Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 60
Rewarding Provider Performance: Aligning Incentives in Medicare 3 Alternative Funding Sources CHAPTER SUMMARY Several possible funding sources can be tapped to generate the resources needed for a Medicare pay-for-performance initiative. This chapter examines the process of choosing initial funding sources, including the strengths and weaknesses of each source as a strategy for moving toward a sustainable payment system that rewards care characterized by high clinical quality, patient-centeredness, and efficiency. For each option, the committee considered several important criteria: the adequacy of the source, its stability or reliability, the extent to which different types of providers would consider it to be fair, and the impact of the mechanism on the overall quality of care and health care spending. While this chapter focuses explicitly on sources of funding, the analysis of these options is unavoidably intertwined with distribution and implementation issues (such as whether reward pools are to be divided by provider setting or aggregated into one large pool). These issues are noted where appropriate; however, more detailed discussion of distribution and implementation is presented in Chapters 4 and 5. The establishment of a funding pool for any pay-for-performance program has major policy implications. Because funds must be available before any reward payments can be made, creation of a funding pool signals the start of the pay-for-performance program and will send a strong message that the overall quality of care, in addition to the events of care, is about to acquire importance in the payment system. Resistance to any pay-for-performance program is likely to start when the funding sources (the Medicare Trust Funds, purchasers, providers) become apparent. Therefore, decisions about the source(s) require careful consideration of several major design questions. For example, will there be one large reward pool or a separate pool for each provider setting? What pool size will be needed to
OCR for page 61
Rewarding Provider Performance: Aligning Incentives in Medicare support rewards of sufficient magnitude to motivate various types of providers? If there is one large pool, what is to be done with those who, because of inadequate performance measures, cannot initially qualify for rewards? While many of these matters are discussed in Chapters 4 and 5, which focus on the distribution of funds and implementation issues, respectively, these questions must be considered when basic decisions on the initial funding source(s) for the rewards are made since all subsequent program decisions will hinge on those choices. BASIC FUNDING MODELS The resources needed to support a pay-for-performance program can be obtained from existing funds, from generated savings, or from direct investment (new money). Existing funds are monies that are already part of the payment system. This model reduces payments to all or selected types of providers for redistribution to those exhibiting higher quality in examined areas. The reward pool that is divided among those providers who reach specified quality goals may be created by reducing planned fee schedule increases (referred to as “shaving the update”), by withholding a portion of the base payment, or by enacting an explicit set of Medicare program cuts. The generated-savings model creates a reward pool from the money saved as a result of the adoption of cost-reducing reforms and efficiencies associated with the effort to improve quality. The direct-investment model adds new money from either Medicare’s Hospital Insurance trust fund or general revenues and distributes it as bonuses over and above a scheduled payment. Variations on each of these funding sources are discussed in more detail later in this chapter. Although these models represent three distinct approaches to funding, mixed approaches are also possible. The decision as to the source of funding is important for several reasons. Most important, there will be concern over whether the program is budget neutral—that is, whether it will add to government spending. In an era of high and escalating budget deficits, lawmakers are likely to object to spending more on a program that is already very costly, with expenditures growing rapidly. Provider groups, on the other hand, will want new funds to be used, arguing that payment rates are already too low and that redistributing a portion of these inadequate amounts will leave some with insufficient resources to do their jobs well and respond to new demands. In addition, some types of providers who are adversely affected by the funding mechanism selected will initially not have the opportunity to receive performance rewards. For example, performance measures are insufficiently developed for some specialties and provider types to allow for participation in a pay-for-performance program. This disconnect will cause understandable dissension. Overall, defining the mechanism for the creation of an initial
OCR for page 62
Rewarding Provider Performance: Aligning Incentives in Medicare funding pool for a pay-for-performance program is likely to have very real and contentious policy implications. Models in the Private Sector The pay-for-performance programs that have emerged in the private sector are frequently characterized as new-money models, whereby providers are paid a bonus on top of the regular fee schedule. However, some of these programs anticipate that the use of performance measures and the bonus structure will generate savings through improved efficiency, as well as long-term savings due to increased use of preventive services. For example, the Bridges to Excellence program invested money up front for bonuses using actuarial models (de Brantes et al., 2003). Box 3-1 describes how the Bridges to Excellence program predetermined an adequate reward pool encompassing both new initial funds and generated savings. The program expected to devote 50 percent of anticipated savings to the reward pool, with the other 50 percent being considered a return on investment to the purchasers. Other private-sector models include the Integrated Healthcare Association program in California, which rewards physician groups on the basis of clinical performance, patient experience, and use of information technology. Anthem Blue Cross and Blue Shield of New Hampshire pays bonuses to physician groups based on their provision of preventive services, thus making this a prospective generated-savings model. Some providers are skeptical of these strategies, asserting that the bonuses are financed by redirecting full payment updates rather than by providing new money (Bailit Health Purchasing LLC, 2002). Other bonus programs, including Blue Cross Blue Shield of Massachusetts and Harvard Pilgrim Health Care, generate revenues from a percentage of an annual withhold (Rosenthal et al., 2004). According to stakeholder testimony and a longitudinal study performed by ViPS and Med-Vantage from 2003 through 2005, provider groups favor strategies that generate new and increased income to providers, while purchasers favor budget-neutral approaches (ACHP, 2004, 2005; AAFP, 2005; ACOG, 2005; ACP, 2005; AMA, 2005; Baker and Carter, 2005). Models in the Public Sector As discussed in Chapter 2, the Centers for Medicare and Medicaid Services (CMS) has undertaken several pay-for-performance initiatives for multiple provider settings. Because the Office of Management and Budget, in general, insists that demonstrations be budget neutral overall, these initiatives are funded largely on the basis of anticipated savings attributable to
OCR for page 63
Rewarding Provider Performance: Aligning Incentives in Medicare BOX 3-1 Bridges to Excellence: Combination of an Adequate Reward Pool and Generated Savings The Bridges to Excellence program was developed by a coalition of providers, health plans, and employers that worked with General Electric to apply that organization’s quality improvement methodology, six sigma, to the health care field (de Brantes et al., 2003). The program’s mission included rewarding providers for clinical quality performance. In the initial stages of program development, the coalition identified attributes that would define the needs of providers, consumers, and purchasers. The coalition recognized in particular the need to identify what level of funding would be sufficient to motivate providers to change their behaviors and accept the program (Personal communication, F. de Brantes and R. Galvin, General Electric, January 30, 2006). Within a generated-savings model, actuarial studies were done to quantify generated savings. Providers participated in focus groups held to determine the levels and types of incentives that would be necessary to motivate change. Other studies on incentive programs were also considered (Bailit Health Purchasing LLC, 2002). Thus the Bridges to Excellence program is an example of how a funding pool for incentives based on the needs of providers was predetermined. The program may also be viewed as having elements of the direct-investment and generated-savings models in that theoretically, new money was put into the program to work on quality improvement in specific areas, and the program expected to reinvest savings into the reward pool. SOURCE: Personal communication, F. de Brantes and R. Galvin, General Electric, January 30, 2006. improved efficiency. However, most of these initiatives received some new funds to initiate their implementation. The Experience of the United Kingdom As described in Chapter 2, the National Health System in the United Kingdom initiated a pay-for-performance program for general health in April 2004. The program links a major portion of payment to performance on clinical indicators, organizational indicators, and patient experience. The British government invested more than $1.8 billion—a more than 20 percent increase over the previous health budget—in new money for bonuses for general practice (Roland, 2004). Previous incentive programs in the United King-
OCR for page 64
Rewarding Provider Performance: Aligning Incentives in Medicare dom had demonstrated positive responses to financial incentives. A “fund holding experiment” from 1991 to 1998 gave practitioners fixed budgets to provide secondary care and pharmaceuticals to their patients; this program ultimately resulted in fewer inpatient procedures and reduced patient waiting times (Smith and York, 2004). A program in East Kent from 1998 to 2000 established disease management targets and required that practitioners repay funds if those targets were not met (Smith and York, 2004). These two initiatives required new money initially; however, the first created incentives for efficiency, which generated savings, while the second relied on a reverse withhold to encourage quality improvement. GUIDING PRINCIPLES FOR SELECTION OF A FUNDING SOURCE Evaluation Criteria The committee considered four criteria when evaluating the merits of alternative funding sources for pay for performance in Medicare: adequacy, stability, fairness, and impact (see Table 3-1). Adequacy refers to the ability to generate a pool of sufficient size to support (on an ongoing basis) bonuses that are large enough to create meaningful incentives for improved performance. Stability refers to the predictability of the source. Will it depend on the annual appropriation decisions of Congress, and therefore be subject to political fluctuations and broader budgetary concerns? Will the size of the pool vary with fluctuations in the strength of the economy? Fairness involves the balance between those who are asked to contribute to the reward pool and those who have an opportunity to receive bonuses. This concern relates directly to providers or settings whose payments might be reduced to fund a reward pool but for whom good performance measures do not currently exist, such as specialists, rehabilitation facilities, and long-term care hospitals. Finally, impact is related to how each option might affect the Medicare program and its beneficiaries positively or negatively. To what extent might the funding mechanism in TABLE 3-1 Criteria for Evaluating Alternative Funding Sources Adequacy Stability Fairness Impact Size Length of time to establish Ability to influence behavior Predictability of source Sustainability Complexity of implementation Winners and losers Ability to participate Effect on program savings Effect on quality of care
OCR for page 65
Rewarding Provider Performance: Aligning Incentives in Medicare and of itself encourage efficiency or undermine the quality of care provided to beneficiaries? Other Considerations Other factors considered by the committee include the likelihood of stakeholder acceptance of each alternative, its administrative complexity, and its budgetary impact. The committee assigned significant weight to this last dimension, favoring funding approaches that would be budget conscious or, preferably, budget neutral. Budget-neutral approaches are those that do not initially lead to an increase in overall government spending (exclusive of administrative costs). The committee recognizes that there will be added expenses at the provider level associated with data collection and reporting or the implementation of process redesign. The committee agreed that if a funding alternative could not be budget neutral initially, it should at least be budget conscious, ensuring that budget concerns are explicitly recognized and addressed. Scope When one is considering how best to fund a pay-for-performance program, overarching decisions must be made concerning the scope of the reward pool. Separate pools could be created for each provider setting, or one aggregate pool could fund rewards for all provider types. The pool could be national or regional in scope. The program could also start with regional pools that were specific to provider types and move over time to a single national pool. However, the creation of regional pools could be too complex from an administrative standpoint and lead to variations in treatment across regions and types of providers that some would consider inequitable. Moreover, multiple pools could undermine efforts to encourage care coordination and lead some categories of providers to attempt to avoid participation altogether. SHORT-TERM MODELS In evaluating alternative funding sources for pay for performance, the committee was cautious, realizing that there is no strong evidence base to guide its recommendations. Ideally, CMS would mount a series of staged demonstrations in selected regions of the country, encompassing a well-structured evaluative component, which would allow for systematic evaluation of the effectiveness of different approaches. However, sensing the urgent need for reform, the number of years such demonstrations would take, and the opposition the demonstrations might generate in affected regions,
OCR for page 66
Rewarding Provider Performance: Aligning Incentives in Medicare the committee concluded that it would be best to move forward with both short- and long-term strategies. For the short term, the committee considered the three basic models noted above: Use of existing funds—pool formed from money already in the health care system. Generated savings—pool based on savings (relative to projected spending). Direct investment—introduction of new money to work on specific conditions. Use of Existing Funds Models based on the use of existing funds create an initial reward pool from the current funding base. Hence, funding of a pay-for-performance program is budget neutral from the start. This type of pool could be created by: Using a portion (or all) of scheduled payment updates (known as “shaving the update”). Reducing base payments by a certain percentage. Having Congress establish a reward pool of a predetermined size, spending for which would be offset by enacting a package of specific cuts in the Medicare program that might or might not affect payment rates. Prominent physician groups in particular have expressed concern about approaches that are budget neutral or provide no new money to fund pay for performance because they are aware that under the sustainable growth rate (SGR) mechanism, physician fees are already scheduled to decline significantly over the next few years, which makes reductions in existing payment rates especially concerning for this provider group (ACHP, 2004, 2005; AAFP, 2005; ACOG, 2005; ACP, 2005; AMA, 2005). An initial reward pool based on shaving the update might be generated from a tax or assessment that reduced the size of the annual update of Medicare payment to providers by one or two percentage points. Alternatively, base payment rates could be reduced by a percentage point or two. The total pool would be paid out annually. Adequacy If the updates were shaved or the base payment rate reduced only in the first year, the size of the reward pool would amount to only 1 or 2 percent-
OCR for page 67
Rewarding Provider Performance: Aligning Incentives in Medicare age points of payments year after year. Substantial uncertainty exists as to whether a pool created by a one-time shaving of the update or a single reduction in the base payment of such a small magnitude would be sufficient to motivate the desired behavioral changes among all types of providers. While institutional providers, such as hospitals, might be motivated by relatively small bonuses, it is doubtful that rewards of one or two percentage points of base payments would be sufficient to motivate physicians to adopt the infrastructure supports, such as data registries, needed to track and monitor patients with chronic conditions so as to ensure that evidence-based care is being delivered. Initial experience in the private sector suggests that reward thresholds are within the range of 5–15 percent of earnings for physicians and 1–2 percent of gross revenues for hospitals (Personal communication, F. de Brantes and R. Galvin, General Electric, January 30, 2006) (Baker and Carter, 2005; Nussbaum, 2005). Box 3-1, presented earlier, describes how Bridges to Excellence determined an adequate funding pool. To provide a rough estimate of what this would mean in the Medicare setting, the committee consulted with MedPAC to perform data runs on the total payments that are associated with the three conditions for which a majority of Medicare payments are made—chronic heart failure, coronary artery disease, and diabetes (see Appendix D). To determine the average reward per unique physician identification number (UPIN), the committee made the following assumptions: (1) one-quarter of physicians would not be eligible based on the lack of available adequate measures, (2) only half of physicians would achieve the level of performance required to receive rewards, and (3) 2 percent would be taken from base payments. This results in a denominator of 75,000 physicians receiving rewards. To calculate the numerator, the committee took 2 percent of the total physician fee schedule payments made by Medicare for services associated with treatment of the above-named conditions, or about $6.61 million. Dividing numerator by denominator results in rewards approximating $88 per physician per year (see Table 3-2). While there are other important variables to consider, and this example only uses three conditions, the committee used this calculation to demonstrate that either adding new money or putting a larger proportion of the base payment at risk may be necessary to motivate providers adequately. The updates or base payments could be reduced by an additional 1 or 2 percentage points in each of the first few years. If this were repeated for each of the first 5 years, the pool available for bonuses would grow gradually, reaching between 5 and 10 percent of baseline base payments by year 5. This pool would support bonuses of 10 to 20 percent of base payments for the top-performing half of physicians.
OCR for page 68
Rewarding Provider Performance: Aligning Incentives in Medicare TABLE 3-2 Example of Rewards per Eligible UPIN Based on Percentage of Total Payments Percentage of Payment Rewards in Dollars per Eligible UPIN (based on total fee schedule payments) 1 44.08 2 88.17 10 440.85 15 661.20 NOTE: Total physician fee schedule payments = $330,627,588.35; UPINs in above examples = 75,000. Numbers are based on services associated with treatment of chronic heart failure, coronary artery disease, and diabetes. Stability This option is a budget-neutral strategy that requires no new money and is fiscally prudent in an environment of scarce resources. It could be implemented immediately. The variant of this option that involves shaving the updates is more complex and uncertain than the variant that relies on reducing base payments because not all provider types receive an automatic update, and Congress often reduces the updates called for under law for budgetary or other reasons. In fact, the SGR currently calls for the physicians’ fee schedule to undergo negative updates of more than 4 percent annually from 2007 to 2011 (MedPAC, 2006). Approaches based on the use of existing funds are less complex than those based on generated savings because they do not depend on uncertain actuarial estimates of savings generated by policy changes. Fairness A major disadvantage of the existing-funds model initially is the likelihood that some providers who would have their payments reduced to generate the reward pool would be unable to compete for bonuses. For example, an across-the-board reduction in the update or the base payments for physicians would affect all physicians, whereas certain specialists or others for whom accurate and reliable performance measures do not yet exist would be precluded from participating initially in a pay-for-performance program. This problem would be compounded among physicians who might face unrecoverable cuts in payment while concurrently facing a negative update. Many would regard this to be unfair, even if it would be a
OCR for page 69
Rewarding Provider Performance: Aligning Incentives in Medicare transitional situation. On the other hand, this inequity could create incentives for providers lacking measures to develop them more quickly to become eligible for rewards. If the reward pool were generated from an explicit set of Medicare cuts that did not include reductions in payment rates or updates, however, the program might be more palatable to providers. Impact The existing-funds model could have several undesirable impacts. First, cutting payment rates could leave some providers who already had negative Medicare margins with insufficient resources to maintain their current quality of care. Second, some might feel that Medicare rates had become insufficient and thus be less willing to see Medicare beneficiaries, creating access problems. Finally, in an effort to maintain revenues in the face of lower payment rates, providers might increase their rate of services, leading to more low-value or unnecessary care. Generated Savings Another budget-neutral approach is to build the reward pool solely from savings generated by efficiencies providers adopt in attempting to improve the overall quality of the care they deliver. Under this model, if spending grows at a slower pace than was projected, the difference between projected and actual spending is made available for performance-based awards. This approach depends on reliable projections of baseline Medicare spending; in the past, such projections have been difficult to make. CMS’s Physician Group Practice Demonstration is an example of the generated-savings model. It provides a framework for how such a model can be used for subsets of providers (see Box 3-2). Rather than comparing actual spending with a projection, the demonstration compares actual spending on the beneficiaries served by participating providers with that on beneficiaries who receive care from similar providers not participating in the demonstration. However, this method of overcoming the difficulties involved in developing accurate projections would not be available if all providers were included in the program. Adequacy Whether the generated-savings model can produce adequate resources for a reward pool is largely unknown. A conundrum could develop: rewards of a certain size might be required to motivate the implementation of meaningful efficiency initiatives, but until significant efficiencies had been realized, the reward pool might be inadequate to support significant bonuses.
OCR for page 70
Rewarding Provider Performance: Aligning Incentives in Medicare BOX 3-2 The Physician Group Practice Demonstration: Example of a Generated-Savings Reward Pool The Physician Group Practice Demonstration is Medicare’s first pay-for-performance program for physicians. The program was mandated by the Medicare, Medicaid, and State Children’s Health Insurance Program Benefits Improvement and Protection Act of 2000 and became operational in April 2005. Under this project, ten large group practices (consisting of at least 200 physicians each) are encouraged to improve both clinical quality and efficiency in health care services for Medicare fee-for-service beneficiaries. The 3-year program is aimed at improving coordination of Part A and Part B services, investing in structural and process efficiencies, and rewarding improved outcomes (CMS, 2005). Physicians will continue to be paid on a fee-for-service basis; they may earn additional rewards based on their results, but only if they achieve savings as compared with a projected annual target. By law, the demonstration is budget neutral. Accordingly, the performance rewards are derived from the savings. SOURCE: CMS, 2005. Stability This model could prove fairly unstable in that the difference between projected and actual spending could vary significantly from year to year. It is difficult to predict year-to-year spending increases with any degree of accuracy because so many factors affect spending, often in ways that are not fully understood. Fairness Providers would probably consider this approach more equitable than the existing-funds model. They would reap the benefits of their actions directly, in that the savings generated by their efforts to improve would be distributed back to them. However, the generation of savings might depend critically on the cooperation of multiple providers across settings within a community, which might be difficult to achieve. Initially at least, this could be problematic if the savings generated by all provider settings were consolidated into one reward pool, and adequate performance measures that could be used to allocate rewards were lacking for some types of providers. Accrued savings could be attributed and awarded to individual providers, but this would be an extremely complex undertaking fraught with difficul-
OCR for page 71
Rewarding Provider Performance: Aligning Incentives in Medicare ties, including those associated with the calculation of savings and attribution of care to a single provider. Under this model, multiple smaller reward pools might be considered fairer for properly attributing savings. Variations across regions in levels of spending and the pace of spending growth also relate to perceptions of fairness. Providers in regions with relatively low spending levels might argue that they had already achieved a desirable level of efficiency and had little scope for reducing their rate of spending growth. If the reward pool were generated largely from high-spending regions but the bonuses were concentrated in more efficient regions, many of which deliver relatively high-quality care (Fisher et al., 2003), provider dissatisfaction could emerge in the former areas. Impact The generated-savings model places great emphasis on efficiencies and program savings. This emphasis could have a negative impact on clinical quality. It is therefore important that the bonus or reward system under such a model emphasize clinical quality and patient-centeredness. This could be accomplished through the reward distribution design by establishing thresholds for these two domains that would have to be exceeded before a provider received a reward for either the level of or improvement in efficiency (see Chapter 4). To the extent that efficiencies can best be achieved by providers working together, this approach could indirectly encourage providers to collaborate to generate these savings and individual physicians to form larger group practices through formal or informal affiliations. This approach would also allow communities to self-organize at the market level to work toward a common objective. While this approach may appear plausible in any care setting, however, in reality such a model has been implemented only at the hospital level (see Box 3-3). Sparse empirical evidence exists to support the transfer of this approach to a larger-scale effort. Direct Investment The options discussed thus far are initially budget neutral, involving no new money up front. The direct-investment model differs in this regard. The Institute of Medicine’s (IOM’s) Crossing the Quality Chasm report recommended that a discrete number of common, high-burden chronic conditions serve as leverage points for achieving rapid and widespread improvements in the six quality aims of health care (IOM, 2001). A subsequent IOM report, Priority Areas for National Action: Transforming Health Care Quality, recommended 20 priority conditions or areas in which to proceed (IOM, 2003). Continuing with this path, the committee considered whether initial pay-for-performance strategies should be directed at these priority
OCR for page 72
Rewarding Provider Performance: Aligning Incentives in Medicare BOX 3-3 Excellus Blue Cross Blue Shield: Combination of Generated Savings and Withholds for Reward Pools In an example of funding based partially on cost savings, Excellus Blue Cross Blue Shield of Rochester, New York, partnered with 900 physicians of the Rochester Individual Practice Association to reward providers who met standards of clinical quality, cost savings, and patient experience of care. Excellus looked at the years 2001 and 2002 as a base. It then invested $1.1 million into the program each year, with a positive return on investment of 2–3 dollars for every dollar invested. The program also created a bonus pool of up to $15 million based on a combination of withholds and shared savings. Thus this program shows elements of all three short-term models discussed in this chapter, although in this example, it is used specifically to illustrate how cost savings can be employed in the funding of a reward pool. With this combination of withholds and shared savings, 8 percent of physicians’ reimbursement was at risk, but they could receive a return of 50–150 percent depending on their performance. SOURCE: AIS, 2004. areas. The assumption behind doing so would be that investments in upstream preventive services, such as cancer screenings, and high quality for individuals in the initial stages of a chronic condition could generate significant downstream savings that could fund future pay-for-performance initiatives for other health conditions. The goal would be to target high-value interventions and to reward providers for delivering evidence-based services. Adequacy An investment of any size in a pay-for-performance reward pool would constitute adding new money to the health finance system. The size of the investment, however, could vary and would have to be sufficient to stimulate the desired level of activity. Providers could be given funds up front to work on specific conditions, such as the priority areas noted above, with the payments not being linked to performance, so as to initiate the program and encourage providers to focus on these areas. Subsequent payments could be tied to the achievement of specific treatment goals for that limited number of conditions. (See Box 3-4 for an example of direct investment for specific clinical conditions.)
OCR for page 73
Rewarding Provider Performance: Aligning Incentives in Medicare BOX 3-4 HealthPartners: Use of Comprehensive Quality Standards in Direct-Investment Models HealthPartners is a consortium of nonprofit, consumer-run health care organizations in Minnesota, including a health plan that covers nearly one-fourth of the residents of the Minneapolis–St. Paul metropolitan area. In 1997, HealthPartners began rewarding primary care physicians for achieving specific clinical quality performance goals through its Outcomes Recognition Program. In contrast to some programs, HealthPartners did not attempt to create a budget-neutral approach—any provider achieving goals on health outcomes received awards. Most of the outcome goals depended on an “all-or-none” approach. That is, for a physician to receive a reward for one condition, the individual patient had to meet standards for multiple outcomes. For example, “Optimal Diabetes Care” required that at least 30 percent of a group’s HealthPartners adult (aged 18–75) patients with diabetes have all cardiovascular risk factors optimally managed. To meet this comprehensive standard, patients had to have blood sugar (HbA1c) levels less than or equal to 8 percent, LDL cholesterol less than 130 mg/dl, and blood pressure under 130/85 mmHg. Moreover, patients had to be nonsmokers, and those older than age 40 had to be taking aspirin daily. In 2004, 3 of the 26 eligible primary group practices were able to achieve this standard for diabetes care. SOURCE: Apland and Amundson, 2005. Stability In and of itself, the pool under a direct-investment model is initially stable in that it represents a set amount of funds calibrated to meet the needs of the program. The pool size could expand if savings were reinvested into the program, much as in the generated-savings model. Caution is in order, however, because the current evidence base provides few examples of large positive returns on investments of this sort. There has been some experience with this general model in the private sector that offers insight into the process of implementation, operation, and lessons learned. However, a significant challenge remains because this strategy requires new money from the outset, which would be difficult to obtain in the current fiscal climate. Realistically, if this model were to be financially sustainable in the long run, it would require that both clinical effectiveness (i.e., meeting clinical quality targets) and efficiency be achieved.
OCR for page 74
Rewarding Provider Performance: Aligning Incentives in Medicare Fairness This option is characterized by a high degree of fairness because while not all providers would be eligible for initial rewards, the funds used to support the program would not come from any one provider group. Impact The impact of this model is unclear. If successful, it would directly affect the clinical quality of care delivered to beneficiaries for the targeted priority conditions because specific care processes for these conditions would attract more attention. Use of this model is supported by the fact that Medicare payments are highly concentrated on a few conditions. For example, 70 percent of Medicare inpatient spending is made for beneficiaries who have, either singly or in combination, three chronic conditions— chronic heart failure, coronary artery disease, and diabetes (MedPAC, 2005) (see the discussion of rewarding by condition in Chapter 4). By focusing on a few high-cost conditions, providers could have a significant impact on both savings and clinical quality. However, one serious concern with this approach is the possibility that certain providers might choose to avoid patients with these conditions so they would not be compelled to participate in a pay-for-performance program. Of course, if the rewards were substantial, providers might be attracted to the program, thereby improving access for those with these chronic conditions. A direct-investment model in which savings went into a common reward pool that was allocated only in cases in which the patient received all necessary care from all providers could promote shared accountability for and coordination of patient care since each provider’s bonus would be tied to the performance of all. However, there are few established models for distributing rewards in this manner across multiple providers. Reward pools might be diluted if the money were allocated across a large number of providers. (Distribution issues are discussed in greater detail in Chapter 4.) Evaluation of Short-Term Alternatives As described above, the committee examined a number of alternatives for funding an initial pay-for-performance program in Medicare, weighing the strengths and weaknesses of each. A good deal of uncertainty surrounds all of these alternatives. Moreover, the committee recognizes that these models have often been used in combination, and that new strategies may emerge as experience enriches the knowledge base. The committee evalu-
OCR for page 75
Rewarding Provider Performance: Aligning Incentives in Medicare ated each with respect to the four overarching criteria it deemed important to the choice of a funding source: adequacy, stability, fairness, and impact. A summary of the alternatives and their characteristics according to these four criteria appears in Table 3-3. In light of this comparison, the committee makes the following recommendation: Recommendation 2: Congress should derive initial funding (over the next 3–5 years) for a pay-for-performance program in Medicare largely from existing funds. Congress should create provider-specific pools from a reduction in the base Medicare payments for each class of providers (hospitals, skilled nursing facilities, Medicare Advantage plans, dialysis facilities, home health agencies, and physicians). Congress should ensure that these pools are large enough to create adequate motivation for improved performance on selected measures. Because of unique challenges of physician payment relating to the sustainable growth rate (SGR), investment dollars may be necessary to create adequate resources to effect change. Initial funding should be budget conscious in taking into account the resources needed for both funding the pools and implementing the program. The committee recognizes that generation of a reward pool with existing funds is useful for an initial pay-for-performance program, but should be only be a short-term solution while alternative funding sources are explored. The feasibility of other funding approaches and the savings realized from efficiency improvement should be evaluated over the next 3–5 years. The committee concluded that the generated-savings model on its own is currently limited by difficulties with prediction and is not sustainable, but has great potential. As a result, the committee proposes that efforts to test and demonstrate ways of making this source of funds work be mounted and aggressively pursued. Based upon the findings from those efforts, funds generated by increased efficiency should be used as a supplemental source in the creation of subsequent funding pools. The committee also concluded that the direct-investment model is inadequate on its own because of difficulties in finding sufficient new money for initial investment, as well as related concerns regarding the attribution of care. However, the committee did recognize that some new funds may be needed in addition to existing sources to initiate the program up front.
OCR for page 76
Rewarding Provider Performance: Aligning Incentives in Medicare TABLE 3-3 Comparison of Initial Funding Options Criterion Existing Funds Generated Savings Direct Investment Adequacy Mixed—Updates and base payment reductions are too small to motivate some providers and will need time to accumulate to viable size. Will be adequate if pool is created to a level that is predetermined. Low—Amount required to motivate change is unknown, and amount of savings is largely unpredictable. Also, no start-up funds present to initiate changes. Low—Places new money into the system, but finding a source of new money in the current fiscal climate that will be adequate to motivate all providers is highly unlikely. Stability Moderate—Low complexity and possibility of immediate implementation, but high variability in size of update. Confounded in light of negative updates. Low—Difficult to predict spending increases, requires complex formulas to calculate savings, and can fluctuate greatly from year to year. Low—Pool initially stable since has known source and size, but return on investment is unknown, and sustainability is uncertain. Fairness Moderate—Some providers are precluded from program, and not all get updates (or equivalent updates). Moderate—Some providers are precluded, but others may directly reap the benefit of their efficiencies. Complicated by sharing of pool, attribution, and comparison groups. High—Fair in that source does not affect those who cannot participate. Impact Reduces already low payment rates. May create access problems. Greater incentive for savings and efficiency. May encourage care coordination. Specific conditions are targeted for attention. Potential exists for adverse selection. May promote care coordination.
OCR for page 77
Rewarding Provider Performance: Aligning Incentives in Medicare In considering geographic levels for funding pools, the committee concluded that initial reward pools should be established by provider setting at the national level because of concerns about complexity and attribution at the regional level, as well as the fairness of creating an aggregate pool for all provider settings. While the committee acknowledges that pools created at the provider setting level would do little to foster increased care coordination, it also recognizes that the current payment system does even less in this regard. Furthermore, the creation of a single aggregate pool would likely not be acceptable to major stakeholders because of perceptions of unfairness, the inability to attribute care accurately, and the need for the development of more sophisticated measures that would specifically address care coordination. The existing-funds model is consistent with the goal of linking larger proportions of payment to performance. Again, while recommended as an initial approach to funding, the committee recognizes that this model represents a short-term solution and that other strategies will have to be tested and considered as new challenges arise in Medicare pay-for-performance programs. LONG-TERM FUNDING The prior discussion of alternative models for funding pay-for-performance programs constitutes a plan for incremental reforms. The committee has presented these alternatives and made its final recommendation while fully acknowledging the strengths and weaknesses of each model. The committee also recognizes that each of these alternatives may be a transitional step while long-term funding options are explored. In essence, these short-term strategies should be viewed as initial building blocks that could help move Medicare toward future funding alternatives and payment policies that would represent more robust approaches for aligning payment incentives with higher-quality care. In considering the scope of the reward pools, the committee has recommended separate pools for each major provider setting. In the long term, however, the committee believes that as measurement becomes more sophisticated and it becomes possible to evaluate performance for episodes of care and for all care given to those with significant chronic conditions, it will make sense to consolidate the multiple reward pools into a single national pool. However, the committee recognizes the need to reconcile this approach with the fact that provider payments in Medicare (such as from Part A and Part B) are currently unlinked.
OCR for page 78
Rewarding Provider Performance: Aligning Incentives in Medicare Recommendation 3: Congress should give the Secretary of DHHS the authority to aggregate the pools for different care settings into one consolidated pool from which all providers would be rewarded when the development of new performance measures allows for shared accountability and more coordinated care across provider settings. CONCLUSION Pay for performance represents a major shift from the status quo, one that is likely to be met with marked resistance. The committee has recommended that the initial pool for pay for performance come from a reduction in base Medicare payments, with the current update mechanism being retained. CMS should also experiment with the generated-savings model as soon as possible. In certain areas, investments of new funds may be unavoidable. The committee has also recommended the initial creation of multiple reward pools by care setting that, in the long term, would be aggregated into one large pool from which all providers could earn rewards. The issues examined in this chapter related to the funding of a pay-for-performance program in Medicare set the stage for discussion of many of the policy implications of subsequent decisions about the implementation of such a program. While some of these considerations have been touched upon in this chapter, separate discussion of the distribution of these funds is presented next in Chapter 4, and overarching concerns regarding the implementation of the program are addressed in Chapter 5. REFERENCES AAFP (American Academy of Family Physicians). 2005 (July 1). Position Statement from the American Academy of Family Physicians: Per Family Physicians: Medicare Value Purchasing Act Misses the Mark. Unpublished. ACHP (Alliance of Community Health Plans). 2004. Position Statement of the Alliance of Community Health Plans: Four Principles for “Payment-for-Performance” in Medicare. [Online]. Available: http://www.achp.org/news/article.asp?nyear=2004&article_id=97 [accessed June 12, 2006]. ACHP. 2005 (June 28). Position Statement of the Alliance of Community Health Plans. Unpublished. ACOG (American College of Obstetricians and Gynecologists). 2005. Position Statement of the American College of Obstetricians and Gynecologists: Pay for Performance (P4P). Unpublished. ACP (American College of Physicians). 2005 (March 15). Position Statement of the American College of Physicians for the Record of the Hearing on Measuring Quality and Efficiency of Care in Medicare. Unpublished.
OCR for page 79
Rewarding Provider Performance: Aligning Incentives in Medicare AIS (Atlantic Information Services, Inc.). 2004. Case Studies in Health Plan Pay-for-Performance Programs. Washington, DC: AIS. AMA (American Medical Association). 2005. Position Statement from the American Medical Association: Pay-for-Performance Principles and Guidelines. [Online]. Available: http://www.ama-assn.org/meetings/public/annual05/bot5a05fin.doc [accessed June 12, 2006]. Apland BA, Amundson GM. 2005. Financial incentives: An indispensable element for quality improvement. Patient Safety and Quality Healthcare 2(5):6–8. Bailit Health Purchasing LLC. 2002. Provider Incentive Models for Improving Quality of Care. Washington, DC: Academy for Health Services Research and Health Policy. Baker G, Carter B. 2005. Provider Pay-for-Performance Incentive Programs: 2004 National Study Results. San Francisco, CA: Med-Vantage, Inc. CMS (Centers for Medicare and Medicaid Services). 2005. CMS Fact Sheet. [Online]. Available: http://www.premierinc.com/advocacy/issues/medicare/05/physician-group-practice-fact-sheet-0105.pdf#search=’Physician%20Group%20Practice%20Demonstration’ [accessed March 2, 2006]. de Brantes F, Galvin RS, Lee TH. 2003. Bridges to excellence: Building a business case for quality care. Journal of Clinical Outcomes Management 10(8):439–446. Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. 2003. The implications of regional variations in Medicare spending. Part 1: The content, quality, and accessibility of care. Annals of Internal Medicine 138(4):273–287. IOM (Institute of Medicine). 2001. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press. IOM. 2003. Priority Areas for National Action: Transforming Health Care Quality. Adams K, Corrigan JM, eds. Washington, DC: The National Academies Press. MedPAC (Medicare Payment Advisory Commission). 2005. MedPAC Data Runs. Washington, DC: MedPAC. MedPAC. 2006. Report to the Congress: Medicare Payment Policy. Washington, DC: MedPAC. Nussbaum S. 2005. Testimony to Medicare Payment Advisory Commission. [Online]. Available: http://www.medpac.gov/public_meetings/transcripts/0905_allcombined_transc.pdf [accessed August 2, 2006]. Roland M. 2004. Linking physicians’ pay to the quality of care: A major experiment in the United Kingdom. New England Journal of Medicine 351(14):1448–1454. Rosenthal MB, Fernandopulle R, Song HR, Landon B. 2004. Paying for quality: Providers’ incentives for quality improvement. Health Affairs 23(2):127–141. Smith PC, York N. 2004. Quality incentives: The case of U.K. general practitioners—An ambitious U.K. quality improvement initiative offers the potential for enormous gains in the quality of primary health care. Health Affairs 23(3):112–118.
Representative terms from entire chapter: