Valuation of Alternative Payment Models

Transcription

1 Valuation of Alternative Payment Models No portion of this white paper may be used or duplicated by any person or entity for any purpose without the express written permission of PYA.

2 I. Introduction: The Intersection of New Payment Models and the Fraud and Abuse Laws With the goal of improving healthcare quality and patient outcomes while reducing total costs of care, payers now are promoting alternative payment models, or APMs, which compensate providers based on the value of the care they deliver, rather than the volume of services they perform. This transition from fee-for-service payments to value-based reimbursement, however, presents many challenges for providers, including compliance concerns. As directed under Section 512(b) of the Medicare Access and CHIP Reauthorization Act, the Centers for Medicare & Medicaid Services (CMS) recently issued a report to Congress on the impact of the current fraud and abuse laws on providers willingness and ability to participate in APMs. CMS concluded these laws may serve as an impediment to robust, innovative programs that align providers by using financial incentives to achieve quality standards, generate cost savings, and reduce waste. To encourage provider participation in Medicare APMs, CMS and the Office of Inspector General (OIG) have promulgated specific waivers of the fraud and abuse laws as part of Medicare shared savings and episodic payment programs. Generally speaking, these waivers insulate from challenge financial arrangements among model participants, provided specific requirements are satisfied. Those requirements vary by program, and strict compliance is required to invoke waiver protection. In broad terms, the waivers require the following: (1) the terms of the financial arrangement are memorialized in writing prior to the relevant performance period, and (2) the arrangement is structured in a manner that promotes the APM s underlying purposes (i.e., improving healthcare quality and outcomes and lowering costs). 2 Valuation of Alternative Payment Models

3 Unless and until these laws change, however, financial arrangements among providers under APMs outside of waiver protections must be structured to comply with the Anti-Kickback Statute and the Stark Law. This includes demonstrating that any payment made to a physician participating in an APM is consistent with fair market value. Even if a financial arrangement is protected under a Medicare APM waiver, demonstrating fair market value may be prudent for three reasons: (1) the waivers remain untested, and thus some degree of risk remains; (2) the agencies may revise the waivers at any time; and (3) if one or more of the parties is a not-for-profit entity, that entity may be at risk under tax-exempt organization regulations if the financial arrangement is not commercially reasonable. With fee-for-service reimbursement, valuation has been the means by which providers have proven payments represent fair compensation for the volume of work performed, and not rewards for patient referrals. Under APMs, however, patient referrals are less of a prized commodity; instead, providers will assign more value to quality and efficiency in care delivery. Regulators focus on improper care, poor quality of care, and patient steerage (also known as lemon dropping and cherry picking ). These new and different concerns necessitate new and different measures to evaluate and guard against improper utilization of resources and demonstrate fair market value. INCENTIVES The requirements of the aforementioned Medicare APM waivers provide a starting point with regard to those measures. The more the APM and associated financial arrangements in question are structured in a manner similar to governmental programs and their associated waivers, the better the argument that payments made under those models are incentives and rewards for physicians and other providers to deliver high-quality, efficient care, and not to influence referrals. While a starting point, the Medicare APM waivers are only one tool available to those evaluating the fair market value of financial arrangements relating to APMs. In the following sections, we describe how different APMs work and analyze how to demonstrate value within each model. Valuation of Alternative Payment Models 3

4 II. APM Categories From the least to the most dramatic shift from fee-for-service reimbursement, there are four general categories of APMs: (1) pay-for-performance arrangements (i.e., defined bonuses or adjustments to fee-for-service payments based on providers performance on quality and efficiency metrics); (2) shared savings arrangements; (3) episodic payments (also known as bundled payments); and (4) global budgets. While each APM that falls under a specific category operates differently, those APMs share common characteristics with regard to incentives and degrees of risk. Rewards the achievement of specified performance standards Incentive: Upward or downward adjustments to fee-for-service payments based on scores on objective performance measures Structure: Model co-exists with fee-for-service Pay-for- Performance (P4P) Examples: Hospital Value-Based Purchasing Program; Hospital Readmissions Reduction Program; Hospital-Acquired Condition Reduction Program; Physician Value Modifier Program; Medicare Quality Payment Program Rewards providers for working together to reduce payer s total cost of care for a defined population Incentive: Portion of the realized savings, in addition to fee-for-service payments Shared Savings Arrangements Structure: One- or two-sided models, depending on risk tolerance Examples: Medicare Shared Savings Program, Next Generation ACOs Rewards coordination and efficiency among all providers within a specific episode of care Incentive: Retention of overage of payment if costs are managed below target Episodic Payments Structure: Single payment rate for all services furnished during an identified episode of care, prospective or retrospective models, depending on risk tolerance Examples: Bundled Payment for Care Improvement; Episodic Payment Models Rewards provider network for managing a defined patient population within a specified budget Incentive: Reduction in unnecessary and avoidable services to remain within budget Structure: Advance payment for provider network to assume full responsibility for defined population Global Budgets Examples: Comprehensive ESRD Care Model; Direct Primary Care; Provider-Sponsored Medicare Advantage Plans 4 Valuation of Alternative Payment Models

5 III. Key Principles of APM Valuation A. P4P Payments In many instances, a provider receives P4P payments directly from a payer based on the individual provider s scores on pre-determined performance measures. Other arrangements involve payments to a group practice, health system, or provider network for distribution to participating providers. The manner in which such funds are distributed to those providers may be subject to challenge if the formula or payments are not consistent with fair market value or commercially reasonable. At present, there is little market data or consistent practice concerning such distributions of P4P payments to individual providers. Absent such benchmarks, one must consider other approaches to valuing these incentive payments. The traditional method of measuring a provider s time and effort to produce a certain result makes little sense in allocation of P4P payments. Adherence to clinical guidelines may result in a physician expending less time and effort, as several guidelines address overutilization of services. In other cases, the incentivized behaviors will require more work and resources for achievement. Importantly, physicians may be expending additional effort not readily measured by hours or work relative value units (wrvus), such as patient engagement, communication and coordination with other providers, or efficient management of clinical and administrative staff. One key factor in valuing P4P compensation is determining whether the measures for the compensation are appropriate and legitimate. For measures to be appropriate, they must require meaningful levels of achievement, and in many cases, improvement over historical levels, to justify the payment to an individual physician. One should also consider whether there is a reasonable relationship between the behavior that the group practice, health system, or provider network desires to incentivize and the value of the P4P payment made to physicians. If there is not, one can argue that the payment received by an individual physician serves another purpose, such as influencing referrals. The criteria for receiving remuneration and the potential amount of the P4P payment must be communicated to the physician in advance of his or her providing the services to which the payment relates. Simply put, the offer of a reward does not create an incentive to behave in a certain way if the physician is not aware of the award prior to the performance period. Valuation of Alternative Payment Models 5

6 B. Shared Savings Distributions Shared savings programs usually involve a payer agreeing to compensate a provider network a portion of the difference between the payer s actual total cost of care for a specified population and a pre-determined benchmark (one-sided), so long as the network achieves a certain score on defined quality measures. A small, yet growing, number of shared savings programs involve downside risk, i.e., the network must repay certain amounts to the payer if the actual total cost of care exceeds the benchmark (two-sided). Both one-sided and two-sided shared savings arrangements are intended to incentivize the network to identify and implement strategies to reduce costs while maintaining quality. Upon receipt of a shared savings payment, a network then makes categorical distributions. Typically, a portion of the payment is reserved by the network to repay initial infrastructure investments and/or fund ongoing operations. The remaining dollars then are divided among categories of participating providers based on a pre-determined formula. For example, the network may allocate a certain percentage of shared savings to fund a specialist pool and another percentage to fund a primary care physician pool. The funds in these pools then are distributed to individual participants. These categorical and individual distributions must be analyzed because different considerations will often apply to each. 1. Categorical Distributions Market Data. The approach that is likely the most amenable to valuing shared savings distributions is currently the analysis of market data to determine what the marketplace pays for comparable ownership interests or services. However, given the recent origins of the shared savings model and scarcity of available market data, even the market approach can be difficult to apply to categorical distributions. During this period of transition from fee-for-service to value-based payments, available survey data lags behind by one or two years, and generally reflects information from a limited number of respondents regarding somewhat limited data categories, with most participants predominantly still functioning under fee-for-service reimbursement. As value-based payments increase as a proportion of total compensation, the compensation data based on fee-for-service payments of past years becomes less and less comparable. Until reliable survey data related to value-based payments is available, appraisers likely will have to be flexible, using and appropriately adjusting proxy data, and understanding and documenting appropriate rationales and parameters for assessing value-based compensation. Currently, published data derived directly from value-based compensation models is sparse. Available data sources include the National Association of ACOs 2016 ACO survey (network-level expenses, accountable care organization [ACO] start-up and operating costs) and the Brookings Institute 2014 ACO implementation guide (estimates of ACO operating costs). 6 Valuation of Alternative Payment Models

7 Data is available from the public websites of the ACOs participating in the Medicare Shared Savings Program (MSSP). These ACOs are required by MSSP regulations to publish information such as the ACO s composition and anticipated shared savings distribution formula. Mining this data can determine what percentages of shared savings are planned for distribution among participants and what portion will be retained by the provider network to cover start-up and administrative expenses. Additionally, Medicare publishes data regarding the actual payments earned by ACOs under the MSSP. This data can be used to further investigate not only whether a participating network has earned bonus payments (and the associated factors leading to that success), but also the range of payments received by the different networks. Appraisers can further assess these distributed amounts to estimate the implied categorical distributions based on the ACO s self-reported allocation percentages. Repayment of Investments and Operating Costs. Some MSSP ACOs have reported a policy of recouping all costs incurred in establishing the provider network before distributing shared savings. However, based on the publicly reported MSSP information, the majority of ACOs currently do not require full repayment of start-up costs prior to these distributions. In the case in which a health system incurs the start-up costs for an ACO, one may argue those costs should be repaid prior to any other distribution of shared savings. However, a strong argument can be made that the health system s return on investment is realized through its enhanced ability to succeed under new healthcare payment and delivery models. The health system s ACO investment, therefore, is analogous to an investment in a new facility in which members of its medical staff will provide services. There is limited market data available on the portion of shared savings that should be reserved for reinvestment in network infrastructure (e.g., staffing, technology). The diverse approaches reported by MSSP ACOs range from systems requiring full investment expense recoupment prior to any distributions, to ACOs retaining more than 20% of shared savings. Relative Contribution to Success. Generally speaking, primary care physicians are best positioned to impact total cost of care through patient engagement, preventive care, referrals to other healthcare providers, and care coordination. However, specialists and hospital-based physicians may have a greater impact with high-complexity, high-cost patients through shared decision-making and care management. Assumption of Risk. If a shared savings arrangement includes down-side risk, the degree to which a category of providers assumes risk of repayment also would impact the appropriateness of any shared savings allocation. For example, a hospital participating in an ACO with community physicians may be willing to accept responsibility for repayment to incentivize physicians participation in such an APM. In such case, it may be appropriate for the hospital to receive a higher percentage of any shared savings. Valuation of Alternative Payment Models 7

8 In evaluating shared savings distribution models, the consideration of risk may not be limited to any repayment obligation. It may be appropriate, for example, to consider providers potential loss of fee-for-service revenue due to participation in a shared savings arrangement in evaluating a distribution formula. 2. Individual Distributions An evaluation of the manner in which shared savings are distributed to individual providers requires careful consideration and identification of the goals of the clinically integrated network (CIN), the interventions that will achieve those goals, the metrics that accurately will measure the interventions and outcomes, and the direct connection between the physicians actions and shared savings. Without extensive market data to assess the fair market value of distributions to individual physicians, the rationale for the distributions becomes paramount. Individual distributions may be subject to challenge if they do not bear some relationship to the network s goals of improving quality and outcomes and reducing total costs of care. For example, a significant distribution rendered to a physician who made no real contribution to the network s success may raise the inference of improper payments for referrals. Gatekeeper Measures. Gatekeeper measures provide a minimum threshold of performance necessary for a participant to receive, or in some cases, be further evaluated for the receipt of, shared savings. For example, an ACO may require that a participating provider achieve a specified score on identified performance measures as a condition of receiving any shared savings distribution. The use of gatekeeper measures provides one reliable means for demonstrating the relationship between any payment to a participating provider and the network s achievement of its goals. Distributions Based on Patient Attribution. For primary care physicians, patient attribution total or panel size may be a reasonable substitute for wrvu production (assuming any necessary adjustments are made to account for variances in patient acuity and other factors). Patient attribution also can provide a reasonable, if not entirely nuanced, approximation of an individual physician s contribution to the network s cost-savings efforts for the purpose of allocating shared savings on an individual basis. Unfortunately, the volatility of patient attribution data in the current market benchmarks suggests that this information may not be consistently reported by respondents. Benchmark data regarding patient attribution should be carefully considered for the purpose of setting and evaluating compensation. As more market participants move toward a global budget model, and a growing number of systems depend upon this measure, more reliable data should become available. 8 Valuation of Alternative Payment Models

9 Distributions Based on Performance on Selected Measures. To the extent scores on certain performance measures are tied to reductions in costs, a portion of the shared savings could be distributed to reward the best performers on those metrics. For example, distributions may be tied to the percentage of a physician s patients who receive annual wellness visits or other preventive services. Such a formula would provide a more direct reward for highly effective care, closely linking incentives to desired outcomes. While the inclusion of quality metrics in determining any value-based payment distribution is prudent, and it makes sense to include higher rewards for higher levels of performance, networks should still test the distribution model to ensure the structure is not likely to result in an inappropriate windfall to participants. Cap on Individual Distributions. One way to prevent such windfalls is to include a cap on individual distributions. For example, the amount of an individual physician s shared savings distribution may be capped at a percentage of that physician s total fee-for-service payments. The inclusion of a cap in a formula that otherwise calls for equal distribution of funds among participating physicians, for example, may prevent significant inequality between effort and reward. 3. IRS Private Letter Ruling On April 8, 2016, the IRS released Private Letter Ruling (PLR) denying tax-exempt status to a non-mssp ACO comprised of a tax-exempt health system and independent community physicians. The IRS concluded that because the ACO did not exclusively promote community health, but benefited the ACO s physicians, it could not qualify for tax-exempt status. While recognizing that MSSP ACOs further the charitable purpose of lessening the burden of government, the IRS determined this purpose was not applicable to commercial ACOs. The IRS noted that not all activities that promote health are necessarily charitable activities. In the case of a commercial ACO at least according to the IRS the primary benefit goes to unrelated healthcare providers, not the general public. In light of the PLR, those ACOs pursuing commercial contracts must be prepared to demonstrate the manner in which distributions made to participating providers will promote the federal government s stated purposes of improving the quality of care, enhancing patients experience of care, and reducing healthcare costs. In making this case, an ACO then can equate its distribution to providers with the productivity-based compensation paid by a tax-exempt health system to its employed and contracted providers the payment is consistent with the value provided by the physician. Valuation of Alternative Payment Models 9

10 C. Episodic Payment Distributions Under a retrospective bundled payment arrangement (the most common arrangement in the current market), a single entity (e.g., a provider network or a hospital) is accountable to a payer for the total cost of all care furnished to a patient during a specific episode of care, regardless of provider. The mechanics of these arrangements are relatively straightforward: (1) a discounted target price is established for a specific episode of care (e.g., date of admission for knee replacement surgery through 90 days post-discharge), based on historical data; (2) participating providers furnish services throughout the episode and receive fee-for-service payments; (3) if the payer s total cost of care is less than the target price, the payer compensates the difference to the entity; (4) if the payer s total cost of care is higher than the target price, the entity pays the difference to the payer; and (5) participating providers share the payment or costs with the entity, depending on the parties prior agreement. For such an arrangement, an appraiser may be called upon to evaluate the payments made between the entity and its participating providers. In that case, the principles discussed in the prior section about shared savings arrangements would be applicable. Under a prospective episodic payment arrangement, the payer makes a single payment to the entity for a specific episode of care, and the entity then distributes that payment among the providers involved in the patient s care. Again, an appraiser may be asked to evaluate the payments made by the entity to the providers. In that case, the principles discussed in the following section on global budgets would be applicable. D. Global Budget Distributions Unlike P4P, shared savings, and episodic payments all of which are expanding rapidly global budgets have not yet achieved widespread adoption, as this payment model is the most removed from fee-for-service reimbursement. Under this model, the payer agrees to pay a budgeted amount to a provider network to provide a specific range of services for a defined population for a certain period of time. As a condition of coverage, the individuals within the population are required to receive specified services from providers within the network. Two key factors determine the amount of the global budget: the scope of services to be provided (e.g., preventive and primary care services vs. all medically necessary services with specified exclusions) and the patient population (e.g., individuals with a specific highcost diagnosis vs. residents of a geographic area). Like episodic payments, global budgets may be retrospective (network providers continue to receive individual fee-for-service payments subject to reconciliation at the end of 10 Valuation of Alternative Payment Models

11 the performance period) or prospective (with periodic payments made to the network for distribution to its participants). Under these arrangements, the provider network bears the risk of increasing healthcare costs, and thus its participants must be incentivized to maintain their patients health, as opposed to providing additional (or more costly) healthcare services. At the same time, the network must monitor its participants to prevent lemon dropping or cherry picking, as the payer will condition payments for satisfactory scores on performance measures to restrict such activities. All of these factors come under consideration in evaluating the fair market value of network payments to participating providers. Presumably, global budget payments should be distributed among providers according to their relative roles in providing care to the patient, as opposed to the sheer volume of services provided. The significant challenge is in the appropriate measurement and representation of these relative efforts. In place of the relative value unit used today to measure the amount of work performed by a physician, a network will require a relative efficiency measure to quantify the physician s role in the network s success (or lack thereof). CMS now is developing patient relationship categories and codes that define and distinguish the relationship and responsibility of a physician with a specific patient at the time of furnishing an item or service. As these codes become available, they will be useful in distributing payments among participating providers based on their relative involvement in patients care. Valuation of Alternative Payment Models 11

12 The value of a physician to a provider network, however, is not defined solely by the role he or she plays in patients care. The network also will want to evaluate and reward physicians based on specific performance measures tied to quality of care and controlled costs. In the context of global payments, fair market value is a more elusive concept, as it becomes intertwined in the network s success in establishing a reward system that incentivizes certain physician behaviors. If the network is not successful in this regard, there will be less money available for distribution. At this point, the concerns underlying the fraud and abuse laws improper payment for referrals and overutilization of services should have limited relevance given the incentive structure on which global payment distributions are built. The role of the appraiser, therefore, likely will shift to evaluating the manner in which payments are calculated and distributed. IV. Conclusion The health of our economy is tied directly to the success of the transition from volume-based to value-based reimbursement, given these APMs potential for driving down healthcare spending. The appraiser will play a key role in this transition, giving providers the confidence to pursue new financial relationships without risk under the fraud and abuse laws. Although valuation of APMs is less straightforward than valuation of productivity-based compensation models, appraisers can identify and employ reliable standards by which to evaluate fair market value. With our extensive experience and expertise in the valuation of physician compensation, coupled with our consultants comprehensive understanding of alternative payment models, PYA can assist your organization meet the challenges of participating in these new models. To discuss how we can be of service to your organization, please contact one of the following: Lyle Oelrich Principal loelrich@pyapc.com (800) Martie Ross Principal mross@pyapc.com (800) Carol Carden Principal ccarden@pyapc.com (800) No portion of this white paper may be used or duplicated by any person or entity for any purpose without the express written permission of PYA. 12 Valuation of Alternative Payment Models