our report is that beneficiaries who use certain services with high cost sharing in MA plans could have higher overall out-of-pocket costs than under Medicare FFS. CMS provided several additional comments. CMS commented that it did not disagree with our finding that 16 percent of beneficiaries were in plans with higher inpatient cost sharing than Medicare FFS. However, it noted that our discussion of the issue and accompanying table and figure did not account for several factors that would have mitigated the impact of the finding. Specifically, CMS commented that we should have considered that MA plans generally combine physician cost sharing in the hospital with inpatient hospital cost sharing, which would have decreased the difference in cost sharing between MA plans and Medicare FFS. Although we had noted this in table notes in the draft, we agree that this should be clearer and modified our text and accompanying figure comparing MA and Medicare FFS cost sharing, and clarified existing table notes. We also modified the text and accompanying figure to differentiate between first and subsequent admissions within the same benefit period, in response to CMS comments. These changes did not affect our finding that some beneficiaries could have cost sharing that was considerably higher than in Medicare FFS. CMS also commented that we should have discussed the mitigating impact of particularly long hospitalizations because beneficiaries with long inpatient hospitals stays in MA plans are likely to have lower cost sharing than under Medicare FFS. We acknowledged CMS's point and addressed this issue in the finding and modified the accompanying figure. However, most beneficiaries have relatively short lengths of stay. For example, in 2005, the average length for an inpatient stay was 5.4 days. This modification did not change our message that some beneficiaries in MA plans could have higher out-of-pocket costs. In addition, CMS commented that we should have noted that many plans have "effective" out-of-pocket maximums for inpatient stays even if they are not specified as such in the plan benefit package. For example, plans may require copayments for specific days of an inpatient stay, such as days 1 through 6, but not for any days beyond the sixth day, thereby capping maximum cost sharing for the stay. We agree that most plans have "effective" or actual out-of-pocket maximums for inpatient hospital services. We also agree that in many cases these maximums can limit beneficiary inpatient cost sharing to levels below inpatient cost sharing under Medicare FFS. However, MA plans projected that about 16 percent of beneficiaries were enrolled in plans that projected higher cost sharing than under Medicare FFS even after accounting for "effective" or actual out-of-pocket maximums. While some of the 16 percent of plans may have bundled physician services with their inpatient estimates, we also showed that 80 plans with high out- of-pocket maximums for inpatient services could have higher cost sharing than Medicare FFS even with "effective" out-of-pocket maximums for inpatient hospital services. CMS raised other concerns about our out-of-pocket maximum analysis, specifically stating that we overestimated the impact of the exclusion of Part B drugs from out-of-pocket maximums. It noted that Part B drugs administered in a physician's office would be included under an out-of- pocket maximum and that only a subset of plans excluded Part B drugs obtained from a pharmacy from the out-of-pocket maximum. We relied on the Plan Benefit Package for information regarding the analysis of Part B drug exclusions from out-of-pocket maximums. According to these data, there were 1.1 million beneficiaries in plans that reported such exclusions in 2007. We noted that the exclusions applied to Part B drugs obtained from a pharmacy and that the plans did not indicate the coverage for Part B drugs administered by a physician. We sought clarification from CMS for which Part B drugs were excluded from the out-of-pocket maximum and were told by a CMS official that plans excluded spending on Part B drugs from the out-of-pocket maximum if beneficiaries received them on an outpatient basis. We added this point of clarification to a footnote in the draft. Given CMS's subsequent agency comments on this issue, we clarified in the text that the exclusions applied to Part B drugs obtained from a pharmacy and do not typically apply to Part B drugs administered by a physician. However, we are concerned that the information in the Plan Benefit Package-- information that beneficiaries rely on when they are seeking benefit coverage information--does not indicate whether chemotherapy drugs are included or excluded under the out-of-pocket maximums. CMS also provided technical comments and clarifications, which we incorporated as appropriate. AHIP Comments: AHIP representatives stated that they agreed with our methodology, but raised certain points that they thought the report should have made or emphasized. AHIP representatives said that while they understood why we made a distinction between additional benefits and cost-sharing reductions, they believed that we characterized additional benefits as being the more valuable of the two. We disagreed with AHIP's assessment. While we did include a discussion of how MA plans projected they would allocate their rebates to additional benefits, premium reductions, and cost- sharing reductions, it was beyond the scope of our work to assess the relative value of the allocation options. With regard to our cost-sharing finding, AHIP stated that while MA beneficiaries may have higher cost sharing for some categories of services, these may be offset by lower cost sharing for other categories of services. Like CMS, AHIP contended that our example of an MA plan with higher cost sharing for inpatient services, relative to FFS, did not account for the additional cost sharing Medicare FFS beneficiaries would pay for physician services during their inpatient stays. As both CMS and AHIP pointed out, most MA plans do not charge extra for physician services during inpatient stays. We have made changes to the text of our report and the accompanying figure to clarify this point. However, as our report noted, beneficiaries who frequently use high cost-sharing services could have overall cost sharing that would be higher than under Medicare FFS. AHIP stated that although some beneficiaries may face higher cost sharing under an MA plan than if they were enrolled in Medicare FFS, their out-of-pocket costs could be lower if their MA plan has a lower premium than Medicare FFS. While this may be true in some cases--we found that, on average, plans used 3 percent of their rebates to reduce Part B premiums--it was beyond the scope of our work to make such a determination. AHIP further stated that MA plans provide beneficiaries with options. Beneficiaries who prefer more predictable expenses can choose MA plans with higher premiums and lower cost sharing, while beneficiaries who are less averse to risk can choose MA plans with lower premiums and higher cost sharing. We agree that the MA program provides beneficiaries with options and have added this point to the text of our report. With regard to our reporting on MA plan medical loss ratios, AHIP representatives indicated that our point was fairly stated, but they asked us to mention this point in the Results in Brief section of the report. We believed that we made this point clear in our discussion of medical loss ratios and that the issue did not warrant mentioning in our high-level summary. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from its date. At that time we will send copies to the Administrator of CMS and interested congressional committees. We will also make copies available to others upon request. The report will also be available at no charge on the GAO Web site at [hyperlink, http://www.gao.gov]. If you or your staffs have any questions about this report please contact me at (202) 512-7114 or cosgrovej@gao.gov. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix VI. Signed by: James C. Cosgrove: Acting Director, Health Care: List of Requesters: The Honorable John D. Dingell: Chairman: Committee on Energy and Commerce: House of Representatives: The Honorable Frank Pallone, Jr. Chairman: Subcommittee on Health: Committee on Energy and Commerce: House of Representatives: The Honorable Henry A. Waxman: Chairman Committee on Oversight and Government Reform: House of Representatives: The Honorable Charles B. Rangel: Chairman Committee on Ways and Means: House of Representatives: The Honorable Pete Stark: Chairman: Subcommittee on Health: Committee on Ways and Means: House of Representatives: [End of section] Appendix I: Example of a Rebate Calculation: For most Medicare Advantage (MA) plan types, Medicare provides plans with a rebate if the plan's bid is below the benchmark, but provides no rebate if the plan's bid exceeds the benchmark.[Footnote 32] Table 6 is an example of rebate calculations for two hypothetical plans, both in the same county. Table 6: The Calculation of the Rebate for Two Hypothetical MA Plans: County's fee-for-service spending: Plan A dollars per member per month: $720; Plan B dollars per member per month: $720. County's benchmark: Plan A dollars per member per month: $800; Plan B dollars per member per month: $800. Plan bid: Plan A dollars per member per month: $700; Plan B dollars per member per month: $840. Amount by which bid is lower than benchmark; Plan A dollars per member per month: $100; Plan B dollars per member per month: $0. Plan's rebate (75 percent of amount by which bid is lower than benchmark): Plan A dollars per member per month: $75; Plan B dollars per member per month: $0. Medicare payment: Plan A dollars per member per month: $775; Plan B dollars per member per month: $800. Mandatory plan premium: Plan A dollars per member per month: $0; Plan B dollars per member per month: $40. Additional benefits, reduced premiums, and reduced cost sharing to beneficiary: Plan A dollars per member per month: $75; Plan B dollars per member per month: $0. Source: GAO. Note: All numbers in this example are standardized to represent a beneficiary of average health status. [End of table] Both plans have the same benchmark because they are in the same county. Plan A submits a bid of $700 per member per month (PMPM). Because the plan's bid is $100 PMPM below the benchmark, it receives a rebate equal to 75 percent of that amount, or $75 PMPM. Plan A must use the $75 PMPM rebate to provide additional benefits, reduced premiums, reduced cost sharing, or any combination of the three. Plan B, however, submits a bid that is $40 PMPM above the benchmark. As a result, the plan receives no rebate. Medicare's payments to plans cannot exceed the benchmark, so Medicare's payment to Plan B is set at $800 PMPM, the amount of the benchmark. Plan B must make up the remainder of the bid by charging its beneficiaries a mandatory plan premium of $40 PMPM. Since Plan A has extra benefits and no additional premium, while Plan B has no extra benefits and an additional premium, Plan A may attract more beneficiaries. If most beneficiaries choose Plan A over Plan B, Plan B is given an incentive to become more efficient in the following year and lower its bid. [End of section] Appendix II: Scope and Methodology: This section describes the scope and methodology used to analyze our four objectives: (1) how MA plans projected they would allocate the rebates they receive, (2) what additional benefits MA plans commonly covered with the rebates and additional premiums, and the projected costs of these additional benefits, (3) how MA plans' projected beneficiary cost sharing, overall and by type of service, compared to Medicare fee-for-service (FFS), and (4) how MA plans projected they would allocate their revenue to medical and other expenses. We used two primary data sources to analyze our four objectives: the 2007 Bid Pricing Tool data and the 2007 Plan Benefit Package data. These data are submitted by MA plans to the Centers for Medicare & Medicaid Services (CMS), the agency that administers Medicare. The bid pricing data contain MA plans' projections of their revenue requirements and revenue sources. Specifically, the bid pricing data include MA plans' projections of revenue requirements--spending on medical expenses, spending on non-medical expenses, and the margin. The bid pricing data also contain information on the benefits and cost- sharing arrangements of plans, including how MA plans' projected cost sharing compares to cost sharing in Medicare FFS. In addition, the bid pricing data contain information on the amount of rebates and additional premiums plans project they will require to fund additional benefits, reduced premiums, and reduced cost sharing. The benefit package data contain detailed information about the benefits and cost- sharing requirements that MA plans offer to Medicare beneficiaries. For our objectives, we focused our analysis on plan types that account for 98 percent of MA enrollment: Health Maintenance Organizations (HMO) (71 percent), Private Fee-for-Service (PFFS) Plans (21 percent), Preferred Provider Organizations (PPO) (5 percent), and Provider- Sponsored Organizations (PSO) (1 percent).[Footnote 33] We excluded Medical Savings Account plans and regional PPOs from our analysis because they follow a different bidding process. We excluded plan types that have unique restrictions on enrollment--such as employer plans, Special Needs Plans (SNP), and demonstration plans--and bids for plans that only cover Part B services. We also excluded plans with service areas that are exclusively outside the 50 states and the District of Columbia. Plans sponsors are permitted to submit separate bids for a single package of benefits by dividing the service area into segments; in these cases, benefits would be the same for each segment, but each segment's cost sharing and premiums may differ. We counted each segment as a separate plan. We used August 2007 enrollment numbers to weight our results. As a result of our methodology, we included 2,055 plans and 5,764,368 beneficiaries (71 percent of total MA enrollment) in our analysis--these numbers apply to all tables and figures in the report, unless otherwise noted. Because there were only 22 PSOs after the exclusions, and enrollment in those plans was 1 percent of MA enrollment, we do not report results separately for PSOs, but we include them in the aggregated results we report for all MA plans. To determine how plans projected they would allocate the rebate to additional benefits, reduced premiums, and reduced cost sharing, we used the bid pricing data. The bid pricing data contain the total amounts plans projected they would spend on additional benefits, reduced premiums, and reduced cost sharing. However, since MA plans use both rebates and additional premiums as a funding source for these additional benefits, reduced premiums, and reduced cost sharing, we calculated the proportion of total funding plans projected they would spend on additional benefits, reduced premiums, and reduced cost sharing and applied these projections to the projected rebate. We restricted our analysis of rebate allocations to the 1,874 plans that received a rebate. To identify the additional benefits that MA plans commonly covered with rebates and additional premiums, we used the benefit package data. The benefit package data provide the most detailed and accurate information about benefits offered, including additional benefits. We used the crosswalk CMS recommended--but did not require--plans to use to match service categories in the benefit package data to categories in the bid pricing data, and identified the percentage of beneficiaries in plans that offered additional benefits using bid pricing categories.[Footnote 34] To identify the costs associated with these additional benefits, we used the bid pricing data. Plans did not use consistent categories for their additional benefits in the bid pricing data. For example, some plans categorized additional vision benefits under the category of other non-covered services. Therefore, our estimates of the costs of additional benefits do not include all plans that offer those benefits, but are based on a smaller number of plans that specified that additional benefit and the associated cost of providing that benefit. In addition, some categories, such as professional services and other non-covered services, were identified by CMS as unreliable because they likely included a variety of services, and we excluded these categories from our analysis. Other categories of additional services may include some inconsistent services, and the cost estimates for additional benefits should therefore be considered as approximations. To calculate estimated costs for each of the additional service categories, we identified plans that offered the additional benefit and that had projected a cost of at least $0.01 PMPM. The projected amounts of plans' additional benefits were adjusted for the health status of the plans' projected population by dividing the amount of the plans' additional benefits by the plans' projected risk scores--a number representing how a plan's beneficiaries' health expenditures are predicted to differ from the average beneficiary in Medicare FFS.[Footnote 35] We then calculated the average amount of the additional benefit, weighting the average by the number of enrollees in the plans. If we had estimated the amount of additional benefits funded only by the rebates, the PMPM amounts of additional benefits would be lower. To compare projected beneficiary cost sharing in MA plans and Medicare FFS, we analyzed plans' cost sharing for Medicare-covered services as reported in the bid pricing data and the equivalent Medicare FFS cost- sharing amounts, also included in the bid pricing data. The equivalent Medicare FFS cost sharing represents an MA beneficiary's expected cost sharing under Medicare FFS if the beneficiary's MA plan had the same pricing and utilization as Medicare FFS. The Medicare FFS equivalent cost sharing for each service category was calculated by applying the average cost-sharing percentage under Medicare FFS for a given service category to each plan's total cost estimates for providing benefits in that service category. For example, if the cost-sharing percentage under Medicare FFS for inpatient services is 10 percent for a given county, and an MA plan in that county projects spending on inpatient services at $200 PMPM, then the equivalent inpatient cost sharing is 10 percent of $200, or $20 PMPM. For Part A services, the cost-sharing percentage under Medicare FFS is calculated for each county--one county may have an equivalent inpatient cost-sharing percentage of 10 percent, while another county may have a percentage of 8 percent. For Part B services, however, the cost-sharing percentages are a national average, so the same percentages were applied to all counties. We divided each plan's estimated cost sharing and the Medicare equivalent cost sharing by each plan's projected risk score to get estimated cost sharing for a beneficiary with average Medicare health spending. We reported the percentage of plans that had cost sharing higher than the estimated Medicare cost sharing for a given service category. When we calculated the amount of reduced cost sharing, we used the total amounts reported in the bid pricing data. We included both rebates and additional premiums because this provided the accurate amount of cost-sharing reductions that MA plans projected their beneficiaries will receive. The amounts of the additional benefits and cost-sharing reductions in our analyses would be lower if we had restricted our analysis to rebates as the sole funding source. To determine plans' out-of-pocket maximums, we examined the in-network out-of-pocket maximum and the combined out-of-pocket maximum (a maximum that applies to both in-network and out-of-network services) fields in the benefit package data. If the two fields were the same value, then we defined the out-of-pocket maximum as equal to that value. If one of the fields was blank, and the other field was a positive number, then we defined the out-of-pocket maximum as equal to the value of the field with the positive number. If both fields had a positive number, but they were not equal, then we defined the out-of-pocket maximum as equal to the value of the field with the lower value. We categorized a plan as having an out-of-pocket maximum even if the plan excluded certain categories of service from that maximum. We did not categorize a plan that had only a service-specific maximum as having an out-of-pocket maximum. To determine the percentage of total revenue allocated to medical expenses and other expenses, we used the bid pricing data and calculated the projected values of medical expenses, non-medical expenses, and margin as a percentage of revenue for all plans and by plan type.[Footnote 36] We reported the percentages of beneficiaries in plans that projected medical expenses less than 85 percent. We also analyzed the percentage of revenue projected to go to sales and marketing from the bid pricing data. [End of section] Appendix III: Plan Variation in Rebate Amounts: Rebate amounts, as well as the allocation of rebates, varied considerably from plan to plan. To provide a measure of this variation, we calculated rebate amounts and the amounts of additional benefits, reduced premiums, and reduced cost sharing at the 25th and 75th percentiles, weighted for enrollment. A percentile is the value below which a certain percentage of beneficiaries fall. For example, the value of the cost-sharing reduction at the 25th percentile was $39.02 PMPM and at the 75th percentile was $78.90 PMPM, meaning that at least 25 percent of beneficiaries were in plans that projected a cost-sharing reduction of $39.02 PMPM or less, and at least 75 percent of beneficiaries were in plans that projected a cost-sharing reduction of $78.90 PMPM or less. (See table 7.) Table 7: Rebate Amount Allocated to Additional Benefits, Premium Reductions, and Cost-Sharing Reductions by Plan Type, 2007: Average: Rebate total; 25th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $57.81; PFFS Plans = 367 Beneficiaries = 1,361,668: $59.70; PPO Plans = 306 Beneficiaries = 268,460: $37.33; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $56.32. Average: Rebate total; 75th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $118.19; PFFS Plans = 367 Beneficiaries = 1,361,668: $83.30; PPO Plans = 306 Beneficiaries = 268,460: $69.83; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $108.55. Amount of rebate allocated to Additional benefits[B]; 25th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $4.14; PFFS Plans = 367 Beneficiaries = 1,361,668: $0.00; PPO Plans = 306 Beneficiaries = 268,460: $3.56; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $2.75. Amount of rebate allocated to Additional benefits[B]; 75th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $15.51; PFFS Plans = 367 Beneficiaries = 1,361,668: $11.41; PPO Plans = 306 Beneficiaries = 268,460: $13.96; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $13.70. Amount of rebate allocated to Part D premium reduction[C]; 25th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $0.21; PFFS Plans = 367 Beneficiaries = 1,361,668: 0.00; PPO Plans = 306 Beneficiaries = 268,460: 0.00; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: 0.00. Amount of rebate allocated to Part D premium reduction[C]; 75th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $24.04; PFFS Plans = 367 Beneficiaries = 1,361,668: $24.12; PPO Plans = 306 Beneficiaries = 268,460: $7.25; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $24.12. Amount of rebate allocated to Part B premium reduction; 25th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $0.00; PFFS Plans = 367 Beneficiaries = 1,361,668: $0.00; PPO Plans = 306 Beneficiaries = 268,460: $0.00; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $0.00. Amount of rebate allocated to Part B premium reduction; 75th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $0.00; PFFS Plans = 367 Beneficiaries = 1,361,668: $0.00; PPO Plans = 306 Beneficiaries = 268,460: $0.00; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $0.00. Amount of rebate allocated to Cost-sharing reduction[B]; 25th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $42.89; PFFS Plans = 367 Beneficiaries = 1,361,668: $39.02; PPO Plans = 306 Beneficiaries = 268,460: $26.79; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $39.02. Amount of rebate allocated to Cost-sharing reduction[B]; 75th percentile; HMO Plans = 1,179 Beneficiaries = 3,747,087: $84.88; PFFS Plans = 367 Beneficiaries = 1,361,668: $68.95; PPO Plans = 306 Beneficiaries = 268,460: $52.60; All plans[A] Plans = 1,874 Beneficiaries = 5,454,573: $78.90. Source: GAO analysis of 2007 CMS Bid Pricing Tool data. Notes: Values are weighted by August 2007 plan enrollment and are standardized to represent a beneficiary of average health status. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. There were 1,874 plans that received a rebate. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted1 percent of total MA enrollment. [B] The rebate amounts allocated toward cost sharing and additional benefits included some non-medical expenses, such as administrative costs and plans' margins. [C] Of 1,874 plans that received a rebate, 1,423 offered Part D benefits to their beneficiaries. Of those that offered Part D, 1,037 reduced Part D premiums. [End of table] [End of section] Appendix IV: Plan Variation in the Out-of-Pocket Maximum: In 2007, about half of MA beneficiaries were in plans that had an out- of-pocket maximum, a dollar limit on a beneficiary's cost sharing. The out-of-pocket maximum varied from plan to plan. To provide a measure of this out-of-pocket maximum variation, we calculated the out-of-pocket maximum at the 25th and 75th percentiles, weighted for enrollment. A percentile is the value below which a certain percentage of beneficiaries fall. For example, the out-of-pocket maximum at the 25th percentile was $2,750, and at the 75th percentile it was $4,600, meaning that at least 25 percent of beneficiaries were in plans with an out-of-pocket maximum $2,750 or less, and at least 75 percent of beneficiaries were in plans with an out-of-pocket maximum of $4,600 or less. (See table 8.) Table 8: Variation in Values of Out-of-Pocket Maximum by Plan Type, 2007: Value of out-of-pocket maximum: Average; Plan type: HMO, Plans = 444, Beneficiaries = 1,436,148: $3,204; Plan type: PFFS, Plans = 350, Beneficiaries = 1,087,383: $4,026; Plan type: PPO, Plans = 219, Beneficiaries = 205,713: $2,377; Plan type: All plans[A], Plans = 1,016, Beneficiaries = 2,738,531: $3,463. 25th percentile; Plan type: HMO, Plans = 444 Beneficiaries = 1,436,148: $2,750; Plan type: PFFS, Plans = 350 Beneficiaries = 1,087,383: $3,000; Plan type: PPO, Plans = 219 Beneficiaries = 205,713: $1,000; Plan type: All plans[A], Plans = 1,016, Beneficiaries = 2,738,531: $2,750. Value of out-of-pocket maximum: 75th percentile; Plan type: HMO, Plans = 444 Beneficiaries = 1,436,148: $4,000; Plan type: PFFS, Plans = 350 Beneficiaries = 1,087,383: $5,000; Plan type: PPO, Plans = 219 Beneficiaries = 205,713: $3,100; Plan type: All plans[A], Plans = 1,016, Beneficiaries = 2,738,531: $4,600. Source: GAO analysis of 2007 CMS Plan Benefit Package data. Notes: Values are weighted by plan enrollment. If a plan had two out- of-pocket maximums--one for in-network services and one for combined in-and out-of-network services, then we used the lower value for this analysis. Determination of a plan's overall out-of-pocket maximum did not take into account whether a plan had a maximum for a specific category of service. Employer plans, Part B only plans, SNPs, regional PPOs, and plans with service areas that are exclusively outside of the 50 states and the District of Columbia were excluded from the analysis. [A] The "All plans" category includes HMOs, PFFS plans, PPOs, and PSOs. Results are not reported separately for PSOs because there were only 22 PSO plans and enrollment in those plans constituted 1 percent of total MA enrollment. [End of table] [End of section] Appendix V: Comments from the Centers for Medicare & Medicaid Services: Department Of Health & Human Services: Centers for Medicare & Medicaid Services: Office of the Administrator: Washington, DC 20201: Date: February 1, 2008: To: James C. Cosgrove: Director, Health Care: Government Accountability Office: From: [Signed by] Kerry Weems: Acting Administrator: Subject: Government Accountability Office (GAO) Draft Report: Medicare Advantage: Increased Spending Relative to Medicare Fee for Service May Not Always Reduce Beneficiary Out of Pocket Costs (GAO-08-359). Thank you for the opportunity to review and comment on the GAO draft report referenced above. In this draft report. the GAO examined (1) Medicare Advantage (MA) plans' projected rebate reallocations; (2) additional benefits MA plans commonly covered and their costs: (3) MA plans' projected cost-sharing; and (4) MA plans' allocation of projected revenues and expenses. General Comments: Inpatient Cost Sharing: The report finds that out of pocket costs for inpatient services arc more than double for 16 percent of MA enrollees than in Fee-For-Service (FFS). We do not disagree with the findings: however the report does not address a number of areas that mitigate the impact of the finding: (1) MA bundling of cost sharing versus unbundled FFS cost sharing: We believe that GAO did not take into consideration that an inpatient stay under FFS is not comprised of just the deductible amount for each hospitalization. It also contains a professional service charge which is billed to the beneficiary. As an example of the type of expense incurred by a FFS beneficiary for a 90-day hospital stay. the average professional service co-pay under FFS alone is equal to S5.286. MA enrollees do not receive a separate bill for professional services. The cost of these Part B services were not incorporated into the GAO analysis and therefore overstate the effect of inpatient cost sharing on the possible MA enrollees cited in the draft report. 2) MA is more protective for longer hospitalizations than FFS: The report also does not contemplate the FFS costs for a long hospitalization (e.g., more than 15 days) compared to a long hospitalization under MA. The Centers for Medicare & Medicaid Services (CMS) has data to show that MA plans have lower cost sharing for an inpatient stay than an FFS stay for longer stays — for example 95 percent of MA plans have a lower cost share than FFS for a 90-day stay, and 85 percent of MA plans have a lower cost share than FFS for a 25- day stay. Figure 4 on Page 21 is misleading in that it only compares out-of-pocket costs for 15 days of hospitalization or less. Less healthy beneficiaries may have longer than 15-day hospitalizations in a year. We would recommend providing longer inpatient hospitalization information and a figure demonstrating it to give a more balanced perspective on inpatient hospitalizations. 3) The report does not address effective out-of-pocket maxes for inpatient stays: Maximum out-of-pocket values in the plan benefit package (PBP) data are not the only way to specify the maximum amount an enrollee will pay for a specific benefit. For example, a plan showing no out of pocket maximum value in its PBP but having a $200 per day copay for days 1 through 6 of inpatient stays has an effective maximum enrollee out-of-pocket cost of $I,200 for inpatient stays regardless of the length of stay. In fact, most plans "max out" their out of pocket costs for inpatient stays at values far less than those for FFS. Our data show that, for the same population of plans GAO includes, 74 percent of enrollees in 2007 were in plans having an effective maximum out of pocket for inpatient stays of $1,000 or less and nearly 100 percent of enrollees were in plans having an effective maximum out of pocket costs less than the FFS Medicare value. 4) Misrepresentations of data in text, table and figures on inpatient cost sharing: * Footnote b to Table 4 states: "In some cases, MA plans include cost sharing for professional services, such as physician visits in, in the inpatient category." We believe that because plans include professional services in the bundled inpatient cost sharing, the data in Table 4 is a misrepresentation of a true comparison between MA inpatient cost sharing and FFS inpatient cost sharing. That is, inpatient cost sharing for MA will naturally look higher than FFS in your table because of the inclusion of professional services under the inpatient services category for MA. * Also, about 1/3 of Medicare FFS inpatient admissions fall within an existing benefit period resulting in no beneficiary liability for the Part A deductible. Also, FFS beneficiaries are required to pay cost sharing beginning on day 61 of an inpatient admission. Further, FFS beneficiaries are required to pay certain Part B cost sharing for care provided in inpatient facilities. Accordingly, the FFS representation in Figure 4 has an incorrect level and slope, and we suggest that the chart be modified accordingly, or removed from the report. 5) Part B Drugs and out-of-pocket exclusion and more than 1 million affected enrollees: We believe that the number and impact contained in the report is overstated. Part B drugs administered in a physician's office would be included under an out-of-pocket cap because the bill for the Part B coinsurance would be incurred as a physician office visit. The exclusion of Part B drugs from out of pocket maximums is for those drugs obtained from a pharmacy (as opposed to at the physician's office). We believe Part B drugs obtained at the pharmacy represents a subset of the 2007 plans with a Part B drugs out-of-pocket maximum exclusion. 6) We believe that the report misses the fact that the majority of MA benefit packages in 2007 are better and are more protective for out-of- pocket costs than FFS. Overall, MA benefit packages in 2007 are richer than FFS. Some examples include: * MA plans are not required to cap out-of-pocket expenses for enrollees; however, almost one-half (48 percent in 2007) cap out-of- pocket expenses for enrollees, making them far more protective than an FFS beneficiary who has no cap on out-of-pocket costs. * 72.8 percent of enrollees are in plans with a 21-day inpatient psychiatric stay cost less than that of FFS. * 93.6 percent of enrollees are in plans with an out-of-pocket cost for annual renal visits (dialysis) less than that of FFS. * 98.4 percent of eligibles have access to a plan with a maximum out-of- pocket value of $3.150 or less that includes key specific benefit categories. * 87.3 percent of the plans cover additional inpatient hospital days beyond the FFS allowed 90-day maximum. 7) We believe that the report fails to acknowledge that MA plans provide beneficiaries with a variety of choices in benefit structures other than original Medicare. These choices allow beneficiaries to select a benefit structure that best suits their medical needs while taking their financial considerations into account. For example, some beneficiaries may prefer to pay lower cost-sharing for some Medicare benefits that they anticipate using during the year, and are willing to accept the risk of paying higher cost-sharing on other benefits should they need them. 8) We are also concerned that this report is unfairly skewed. To name a few that we believe should he addressed: * The title is misleading and does not reflect the content of the report. GAO report titles should be neutral, or, at least, not misrepresent the contents of the report. * On page 1 of the report, the discussion of the payment increases based on the Medicare Prescription Drug, Improvement and Modernization Act of 2003 fails to address the intention of Congress to expand access to MA plans in rural areas. * The presentation of MA plans that offer cost sharing that is higher than Medicare FFS does not provide a similarly structured presentation of examples of plans offering cost sharing lower than FFS. As such, the report relies on outlier plan designs (high inpatient cost sharing) with outlier plan services (e.g. long inpatient lengths of stay) to make its point. [End of section] Appendix VI: GAO Contact and Staff Acknowledgments: GAO Contact: James C. Cosgrove, (202) 512-7114 or cosgrovej@gao.gov: Acknowledgments: Other contributors to this report include Christine Brudevold, Assistant Director; Jennie Apter; William Black; Alexander Dworkowitz; Gregory Giusto; Dan Lee; Hillary Loeffler; and Christina C. Serna. [End of section] Footnotes: [1] Medicare is the federally financed health insurance program for persons aged 65 and over, certain individuals with disabilities, and individuals with end-stage renal disease. Medicare Part A covers hospital and other inpatient stays. Medicare Part B is optional insurance, and covers hospital outpatient, physician, and other services. Medicare Parts A and B are known as original Medicare or Medicare FFS. Medicare beneficiaries have the option of obtaining coverage for Medicare Part A and B services from private health plans that participate in Medicare's MA program--also known as Medicare Part C. All Medicare beneficiaries may purchase coverage for outpatient prescription drugs under Medicare Part D. [2] Pub. L. No. 108-173, § 201, et. seq., 117 Stat. 2066, 2176. [3] Private health plans had previously provided heath coverage to Medicare beneficiaries through the Medicare + Choice program. MMA renamed the program "Medicare Advantage" and changed certain payments and other aspects of the program. [4] Medicare Payment Advisory Commission, The Medicare Advantage Program and MedPAC Recommendations (Washington, D.C.: April 2007). [5] Congressional Budget Office, The Medicare Advantage Program: Enrollment Trends and Budgetary Effects (Washington, D.C.: April 2007). [6] For a discussion of Medicare's long-term financial challenges, see GAO, Long-Term Budget Outlook: Saving Our Future Requires Tough Choices Today, GAO-07-342T (Washington, D.C.: Jan. 11, 2007). [7] Medicare compares a plan's bid to the benchmark after adjusting the benchmark to reflect the health status of the plan's enrollees. [8] About 95 percent of MA beneficiaries are in plans that receive rebates and 41 percent of MA beneficiaries are in plans that charge additional premiums. Some plans also offer optional benefits, which beneficiaries can purchase with the standard benefit package. Rebates can not be used for optional benefits. [9] Margins, sometimes referred to as profits, refer to plans' remaining revenue after medical and non-medical expenses are paid. In certain circumstances, such as for new market entrants, CMS allows a plan to have a negative margin, meaning that the plan's revenue is less than its combined medical and non-medical expenses. [10] HMOs account for 71 percent of total MA enrollment; PFFS plans 21 percent; PPOs 5 percent; and PSOs 1 percent, totaling to 98 percent of enrollment. The remaining 2 percent of beneficiaries were enrolled in Medical Savings Accounts and regional PPOs. Beneficiaries in HMOs are generally restricted to seeing providers within a network, while PFFS beneficiaries can see any provider that accepts the plan's payment terms. Beneficiaries in PPOs can see both in-network and out-of-network providers but must pay higher cost-sharing amounts if they use out-of- network services. PSOs are MA plans that are operated by a provider or providers. [11] Some MA plans only cover Medicare Part B services. [12] MA plans do not cover hospice care, a benefit that is provided under Medicare FFS. [13] U.S. citizens and permanent residents meet Medicare's work requirement if they worked for at least 10 years in Medicare-covered employment or if their spouse worked for at least 10 years in Medicare- covered employment. [14] Beneficiaries who are also eligible for Medicaid can have their Part B premium paid for by their state Medicaid program. [15] Individuals with end-stage renal disease are not eligible for most MA plans, unless they develop the disease while enrolled in an MA plan. 42 U.S.C. § 1395w-21(a)(3)(B)(2000). [16] Many Medicare FFS beneficiaries pay premiums for a type of supplemental insurance known as Medigap, which limits beneficiary cost sharing for Medicare-covered services. Medigap policies do not cover the cost sharing of MA beneficiaries. [17] CMS officials said that the thresholds that trigger further review by CMS are at or above Medicare FFS cost-sharing levels. For example, in 2007 Medicare FFS beneficiaries were charged a $992 deductible for hospital services, so the cost-sharing threshold was at or above $992. [18] CMS officials indicated that in evaluating 2008 plans, they stratified plans based on having an out-of-pocket maximum of $3,250, instead of $3,100. [19] The rebate amounts allocated to cost sharing include some non- medical expenses, such as administrative costs and plans' margins. [20] The average additional premium has been standardized to represent a beneficiary of average health status. [21] The 888 plans that received a rebate and did not charge an additional premium projected that they would allocate 11 percent ($11 PMPM) of their rebate to additional benefits, 21 percent ($22 PMPM) to Part D premium reductions, 3 percent ($4 PMPM) to Part B premium reductions, and 65 percent ($70 PMPM) to cost-sharing reductions. The 986 plans that charged additional premiums and received a rebate projected that they would allocate 14 percent ($8 PMPM) of their rebate to additional benefits, 3 percent ($2 PMPM) to Part D premium reductions, 0 percent ($0 PMPM) to Part B premium reductions, and 83 percent ($44 PMPM) to cost-sharing reductions. These numbers are enrollment weighted. [22] Some categories were identified by CMS as unreliable and were excluded from our analysis. [23] Average cost sharing reflects expenditures for the entire population and includes both beneficiaries who are projected to use a certain category of service and beneficiaries who are not projected to use that service. [24] Medicare FFS beneficiaries could have paid the deductible more than once for multiple visits under some circumstances. The 2007 deductible was $992 for each benefit period. Under Medicare FFS, a benefit period begins the day a beneficiary enters a hospital, skilled nursing facility, or critical access hospital, and it ends when the beneficiary has not been an inpatient of a hospital, skilled nursing facility, or critical access hospital for 60 consecutive days. A Medicare FFS beneficiary who had three hospital stays in one benefit period in 2007 would have paid a $992 deductible, while a beneficiary who had three hospital stays in three separate benefit periods would have paid a $992 deductible for each hospital stay, or $2,976. [25] The plans either had no out-of-pocket maximum or had a maximum that was above $3,100. In addition, the plans had no service-specific maximum for inpatient services. [26] The average length of stay in Medicare FFS was 5.4 days in 2005, according to a MedPAC analysis of Medicare cost report data. For plans with no out-of-pocket maximum and a per day copayment of $200 or more for the first 10 hospital days, beneficiaries would have been billed at least $2,000 for a 10-day hospital stay and at least $3,000 for three stays that are each 5 days long. However, beneficiaries in plans with an out-of-pocket maximum and a per day copayment of $200 or more could have been billed less than these amounts if they had already paid cost sharing for other categories of services. About 15 percent of hospital stays under Medicare lasted 10 days or more in 2004, according to CMS data. [27] Medicare FFS does not have an out-of-pocket maximum. However, Medicare FFS beneficiaries who have supplemental insurance can have some or all of their cost sharing paid for. Medicare FFS beneficiaries who buy Medigap insurance have their Part A and Part B cost sharing paid for by their Medigap plan, although they still may pay deductibles. Medicare FFS beneficiaries with Medicaid and with employer plans can also have some or all of their cost sharing paid for by their plan. As of 2004, 26 percent of Medicare beneficiaries had Medigap insurance, 17 percent had Medicaid, and 38 percent had employer insurance, with some beneficiaries having more than one type of supplemental insurance. Data are based on MedPAC's analysis of the 2004 Medicare Current Beneficiary Survey. [28] A plan was considered to have excluded a service category from the out-of-pocket maximum if the out-of-pocket maximum did not cover that service category and if the plan had no service-specific maximum for that category. [29] Non-medical expenses include administration, marketing, and sales. Margin is the amount of revenue above or below the revenue needed to cover medical and non-medical expenses. Allocations to medical expenses, non-medical expenses, and margins do not add to $783 PMPM due to rounding. [30] There is no definitive standard for what a medical loss ratio should be. For example, the CHAMP Act, H.R. 3162, 110th Cong., § 414 (2007), which was passed in the House of Representatives on August 1, 2007, included a medical loss ratio threshold of 0.85. In contrast, individual Medigap policies are currently required to achieve a medical loss ratio of at least 0.65, while group Medigap policies are required to achieve a medical loss ratio of at least 0.75. AHIP reported that from 1960 to 2003, the medical loss ratio for private plans averaged about 0.88. [31] Direct administration accounts for functions that are directly related to the administration of the MA program, such as customer service and medical management. Indirect administration accounts for functions that may be considered "corporate services," such as accounting operations and human resources. Private reinsurance is the insurance provided by another company that assumes financial risk previously assumed by the MA plan. The net cost of private reinsurance is equal to the reinsurance premium less projected reinsurance recoveries. [32] For Medical Savings Account (MSA) plans, Medicare makes a deposit into a beneficiary's savings account if the bid is lower than the benchmark, instead of providing the plan with a rebate. Regional Preferred Provider Organizations (PPO) can receive rebates, but their benchmarks are determined differently than local plans. Due to these differences, the example in this appendix does not refer to MSA plans and regional PPOs. [33] Percentage of MA enrollment by plan type is based on August 2007 enrollment. [34] Centers for Medicare & Medicaid Services, Instructions for Completing the Medicare Advantage Bid Pricing Tool For Contract Year 2007 (Baltimore, Md.: May 2006). [35] If a plan has a population with health expenditures that on average are the same as those for Medicare FFS, then the plan would have a risk score of one. If a plan has a population with projected health expenditures that on average are greater than or less than those for an average beneficiary in Medicare FFS, then the plan's risk score would be greater than or less than one, respectively. [36] The bid pricing data exclude the additional revenue requirements for beneficiaries with end-stage renal disease from this calculation. [End of section] GAO's Mission: The Government Accountability Office, the audit, evaluation and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO's commitment to good government is reflected in its core values of accountability, integrity, and reliability. Obtaining Copies of GAO Reports and Testimony: The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each weekday, GAO posts newly released reports, testimony, and correspondence on its Web site. To have GAO e-mail you a list of newly posted products every afternoon, go to [hyperlink, http://www.gao.gov] and select "Subscribe to Updates." Order by Mail or Phone: The first copy of each printed report is free. Additional copies are $2 each. A check or money order should be made out to the Superintendent of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a single address are discounted 25 percent. Orders should be sent to: U.S. Government Accountability Office: 441 G Street NW, Room LM: Washington, D.C. 20548: To order by Phone: Voice: (202) 512-6000: TDD: (202) 512-2537: Fax: (202) 512-6061: To Report Fraud, Waste, and Abuse in Federal Programs: Contact: Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: E-mail: fraudnet@gao.gov: Automated answering system: (800) 424-5454 or (202) 512-7470: Congressional Relations: Ralph Dawn, Managing Director, dawnr@gao.gov: (202) 512-4400: U.S. Government Accountability Office: 441 G Street NW, Room 7125: Washington, D.C. 20548: Public Affairs: Chuck Young, Managing Director, youngc1@gao.gov: (202) 512-4800: U.S. Government Accountability Office: 441 G Street NW, Room 7149: Washington, D.C. 20548: