istockphoto/ spxchrome

Bundling Payment for Episodes of Hospital Care

Issues and Recommendations for the New Pilot Program in Medicare
Harriet L. Komisar, Judy Feder, and Paul B. Ginsburg July 2011

w w w.americanprogress.org

Bundling Payment for Episodes of Hospital Care

Issues and Recommendations for the New Pilot Program in Medicare
Harriet L. Komisar, Judy Feder, and Paul B. Ginsburg July 2011

Contents

1 Introduction and summary 4 Payment bundling and the Affordable Care Act 6 The case for bundled payments for episodes of care around hospitalization 11 Issues and recommendations 22 Conclusion 23 Endnotes 25 About the authors and acknowledgements

Introduction and summary

At the heart of health reform is the fundamental challenge to simultaneously improve the quality of our health care and lower its costs. And at the heart of meeting that challenge is changing the way we use and pay for care. The Affordable Care Act is replete with measures aimed at this goalincluding initiatives to promote prevention and primary care, to reward good (and penalize poor) provider performance, and to combine now-separate payments to doctors, hospitals, and other providers into collective payment arrangements for multiple services, thereby promoting better-coordinated, more accountable care. In the middle of the mix is the requirement that the Department of Health and Human Services launch a pilot project to bundle Medicare payments around hospital episodes of carethat is, pay collectively for the services an individual receives during a hospital episode (which includes a period of time after discharge), rather than paying separately for each service delivered by each health care provider at the hospital.1 By paying for an episode of care as a whole, bundling offers providers the flexibility and financial incentive to coordinate care within an episode and avoid preventable complications and readmissions. Bundling boasts the potential to benefit: Patients through better care Health care providers through financial rewards for delivering that care more efficiently The Medicare program through lower costs. Bundling, in short, can be a win-win-win for everyone involved in episodes of care, including taxpayers. Hospital episode bundling is currently receiving less policy attention than a broader payment reform known as accountable care organizations, which would create new payment incentives for all services a person receives during the year

center for American progress | Bundling payment for episodes of hospital care

that is, pay on a per-person basis rather than on a per-episode basis. But given the urgency as well as the uncertainties of efforts to improve our health care system, few would suggest we put all our eggs in one basket. With its potential to improve patient care by increasing coordination and reducing unnecessary services as well as reducing complications, errors, and hospital readmissions, hospital episode bundling offers a promising opportunity to promote efficient, coordinated care that should be actively pursued. The goal of this report is to offer guidance on key choices in designing a pilot program to most effectively explore episode bundling to meet health reforms twin goals of better quality care at lower costs. Specifically, an effective bundling pilot program would: Encourage the broadest possible provider participation in nationally scalable payment methods, with a payment design that sets broad conditions for participation but leaves operational details to participating health care providers and is open to all providers who satisfy the conditions. This new model should build on current payment methods to simplify implementation. Target the pilot program to diagnoses with the greatest potential to improve both quality and efficiency by focusing on high-volume conditions for which interventions are well established and supported by clinical guidelines, and for which, despite those guidelines, actual treatments (and related costs) vary substantially. As experience develops, bundling can be applied to a broader array of conditions. Design payment methods to promote collaboration among providers, attract participants, and assure quality. To facilitate collaboration, offer providers the option of either a single bundled payment amount that they would divide among themselves, or an alternative payment method that pays each individual provider involved in the episode an amount that blends existing payment methods with financial incentives based on the combined performance of all providers involved in the episode. Set initial payment levels to reflect the current costs of care, to attract participants, limiting risks and offering health care providers up-front resources and rewards to efficient delivery. In subsequent years, constrain annual rate increases to yield Medicare savings over the life of the pilot. And to assure quality care and protect patients, vary payments to reflect patients complexities, tie payments to quality performance, and require public reporting of quality measures.

center for American progress | Bundling payment for episodes of hospital care

 Engage and protect Medicare enrollees by requiring participating providers to inform beneficiaries about the pilot program, providing patient advocacy support to beneficiaries, and allowing beneficiaries to retain the option of seeking care from nonparticipating providers. In the pages that follow, we will describe the pilot program mandated by Congress, examine the reasons to develop episode-of-care payments involving hospitalizations, and then explore the best ways we believe this pilot program could be set up and run. We then close the paper with our detailed set of recommendations that we believe can best test the efficacy of episodes of care as a payment model to lower our nations health care costs while improving the quality of care.

center for American progress | Bundling payment for episodes of hospital care

Payment bundling and the Affordable Care Act

Broadly defined, payment bundling means paying for health care services with a single, comprehensive payment amount that covers multiple services and items received by a patientinstead of making separate fee-for-service payments for each particular service or item. Medicares prospective payment system for inpatient hospital care already uses bundles to pay for all services provided during a hospital stay; the payment amount varies according to the patients diagnosis and major treatment decisions but does not depend on the specific quantities of specific services received during the stay.2 Medicare also applies a narrowly defined bundle in paying surgeons for an operation and for one day of preoperative and 90days of post-operative care.3 The Affordable Care Act builds on these payment approaches by extending the bundle to cover payment across multiple providers. Although bundles could be shaped in various ways, the Affordable Care Act explicitly requires a National Pilot Program on Payment Bundling in Medicare to pay for episodes of care around hospitalization. The new health reform law specifies an episode as the time period from three days prior to hospital admission through 30 days after dischargebut allows the secretary of Health and Human Services to designate a different timeframe. As specified in the law, the services to be covered by the episode payment consist of: Acute inpatient hospital Physician services delivered in and outside the hospital Outpatient hospital services Emergency room services Post-acute services such as physical therapy and nurse visits at home Appropriate services identified by the secretary such as care coordination and transitional care services

center for American progress | Bundling payment for episodes of hospital care

The pilot will test payment bundling in Medicare for 10 conditions, to be selected by the secretary, with voluntary participation by providers. Medicare will pay a participating provider-entity a bundled amount for each applicable Medicare beneficiarythat is, each beneficiary who is admitted to a participating hospital with one of the pilot programs 10 selected conditions and meets certain Medicare enrollment criteria. Specifically, the beneficiary must be eligible for care under Medicare Parts A (hospital coverage) and B (medical insurance), but not be enrolled in a private health plan through Part C (Medicare Advantage) or PACE (Programs of All-Inclusive Care for the Elderly). The entity receiving the bundled payment could be a formal organization comprising multiple providers (including, for example, a hospital, multiple physicians, and post-hospital care providers) or one of those providers (a hospital or physician group, for example) with contractual arrangements with others. The pilot is scheduled to begin by January 1, 2013, and continue for five years. But an important feature of the law is that the secretary has the option of expanding the duration and scope of the pilot if expansion is expected to reduce Medicare spending while improving, or not reducing, quality and not limiting Medicares coverage or benefits for individuals. Thus, if the pilot is successful, bundling could become a significant element in Medicare payment. In the next section we explain the potential value of bundling around hospitalizations.

center for American progress | Bundling payment for episodes of hospital care

The case for bundled payments for episodes of care around hospitalization
Compared with fee-for-service payment, payments for bundled episodes of care alter the financial incentives in a fundamental way. By paying for an episode rather than for each service, bundled payment encourages providers to determine which services are appropriate within an episode and to eliminate the unnecessary ones, in contrast to rewarding volume of services. Further, paying providers as a group, rather than paying each separately, encourages providers to work together to coordinate care, eliminate duplicative and unnecessary services, and avoid preventable complications. As a result, bundled payments have the potential to deliver better care at lower costs by reducing fragmentation and increasing the coordination of care while also reducing inefficiencies.4 Providers, as well as patients and the Medicare program, can potentially benefit from payment bundling. With bundling, providers have the opportunity to retain financial rewards from finding ways to reduce unnecessary services and avoiding preventable complications and flexibility in finding ways to do so. Hospitalswhich, beginning in October 2012, will be financially accountable for especially high readmissionscan benefit from the flexibility that bundling permits as well as relationship building with physicians. Bundled payments will, for example, offer a financial incentive for physicians to be engaged in helping hospitals reduce complications, avoid re-admissions, and use hospital resources efficiently. Bundling payment around hospitalization, as required in the pilot, will create incentives to improve the coordination and efficiency of care both during the hospital stay and during a post-hospital period. During the hospital stay, it is physicians who direct a sizable portion of the resources. Better aligning the financial incentives of physicians and hospitals with bundled payment could lead to more efficient use of those resources for example, through more cost-effective choices of medical devices and pharmaceuticals. Evaluations of the Medicare Participating Heart Bypass Center Demonstration project, which tested bundled payment for inpatient hospital and physician services in the 1990s, offer some evidence that such savings can be achieved (see box).

center for American progress | Bundling payment for episodes of hospital care

Previous experience with bundled payments for hospital episodes

Although previous experience with payment bundling is limited, there are several public- and private-sector initiatives that can inform the pilot program. Two Medicare demonstration projects involve bundled payment for inpatient hospital episodesthe Participating Heart Bypass Center Demonstration, completed in 1996, and the Acute Care Episode Demonstration, begun in 2009. In addition, some private organizations have developed bundled payment initiatives. Among these, two examples that may be the most helpful in developing the pilot program are the PROMETHEUS Payment model and Geisinger Health Systems ProvenCare program.5 So lets examine each of these programs briefly in turn. Medicare Participating Heart Bypass Center Demonstration In the 1990s, the Medicare Participating Heart Bypass Center Demonstration tested the application of a single, negotiated bundled price for inpatient hospital and physician care for coronary artery bypass graft patients. The demonstration included a total of seven hospital sites; four participated for five years, 1991-1996, and three participated for three years, 1993-1996.6 Each hospital received a bundled payment amount for hospital and physician services during the inpatient stay, plus any readmissions within 72 hours of discharge; the bundled amount increased annually based on updates in Medicares hospital and physician payment rates. Sites chose differing methods of dividing the payment among the hospital and physicians. Patients covered by the demonstration were responsible for a single preset cost-sharing amount in place of the usual deductible and co-insurance amounts, with the amount set to be less than expected for a typical admission. Overall, the demonstration was estimated to have reduced Medicare spending by about 10 percent compared with what it otherwise would have been for covered patients, without adversely affecting patients.7 An evaluation of the first four hospitals in the demonstration found that the hospitals achieved lower costs mainly through reductions in costs of intensive care unit and routine nursing, pharmacy, and laboratory.8 Physicians, for example, became more involved in reviewing hospital pharmacies drug formularies and making substitutions to reduce costs. Medicare Acute Care Episode Demonstration A second Medicare demonstration, the Acute Care Episode Demonstration, is currently testing bundled payment for several cardiac procedures (such as coronary bypass procedures and cardiac pacemaker procedures) and orthopedic procedures (such as knee replacement and hip replacement surgeries).9 Participating physician-hospital organizations receive a single global payment that covers Medicare hospital and physician services provided during the hospital stay. After the demonstrations first year, the Centers for Medicare and Medicaid Services, the federal agency that administers these two health care programs, may consider broadening the scope of services in an episode to encompass some post-acute care. Each site decides for itself how the payment is divided among providers; financial incentives for providers to promote efficiency are permitted subject to certain rules and limits. The participating sites and their payment amounts are determined through a competitive bidding process. So far, CMS has selected five sites for the demonstration: Two sites began their programs in 2009, a third began on January 1, 2010, and the other two began in November 2010. The demonstration may expand the number of sites up to a maximum of fifteen. The demonstration is also testing the effects of offering beneficiaries a financial incentive to choose participating sites, referred to as Medicare Value-Based Care Centers. The dollar amount of the incentive varies by procedure and site. Beneficiaries receive half of the estimated amount Medicare saves, up to a maximum equal to their annual Part B medical insurance premium amount (and currently not greater than $1,157).10 PROMETHEUS Payment Model The PROMETHEUS Payment model is part of an ongoing project aimed at developing an episode-of-care approach to paying for health care for chronic and acute conditions.11 The model was designed by a nonprofit organization, PROMETHEUS Payment, Inc. (now part of the Health Care Incentives Improvement Institute), with primary support from the Robert Wood Johnson Foundation. The PROMETHEUS Payment model uses a bundled payment amount, adjusted for the patients severity and complexity, to pay for all services provided during an episode of care. These bundled payment amountscalled evidence-informed case ratesare based on the appropriate services for treating a condition as determined by clinical guidelines and expert opinion, plus

center for American progress | Bundling payment for episodes of hospital care

Previous experience with bundled payments for hospital episodes (continued)

an allowance for the costs of potentially avoidable complications. The allowance for potentially avoidable complications is a portion of the costs of these complications for the condition indicated in data reflecting typical experience. The objective is to give providers an incentive to prevent avoidable complications. And if they reduce the cost of complications, on average, to less than the bundled payment amounts allow, then the revenues will more than cover the costs of delivering appropriate care. So far, the PROMETHEUS model has created 21 evidence-informed case rates that include five inpatient procedures (such as hip replacement and heart bypass surgery), five outpatient procedures (such as knee arthroscopy and colonoscopy), plus acute and chronic medical conditions (such as stroke and diabetes).12 The relevant episode length depends on the condition; for hospital procedures, the episode length includes a rehabilitation period. PROMETHEUS is now being tested at four sites. HealthPartners, a nonprofit health plan in Minnesota, has contracted with local provider networks. Independence Blue Cross and Crozer-Keystone Health System have partnered in Pennsylvania. Employers Coalition on Health is working in partnership with local healthcare providers in Rockford, Illinois. And Priority Health-Spectrum Health is working with PROMETHEUS in Michigan.13 Additional sites are being developed in New York and Colorado. Geisinger Health Systems ProvenCare A second private initiative that involves bundled episode payment for hospital services is the ProvenCare program developed by Geisinger Health System, a large, non-profit, integrated delivery system in Pennsylvania. Beginning in 2006, Geisinger implemented a ProvenCare program for elective coronary artery bypass graft surgery that uses a checklist of 40 processes or benchmarks that should be completed for every elective CABG patient, including a determination that the surgery is appropriate for the patient, based on established guidelines for best practices. As part of the program, Geisinger charges a fixed rate for elective CABG surgery that covers all services related to the procedure and to treating any related complications that occur within 90 days following the surgery. Evaluations provide evidence that ProvenCare has reduced hospital costs, complication rates, and readmissions among CABG patients.14 More recently, Geisinger developed ProvenCare programs for additional types of episodes, many of which are also priced as a bundled episode amount.15

Bundled payment will also affect services in the post-hospital period. The provider organization receiving the bundled payment will be responsible for arranging and coordinating follow-up care after the hospital stayand, importantly, for addressing any complications that arise in the covered post-hospital period. Financial responsibility encourages providers to actively prevent complications and avoid their associated treatment costslowering costs and promoting quality at the same time. In contrast, under current payment incentives, discharge policies at hospitals today typically focus on getting the patient to the next step, with little incentive to see that follow-up care is of good quality. The high rate of hospital readmissions among Medicare beneficiaries is evidence of currently inadequate support as patients

center for American progress | Bundling payment for episodes of hospital care

transition from the hospital. The Medicare Payment Advisory Commission, or MedPAC, which advises Congress on Medicare issues, estimates that 18 percent of Medicare beneficiaries discharged from a hospital in 2005 were readmitted within 30 days, and that about three-quarters of those readmissions (or about 13 percent of total admissions), costing $12 billion, were potentially preventable.16 Bundling is one of several new payment arrangements the Department of Health and Human Services is required to explore under the Affordable Care Act. Broad experimentation makes senseand bundling can be compatible with other initiatives, including Accountable Care Organizations. 17 But bundling has advantages in its own right. First, its focus on hospital episodes, with a bounded set of services and providers, raises fewer organizational challenges than does a population-based payment arrangement such as under accountable care organizations, affecting all services a patient may need over a year. Second, bundled episode payment may actually be a desirable endpoint in itself, preferable to population-based payment. Episode-based bundled payment could support a system of provider organizations that target specific areas, such as orthopedic procedures or services for people with diabetes. Arguably, developing specialized organizations supports more competition, consumer choice, and consumer satisfaction than a payment system relying on large integrated health systems to provide services. Bundling around hospital stays, as in the pilot, may provide a transitional step to a broader set of episode-based bundled payments. including outpatient acute and chronic care episodes. Bundlings potential to improve quality and lower costs does not mean it is a payment policy without challenges or without risks. Organizational challenges are significant, as is defining what services are in and out of a hospital bundle.18 Further, bundlings incentives pose some negative, alongside positive, possibilities. By rewarding physicians as well as hospitals for an efficiently-managed hospital admission, bundling may generate more hospital episodesthus, potentially increasing the number of inappropriate hospital episodes, even though services within each one would be efficiently used.19 Moreover, by rewarding providers for lower costs, episode payments may encourage providers to skimp on services within an episodeespecially on services for which any adverse repercussions occur down the road, outside the time frame (or service scope) of the episodeor avoid patients who are likely to be especially costly within a diagnosis category.20

center for American progress | Bundling payment for episodes of hospital care

In addition, if widely adopted, episode bundling around hospitalizations could potentially lead to increased concentration in health care markets and fewer choices of providers for beneficiaries. For instance, if a surgeon who formerly treated patients at more than one hospital instead enters a contract (or employment arrangement) with a single hospital to provide bundled services, then patients of that surgeon will have fewer choices of hospitals. Similarly, hospital contracts with selected post-acute providers might lead to fewer options if some free-standing providers not entering into arrangements with hospitals no longer have enough clients to stay in businessalthough some closures may be among the lowest quality providers, other providers might also be affected. That bundling poses potential risks as well as benefits does not mean we should not explore it. Rather, it means we should explore it with attention to policy design choices that mitigate the chance of negative outcomes and make the most of the opportunities bundled payment offers. We turn to design in the next section.

10

center for American progress | Bundling payment for episodes of hospital care

Issues and recommendations

In order to contribute to the overarching goals of the new health reform law delivering quality care at lower costsan effective design of a pilot for bundled payment must address the following questions: How can the pilot be designed to lead to national application? What types of conditions should the pilot target? How should bundled services be paid for? How will the pilot engage and protect Medicare beneficiaries? So lets turn to ways the Center for Medicare and Medicaid Services can best answer these questions to achieve health reforms goals.

How can the pilot be designed to lead to national application?

Medicare demonstrations (including those described in the box on page 7) are designed as carefully specified research enterprisesthat is, narrowly defined to test a highly specified policy intervention, involve a small set of providers, adhere to evaluation methods that can constrain adaptation, and produce results that may or may not be practically replicable in national policy. To successfully promote change, implementation of the Affordable Care Act will require a new approach to innovation and experimentation.21 The key to that approach when designing an episode-based bundling pilot should be from the outset to test methods that have the potential to be adapted and adopted widelythat is, to be scaled to national implementation in the future. Achieving this goal requires a design that is both sufficiently simple to attract broad provider participation and can be readily administered by the Centers for Medicare and Medicaid Services, or CMS.

11

center for American progress | Bundling payment for episodes of hospital care

Specific design choices will be discussed below. But the general elements are straightforward. For providers, the key is clear conditions for participation and clear standards for performancewith operational details (for example, the nature of payment allocation, as discussed below) left in large part to the discretion of participating providers. For CMS, the key is employment of payment and monitoring mechanisms that build on CMSs existing capacity and methods. For instance, building on current payment methods can not only facilitate management of the pilot, but simplify its broader adoption in the future if it proves successful. A focus on facilitating adoption also calls for extensive, rather than tightly constrained, participation in the pilot. For accountable care organizations, the Affordable Care Act opens participation to all providers who want to participate and satisfy participation criteria. Although accountable care organizations are technically a program, rather than a pilot, the new Center for Medicare and Medicaid Innovation within CMS should adopt a similar approach for the bundling and other pilots. In so doing, CMS could dramatically speed up the innovation processsimultaneously learning about and promoting widespread change. Unlike the accountable care organization program, a national pilot for bundled episodes of care would be time-limitedwith continuation or modification requiring an explicit decision by the secretary of Health and Human Services, based on the secretarys assessment of its impact on costs and quality of care. Although these pilot characteristics create more uncertainty for providers than having formal program rules, they offer providers as well as policymakers the opportunity to modify policy based on experiencea significant advantage in promoting not only rapid but effective change. The potential for broad adoption of bundling can be further enhanced if private insurers join Medicare in exploring the bundled approach. Although many private insurers do not currently use Medicares so-called diagnosis-related group, or DRG approach to hospital payment, introduction of a new and broader payment bundling in Medicare may provide an opportunity to align public and private payment methods and the incentives providers face. Private payer participation in a bundling initiative (paying in the same way as Medicare for similarly defined sets of services, though not at the same rate) will give health care providers more incentives to change their pricing behavior, extend that behavioral change to a larger share of the health care system, and significantly increase the impact of this payment innovation on the efficiency and quality of care.

12

center for American progress | Bundling payment for episodes of hospital care

What types of conditions should the pilot target?

The Affordable Care Act requires the pilot program to test bundled payments for a limited number of conditions; the law specifies 10. Although the new health law lists several factors for the secretary of Health and Human Services to consider in selecting the specific conditions, it does not set priorities among factors or establish specific requirements for selected conditions.22 The following criteria are the most important to address. First, a key criterion in selecting conditions is to choose diagnoses for which medical interventions are well-established and supported by evidence.23 The availability of accepted treatment standards can guide providers in developing care plans, and be incorporated into quality assurance measures. They also provide the basis for shared patient and provider decision-making, to assure the appropriateness of hospitalization. A second criterion in choosing diagnoses for the pilot is volume. Choosing conditions with a relatively high volume enables providers to average costs over a large number of patients and makes costs more predictable and stable. Largevolume conditions will also assist in attracting providers to participate in the pilot. Higher volume makes it worthwhile for providers to establish needed contractual relationshipsfor example, for a hospital to establish contracts with physicians and post-acute providersand to invest in organizing and staffing care coordination activities, such as providing care continuity and assistance to individuals and their families when a patient transitions from hospital to home or other setting.24 Similarly, the larger the proportion of Medicare hospitalizations the pilot addresses, the greater will be its impact on quality and efficiency in the program. Third, it is desirable to focus initially on conditions for which there is substantial variation in treatment patterns and expenditures, even after controlling for patients characteristicsas long as there are practice guidelines to inform practitioners. Conditions with large variation from clinical best practices provide the greatest opportunity to reduce unnecessary services and produce savings.25 Consistency of payment reforms with professional norms of practice can also generate consumer confidence in bundling as a means to improve the quality and value of services. Over time, use of bundling may stimulate the development of guidelines for a broader set of conditions, facilitating the reach of bundling as its success is demonstrated.

13

center for American progress | Bundling payment for episodes of hospital care

How should bundled services be paid for?

A number of fundamental decisions need to be made in developing the specific design for bundled payment.26 For the pilot, it will be crucial to be mindful of practical considerationsincluding the need to attract health care providers to participate, the readiness of providers to respond to the new opportunity to change practice patterns, and the types of data it will be feasible to obtain. The pilot can begin with a design that is feasible to implement now but that can, over time, be enhanced to achieve broader results. For instance, over time, the payment design can be expanded and refined to encompass a larger array of medical conditions, incorporate more quality and outcome measures as they become available, and achieve greater cost savings as providers learn how to increase efficiency in the program to assure adequacy.

Offer both single bundled payments and an alternative design

In a basic bundled payment designwhich we refer to as single bundled paymentMedicare would make a single payment for each episode of care. A provider entity, such as a hospital, a physician-hospital organization, a physician group, or another type of provider organization, would receive the payment and be responsible for organizing the range of services included in the episode and dividing the payment among the various providers and suppliers. The entity receiving the payment, for example, could work out contractual arrangements governing how providers would work together and how payment and financial risk would be shared. For bundled episode payment, Medicare needs a method of defining an episode that is, an episode grouper that identifies which services in the time period are related to the hospital episode and which are unrelated.27 Medicare also needs to determine an appropriate level of total payment for an episode. The payment amount would vary based on the condition being treated and be adjusted for additional health conditions of the patient that affect care needs for the episodethat is, adjusted for the patients severity and complexity. Payments could also be adjusted for other factors affecting the cost of providing services, such as input costs in the geographic area, like they are in the current prospective payment system for hospitals. Similar to Medicares current hospital

14

center for American progress | Bundling payment for episodes of hospital care

payment system, any outlier payments for unusually costly episodes would also be included, so thatalong with adequate adjustments for patient complexity providers would be adequately paid for treating patients with the highest needs. The provider entity receiving the payment would in turn need to figure out a method of sharing payment and risk among the participating providers which might consist, for example, of a hospital, numerous physicians providing services to the hospitals Medicare patients, and several post-acute service providers participating in the organization. In a simple example, a hospital could take on the full risk by receiving the single bundled payment and paying other providers on a fee-for-service basis using contracted rates (so the hospital, in this example, would face a loss if the total costs of care exceeded the bundled payment, or gain if costs were less than the payment). But in order to limit the magnitude of risk to each organization and to engage all the providers that are involved, it is likely that hospitals or provider organizations would work out arrangements for sharing losses or gains among providers. Although the single bundled payment design is conceptually simple, provider organizations would face complex issues in establishing contractual relationships and figuring out how to distribute payments and share risks.28 It may be costly for providers to make the investments of time and resources needed to resolve these issues. A single bundled payment model might therefore attract relatively few providers, limiting participation to larger organizations most able to make the necessary investments. To attract a wider range of providers, it may be beneficial for the pilot program to offer potential participants the option of an alternative payment model that does not rely on a single provider entity receiving a single, fully bundled payment and arranging all care in the episode. In particular, this option would use a payment model in which each separate provider involved in an episode receives payments through a method they are already familiar withthe DRG or fee-for-service basisbut also has financial incentives based on the combined performance across all services in the episode. The idea is to include financial incentives for efficiency and better coordination among providers without requiring there to be a single entity that has the ability to receive a single bundled payment and distribute it among individual providers. Similar ideas have been suggested by MedPAC and others.29 Financial rewards or penalties could be computed by comparing actual Medicare spending for a set of covered episodes with a benchmark based on what would have been paid using single bundled payment for this set of episodes. A set of

15

center for American progress | Bundling payment for episodes of hospital care

episodes, for example, could consist of all covered episodes of patients treated at a participating hospital during a year. If total spending for all services in these episodes (hospital, physician, and post-acute services) is below the benchmark, then each of the providers involved in the episodes would receive a financial reward. Similarly, if total spending exceeded the benchmark, then providers payments would be reduced. Initially, as MedPAC suggests, the financial penalties and rewards could apply to each participating hospital and the physicians providing services in the hospitals covered episodes (and not other types of providers) although total spending for all service types in the episode would be used to determine the penalty or reward. 30 The financial rewards, or penalties, could be applied to the hospital and physicians in proportion to each provider types share of baseline spending (that is, based on historic spending for hospital and physician services for the same types of episodes). As experience with this type of payment approach grows, however, other providers involved in the covered episodes should also be subject to financial rewards and penalties.

In setting prices for the pilot, start with current amounts

How initial payment rates are set will have a significant impact on providers willingness to participate in the pilot. To promote participation, the pilot can use current patterns of care to set its initial rates and benchmarks, and then use the potential savings to attract providers to participate.31 This would enable providers to invest resources in making changes to improve care coordination. The rates in the first year would therefore not be intended to achieve aggregate savings. To achieve the broader goal of savings over time, payment increases would be constrained to reduce spending over time compared with what Medicare would otherwise pay. By limiting risk and offering rewards up front, this approach will help overcome providers reluctance to invest staff time and other resources in establishing a program and developing necessary contracts and arrangements. And by constraining rate increases in subsequent years (for example, by holding rates at their initial levels for the three years of pilot or by holding annual increases below the average increase in Medicares payment rates), the bundled rates would yield Medicare savings over the life of the pilot. And if providers achieve cost reductions, the Medicare program can reap larger savings down the roadfor example, by constraining episode rates further over time and by implementing bundled payments program-wide.

16

center for American progress | Bundling payment for episodes of hospital care

For the pilot, payment rates could be calculated that are hospital-specific, but also draw on regional information for post-acute services. A base bundled payment for an episode could be computed for each hospital and episode type using facility-specific data for the specific hospital and its affiliated physicians plus regional data for post-acute providers (because hospitals may refer to numerous post-acute providers). This approach has the advantage of creating an incentive for each participating hospital (or other entity receiving bundled payment) to change their behavior relative to historical patterns to reduce costs. Using DRG payments for hospital services as the core in determining payment makes sense because hospital services are the largest component of spending for hospital episodes. In an analysis of average Medicare spending for episodes around hospital stays that include 30days after discharge, MedPAC found hospital services accounted for more than half of episode spending for three selected, relatively prevalent, conditions (see Figure 1).

Figure 1

Most spending in episodes is for hospital services

Distribution of Medicare spending for hospital episodes by type of service, selected conditions
100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 13% 4% Chronic obstructive pulmonary disease 15% 7% 20% 7% 21% 10% 6% 5% 5% 3% Congestive Coronary artery heart failure bypass graft with cardiac catheterization 56% 52% 76% Hospital Physician Hospital readmission Post-acute care Other

Notes: Based on average risk-adjusted Medicare expenditures during and 30 days following a hosptial stay in 2001-2003. Readmission includes hospital and physician spending during a hospital readmission. Other includes outpatient services and physician services outside the hospital. Source: Medicare Payment Advisory Commission, Report to the Congress: Reforming the Delivery System, (June 2008), available at http://www.medpac.gov/documents/Jun08_EntireReport.pdf.

Using payment rates set by Medicare in the pilot program has important advantages over competitive bidding or negotiated price approaches, both of which have been used in Medicare demonstrations and suggested for private and public bundled payment initiatives. In these two approaches, each provider organization interested in participating would propose bundled payment amounts for the covered diagnoses that Medicare could accept or reject (the competitive bidding approach), or use in negotiating with the provider to reach agreement on rates (the negotiated price approach). The problem with these approaches, however, is that competitive bidding and negotiation work best in situations where providers use price either to compete in a selection process (such as for a demonstration) or to compete in attracting patients. Neither of these situations applies in the pilot program.

17

center for American progress | Bundling payment for episodes of hospital care

Although bidding can work well in Medicare when providers are competing to be selected to participate in a demonstration, it is not as well suited to situations where the goal is potential widespread participation, such as in the bundling pilot. A bidding approach relies on selecting winning bidders, and rejecting others, to achieve cost-savings. In contrast, a payment-rate approach would enable Medicare to allow widespread participation and achieve savings over time by constraining rates. Bidding or negotiated prices can also work well when they provide a financial incentivesuch as through different cost-sharing amountsfor clients to choose among different providers. This approach has the potential to be an effective strategy in private insurance situations. But it is difficult to apply these types of financial incentives to Medicare beneficiaries, as discussed in more detail below. Rates based on historical hospital-specific costs have advantages relative to two other potential approaches to bundled ratesspecifically, rates or benchmarks based on national or regional averages among hospitals, and rates based on evidence-informed protocols. Use of national or regional averages would make the pilot especially attractive to providers who already have costs below the average; these providers would be rewarded under the pilot even without reducing costs (and indeed, this approach could draw participants who were unprepared to make the investments in care coordination and other desired changes in delivery). Selected participation of this type could lead to both an overall increase in Medicare costs (because providers previously below average would now get average payments) and yield less change in service than the hospital-specific payment design suggested here. Basing rates on evidence-based, rather than historical, costs clearly has theoretical appeal. The exploration of evidence-informed case rates in the PROMETHEUS Payment model, which is currently being tested by several health plan-provider partnerships, will provide valuable guidance to future payment development. As currently implemented, PROMETHEUS rates are a blend of estimated costs based on evidence-informed protocols and historical costs reflecting actual experience. The PROMETHEUS partnerships have the flexibility to tailor their approach to the specific circumstances of participating providers. But that tailoring would be difficult to replicate on a national scale. And without it, payment rates tied to specific protocols would place too much weight on the judgments of a panel of experts and likely be too rigid to allow providers enough dis-

18

center for American progress | Bundling payment for episodes of hospital care

cretion to deliver appropriate care to individual patients. In addition, basing rates on a broadly-applied standard, rather than a hospital-specific one, would raise the same challenges as noted above for national or regional averagesnamely that providers with costs that were already below the rate would be attracted to the pilot and rewarded even if their behavior is unchanged. Further, keeping evidence-based case rates up-to-date would pose an enormous administrative challenge given the rapidity of change in medical practice and technology. And the broader the application of bundles across conditions, the greater the burdenimpeding rather than facilitating national application. Although experimentation with evidence-based bundled payment should continue, an aggressive national pilot would do well to start with something simpler.

Include financial incentives to promote quality

A goal of bundling is to improve efficiency and appropriate services while reducing unnecessary services. But a concern is that bundlings incentives may also encourage providers to avoid treating the sickest patients or to fail to provide costly but beneficial services to the patients they treat. Adjusting payments to reflect patients conditions and complexities addresses reluctance to treat patients with the greatest needs. Additional measures are needed to assure quality care. Follow-up services after a hospital stay, including medical and post-acute services, require particular attention because the current inadequacy and lack of coordination in these services contributes to preventable complications and re-admissions. By extending the hospital episode beyond hospital discharge, the pilot aims to address this problemcreating a financial incentive for providers to pay attention to the care an individual receives after leaving the hospital. But some servicesfor example, physical therapy after surgerymight contribute greatly to a patients recovery and ability to resume regular activities, but may not make much of a difference in the risk of costly post-hospital complications. An important concern, then, is that the provider organization responsible for receiving the bundled payment might provide only what it considers the minimum post-acute care necessary to avoid a costly readmission, and fail to provide additional services that would be beneficial to a patient.

19

center for American progress | Bundling payment for episodes of hospital care

Concern that aligning provider incentives to reduce care might harm beneficiaries was reflected in the Department of Health and Human Services Office of the Inspector Generals judgment that terminated experiments with gainsharing arrangements between hospitals and physicians in 1999. The Office of the Inspector General concluded that any arrangement in which a hospital makes a payment, directly or indirectly, to induce a physician to reduce or limit services to Medicare or Medicaid patients is in violation of the Social Security Act.32 Policy perspectives and prescriptions have changed, however. An explicit objective of the Affordable Care Act is to reduce unnecessary services in ways that actually improve quality of care. To assure that bundled payment promotes, rather than undermines, good quality care, payments should be tied to quality performance. Bundled payments could be made contingent on meeting specified performance thresholdsincluding outcome measuresspecific to patients diagnoses. This is the approach the Affordable Care Act requires for accountable care organizations. Further, quality improvement would be encouraged by requiring public reporting of outcome and other quality measures. To promote patient awareness and instill quality-based competition, that reporting must be timely, easily obtainable, and understandable to beneficiaries.

How will the pilot engage and protect Medicare beneficiaries?

The pilot program is driven by the goal of promoting changes in the way care is delivered to yield improvements in the continuity and quality of care that patients experience, alongside savings from reducing avoidable complications and unnecessary services. Patients satisfaction with care received under the pilot is central to its success and potential for expansion. An important design question for the pilot, then, is what information and choices patients will have. Provider participation in the pilot program is voluntary, but decisions are needed as to the information and choices that patients have in the program. It makes sense that beneficiaries of bundled episodes of care would automatically be covered by the pilot if their primary physician for the intervention (say, the surgeon) or hospital is a pilot participant. But patients should be informed of the hospitals and physicians participation and its implications, early enough (except in emergencies) to allow consideration of switching provid-

20

center for American progress | Bundling payment for episodes of hospital care

ers. Information that patients receive should include specifics on appropriate services for their condition and on providers obligations to assure quality care, not only during the hospital stay but through the entire episode. Information should be supplemented with the availability of patient advocacy support. A beneficiarys decision to receive bundled services, however, should not limit all patient choices to providers participating in the pilot. Beyond the hospital and primary physician, beneficiaries should retain the ability to select nonparticipating providers without financial penalty. If a patient receiving a bundled episode of care wants services from a post-acute provider or a consultation from a physician that is not affiliated with the hospital for the bundling pilot, then the beneficiary would be able to obtain those services as covered under their regular Medicare benefits. The costs of these services would be attributed to the organization receiving the bundled payment when measuring financial performance. Some experts propose encouraging patient participation in new payment mechanisms by enabling patients, along with providers, to benefit financially from savings achieved.33 Although financial incentives may make sense in some circumstances, their use is problematic for Medicare hospital episodes. One reason is that because Medicare enrollees supplemental coverage (Medigap or Medicaid) covers cost-sharing, it is difficult or impossible to reduce cost sharing as an incentive to participate. Medicares Acute Care Episode demonstration (see box on page 7) uses an alternative approach to offer a financial incentive, paying beneficiaries a share of the savings providers achieve. Early evidence from one of the participating sites, however, suggests that this financial incentive has had little effect on patient choices.34 Further, because financial rewards associated with hospital episodes have the perverse effect of providing patients with financial gain from seeking hospital care, information and education seem preferable as strategies to engage beneficiaries in the new payment arrangement.

21

center for American progress | Bundling payment for episodes of hospital care

Conclusion
The National Pilot Program on Payment Bundling is one of several initiatives in the Affordable Care Act aimed at improving health care quality and slowing growth in health care costs. Pursuing these initiatives aggressively and, ultimately successfully, is not only critical to sustaining coverage supported by the new law, it is also essential to sustaining Medicares commitments and assuring an affordable health care system for the future. Success will require taking full advantage of every possible tool to shift payment from a payment system that promotes volume of services, without regard to their benefits, to a system that rewards high quality care, efficiently delivered. Hospital-episode bundling, effectively designed, is one such tool. Bundling payment around a hospital stay has the potential to give providers the flexibility and incentive to work together to better coordinate care and reduce avoidable complications and unnecessary costs. Achieving that result on a national scale requires a pilot design that is simple and attractive to a broad range of providers, targets the most suitable diagnoses, provides payment incentives that both lower cost growth and improve quality, and assures patient protection and choice. The bundling design offered in this paper can thereby advance urgently needed, successful payment and delivery reform.

22

center for American progress | Bundling payment for episodes of hospital care

Endnotes
1 the pilot program is described in sec. 3023 of the Affordable care Act, public Law 111-148, 111congr, 2 sess. Government printing office, 2009. 2 there are additional adjustments to payments, such as an adjustment for differences in input costs among locations. For details, see medicare payment Advisory commission, hospital Acute inpatient services payment system, Payment Basics (2009), available at http://www.medpac.gov/documents/medpAc_payment_Basics_09_hospital.pdf. 3 robert A. Berenson and eugene c. rich, Us Approaches to physician payment: the Deconstruction of primary care, Journal of General Internal Medicine 25 (6) (2010): 613-18, available at http://www. springerlink.com/content/u512x31ml55j2134/fulltext.pdf. 4 Berenson and rich, Us Approaches to physician payment cheryl L. Damberg and others, exploring episode-Based Approaches for medicare performance measurement, Accountability and payment, (Washington: Us Department of health and human services, 2009), available at http://aspe.hhs.gov/health/reports/09/mcperform/ report.pdf; Gerben DeJong, Bundling Acute and post-acute payment: From a culture of compliance to a culture of innovation and Best practice, Physical Therapy 90 (5) (2010): 658-62, available at http:// ptjournal.apta.org/content/90/5/658.full.pdf+html; stuart Guterman and others, reforming provider payment: essential Building Block for health reform, the commonwealth Fund (2009), available at http://www.commonwealthfund.org/~/media/Files/publications/ Fund report/2009/mar/1248_Guterman_reforming_provider_payment_essential_building_block_FiNAL.pdf; Glenn hackbarth, robert reischauer, and Anne mutti, collective Accountability For medical caretoward Bundled medicare payments, The New England Journal of Medicine 359 (1) (2008): 3-5, available at http://www.nejm. org/doi/full/10.1056/NeJmp0803749; peter s. hussey and others, episode-Based performance measurement And payment: making it A reality, Health Affairs 28 (5) (2009): 1406-17, available at http:// content.healthaffairs.org/content/28/5/1406.full.pdf+html; robert e. mechanic and stuart h Altman, payment reform options: episode payment is a Good place to start, Health Affairs 28 (2) (2009): w262-71, available at http://content.healthaffairs.org/cgi/reprint/28/2/w262; medicare payment Advisory commission, report to the congress: reforming the Delivery system (2008), available at http://www. medpac.gov/documents/Jun08_entirereport.pdf; harold D. miller, From Volume to Value: Better Ways to pay For health care, Health Affairs 28 (5) (2009): 1418-28, available at http://content.healthaffairs. org/content/28/5/1418.full.pdf+html; hoangmai h. pham and others, episode-based payments: charting a course for health care payment reform, (Washington: National institute for health care reform, 2010), available at http://www.nihcr.org/episodeBasedpayments. html; W. pete Welch, Bundled medicare payment For Acute and postacute care, Health Affairs 17 (6) (1998): 69-81, available at http:// content.healthaffairs.org/content/17/6/69.full.pdf. 5 For additional examples, see American hospital Association committee on research, Bundled payment: AhA research synthesis report (2010), available at http://www.hret.org/bundled/resources/ Bundledpayment.pdf. 6 centers for medicare and medicaid, medicare health Bypass summary (1998), available at http://www.cms.gov/Demoprojectsevalrpts/downloads/medicare_heart_Bypass_summary.pdf. 7 Jerry cromwell and others, medicare participating heart Bypass Demonstration, executive summary, Final report (1998), available at http://www.cms.gov/Demoprojectsevalrpts/downloads/medicare_heart_Bypass_executive_summary.pdf. 8 Jerry cromwell, Debra A. Dayhoff, and Armen h. thoumaian, cost savings and physician responses to Global Bundled payments for medicare heart Bypass surgery, Health Care Financing Review 19 (1) (1997): 41-57, available at http://www.ncbi.nlm.nih.gov/ pubmed/10180001; chuan-Fen Liu, sujha subramanian, and Jerry cromwell, impact of Global Bundled payments on hospital costs of coronary Artery Bypass Grafting, Journal of Health Care Finance 27 (4) (2001): 39-54, available at http://www.ncbi.nlm.nih.gov/ pubmed/11434712. 9 centers for medicare and medicaid, medicare Acute care episode Demonstration for orthopedic and cardiovascular surgery, http:// www.cms.gov/Demoprojectsevalrpts/downloads/Ace_web_page. pdf; centers for medicare and medicaid, Acute care episode Demonstration, http://www.cms.gov/Demoprojectsevalrpts/ downloads/AceFactsheet.pdf; centers for medicare and medicaid, Frequently Asked Questions about the Acute care episode (Ace) Demonstration, available at https://www.cms.gov/Demoprojectsevalrpts/downloads/Acemoreinfo.pdf. 10 centers for medicare and medicaid, medicare Acute care episode Demonstration for orthopedic and cardiovascular surgery, available at http://www.cms.gov/Demoprojectsevalrpts/downloads/ Ace_web_page.pdf. 11 Francois de Brantes and others, Building a Bridge from Fragmentation to Accountabilitythe prometheus payment model, The New England Journal of Medicine 361 (11) (2009): 1033-36, available at http://www.nejm.org/doi/pdf/10.1056/NeJmp0906121; health care incentives improvement institute, available at http://www.hci3.org/. 12 health care incentives improvement institute, prometheus payment inc., prometheus Newsletter: issue 5, (January 2010), available at http://www.rwjf.org/files/research/prometheus2009issue5.pdf; health care incentives improvement institute, prometheus payment inc., prometheus Newsletter: issue 6, (June 2010), available at http://www.rwjf.org/files/research/65088.pdf. 13 health care incentives improvement institute, prometheus implementations, available at http://www.hci3.org/. 14 Alfred s. casale and others, provencare: A provider-Driven pay-Forperformance program for Acute episodic cardiac surgical care, Annals of Surgery 246 (4) (2007): 613-21, available at http://www.ncbi. nlm.nih.gov/pubmed/17893498; mechanic and Altman, payment reform options. 15 Douglas mccarthy, kimberly mueller, and Jennifer Wrenn, Geisinger health system: Achieving the potential of system integration through innovation, Leadership, measurement, and incentives, the commonwealth Fund (2009), available at http://www.commonwealthfund. org/content/publications/case-studies/2009/Jun/Geisinger-healthsystem-Achieving-the-potential-of-system-integration.aspx.

23

center for American progress | Bundling payment for episodes of hospital care

16 medicare payment Advisory commission, report to the congress: promoting Greater efficiency in medicare (2007), available at http:// www.medpac.gov/documents/Jun07_entirereport.pdf. see also, stephen F. Jencks, mark V. Williams, and eric A. coleman, rehospitalizations among patients in the medicare Fee-for-service program, The New England Journal of Medicine 360 (14) (2009): 1418-1428, available at http://www.nejm.org/doi/full/10.1056/NeJmsa0803563. 17 pham and others, episode-based payments. 18 Alexis Ahlstrom and others, piloting Bundled medicare payments for hospital and post-hospital care / A study of two conditions raises key policy Design considerations, (Washington: Avalere health LLc, 2010), available at http://www.avalerehealth.net/ research/docs/20100317_Bundling_paper.pdf; Damberg and others, exploring episode-Based Approaches for medicare performance measurement, Accountability and payment; melissa morley and others, post Acute care episodes (Washington: Us secretary of health and human services, 2009), available at http://aspe.hhs.gov/ health/reports/09/pacepifinal/report.pdf. 19 mechanic and Altman, payment reform options. 20 harold D miller, From concept to reality: implementing Fundamental reforms in health care payment systems to support Valuedriven health care, (Network for regional healthcare improvement payment reform summit, 2008), available at http://www.ncbi.nlm. nih.gov/pubmed/19738259. 21 the Brookings institution, Accelerating health care innovation to Achieve system-wide impact, transcript (2010), available at http:// www.brookings.edu/~/media/Files/events/2010/1018_health_innovation/20101018_health_innovation_panel_one.pdf; David cutler, how health care reform must Bend the cost curve, Health Affairs 29 (6) (2010): 1131-35, available at http://content.healthaffairs.org/ content/29/6/1131.full.pdf+html. 22 the law lists six factors for the secretary to consider in selecting the 10 conditions for the pilot. the factors are: whether the selected set includes a mix of chronic and acute conditions; whether the set includes a mix of surgical and medical conditions; whether the selected set includes a mix of chronic and acute conditions; whether there is evidence that providers and suppliers could improve quality of care for a condition while reducing spending; whether a condition has significant variation in readmissions and in spending for post-acute services; whether a condition has a high volume of cases and high post-acute spending; and which conditions are most amenable to bundling given practice patterns for medicare patients.

23 pham and others, episode-based payments. 24 carol Levine and others, Bridging troubled Waters: Family caregivers, transitions, And Long-term care, Health Affairs 29 (1) (2010): 116-24, available at http://content.healthaffairs.org/cgi/reprint/29/1/116. 25 pham and others, episode-based payments. 26 Damberg and others, exploring episode-Based Approaches for medicare performance measurement, Accountability and payment. 27 pham and others, episode-based payments. 28 medicare payment Advisory commission, report to the congress: reforming the Delivery system; harold D miller, From concept to reality: implementing Fundamental reforms in health care payment systems to support Value-driven health care (pittsburgh: Network for regional healthcare improvement payment reform summit, 2008), available at http://www.ncbi.nlm.nih.gov/ pubmed/19738259. 29 medicare payment Advisory commission, report to the congress: reforming the Delivery system; miller, From Volume to Value. 30 medicare payment Advisory commission, report to the congress: reforming the Delivery system. 31 mechanic and Altman, payment reform options. 32 Us Department of health and human services office of inspector General, Gainsharing Arrangements and cmps for hospital payments to physicians to reduce or Limit services to Beneficiaries (1999), available at http://oig.hhs.gov/fraud/docs/alertsandbulletins/gainsh.htm; Gail r. Wilensky and others, Gain sharing: A Good concept Getting A Bad Name? Health Affairs (2006): w58-67, available at http://content.healthaffairs.org/content/26/1/w58.full. pdf+html. 33 Aaron mckethan and mark mcclellan, moving From Volume-Driven medicine toward Accountable care, health Affairs Blog, August 20, 2009, available at http://healthaffairs.org/blog/2009/08/20/ moving-from-volume-driven-medicine-toward-accountable-care/ comment-page-1/. 34 c. hund and m. Joshi, early Learnings from the Bundled payment Acute care episode Demonstration project, (chicago: health research & education trust, 2010), available at http://www.hret.org/ reform/projects/resources/acute-care-episode.pdf.

24

center for American progress | Bundling payment for episodes of hospital care

About the authors

Judy Feder is a Senior Fellow at the Center for American Progress and a professor

of public policy at the Georgetown Public Policy Institute, where from 1999 to 2008 she served as dean of the institute.
Paul B. Ginsburg is president of the Center for Studying Health System Change,

which he founded in 1995.

Harriet L. Komisar is a research professor in the Health Policy Institute within the

Georgetown Public Policy Institute at Georgetown University.

Acknowledgements
In identifying issues and developing recommendations for this report, we benefited greatly from the input of a number of experts in the field, in particular Bob Berenson, David Cutler, and Harold Miller. We are also grateful to Nicole Cafarella and Elizabeth Wikler for superb research support. While we are indebted to these colleagues for their many contributions, the views presented here are those of the authors. The preparation of this brief was supported by a grant from the Peter G. Peterson Foundation. Paul Ginsburgs participation as a co-author was supported by the Robert Wood Johnson Foundation.

25

center for American progress | Bundling payment for episodes of hospital care

The Center for American Progress is a nonpartisan research and educational institute dedicated to promoting a strong, just and free America that ensures opportunity for all. We believe that Americans are bound together by a common commitment to these values and we aspire to ensure that our national policies reflect these values. We work to find progressive and pragmatic solutions to significant domestic and international problems and develop policy proposals that foster a government that is of the people, by the people, and for the people.

1333 H Street, NW, 10tH Floor, WaSHiNgtoN, DC 20005 tel: 202-682-1611 Fax: 202-682-1867 WWW.ameriCaNprogreSS.org