Identifying and Balancing Provider Contributions When Developing an ACO Shared Savings Distribution Methodology

The Patient Protection and Affordable Care Act (“PPACA”) was promulgated in 2010 with the express goal of meeting the “Triple Aim” of healthcare— better quality, improved health and lower costs. ¹Donald M. Berwick, Thomas W. Nolan, and John Whittington, The Triple Aim: Care, Health and Cost, Health Affairs. 2008 May/June; 27(3): 759-769. A key initiative of the PPACA was the establishment of the Medicare Shared Savings Program (“MSSP”) for accountable care organizations (“ACOs”). Pursuant to the MSSP, ACOs are responsible for managing the quality and cost of patient care for a defined Medicare patient population. If certain quality and patient experience criteria are met and savings are created compared to a benchmark, then the ACO is awarded a portion of such savings. The intention of the MSSP is to move from a volume to a value based reimbursement model.

To achieve the goals of the MSSP, the methodology by which the ACO distributes shared savings among the providers participating in the ACO (collectively, the “ACO participants”) is of significant importance; a fair and equitable shared savings distribution methodology must be established. This will ensure that each of the ACO participants is incentivized to undertake the appropriate actions to continually improve both the quality and efficiency of care, thereby promoting the ongoing success of the ACO. As a result of a series of regulatory waivers provided by the Centers for Medicare and Medicaid Services (“CMS”) and the Office of the Inspector General (“OIG”), each ACO has significant flexibility in developing its own shared savings distribution model. ²CMS and OIG have issued several waivers of certain healthcare fraud and abuse laws for ACOs that limit their activities to participating in the MSSP, including a Shared Savings Distribution Waiver. 80 Fed. Reg. 66725-66745 (Oct. 29, 2015). Under the Shared Savings Distribution Waiver, Stark, Anti-Kickback Statute and Civil Monetary Penalty laws are waived with respect to the distribution or use of shared savings earned by an ACO if: (1) the ACO is participating in the MSSP, (2) the shared savings were earned during a period that the ACO was participating in the MSSP, (3) the shared savings are distributed solely to ACO participants during such period that created the shared savings or the shared savings are used for activities “reasonably related to the purposes of the MSSP,” and (4) shared savings payments are not made by hospitals, directly or indirectly, to participating physicians in order to induce the physician to reduce or limit medically necessary care to patients under the direct care of the physician. Rud M. Blumentritt and Gregory D. Anderson, “ACO Shared Savings Distribution Models”, American Health Lawyers Association available at https://www.healthlawyers.org/Events/Programs/Materials/Documents/HCT13/h_anderson.pdf. In addition, the Internal Revenue Service has indicated that an ACO’s shared savings distribution methodology will not put a tax-exempt participant’s tax-exempt status at risk so long as: (1) the terms of participation in the ACO are defined in a written negotiated agreement, (2) the ACO’s activities are limited to the MSSP, (3) the tax-exempt organization’s share of economic benefits derived from the ACO is proportional to the benefits or contributions the tax-exempt organization provides to the ACO, (4) the tax-exempt organization’s share of ACO losses does not exceed the share of ACO economic benefits to which the tax-exempt organization is entitled, and (5) all contracts and transactions entered into by the tax-exempt organization with the ACO and the other ACO participants, and by the ACO with the other ACO participants and other parties, are at fair market value. Internal Revenue Service FS-2011-11, October 20, 2011. A distribution methodology needs to achieve the goals of the MSSP to encourage better care for individuals, improved health for populations and lower growth in expenditures. ³42 CFR 425.204(d).

In developing an ACO shared services distribution methodology, we believe that ACO participants should be properly rewarded for the contributions they make to the ACO, which help attain shared savings. ⁴J. Bobbitt, et al, “Distribution Based on Contribution: A merit-based ACO shared savings distribution model”, Toward Accountable Care Consortium (2013). However, each ACO participant may contribute to the success of the ACO in a variety of different ways.

The central thesis of this article is that to effectively align the shared savings distribution methodology with the incentives driving ACO participants, it is critical to identify and balance the different types of contributions they make. In addition the shared savings distribution methodology must be flexible enough to account for the shift over time in the contributions received from ACO participants, as the ACO moves from the start-up phase to maturity. We believe that both the clinical and financial contributions from each ACO participant should be taken into account. If only the clinical contributions are taken into account, primary care physicians may reap a significantly disproportionate allocation due to the primary care physician’s central role in “managing the care of the patient and coordinating the movement of the patient along the transition points in the health care system,” when, in fact, reduced hospital and specialist utilization may result in much of the shared savings produced. ⁵Rud M. Blumentritt and Gregory D. Anderson, “ACO Shared Savings Distribution Models”, American Health Lawyers Association (2013) available at https://www.healthlawyers.org/Events/Programs/Materials/Documents/HCT13/h_anderson.pdf. Consequently, we believe the financial impact of ACO participation on certain providers must be factored into the shared savings distribution methodology.

For purposes of this article, we examine the contributions stemming from ACO participants falling within the following classes: hospitals, specialists and primary care physicians. ⁶Healthcare providers that do not fall within the category of hospitals, specialist physicians, or primary care physicians may also constitute ACO participants, such as post-acute care providers. We believe the contributions from these providers must also be properly taken into account in determining shared savings distributions. This article is limited to a discussion of the distribution of shared savings among ACO participants after sufficiently taking into account ongoing ACO working capital requirements, ACO infrastructure reinvestment needs, and ACO investors’ cost of capital. ⁷This article assumes that, in order to secure sufficient capital, as well as ensure ongoing operational capabilities, the ACO will make a first tier allocation of shared savings to working capital and infrastructure needs, such as upgraded or enhanced health information technology, hiring additional care coordinators, and hiring additional support staff to enhance patient care and transition. Rud M. Blumentritt and Gregory D. Anderson, “ACO Shared Savings Distribution Models”, American Health Lawyers Association (2013) available at https://www.healthlawyers.org/Events/Programs/Materials/Documents/HCT13/h_anderson.pdf; Mona Siddiqui, MD, MPH and Scott A. Berkowitz, MD, MBA, “Shared Savings Models for ACOs-Incentivizing Primary Care Physicians” J. Gen Intern Med. 2014 Jun, 29(6):832-4. Furthermore, ACO investors have taken the risk of providing capital to the ACOs and will expect a return on their investment. To determine the return on investment to be expected by ACO equity owners, the cost of capital associated with the ACO investment should be calculated utilizing a weighted average cost of capital (“WACC”) analysis (an in-depth analysis of the factors applied in a WACC analysis is beyond the scope of this article). Our suggestion would be for the return on investment to be viewed in the long-term for investors. When the ACO begins operations, investors should assume that significant shared savings will be allocated to infrastructure and other operating costs of the ACO, which should improve the value of the investment in the future. As the ACO reaches maturity, shared savings should specifically be allocated to reimbursing the investor for its cost of capital before any shared savings are distributed to the non-investor ACO participants. This article is designed to provide a conceptual framework for developing a shared savings distribution methodology, rather than to provide a technical analysis for how a particular ACO might determine its own, specific methodology.

Part 1 of this article examines clinical performance metrics that help measure contributions to the production of shared savings and provides an opinion as to those metrics that should be weighed most heavily when determining shared savings distributions. Part 2 of this article identifies the financial contributions that have been made by certain ACO participants and broadly outlines how such financial contributions should be quantified as a measurement for shared savings distributions. In Part 3, we set forth a potential framework for appropriately balancing clinical and financial contributions to the ACO when developing a shared savings distribution methodology.

Part 1: Clinical Quality Metrics—Identification and Weighting

Performance on quality measures helps to evaluate the relative clinical contributions of ACO participants with respect to the attainment of shared savings. ⁸“Distribution based on contribution: The merit-based ACO shared savings distribution model,” Julian D. “Bo” Bobbitt, Jr., September 4, 2013. Since CMS has already set forth 34 quality measures for the 2016 MSSP, these measures may serve as a useful starting point. ⁹There is much debate within the health care provider community as to utility of these 34 quality measures and their ability to successfully measure health care quality and improvement. We believe that each ACO should make its own determination as to the precise quality measures to utilize. The 34 quality measures fall under four quality domains:

1. Patient/Caregiver Experience Measures

• 2.  Care Coordination/Patient Safety Measures

• 3.  Preventive Health Measures

• 4.  At Risk Population Health Measures

A detailed analysis of these 34 quality measures is beyond the scope of this article. ¹⁰For a detailed narrative of these measures and their specifications, please see: https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/sharedsavingsprogram/Quality-Measures-Standards.html. The purpose of this Part 1 is to make the case that, while all of the measures should be evaluated in determining clinical quality, benchmark performance on only a few may likely be associated with demonstrably superior value results. Consequently, optimal performance on those measures should be accorded disproportionate weight when determining providers’ claims on distributable shared savings.

Performance on the metrics that should carry the greatest weight in determining shared savings distributions are those that best meet the following three criteria (the “Three Factor Test”):

1. The total costs of care;

• 2. Improvements in the health status of the greatest number of patients; and

• 3. Closure of the most significant specifically identifiable gaps in care delivery (having the highest degree of variation) to help achieve measurable outcomes-based enhancements in the results associated with the first two criteria.

The following is an analysis of how comparatively well the various quality measures of the four quality domains meet the Three Factor Test:

Quality Measures in the Patient/Caregiver Experience Domain

The purpose of the “Patient and Caregiver Experience” domain of quality measures is that quality constitutes not only the production of care that results in desired patient outcomes, but also the way in which the overall process of delivery is perceived by patients and caregivers.

Benchmark level performance in all of the measures in this domain should continue to be essential in determining clinical performance and basic eligibility for shared savings. However, these are not measures that should be weighted disproportionately when determining the distribution of shared savings. These measures are surrogate or proxy in nature when it comes to linking performance on them to actual improvements in patient outcomes and decreases in overall costs. Consequently, these measures do not sufficiently satisfy the third criterion of the Three Factor Test.

Preventive Health Measures:

All of the “Preventive Health Measures” are important in improving the overall health of both individuals and populations. However, it usually takes years before the effects of improved preventive health measures manifest as better (and more cost effectively produced) patient outcomes.

Even for some of these measures that work in a shorter timeline, there may be a problem in attributing interventions to results. For instance, it may be possible to show a statistically significant correlation between the rate of influenza and pneumonia vaccines and decreased hospitalizations for these two ailments. However, the decrease in the incidence of these conditions during one period of time may be attributable to many other causes. So, while doing well on these measures is very important to achieving the goals of the Triple Aim, their near term impact on the production of shared savings may be quite difficult to evaluate accurately, thus not meeting the third criterion.

At Risk Population Health Measures:

Similarly, the measures dealing with the “At Risk Population” are essential in evaluating whether patients have received excellent care. Measures such as controlling blood pressure and Hemoglobin A1C levels for patients with diabetes, the use of aspirin for patients with ischemic cardiovascular disease, beta blocking agents for heart failure (HF), and angiotensin converting enzyme (ACE) inhibitors or angiotensin receptor blockers for coronary artery disease are going to impact patient outcomes. However, it is difficult to predict when and to what measurable extent these processes impact outcomes, thus also not meeting the third criterion.

Interestingly, performance on the only true outcome measure in this domain (i.e. achieving remission of depression within 12 months) is only evaluated based on an ACO’s having reported that this remission occurred, not for having actually produced it. If ACOs could make this a true pay for performance outcomes-based measure, benchmark performance on this measure could impact shared savings distributions.

Care Coordination/Patient Safety Measures:

It is within this domain that there are quality measures that meet the Three Factor Test.

There are two quality measures pertaining to “Ambulatory Sensitive Unplanned Admissions.” One is for Chronic Obstructive Pulmonary Disease (COPD) and Asthma (ACO # 9 Measure). The other is for Heart Failure (ACO # 10 Measure). Both of these measures deal with conditions that often result in preventable hospitalizations that meet the first two criteria in the Three Factor Test. Unlike the previously discussed measures, these two could also meet the third criterion in the Three Factor Test. Both are sufficiently specific outcomes-based measures of significant magnitude and frequency. There are also substantial gaps in caring for patients with these two prevalent and potentially costly conditions. The signal ratio for both of these measures, revealing the proportion of total observed variation in results that is related to systematic differences rather than just random variation, is 93% . ¹¹Agency for Healthcare Research and Quality: Guide for Prevention Quality Indicators (PQIs); AHRQ; Rockville, MD 2007

The quality measures that pertain to “All Cause Unplanned Admissions” for patients with (previously diagnosed) Diabetes (ACO # 36 Measure), Heart Failure (ACO # 37 Measure) and Multiple Chronic Conditions (ACO # 38 Measure) also meet the Three Factor Test. First, as outcomes-based performance measures, all three of these meet the first two criteria of the Three Factor Test. These measures also concern the outcomes of patients previously diagnosed with diabetes, heart failure, or two or more conditions such as diabetes, past myocardial infarction, Alzheimer’s disease, atrial fibrillation, chronic renal disease, COPD/Asthma, heart failure, past stroke or Depression. These are patients with high volume, high cost, and significant care variability conditions known to result in preventable admissions. ¹²Levine, S., et. al., Home Care Patients at High Risk of Hospitalization; Am. J. of Managed Care 18(8): e269-76 (2012) Consequently, ACO providers who proactively participate in team-based coordinated care initiatives shown to have a substantial positive impact on decreasing the rate of preventable admissions of patients with these conditions should be differentially rewarded.

Not all of the quality measures in the “Care Coordination/ Patient Safety Measures” domain meet the Three Factor Test. For example, the “All Condition, Risk Standardized Readmissions” quality measure meets the first two, but not the third criterion. Since this measure pertains to “all conditions,” it is difficult to specifically incent particular provider partners to close care gaps. Additionally, there is significant controversy concerning how much “risk standardization” has taken into account the socioeconomic determinants of health that may be the real drivers of readmissions.

Similarly, the “All Cause Readmissions from Skilled Nursing Facilities (SNFs)” measure does not provide enough specificity to evaluate and reward particular providers through differentiated distributions. The main program to decrease readmissions from SNFs (INTERACT) is focused on detecting conditions in SNF patients to prevent readmissions. There might be a rationale for incenting those clinicians working most directly with these patients in SNFs. ¹³Please see https://americannursetoday.com/transitional-care-can-reduce-hospital-readmissions/. However, this group usually does not include the independently contracting primary care providers of these patients in an ACO.

The “Meaningful Use Requirements for the Use of Electronic Health Records” measure is getting subsumed by the Medicare Access and CHIP Reauthorization Act or MACRA) and is more of a process vs. an outcome quality measure. The “Documentation of Current Medications in the Medical Record,” while extremely important for clinical quality, does not meet the third criterion well enough to justify differentiated distributed savings. This is a process type quality measure, and current benchmark performance data is scanty.

The “Preventable Falls” measure is a major risk factor and cause of mortality and morbidity especially in the elderly. As a result, performing well on this measure can and should be very important for awarding shared savings to ACOs in the first place. However, as a process measure, it likely should not be a measure that justifies differential shared savings distributions.

Part 2: Financial Contributions – Identifying and Measuring

Although it is critical to measure the clinical contribution each ACO participant has made to the achievement of shared savings, there are also significant financial contributions made by ACO participants that should be identified, measured, and properly weighed to determine the ACO’s shared savings distribution methodology. In particular, as a direct result of successful ACO participation, certain hospitals and specialist physicians will incur materially lower patient or procedural volumes across the ACO patient population, which will result in decreased payments to such providers. ¹⁴Primary care physicians are unlikely to take the brunt of this reduced utilization because of the renewed emphasis on preventative care and care management by primary care physicians. The reason for this is that a Medicare ACO that receives shared savings must first have reduced utilization in a manner that reduces payments to providers in excess of the amount of shared savings. ¹⁵Jonathan Pearce, “Calculating Provider Revenue Loss in an ACO”, Becker’s Hospital Review, September 14, 2012. It is hospitals and specialist physicians, not primary care physicians, which will likely incur the brunt of this reduced utilization. Renewed emphasis on preventative care and care management reduces hospitalizations and minimizes expensive procedures and costly drugs provided by specialists.

The financial impact of the reduction in payments to hospitals and specialists as a result of ACO participation should be properly assessed as a valuable contribution to the ACO. As the ACO moves from the start-up phase to maturity, hospitals and specialists should begin to see the offsetting benefits of increased referrals to them as high-value, low-cost providers in the community. ¹⁶“Although a few hospitals may see enough shared savings from ACO arrangements to offset some of the pain from reduced utilization, most still will take a hit. But a well-designed ACO can use education and the prospects for shared savings to encourage primary care physicians and specialists to admit patients to the ACO’s in-network hospital, and that could offset some or all of the decreased utilization. Hospitals will see reduced costs with reduced utilization, partly offsetting the negative impact.” Hospitals Look for Ways to Fill in Lost Volume from ACOs, Jill Brown, August 22, 2013, AIS Blogs, Health Plan Business. The shared savings distribution methodology should include a mechanism to account for such offsetting benefits to hospitals and specialists.

Hospitals

Hospitals that participate in successful ACOs are likely to see their inpatient and outpatient volumes drop as a result of increased care coordination, preventative care services, and efforts to provide care in the least costly settings possible. Some hospitals may be able to offset such losses over the long-term through increased market share by acquiring a reputation as an efficient provider. Still, most hospitals will take a material financial hit until ACOs reach a greater level of maturity in the market. ¹⁷Id. To keep these hospitals motivated to continue to assist the ACO with its goals, while securing their financial viability at the reduced level of activity, hospitals should receive credit for the financial impact of reduced volume with respect to the population covered by the ACO when determining the distribution of shared savings. ¹⁸This concern has been taken into account by the State of Maryland with respect to the Global Budget Revenue methodology. Subject to certain limitations, hospitals’ reimbursement from third party payers, including Medicare and Medicaid, will not be reduced as a result of a decline in volume, thereby heavily incentivizing hospitals to work together with other providers to reduce hospital utilization.

To determine the financial impact to the hospital(s) participating in an ACO, there needs to be a methodology to determine the link between the ACO’s specific initiatives and a reduction in volume at such hospitals. For example, if the ACO is targeting a reduction in preventable emergency room visits, all provider services related to this initiative must be identified and the impact on patient volume calculated. ¹⁹Jonathan Pearce, “Calculating Provider Revenue Loss in an ACO,” Becker’s Hospital Review, September 14, 2012. It is often difficult to determine provider losses stemming from ACO activities, as they relate to the omission, rather than the commission, of services.

Once a determination has been made as to the estimated lost volume stemming from ACO initiatives, the financial impact to the hospital(s) from such lost volume must be assessed. To do this, first, the lost net patient service revenue associated with such volume should be computed. Then, the avoidable costs to the hospital as a result of the reduction in volume, which can be eliminated without negatively impacting the hospital’s readiness to serve other patients, should be determined and subtracted from net patient service revenue. Generally, the avoidable costs to the hospital will be the direct, variable costs associated with each admission or outpatient procedure that has been lost. ²⁰Although direct fixed costs associated with the hospital departments impacted by the reduction in volume may be reduced over time if volumes stay lower, the impact would be difficult to ascertain in the short term. In addition, any improvements in Medicare reimbursement as a result of penalties avoided or improvements in value based care metrics associated with care to the ACO patient population should be subtracted from the net patient service revenue that has been lost. ²¹Siva Subramanian, “Project Boost: A Return on Investment Analysis,” Society of Hospital Medicine, http://www.hospitalmedicine.org/Web/Quality_Innovation/Implementation_Toolkits/Project_BOOST/Web/Quality___Innovation/Implementation_Toolkit/Boost/BOOST_Intervention/ROI.aspx.

Specialists

Although not discussed as frequently, specialist physicians may also see a reduction in visits and procedures from the population covered by the ACO. ²²Emergency room physicians, cardiologists, pulmonologists, orthopedists and radiologists have been identified as specialists that are most likely to be affected negatively from a financial perspective by ACO participation. Robert Kocher and Anuraag Chigurupati, “The Coming Battle over Shared Savings—Primary Care versus Specialists,” N Engl J Med 2016; 375:104-106. Specialists must undertake greater care coordination, particularly with certain high risk patients, thereby reducing the amount of time for higher reimbursing procedures. Additionally, ACOs will be targeting higher cost drugs and procedures undertaken by specialists for cost reduction. Finally, specialists that provide coverage in ACO participating hospitals may find that the reduction in volume at the hospital has a financial impact on them as well.

As with hospitals, we must identify and quantify the link between ACO specific initiatives and a reduction in visits and procedures for specialists. The net patient service revenue lost minus any reductions in avoidable costs and improvements in Medicare reimbursement associated with higher quality care through the ACO, should be considered a financial contribution. Similar to hospitals, the financial impact of lost procedural volume by specialists may be offset over the long-term through increased market share. However, increased market share deriving from a positive reputation as an efficient provider will generally take time to develop, likely coinciding with the maturity of ACOs.

Part 3: Weighing Clinical and Financial Contributions to the ACO

In measuring each ACO participant’s unique contributions to the achievement of ACO shared savings, we recommend a model that provides fair relative weights to both clinical and financial contributions. One method for balancing financial and clinical contributions to the ACO is to treat the total contributions by an ACO participant at the end of each year as the equivalent to a capital contribution to a hypothetical partnership. Based upon the equivalent capital contributions received from each ACO participant by the hypothetical partnership, each ACO participant would receive a percentage equity stake in the hypothetical partnership equal to its equivalent capital contribution divided by the total equivalent capital contributions made to the hypothetical partnership for the year. The ACO participant would then receive distributions from the ACO equal to his or her percentage equity stake in the hypothetical partnership multiplied by the total amount of shared savings earned by the ACO for the year. By adding financial contributions to the clinical contributions associated with garnering ACO shared savings, hospitals and specialists will be partially reimbursed for the loss incurred as a result of decreased utilization.

Pursuant to this model, the total amount of clinical contributions among the ACO participants would equal the total amount of ACO shared savings for the year after sufficiently taking into account ongoing ACO working capital requirements, ACO infrastructure reinvestment needs, and ACO investors’ cost of capital. In determining the allocation of clinical contributions among the ACO participants, we would determine how well each ACO participant performed relative to benchmarks on each quality measure (assuming sufficient patient care results for statistically significant evaluations) to provide a score for each such measure.

By way of example, if an ACO participant adhered to the best evidence based practices for treatment of patients with heart failure 60% of the time (and the benchmark was 80%) that ACO participant’s score for that specific quality measure would be 60% divided by 80% or 0.75. Each ACO participant’s score would then be compared to that of the other ACO participants. Those ACO participants who achieved top quintile results compared to the group would receive 1 point for the measure, the second quintile would receive 0.75 points, third quintile .50 points, fourth quintile .25 points, and the lowest quintile performers 0 points. ²³This is a similar scoring methodology to that used in by the federal government with various ACO programs it oversees, including the most recent Oncology Care Model, see: https://innovation.cms.gov/initiatives/oncology-care/. We would repeat this process for each ACO participant’s performance on all 34 measures.

We would then double weight the points of each participant on the five measures that best meet the Three Factor Test described in Part 1. This would permit a physician to receive a “bonus” point total of up to 15% based upon superior performance on these measures (i.e. calculated by accumulating five extra points/34 total measures). We believe this should be a sufficient incentive for physicians to emphasize these quality measures. Groups involved in value based purchasing the longest (e.g. the Integrated Healthcare Association) have found that bonuses of 10% or more are required to produce significant behavioral change among providers. ²⁴Please see http://www.iha.org/.

Finally, we would sum all the points for each ACO participant and divide those totals by the cumulative points scored by all ACO participants. The resulting percentage for each ACO participant would then be multiplied by the total amount of shared savings garnered by the ACO for the year to determine such ACO participant’s particular clinical contribution for the year.

The total amount of the financial contributions from ACO participants would be equal to the measured negative financial impact of ACO participation for the year.

To demonstrate how this model for determining ACO shared savings might work in practice, we have provided an example below:

• In year 1 of operations, an ACO achieved $10 million in shared savings, after payment for ongoing operating costs and reinvestment in infrastructure. The ACO’s primary care physicians’ clinical contributions are measured at $6 million, specialists at $2.5 million, and hospitals at $1.5 million. In addition, the specialists’ financial loss from ACO participation is $2 million and hospitals’ financial loss is $3 million.

Using the methodology set forth above, the total amount of equivalent capital contributions to the hypothetical partnership would be $15 million ($10 million clinical contributions plus $5 million financial contributions). The percentage equity stakes held by each class of ACO participants would be as follows: primary care physicians at 40% ($6 million/$15 million), specialists at 30% ($4.5 million/$15 million), and hospitals at 30% ($4.5 million/$15 million). As a result, the primary care physicians would receive a distribution of shared savings of $4 million (40% of $10 million net shared savings), specialists would receive $3 million (30% of $10 million net shared savings), and the hospitals would receive $3 million (30% of $10 million net shared savings). Individual distributions to a particular ACO participant would be based solely on the amount of equivalent capital contributions provided by each ACO participant, rather than the performance of the ACO participant’s class as a whole.

Conclusion

To promote the enduring success of ACOs, it is crucial to develop a shared savings distribution methodology that adequately accounts for the competing interests of the different ACO participant classes. To accomplish this end, all of the clinical and financial contributions provided by ACO participants should be appropriately identified and measured. A successful shared savings distribution methodology must be flexible enough to adapt over time to take into account the countervailing financial gains achieved by hospitals and specialists as they offset lost contribution margins with respect to certain patient populations with increasing overall market share, which may occur as the ACO moves from the start-up phase to maturity.