The ACO Proposed Rule: A [Skeptical] View From ‘The Street’

By Sheryl R. Skolnick, Ph.D, and Nick Leventis

A first glance at proposed Medicare ACO Rule:  What are we missing?

One of the most eagerly awaited proposed regulations under the Patient Protection and Affordable Care Act (PPACA) as amended (health reform), has been the rule that would define and regulate Accountable Care Organizations, i.e., patient-centered organizations charged with coordinating Medicare beneficiaries’ care, improving their outcomes, controlling the cost of said care and sharing in both the potential cost savings and losses. The idea behind the proposal is inherently appealing: providers would theoretically have the financial incentives to ensure that patients receive the right medical treatment at the right time and in the right place, to improve outcomes and reduce costs.

That proposed rule was finally released yesterday and it took the better part of the day to even begin to digest the complexity of the 429-page proposed rule, mostly because we had trouble believing what we were reading. Indeed, we’re sure of only one thing: that we do not yet fully understand all of the nuances, implications, incentives and even some of the definitions contained in this regulation. But we feel compelled to outline the most important aspects of the proposed rule that we’ve identified thus far in our reading of the rule, the companion proposed rule on anti-kickback waivers and all of the supporting documents provided by CMS today. That leads us to raise a key serious issue that we have with this proposed rule.

Our concern is this: we must be missing something major because, for the life of us, we can’t understand why any rational provider would incur the burdensome costs of establishing and operating an ACO, only to then be at risk for the major moral hazard inherent in the proposed ACO structure: providers who form ACOs (doctors, hospitals and post-acute providers) would be at risk for the cost of health care chosen by patients, yet would be expressly prohibited from requiring that patient to obtain that care within the ACO. More, should the patient not comply with their treatment plans (which must be based on evidence-based, best-practice protocols under the rule) and boomerang back into the hospital, the ACO would be penalized. And, it appears that a patient could be assigned to an ACO retroactively so that the ACO wouldn’t know it has a ‘non-compliant’ patient until after the episode of care is completed. Medicare’s data on a patient’s utilization, however, cannot be obtained by the ACO in real-time: the ACO wouldn’t know that its patients had an ‘unnecessary admission’ to a non-ACO hospital until after the patient is discharged from that facility. The only condition under which we can see an ACO being formed under such rules would be if the ACO had monopoly control over the continuum of care in its market, but the proposed rule puts tight restrictions on the ACO’s market share. Finally, we note that the one organization that best describes an ACO today, i.e., Kaiser Permanente (a vertically integrated delivery system in which care is coordinated) appears not to qualify under the ACO definition. The best thing we can say about the proposed rule is that participation in the ACO program is voluntary.

In summary, the proposed rule appears to do a good job of making ACO’s ‘accountable,’ but appears to do a very poor job of making patients ‘accountable.’ Unless we are wrong (and we could truly have missed something here), we simply can’t see publicly traded hospital companies taking on the risk of forming an ACO, putting some significant amount of their cash flows at risk for arbitrary loss by the actions of non-compliant patients because they are stripped of the single tool that would make ACOs work: the ability to limit patient choices to a continuum of high quality, low cost providers.

For those who haven’t tried to digest the rule, we present the highlights [Editors Note: For the published rule in the Federal Register, click here]:

• The proposal is a three-year pilot program that would implement section 3022 of the PPACA which relates to Medicare payments to providers and suppliers participating in ACOs.
• Patient and provider participation in the ACO is purely voluntary, except it appears that patients can be retroactively assigned to an ACO. The proposed rule states that patients would be assigned to the primary care provider who provided the majority of his/her care.
• Providers would be eligible to receive traditional Medicare fee-for-service payments under Parts A and B and be eligible for additional payments if they meet the new ACO quality and savings requirements.

Risk Models:

• CMS proposes participating ACOs choose either a one-sided or two-sided model: the one-sided model shares only the savings for the first two years and shares savings and losses in the third year, the two-sided model shares savings and risks in all three years.
• There are two risk models because, according to CMS, the Agency wants ‘new’ ACOs to be able to have a period of adjustment, i.e., gaining experience with at-risk reimbursement, before becoming responsible for losses.

Providers eligible to be in the ACO:

• Hospitals and physicians in group practice arrangements, networks of individual  practices of ACO professionals, partnerships or JVs between hospitals and ACO professionals, hospitals employing ACO professionals, and Medicare providers and suppliers determined by the Secretary.
• Private managed care organizations do not appear to be, a priori, allowed to be part of an ACO, although the proposed rule dealing with waivers of Stark Laws indicates that private managed care plans may be part of an ACO under some circumstances. (Some on the Street had believed that health plans could form ACOs, but based on our preliminary reading of the ACO rule, that does not appear to be the case.)

An ACO must take responsibility for a minimum of 5,000 beneficiaries for a period of three years:

• Beneficiaries could be ‘assigned’ to an ACO at the end of a period if they received the majority of their care from a physician-member of an ACO. (See pages 392-393 of the document.)
• Beneficiaries would be informed that the physician is part of an ACO and that should the patient choose to be treated by that physician, said beneficiary would become a ‘member’ of the ACO. Otherwise, the beneficiary could choose another physician.
• But even among those 5,000, patients cannot be limited in their choice of providers and would still receive their full Medicare Part A (inpatient) and Part B (outpatient) benefits.

CMS will develop a benchmark for each ACO against which ACO performance is measured to assess whether it qualifies to receive shared savings, or to be held accountable for losses.

• Benchmarks for quality and cost would be established, but before payment of ‘shared savings’ would be made, the ACO would have to show savings of more than 5% of the benchmark.
• ACOs who fail to deliver minimum quality standards could be terminated.

CMS proposes to establish a minimum sharing rate that would account for normal variations in health care spending, so the ACO would be entitled to shared savings only when savings exceeded the minimum sharing rate. See our comment below: the financial returns may be insufficient to warrant the risks and investments inherent in forming and operating an ACO.

Measuring Quality Improvement

• Patient/caregiver experience of care, care coordination, patient safety, preventive health, and at-risk population/frail elderly health.
• Performance standards would incorporate proposals to prevent providers in ACOs from being penalized for treating complex patients.

Antitrust, Anti-Kickback and Anti-Physician Self-Referral Guidance

• The Department of Justice (DOJ) and the Federal Trade Commission (FTC) have worked to facilitate the creation of ACOs by giving providers guidance so they can operate without violating antitrust laws.
• Anti-Kickback and Physician Self-Referral waivers are proposed in a separate rule and would be limited to very specific circumstances related to sharing in savings and losses.

Questions and Thoughts:

• Is there enough financial incentive to make providers want to participate? We may be wrong here, but we can’t find enough potential return even for the most efficient, sophisticated and integrated providers to justify the upfront and ongoing costs and the moral hazard risks.
• The administrative and ‘set up’ costs appear high, with a heavy governance, compliance, oversight and reporting burden placed on the ACO.
• In determining whether or not savings were achieved, CMS will establish an annual spending target and if actual expenditure is more than 5% below the target, an incentive payment will be made:
• The rule says: “an incentive payment for each year shall be equal to a portion of the amount by which actual expenditures for applicable beneficiaries under Parts A and B for the year are estimated to be less than 5% less than the estimated spending target for the year” (page 31).
• Therefore, the ACO must not only illustrate that costs were reduced, BUT expenditures must be 5% below CMS’ estimate in order to receive an incentive payment.
• ACOs must carry 5,000 beneficiaries in order to be eligible: finding 5,000 willing participants or having them assigned retroactively and saving more than 5% versus a target spending level could be too high a hurdle with too little return to induce participation in an ACO.

We believe that only the most technologically sophisticated physician groups and providers would be capable of participating in the program at the outset.

• The proposal has many complex regulations that providers would have to adhere to including an almost seamless communication network.
• Providers would likely have to have advanced and compatible computer systems to allow them to share patient data.
• Although there is a financial incentive for hospitals to meet meaningful use standards for the HITECH Act, there will likely have to be significant technology investments made by small physician groups if they wanted to participate in the ACO program.

The moral hazard issue:

• Patients participating in an ACO would be able to see or visit any provider they choose and ACOs are expressly prohibited from preventing such free choice.
• ACOs could be penalized if the patient boomeranged back to the hospital for an unnecessary readmission.
• How is the ACO supposed to be able to achieve a better outcome and lower spending if the patient leaves the ACO for care at a provider not affiliated with network?o Why make the investment in the ACO if the provider cannot control the care?o How does the ACO make informed decisions when it cannot get ‘real time’ updates on patient status?
• The potential for malpractice liability:  We are not lawyers, but based on our early read of the rule, we believe physicians in an ACO may potentially face litigation risk if a patient is harmed by another physician participating in the same ACO or, worse yet, outside of the ACO. (The physician is ‘accountable’ as is the ACO, after all.) The rule does not appear to contemplate the malpractice risk.
• It appears that patients can be retroactively assigned to ACOs: the ACO may not even know it is responsible for a particular patient’s care. (See pages 392 to 393.) At least, we think that is what the proposed rule says. We hope that we are wrong.

In other words, the proposed rule appears to make the ACO responsible for ACO behavior AND patient behavior, yet gives the ACO few (if any) tools to retain oversight and direct treatment of the patient. This is in direct contrast to the clinic model (Geisinger, for example) where the patient is seen exclusively by the clinic physicians unless specifically referred to an outside provider.

Sheryl R. Skolnick, Ph.D, is Senior Vice President, and Nick Leventis is an Associate for CRT CApital Group, LLC. For more information contact Dr. Skolnick at 203.569.4359, or via email sskolnick(at)crtllc(dot)com, or 203.569.4326, nleventis(at)crtllc(dot)com.

REQUIRED DISCLOSURES: The recommendations and guidance expressed in this research report accurately reflect the personal recommendations and guidance of the research analyst principally responsible for the preparation of this report. No part of the compensation received by the analyst principally responsible for the preparation of this report was, is or will be directly or indirectly related to the specific recommendations and guidance expressed in this report. Direct or indirect analyst compensation may be based on performance-related considerations associated with the recommendations and guidance expressed by the analyst in this report. The research analyst primarily responsible for the preparation of this report received compensation that is based upon CRT Capital Group LLC’s total business revenues, including revenues derived from CRT’s investment banking business.