It’s Not the ‘Shared Savings’, Stupid: Why ACOs Under the Proposed Rule Will Change Medicine As We Know It

CMS got the Medicare Shared Savings Program (MSSP) proposed rule largely right, but not because of the actual “shared savings” that the ACO model is commonly associated with. Rather, the MSSP will usher in a shift from the practice of medicine as primarily an art, to the practice of medicine as primarily a science.

The explosion over the last 50 years of drugs and devices — and the studies and guidelines concerning their effectiveness — is staggering. Couple this explosion with the lack of effective means for physicians and health care providers to make sense of the information, and it’s not surprising that we have a bloated, inefficient, and costly system that fails to provide value commensurate with our health care budget.

This systemic problem is no secret. The HITECH Act attempts to target the health information technology (HIT) problem with an incentive program, and PPACA attempts to increase evidence-based medicine (EBM) with projects like the Patient-Centered Outcomes Research Institute. But a piecemeal approach does not ensure the necessary integration between HIT and EBM, nor sufficient incentives for industry to embrace them.

Why such faith in the MSSP?

Because if ACOs want to participate in the shared savings they must meet the dual requirements of EBM and HIT. It’s this double-envelopment — combined with the ‘carrot’ of shared savings — that will finally usher in a medical revolution.

Thomas Kuhn, a trained physicist who is better known for his contributions to the philosophy of science , introduced the idea of “paradigm shifts” that occur as science evolves. In “The Structure of Scientific Revolutions,” Kuhn posits that instead of a linear evolution of scientific discovery, the discovery of anomalies can force traditional explanations of natural phenomena to be questioned. If enough anomalies accrue that seriously undermine an accepted explanation, a “crisis moment” occurs. In this circumstance “a scientist’s world is qualitatively transformed [and] quantitatively enriched by fundamental novelties of either fact or theory and a scientific revolution is born.” But as Kuhn notes — with import to our discussion of the MSSP — prior beliefs and experiences can make accepting a new paradigm difficult for scientists.

Kuhn’s theory of the evolution of science helps to explain health reform, or the lack thereof. Our health care paradigm — the spending of significant resources on health care per capita — has accrued significant anomalies, most notably outcomes that do not match up with our spending. We have tried HMOs, PPOs, and every many other types of arrangements, but to no avail. We are in a “crisis moment.” And we have a new paradigm: health care decision making that utilizes EBM at the point of care.

And that, my friends, is where the savings will ultimately be found.

A 2004 study demonstrated that following evidence-based guidelines for the treatment of hypertension in the elderly would save $1.2 billion annually.

There is no shortage of similar studies showing billions of dollars, and better health outcomes, waiting to be unlocked. So why isn’t it occurring?

A new review by Stanford University’s Adam Elshaug, M.P.H., Ph.D., and Alan Garber, M.D., Ph.D. demonstrates that recent studies on complex vertebral spinal procedures point have “cast doubt on the magnitude of any benefits from these procedures and at worst established their ineffectiveness.” The studies have caused payers like Blue Cross to limit or withdraw coverage of the procedure. After analyzing the data, the authors found that a conservative estimate of the savings of scaling down the costly and ineffective procedure would yield between $450 million and $725 million depending on the continued use of the procedures.

But the authors make a crucial point at the end of their piece:

Of course, savings will be derived from [comparative effectiveness research] CER only if practice changes. In the United States, it’s unclear whether these studies are powerful enough to overturn coverage decisions or cut utilization of established procedures. . . ACA features such as bundled payments, shared savings programs, and outcomes-based payments offer mechanisms for stimulating the adoption of practices that are supported by CER and the abandonment of practices that CER calls into question.

I interpret this as an acknowledgment that we have enough data to start saving money and increasing care, but that we are stuck in a rut where the practice of medicine itself is having troubling embracing science, and we are relying on the payers to pick up the slack.

This is not to say that medicine can ever — or should ever — become entirely science-based. There are embedded values in the process of health care decision making that science cannot determine, such as a patient’s desire for aggressive treatment and the risks or costs they are willing to incur. Regardless, there is a baseline degree of science-based medicine that will improve quality, afford greater patient (and physician) autonomy, and decrease cost. Moreover,  studies have shown that better informed patients make more cost-effective choices.

The problem is our inability and/or our unwillingness to embrace the inevitable paradigm shift to a greater science-based medicine even during a crisis moment. That is where the MSSP double-envelopment strategy comes in.

CMS’s Double-Envelopment Strategy: Attract with the Savings, Surround with EBM and HIT

The MSSP allows an ACO, each year, to recoup some of the savings that they have realized in reference to a benchmark cost. There is a fairly complicated procedure for determining the actual savings that the ACO can collect, but the idea is simple: incentivize the health care providers to reduce the cost of care. Health care organizations are racing to form ACOs, but while doing so they are being surrounded by EBM and HIT requirements that will drive a shift in health care delivery.

With respect to EBM, the proposed rule requires ACOs to implement evidence-based medicine or clinical practice guidelines and processes in an effort to improve individual care, improve the health of the population, and lower the growth of health care expenditures. The guidelines and processes must cover diagnoses with “significant potential” for the ACO to achieve quality and cost improvements, taking into account the circumstances of individual beneficiaries. All ACO participants and suppliers/providers must agree to abide by these guidelines and processes, and must be evaluated for their compliance. The rule also states that remedial actions must be a possibility for non-compliance, and ACOs must have policies and procedures for ACO expulsion of participants and/or providers/suppliers.

On the HIT side, ACOs are required to have an infrastructure, such as information technology (which may include EHR technology that is certified for CMS’s incentive-based meaningful use program). This infrastructure must enable the ACO to collect and evaluate data and provide feedback to ACO participants and ACO providers/suppliers across the entire ACO, including providing information to influence decision making at the point of care. Moreover, fifty percent of the primary care providers of an ACO must be “meaningful users” as defined by the HITECH Act by the second year of their ACO contract. As others have noted, the meaningful use requirement is extremely aggressive when considering that the proposed rule allows ACOs to come online as soon as Jan 1st, 2012. Industry has seen the writing on the wall, and has responded with nothing short of an ACO arms race.

The ACO-driven Paradigm Shift

Thus, the proposed rule requires ACOs to leverage HIT to evaluate data and provide feedback to others in the ACO, and do it in such a way that the feedback influences decision making at the point of care.  In other words, it is setting the stage for informed decision making for both physician and patient alike. This is the holy grail of health care reform: that is, an HIT network with users that are reporting data that can be leveraged to enable providers to suggest treatments that are proven to have better outcomes for their specific patient, and to do so at the point of care.

This is in contrast to the current paradigm of managing costs by relying primarily on ex post decision making at the payer level. Often, however, the consumer who has their desired procedure or drug denied (for reasons often opaque to either the physician, insurer, or patient) will decide to pay out of pocket, and can go bankrupt in the process. In this case, no costs have been reduced, rather, they have been shifted to the consumer. While some insurers create and use HIT and EBM, their behind-the-scenes decision making has not been embraced by physicians or patients. That’s because patients trust their physicians, not their insurers. The locus of reform must be on the decision making at the physician-patient level, and that is precisely where the proposed rule places it.

The proposed rule also clearly addresses the fact that you can’t get new practices adopted if physicians have to, for example, minimize their EHR application, fire up their web browser, and start searching the Cochrane Collaboration or some other site for possibly relevant data. They are going to have to do it from within the HIT system.

There is an added benefit politically to this paradigm shift: if the focus is on data-driven doctor-patient decision making, we bypass the political push and pull often associated with determining what treatment is “medically necessary.” This would satisfy the progressive ideal of providing high quality care without overbearing cost-control, while also satisfying the conservative refrain that the doctor-patient relationship remains independent. If the process of creating EBM decision making is HIT-focused, it also encourages the antithesis of cookbook medicine by tailoring the process to the individual patient.

The Long View

Too much focus has been placed on the short term issue of how much money the ACOs can recoup. This is a valid worry for the industry, particularly the smaller practices that can’t afford setting up an ACO. The federal government must do whatever it can to allay these worries so that industry further strives to create the HIT-EBM framework that the shared savings program envisions. If it means increasing the percentage of savings that the ACOs can receive, then so be it. Or perhaps ACOs should come online a year later after the meaningful use stage of EHRs has progressed.

Regardless of how the final rule mitigates industry difficulties, the ACO model is our best chance at creating a true paradigm shift that will better provide the medically necessary and efficient delivery of health care resources. It may take 5, 10, or 20 years to robustly develop the systems and the data, but nobody said a medical revolution would be easy.

The Normative Meets the Practical: Who Should Can Lead ACOs

One of the many $64,000 questions in the accountable care organization (ACO) debate has been who should lead these organizations.  In a policy adopted in November 2010, the American Medical Association (AMA) made clear its view that ACOs must be physician-led.  The American Hospital Association (AHA) refrained (at least in its public letter to CMS) from asserting its entitlement to the ACO helm, based, for example, on its management experience and pools of capital.  Instead, it simply urged CMS to “defer details of the organization, such as leadership and management structure, to each ACO.”

CMS seems to have heeded the AHA’s advice because its recently released proposed rule does not directly take on this normative debate.  (See Summary of CMS Proposed Rule on Accountable Care Organizations recently posted by Jordan T. Cohen for an overview of the proposed rule.)  While “ACO participants must have at least 75 percent control of the ACO’s governing body” to be eligible for participation in the Shared Savings Program (proposed Section 425.5(d)(8)), the definition of “ACO participant” in the proposed rule includes physicians and hospitals, among others (proposed Section 425.4).

Similarly, the proposed rule simply requires that the “ACO’s operations must be managed by an executive, officer, manager, or general partner whose appointment and removal are under the control of the organization’s governing body and whose leadership team has demonstrated the ability to influence or direct clinical practice to improve efficiency processes and outcomes” (proposed Section 425.5(9)(ii)). The proposed rule does not address who or what would make the best such leader.

The proposed rule, however, clearly preserves a role for physicians to form and lead ACOs. For example, it recognizes that ACOs may be comprised of professionals in group practice arrangements and networks of individual practices, independent of hospitals (proposed Section 425.5(b)).

In addition, “[c]linical management and oversight [of the ACO] must be managed by a full-time senior-level medical director . . . who is a board-certified physician . . .,” and “[a] physician-directed quality assurance and process improvement committee must oversee an ongoing action-oriented quality assurance and improvement program” (proposed Sections 425.5(9)(iii) and (iv)).

The proposed rule also builds in a preference for ACOs comprised of all physicians or physician groups with fewer than 10,000 assigned beneficiaries by proposing to exempt them from the 2 percent net savings threshold adjustment under the one-sided model (proposed Section 425.9(c)(4)(i)).  It also proposes to vary confidence intervals, which affect the minimum savings rate, by the size of the ACO in the one-sided model “to improve the opportunity for groups of solo and small practices to participate in the Shared Savings Program” (Preamble to proposed rule at Section II.F.10).

But on a practical level, the specifics of CMS’ proposal may — unintentionally, perhaps — give hospitals the greater chance to take the reins, at least initially.  An apparently leaked CMS internal discussion document reflects some level of concern that physicians may have a hard time taking the lead with ACOs.

The proposed rule’s regulatory impact analysis estimates that the average start-up investment and first year operating expenditures for an ACO in the Shared Savings Program will be $1,755,251.  In addition, the proposed rule uses a 6-months claims run-out (proposed Section 425.7(a)).  Presumably, that means ACOs — assuming they satisfy all program requirements — will not see a dime of shared savings for more than eighteen months.  CMS also proposes to withhold 25 percent of any earned shared savings accrued in a given year to ensure repayment of any losses to the Medicare program in subsequent years of the three-year ACO agreement (proposed Section 425.5(d)(6)(iii)).

Even if private physicians can amass the capital to make these upfront investments, there of course is no guarantee they will regain their outlays. A recent study published online by the New England Journal of Medicine, as reported by the American Medical Association, found that participants in CMS’ Physician Group Practice Demonstration did not recoup, at least in the initial years of the demonstration, all of the money they invested to establish ACOs.  As the AMA summarized:

Early adopters, for the most part, did not recoup their set-up costs in the first three years of operation.  The 10 integrated health systems that were studied spent an average of $1.7 million to take part in the demonstration project.  Eight received no shared savings payments in the first year of the project.  Six got a payment in the second year, and five received a bonus in the third year.

The Everett Clinic in Washington, for example, reportedly spent approximately $1 million on infrastructure for its ACO but recouped only $129,268 in shared savings during the first four years of the demonstration project.

According to a 2007 report from the National Center for Health Statistics (NCHS), in 2003-04, 80.6 percent of office-based medical practices in the United States consisted of one or two practitioners and 94.8 percent had five or fewer practitioners.  The risks associated with forming an ACO are considerable for these smaller practices to absorb, especially when, at best, the ACO will see 75 percent of its portion of any shared savings upwards of eighteen months down the road and could instead be responsible for its share of losses.  It is not clear how many small practices are willing and able to assume these risks without some substantial financial or management support.  Not surprisingly, the AMA’s statement on the proposed ACO rule specifically identifies “the large capital requirements to fund an ACO” as a significant barrier that must be addressed if physicians in all practice sizes and settings will be able to successfully lead and participate in ACOs.

Another aspect of the proposed rule that may present a particular challenge to independent physicians is proposed Section 425.11(b)’s requirement that “[a]t least 50 percent of an ACO’s primary care physicians must be meaningful [Electronic Health Records (EHR)] users, using certified EHR technology as defined in §495.4, in the [Health Information Technology for Economic and Clinical Health (HITECH)] Act and subsequent Medicare regulations by the start of the second performance year in order to continue participating in the Shared Savings Program.”

Physician practices indisputably have increased their use of EHR systems in recent years.  According to the National Ambulatory Medical Care Survey conducted by NCHS (reported here), only 17 percent of physicians in 2008 reported that they had a “basic” EHR system (which is defined as having electronic patient demographic information, patient problem lists, patient medication lists, clinical notes, orders for prescriptions, and laboratory and imaging results).  Recent NCHS data (reported here) show that that number has climbed nearly 50 percent to 24.9 percent of office-based physicians.

But basic use of EHRs is not sufficient under the proposed rule, which requires “meaningful use.”  Survey data from the Office of the National Coordinator for Health Information Technology, as reported here, show that only 41.1 percent of office-based physicians plan to apply for billions of federal dollars in EHR incentive payments that are available to Medicare and Medicaid providers under the HITECH Act, compared with 80.8 percent of acute care non-federal hospitals.  Additionally, as reported here, a recent survey from the Medical Group Management Association (MGMA) found that only 13.6 percent of medical practices that have adopted EHRs and plan to apply for the EHR Meaningful Use incentives currently are able to satisfy the fifteen core criteria necessary to establish that they are meaningful users.   Medical practices have a long row to hoe.

But the news is not all bad for physicians.  The MGMA survey also found something that suggests this issue is far from resolved on a theoretical or practical level.  As reported here, “almost 20 percent of responding independent medical practices that owned EHRs said that they had optimized their uses of EHRs” whereas “[o]nly 8.8 percent of responding hospitals — or [integrated delivery system (IDS)] — owned practices with EHRs said they had optimized their EHR use.”

Almost certainly, it is not just a coincidence that physicians are devoting their energy to becoming meaningful EHR users just as the first EHR Meaningful Use incentive payments are available. If CMS or private foundations develop additional incentive programs to help smaller practices cover the start-up costs associated with forming an ACO, the individual physician could still be in this game.  Notably, the AMA’s brief statement on the proposed ACO rule reiterates its recommendation to CMS to increase access to loans and grants for small practices as part of this puzzle.  It remains to be seen if any such programs are viable in this fiscal climate.

As promised, future posts will address the normative question of who should lead ACOs.

ACOs: OIG Guidance, CMS regulations, and Interpretive Tasks

Tim Greaney has already posted on the FTC/DOJ Joint Policy Statement on antitrust scrutiny of ACO applicants, and Jordan Cohen posted on the  CMS’s draft regulations on the Medicare Shared Savings Program (”MSSP”) for ACOs.   In this post, I’ll describe some interesting structural issues presented by the OIG’s notice on proposed waivers, which cross-references the CMS MSSP draft regulations. The fundamental issue for the OIG is how “virtual organizations” — those not fully integrated in a corporate or financial sense — can serve the integrative goals of the Affordable Care Act while staying on the right side of the web of federal laws prohibiting physician self-referral, kickbacks, and payments to reduce care to Medicare beneficiaries.  The OIG notice proposes to square this circle in two steps.

The OIG two-step; Step One

The Fraud & Abuse problem with ACOs is that they are intended to achieve the efficiency and quality gains of formally integrated delivery systems (Geisinger, Mayo Clinic)  through contract-based aggregation of providers.  This less formal integration, of course, implicates the dealings generally prohibited by the Physician Self-Referral Law (”Stark”), Anti Kickback Statute (”AKS”), and the Prohibition on Hospital Payments to Physicians to Induce Reduction in Services law (which provides for civil money penalties for violations) (”CMP”).  The OIG notice proceeds very incrementally.  In large part, the OIG proposes to affirm that a CMS determination that an ACO qualifies for participation in the MSSP signifies the adoption of quality and fiscal protections sufficient to allow a waiver of Stark, AKS, and CMP enforcement for purposes of the distribution of shared savings.  In addition, and to the extent “necessary for and directly related to” the ACOS’s participation in the shared savings plan, the OIG proposes to waive enforcement under the AKS and CMP provisions with respect to conduct that falls within a Stark exception.  In outline form, then, the OIG proposes as its initial waiver guidance:

• Stark. The OIG proposes to waive enforcement of Stark for
  • The actual distribution of MSSP to and among ACO participants, and ACO providers/suppliers for conduct during a year in which shared savings were earned; and
  • “activities necessary for and directly related to the ACO’s participation in and operations under the” MSSP.

• AKS. The OIG proposes to waive enforcement of AKS under two scenarios:
  • The distribution of shared savings under the MSSP
    • to and among ACO participants, and ACO providers/suppliers for conduct during a year in which shared savings were earned; and
    • for any financial relationships between ACO participants and ACO providers/suppliers necessary for and directly related to the ACO’s MSSP participation that implicate Stark, but fall within a Stark exception.

• CMP. The OIG proposes to waive enforcement of the CMP provisions under two scenarios:
  • Distribution of shared savings from a hospital to a physician so long as
    • “the payments are not made knowingly to induce the physician to reduce or limit medically necessary items or services”; and
    • the hospital and physicians were ACO participants or providers/suppliers during the year in which the shared savings were earned.
  • In the context of financial relationships among ACO participants and/or providers/suppliers necessary for and directly related to the ACO’s MSSP operations, that implicate Stark but that fall within a Stark exception.

The OIG two-step; Step Two

In Step One, the OIG proposes waivers for arrangements that are central to the MSSP gain distributions, or that are central to the ACO enterprise and are structurally within Stark exceptions.  The OIG goes on to solicit input on the need for additional waiver guidance, for conduct that is “beneficial” to ACO participation in the MSSP, but that also protects “patients and programs from harms caused by fraud and abuse.”  The OIG solicits input on:

• Arrangements related to establishing the ACO;
• Arrangements between or among ACO participants or providers/suppliers related to ongoing operations of the ACO and achieving ACO goals;
• Arrangements between the ACO, its ACO participants and/or providers/suppliers and outside individuals or entities;
• Distributions of shared savings or similar payments from private payers;
• Other financial arrangements for which a waiver would be necessary;
• Miscellaneous: duration of waivers, scope of the waivers listed above in “Step One,” and additional safeguards; and
• Arrangements in which providers are subjected to risk, particularly in the “two-sided risk model” in the CMS draft regulations on the MSSP.

“Necessary for and directly related to”

It is pretty clear that the lion’s share of the OIG’s waiver work will be done by determining whether or not an ACO has been qualified by CMS for the MSSP.  There will be waivers, however, even under Step One, that will require the OIG to evaluate the proposed arrangements.  Stark enforcement regarding the actual distribution of shared gains, for example, will be waived under the proposal for CMS-qualified ACOs.  But Step One proposes additional waivers “for activities necessary for and directly related to the ACO’s participation in and operations under the” MSSP.  Similarly, AKS and CMP waivers are proposed for dealings between ACO participants that are within Stark exceptions so long as they are “necessary for and directly related to the ACO’s MSSP participation.”  A standard based on what is “necessary for and directly related to” an ACO’s MSSP participation will, then, do a lot of the waiver work.  It is not a self-defining standard, and further elaboration from the OIG (by providing examples, perhaps) will lend clarity to the OIG’s waiver guidance.

Attribution, assignment, and patient notice

ACO commentators (here (subscription required) and here)  have observed that the method by which patients are “attributed” to ACOs is central to ACOs’ financial and structural planning.  Attribution was the term used in the literature to refer to the formal determination that a particular Medicare patient should “count” for assessing the gain (and possibly the loss) experienced by the ACO.  The CMS draft regulations on the MSSP, consistent with the terms of the Affordable Care Act, use the term “assignment” rather than attribution.  The CMS proposal’s treatment of assignment, or attribution, will engender much discussion here and elsewhere in the coming weeks.

One important issue, however, seems to have been resolved.  In the literature on ACOs, it has not been clear whether Medicare beneficiaries would know whether or not they had been assigned or attributed to an ACO; consistent with continuing commitment to beneficiaries’ right to choose providers within the Medicare fee for service system, the discussion contemplated the possibility that ACOs would organize a beneficiary’s care, gain payments as a result of efficiencies, but never inform the beneficiary of his “attribution” to an ACO.

Donald Berwick highlighted the requirement of patient notice in his Perspective published in the New England Journal of Medicine on the day of the regulations’ release.  The CMS draft regulations on the MSSP explain the requirement of patient notice in the following terms:

[W]hile the statute refers to the assignment of beneficiaries to an ACO, we would characterize the process more as an “alignment” of beneficiaries with an ACO as the exercise of free choice by beneficiaries in the physicians and other health care providers and suppliers from whom they receive their services. . . .  Therefore, an important component of the Shared Savings Program will be timely and effective communication with beneficiaries concerning the Shared Savings Program, their possible assignment to an ACO, and their retention of freedom of choice under the Medicare FFS program.

That piece of consumer protection regulation was absolutely essential.  It would be odd indeed, after decades of struggle with patient protection in managed care systems, were patients to be engaged without their knowledge in a system built on economic incentives to providers directed to care management.  We’ll post more about assignment and other ACO issues in the coming weeks.

Summary of CMS Proposed Rule on Accountable Care Organizations

CMS recently released the proposed rule that will regulate PPACA’s Medicare Shared Savings Program (MSSP). The MSSP relies on the accountable care organization (ACO) model in order to generate and distribute savings. HealthReformWatch.com has discussed the general framework for ACOs before. Clocking in at nearly 500 hundred pages, the proposed rule helps to flesh out what was largely a philosophical exercise in cooperative health care delivery. Below are what I believe to be a number of key pieces of the proposed rule.

Proposed Rule Highlights

The 2 ACO Models – (425.7)

There will be two ACO models. The choice between models appears to be largely geared towards minimizing ACO risk while hospitals and providers are first bringing their ACOs online.

• One-Sided Model: A one-sided ACO shares in the savings, but is not on the hook to share in any of the losses (i.e., costs surpassing the ACO’s benchmark as determined by CMS, see below).
• Two-Sided Model: A two-sided ACO shares in both the savings as well as the losses.

Basic Time frame and Structure

Not surprisingly, ACO hopefuls must form an agreement with CMS directly. ACOs under the MSSP must last for not less than three years after the application has been approved. (425.18).  The performance period will be 12 months. The ACO must have at least 5,000 beneficiaries, and must include a sufficient number of primary care physicians to treat the ACO beneficiary population.

• First 3 years of ACO life: Choose a Track – 425.5(d)(6).
  • Track 1:  ACO operates under a one-sided model for two years, and under a two sided model for the third year. With the exception of quality performance, the third year of this track will be measured using the methodologies that measure the first year of the Track 2 ACOs.
  • Track 2: ACO operates under the two-sided model, sharing both savings and losses with the Medicare program for three years.
• After 3 years
  • ACOs operate under the 2-sided model, thus sharing both gains and losses with Medicare.

Regulating Risk and Payment — 425.5(d)(6)(b)(4)

ACOs must obtain reinsurance, place funds in escrow, obtain surety bonds, establish a line of credit that Medicare can draw upon, or establish other repayment mechanisms that will provide for payment of losses to Medicare under the 2-sided model.

Legal Structure — 425.5(d)(7)

ACO must be constituted as a legal entity for the purposes of, among other things, receiving and distributing shared savings, repaying shared losses, and establishing reporting.

Governance — 425.5(d)(8)

The ACO must establish and maintain a governing body to fulfill and execute ACO functions. It must be comprised of ACO participants or their representatives, as well as representatives of the Medicare beneficiaries in the ACO. At least 75 percent of the governing body must consist of ACO participants. ACO participants and ACO providers/suppliers must have a meaningful commitment to the ACO’s clinical integration, which may consist of a financial investment or a meaningful human investment in the ongoing operations of the ACO, such that potential loss or recoupment is likely to motivate that participant.

Overseeing Quality and Performance — Accountability Internally Enforced by Physician-directed Committee — 425.5(d)(9)(v)

ACOs will be required to have a physician-directed committee tasked with overseeing a quality assurance and improvement program. This program must establish internal performance standards for quality of care, cost-effectiveness, and process and outcome improvements. The committee must hold the ACO providers/suppliers accountable for meeting these standards. The program must have processes and procedures to identify and correct poor compliance.

Evidence-Based Medicine — 425.5(d)(9)(viii)

The ACOs are required to implement evidence-based medicine or clinical practice guidelines and processes in an effort to improve individual care, better the health of the population, and lower the growth of health care expenditures. The guidelines and processes must cover diagnoses with “significant potential” for the ACO to achieve quality and cost improvements, taking into account the circumstances of individual beneficiaries. All ACO participants and suppliers/providers must agree to abide by these guidelines and processes, and must be evaluated for their compliance. Remedial actions must be a possibility for non-compliance, and ACOs must have policies and procedure for ACO expulsion of participants and/or providers/suppliers.

Health Information Technology — 425.5(d)(9)(viii) & 425.11

ACOs are required to have an infrastructure, such as information technology (which may include EHR technology that is certified for the meaningful use program) that enables the ACO to collect and evaluate data and provide feedback to ACO participants and ACO providers/suppliers across the entire ACO, including providing information to influence care at the point of care.

By the second year, at least 50 percent of an ACO’s primary care physicians must be meaningful users of EHR technology. Failure to fulfill this obligation could lead to ACO termination.

Assigning Beneficiaries to ACOs — 425.6

The general approach of the CMS is to assign beneficiaries to ACOs based on the utilization of primary care. 425.6(a). Beneficiaries are assigned based on their utilization of primary care services by a primary care physician who is an ACO provider/supplier during the performance year for which shared savings are to be determined. Assignment to an ACO in no way diminishes or restricts the right of the beneficiaries assigned to an ACO to exercise free choice in determining where to receive health benefits.

More specifically, beneficiaries will be assigned to the ACO where they receive a plurality of their primary care services. 425.6(b). CMS will establish a fixed benchmark which will be adjusted for overall growth and beneficiary characteristics, including health status. This benchmark will be updated annually based on the absolute growth in national per capita expenditures for Medicare Part A and Part B services under the original Medicare fee-for-service program.

Payment and Treatment of Savings — 425.7

• Shared Savings under One-Sided Model — Each year, CMS will determine whether the estimated per capita Medicare beneficiary expenditures under the ACO are below the benchmark for Medicare fee-for-service. To qualify for savings, an ACO in this model must have costs below the benchmark by more than a “minimum savings rate,” as determined by CMS.  ACOs in the one-sided model that exceed this minimum savings rate (as determined by CMS calculations) are eligible to  share savings net 2 percent of its benchmark.  One sided ACOs can share in a maximum of 50 percent of the savings, with an additional 2.5 percent being allowed for rural hospitals or federally qualified health centers–in certain circumstances. Payment is capped at 7.5 percent of the ACOs benchmark. However, ACOs will not be able to blindly slash costs in an effort to obtain savings. Rather, eligibility of the shared savings will be contingent on the reaching of certain minimum quality performance measures. These measures will focus on five areas, including patient/care giver experience, care coordination, patient safety, preventative health, and at-risk population/frail elderly health. In addition to determining general eligibility for savings, the quality performance measures will also determine the actual percentage of savings that the ACO is eligible to take home.
• Shared Savings under Two-Sided Model – In the two-sided model, CMS will also determine a benchmark of Medicare Part A and Part B costs. This benchmark will determine whether the ACO is eligible for savings payments or — as is unique to the two-sided model — whether they are liable for losses. To trigger savings or losses, the ACO must be below or above their benchmark by more than a minimum savings rate, or alternatively, a minimum loss rate when considering losses. Unlike one-sided ACOs whose minimum savings rates are calculated based on the beneficiary population, the minimum savings rates for two-sided ACOs are capped at 2 percent below the benchmark rate. Likewise, to be subject to loss, the ACO’s expenditures must be above 2 percent of its benchmark. Two-sided models also use the quality performance measures to determine eligibility as well as the rate of shared savings. Two-sided ACOs can share in a maximum of 60 percent of the savings, with an additional 2.5 percent being allowed for rural hospitals or federally qualified health centers–in certain circumstances. In addition, whereas the shared savings of the one-sided model caps out at 7.5 percent of the ACO’s benchmark, the two-sided model caps out at 10 percent.

Preventing Cherry Picking — 425.12(b)

CMS will use the methodologies it applies to analyzing ACO performance in an effort to prevent ACOs from “cherry picking” the healthiest individuals for their ACOs. CMS reserves the right to terminate an ACO for avoiding at-risk beneficiaries. Other less drastic options are also at the option of CMS.

Ensuring quality performance — 425.12(c)

CMS will be monitoring ACO compliance by analyzing the data provided by the ACO to determine its eligibility for, and its percentage of, shared savings. If the ACO fails to meet CMS performance standards it will be given a warning. An ACO given a warning will be reevaluated the following year. If the ACO is still failing to meet the performance measures CMS may terminate the ACO immediately or take alternative actions as specified in the rule. If the ACO does not submit the requested quality performance data, CMS will request submission of the data, allow for a correction of the data, or allow for a written explanation of why the data was not provided. ACOs that continue to fail in providing the requested data will be terminated immediately.

Beneficiary Data Sharing – 425.19(d)

CMS will provide ACOs with monthly claims data for potentially assigned beneficiaries. CMS makes clear that HIPAA protections will apply to this data sharing. More notably, ACOs must provide the beneficiary with the opportunity to opt-out of sharing his or her personal health information for the purposes of ACO activities.

Public Reporting and Transparency – 425.23

ACOs will be required to publicly report a variety of information, including quality performance standard scores, shared savings or losses information, the total amount of shared savings distributed among ACO participants, and the total proportion that was used to support quality performance.

Resources:

1. CMS [Proposed Rule]: Medicare Program: Medicare Shared Savings Program:  Accountable Care Organizations on ACOs. For those wanting the rule without the preamble, I have uploaded it here. It is only 59 pages.

2. CMS and HHS Office of the Inspector General [Notice with Comment]: Medicare Program; Waiver Designs in Connection with the Medicare Shared Savings Program and the Innovation Center.

3. DOJ and FTC [Proposed Statement]: Proposed Statement of Antitrust Enforcement Policy Regarding Accountable Care Organizations Participating in the Medicare Shared Savings Program.

4. IRS [Notice]: Notice 2011-20: solicitation for comments regarding what guidance, if any, is needed for tax-exempt organizations participating in Medicare Shared Savings Programs through an ACO.

The Pediatric Accountable Care Organization: Incentivizing a Family-Centered Approach

Late last year, the American Academy of Pediatrics (AAP) issued a report recommending that pediatricians screen new mothers for postpartum depression during the 1-, 2-, 4-, and 6-month well baby visits.  Most pediatricians believe that screening for postpartum depression is within the scope of pediatric practice, because “[a]ddressing maternal depression in a timely and proactive fashion is essential to ensure healthy early brain and child development and readiness to succeed.”

There has been some debate about whether such screening is reimbursable.  The AAP believes that it is (or should be) and recommends that pediatricians seek reimbursement using “[t]he Current Procedural Terminology (CPT) code 99420 [for "Administration and interpretation of a health risk assessment instrument"] … recognizing the [Edinburgh Postpartum Depression Scale (a 10-question screen completed by the mother)] as a measure for risk in the infant’s environment, to be appropriately billed at the infant’s visit.”

Because it would be integrated into a reimbursable well visit, and because of the limited amount of time it takes, reimbursement concerns may not be an insurmountable barrier to postpartum depression screening.  (Even more streamlined than the Edinburgh Scale is a two-question screening tool that, while not specifically designed for use with new mothers, has proved effective at identifying postpartum depression.  The two questions are as follows: “Over the past 2 weeks: 1. Have you ever felt down, depressed, or hopeless? 2. Have you felt little interest or pleasure in doing things?”  Answering yes to one or both of these questions is a positive screening result.)

When the screening identifies a potential problem, though, reimbursement concerns likely are a formidable barrier to treatment delivery.  The AAP acknowledges that pediatricians are not qualified to treat postpartum depression (and many mothers might not want them to in any event) and recommends that pediatricians refer new mothers with positive screening results to qualified providers and other resources in their community.  This recommendation is a good one, as far as it goes, and probably all that can be expected in our current fragmented healthcare system, but it is a long way from ideal, whether you put yourself in the shoes of the provider– or the family.

Might the much-talked-about accountable care organization be a means to the end of fairly compensating pediatricians and their staff for the hard work entailed in quickly and seamlessly moving mothers into care?  Just as Section 3022 of the Patient Protection and Affordable Care Act allows adult medical providers to form ACOs for the purpose of receiving incentive payments tied to savings to Medicare, under Section 2706 states can allow pediatric medical providers to form ACOs to receive payments tied to savings to Medicaid.  While the AAP claims that “[g]reater savings are found in managing care for adults and not children[,]“ Nationwide Children’s Hospital in Columbus, Ohio, the “largest pediatric accountable care organization in America,” disagrees, claiming that “[p]ediatrics offers perhaps the biggest opportunity to bend the long-term cost curve in health care.  While the savings may be less immediate, there is evidence that many of the pervasive, and costly, chronic diseases of adulthood can be successfully prevented in childhood, for example obesity.”

Providers establishing pediatric ACOs will need to confront a number of pediatric-specific structural issues.  As Mark Waxman and Larry Vernaglia point out in a recent Health Law Reporter article, while “children grow out of childhood and, therefore, potentially the [pediatric ACO] structure[,]” many conditions are “best treated by the same team over the life of the individual.”  In addition, “pediatric populations do not exist in vacuums; generally, they live in families and these families also receive medical care.  A [pediatric ACO] may need to coordinate closely with an ACO that provides care to the entire family.  … For example, if wellness programs were targeted at the whole family (e.g. nutrition, exercise, environmental, etc.) they likely would be more effective than if they address the child in isolation.”  Postpartum depression is just one of many conditions that call for care coordination not just on the patient’s behalf but also on the patient’s family’s behalf– and pediatric ACOs should be structured with this in mind.

Jonathan Blum, CMS Deputy Adminstrator, Speaks on ACOs

We’re waiting for the Department of Health and Human Services to release proposed regulations on Accountable Care Organizations.  This site has previously discussed the potential good and bad of ACOS (see here, here, and here).   I attended a conference last week at which an innovative model of “Medicaid ACO” was discussed.  The Medicaid ACO would be authorized in New Jersey under a bill pending before the NJ Legislature.  It is an exciting idea that will attempt to reach the poor and vulnerable who often lose out in health reform programs.  The godfather of the Medicaid ACO project is Jeff Brenner, about whom Atu Gawande recently wrote in the New Yorker.   (subscription required)    I’ll be blogging about the New Jersey bill in a future post.  The conference was funded by the Nicholson Foundation and presented  by the Health Care Quality Institute.

Speaking at the conference was Jonathan Blum, Deputy Administrator and Director of the Center for Medicare at CMS.  In discussing “Accountable Care Organizations and the Affordable Care Act,” Blum was in the difficult position of speaking about a topic of great interest, while not being able to discuss the contents of draft regulations that are no doubt nearing completion. Nevertheless, he made some interesting points that I’ll pass along.

Blum’s talk focused on policy positions that are driving HHS as it drafts the regulations.  The overriding policy positions he described included:

• The ACO regulations will not be “one size fits all.” He emphasized that CMS will be looking for innovative models, with different payment systems, and with different “on ramps” to formation and approval. He emphasized that CMS is interested in models that serve “safety net populations,” as CMS wants to ensure that the poor and underserved get the same opportunities as “suburban” folks. The primacy (at least in order of presentation) was welcomed by the NJ folks, whose model is directed to Medicaid recipients.
• The orientation, consistent with much of the ACO literature, is “patient first.” He distinguished this orientation from one that would see ACOs as a means for powerful interests to gain market share. That tension is, of course, evident in the ACA’s ACO provisions, as has been pointed out most eloquently by Tim Greaney. Blum described CMS as being focused on care systems’ sensitivity to patient and family concerns, and with payment programs oriented to health care “journeys” and not episodes.
• Clinical quality is key. CMS will focus on outcomes measurements “much more” than in the past. It will be interested in particular in quality measurements and patient experience.
• He spent a fair amount of time emphasizing that CMS does not regard the ACO program as static. CMS will constantly review payment and quality issues, with an eye toward updating oversight and program requirements. It will use payment incentives to drive quality improvements. He indicated that there is some tension between CMS’s interest in having quality be data-driven in ACOs with its insistence on protecting patient confidentiality and privacy issues. CMS is interested in encouraging patient advocacy efforts to support continued emphasis on patient privacy and confidentiality.
• In response to a question, Blum recognized the substantial tension between the ACO model’s emphasis on improving quality and reducing cost through organization of care on one hand, and the ACA’s continued embrace of patient choice of provider on the other. He indicated that this tension might best be addressed by ACOs and their constituent providers creating a sufficiently attractive delivery model that patients will want to be involved — exclusively. (Reaching this goal would clearly require unprecedented patient education efforts.)

The Q & A following Blum’s presentation was predictably frustrating on both sides, as could be anticipated in connection with a talk about not-yet-finalized regulations.  He recognized several outstanding issues that he was not at liberty to discuss, but which had been occupying those drafting the regulations, including:

• Will physicians be able to join more than one ACO? CMS is apparently considering different rules for primary care physicians and specialists, although Blum acknowledged that such overlapping provider networks will make the computation of “gain” difficult when gainsharing is implemented.
• Blum, although asked, would not bite on the question of “who will lead” (physicians or hospitals). He anticipates a variety of models, but stressed that no ACO would flourish without physician buy-in.
• The question of geographic exclusivity for ACOs engendered a similarly noncommittal response. Blum acknowledged the conceptual difficulties presented by such overlap, but also pointed to the negative implications of exclusivity on the robustness of competition.

So, it was an interesting discussion of general principles, whetting our appetites to see how HHS will “square the circle” — or circles — in the upcoming regulations.

Reform Rodeo

1. Maggie Mahar at Health Beat discusses a new report published in the NEJM that supports the importance of the individual mandate in combating adverse selection.

2. At the Health Care Blog, Paul Levy writes sardonically about the accountability of accountable care organizations.

2a. On a less sardonic note, Chris Fleming gives an overview of Health Affairs’ special issue covering ACOs.

2b. Thomas Greaney writes in the NEJM about how the federal government’s can help ACOs navigate an already concentrated health care landscape.

3. The Hill reports on the essential items and services that health insurers will have to provide when offering their products in the new exchanges.

4. David Kibbe and Brian Klepper document the federal government’s initiatives in giving the HIT market a much needed shot-in-the-arm.

5. The Commonwealth Fund’s Melinda Abram’s discusses one of the most important facets of health care reform: how the ACA will bolster primary care.

Centripetal Accountability in Health Care: Accountable Care Organizations, A Legal and Practical Overview

Legal scholar Kevin Werbach once observed that the internet has been centripetal, “pull[ing] itself together as a coherent whole.” For Werbach, network formation theory both explains these centripetal tendencies, and some of “the pressures threatening to pull the Internet apart” into balkanized units. Werbach counsels that governments need to “catalyz[e] network formation, and moderat[e] the forces that push towards excessive concentration of power.” These recommendations should also govern new efforts to create “virtual networks” of care in the wake of the new health reform legislation (the ACA).

Critics of the ACA have frequently complained that the legislation does not do enough to improve quality or to cut costs. However, the act did create incentives for new alliances of hospitals and doctors, known as “Accountable Care Organizations.” Now provider lobbies are demanding some pretty dramatic changes to health care regulation in order to implement ACOs. In this post, I want to explain what ACOs are, and why they challenge traditional health care regulatory models.

What’s an ACO?

Elliott Fisher, director of the Center for Health Policy Research at Dartmouth Medical School, describes the “three key attributes” of ACOs: “organized care, performance measurement, and payment reform.” Fisher has argued that insurers are not well-positioned to improve the quality of health care because they “have largely focused on negotiating favorable prices within relatively open networks of providers” instead of trying to improve the health care their members received. (Private insurers have little incentive to keep current subscribers healthy over the long term, since at least half of subscribers on average churn into different plans within three years of signing up with a given plan.) He believes that a “virtual network” of physicians could do a better job, if they teamed up with hospitals. ACO refers to this legal alliance, which would be entitled to receive payments in exchange for cutting costs or improving quality.

In an ACO, an “extended hospital medical staff” (or “a hospital-associated multi-specialty group practice”) can join forces with a hospital and agree to be compensated via a lump sum payment. If the group manages to keep overall costs beneath the lump sum payment, it can share the gains among its members. Each part of the team also has an incentive to work together to keep those they care for healthy. In an ideal world, the ACO responds to the concerns about fragmentation discussed in last month’s symposium on the Elhauge-edited volume recently released by Oxford University Press.

ACO Skeptics

But there are skeptics. Jeff Fisher worries about shadowy new pressures on providers that patients won’t be aware of:

Consumers would not be aware that they were being treated by ACOs. Rather, they would be “attributed” to them: virtual patients of virtual organizations. Aggregate health spending for attributed patients would be tracked, and increases in that spending would be capped using a form of “shadow capitation.” ACOs that lived within the caps would get their fees increased. Those that overspent would see their fees reduced or frozen.

Gail Wilensky believes that hospitals may dominate ACOs, predicting that “if they are the only entities receiving the payment, it will have a bad imbalance between groups of physicians and the hospitals.”

Robert Pear has also reported that a “frenzy of mergers involving hospitals, clinics and doctor groups eager to share costs and savings” worries consumer advocates and antitrust scholars.

“In an environment where health care providers are financially rewarded for keeping costs down,” [a lawyer for the Consortium for Citizens with Disabilities] said, “anyone who has a disability or a chronic condition, anyone who requires specialized or complex care, needs to worry about getting access to appropriate technology, medical devices and rehabilitation. You don’t want to save money on the backs of people with disabilities and chronic conditions.”

“The new law is already encouraging a wave of mergers, joint ventures and alliances in the health care industry,” said Prof. Thomas L. Greaney, an expert on health and antitrust law at St. Louis University. “The risk that dominant providers and dominant insurers may exercise their market power, individually or jointly, has never been greater.” Lobbyists and industry groups are bearing down on the Federal Trade Commission and the Justice Department, which enforce the antitrust laws, and the inspector general’s office at the Department of Health and Human Services, which ferrets out Medicare fraud.

Those agencies are writing regulations to govern . . . accountable care organizations. They face a delicate task: balancing the potential benefits of clinical cooperation with the need to enforce fraud, abuse and antitrust laws. . . . [According to one insurer strategist,] “In some markets, the dominant hospital is like the sun at the center of the solar system. It owns physician groups, surgery centers, labs and pharmacies. Accountable care organizations bring more planets into the system and strengthen the bonds between them, making the whole entity more powerful, with a commensurate ability to raise prices.”

Why do ACOs implicate fraud, abuse and antitrust laws? At a recent government workshop on ACOs, participants addressed “circumstances under which collaboration among independent health care providers in an ACO permits ACO providers to engage in joint price negotiations with private payers without running the risk of engaging in illegal price fixing under the antitrust laws.” HHS also explored “the different ways in which the Secretary may exercise waiver authority or create new exceptions and safe-harbors related to the physician self-referral law, the Anti-kickback statute and the CMP law in order to encourage the creation and development of ACOs.” The AMA has pushed for “explicit exceptions to the antitrust laws” for participating doctors. And, as Pear reports, the president of the Federation of American Hospitals says “the fraud and abuse laws should be waived altogether.”

Evolving Fraud, Abuse, and Antitrust Laws

What is one to make of all this? Many health law scholars have been skeptical of fraud and abuse laws for some time. For a taste of the scholarship, consider this excerpt from Mark A. Hall’s 1988 article in the Journal of Health Politics, Policy, & Law:

Someone uninitiated to the intricacies of health care financing would find it startling to learn that it is potentially a felony punishable by five years imprisonment for a rural hospital to recruit a badly needed specialist to the community, for a doctor to discount his services by waiving insurance deductibles and coinsurance, or for a health care institution to pay its doctors a bonus as a reward for efficient practice. A case can be made that each of these activities falls within the literal terms of the broadly worded Medicare and Medicaid referral fee statute. Enticing a physician to join the medical staff necessarily involves implicit or explicit incentives to refer the physician’s patients to that hospital. Price discounts can be characterized as payments to refer one’s patients to one’s self for treatment. And efficiency bonuses can induce doctors to admit patients to a particular hospital or encourage them to direct patients to a particular insurance plan.

Because these and other absurd applications of the referral fee concept are within a plausible reading of the federal referral fee statute, the statute has been a constant thorn in the side of the health care industry since the 1977 enactment of its current form. However, some relief is now in sight. The Department of Health and Human Services (DHHS) has issued a series of ‘‘safe harbor’’ regulations specifying payment practices that are deemed legal despite their potential referral incentive.

Over the past 20 years, regulation of fraud and abuse has waxed and waned. In 1996, James F. Blumstein concluded that “the modern American healthcare industry is akin to a speakeasy—conduct that is illegal is rampant and countenanced by law enforcement officials because the law is so out of sync with the conventional norms and realities of the marketplace.” Nevertheless, as Joan Krause has shown, there are important public purposes behind these laws, and it’s troubling to see a hospital leader simply advocate for them to be swept away tout court in the case of ACO’s.

Antitrust issues are also complex here, and perhaps help demonstrate the wisdom of delaying the implementation of at least this part of the ACA for a few years. As Tim Greaney has demonstrated time and again, providers have had little to fear from antitrust laws over the past decade. His 2007 article on physician cartels memorably summarizes the situation for doctors:

For over thirty years the United States Department of Justice and Federal Trade Commission (“Agencies”) have confronted bands of businessmen who have steadfastly refused to pay attention to legal precedent, repeated governmental pronouncements, and administrative sanctions imposed on their colleagues. The conduct revealed in these cases evidences a willingness to blatantly disregard the law by repeatedly undertaking arrangements already deemed illegal by the enforcers or by concocting schemes that raise untested but dubious justifications.

[T]hese cases involve physicians, some grouped in associations numbering in the thousands and almost always proceeding with the advice of business consultants and counsel. The conduct challenged by the government involves the formation of loosely-structured organizations, ranging from Independent Practice Associations to Preferred Provider Organizations (PPO) to other kinds of loose “networks” that collectively bargain with employers or managed care organizations for provider contracts.

It’s hard to read Greaney’s work on the topic without concluding that a toxic mix of “doctrinal shortcomings, political pressures, and institutional constraints” have severely compromised antitrust enforcement already. Greaney’s 2004 article on antitrust in health care, Chicago’s Procrustean Bed, also suggests that health care antitrust has, for years, been biased “strongly [in] favor defendants” due to the persistent failures of Chicago-inspired doctrine to reflect “market imperfections” in health care.

Reduced Regulation Should Be Conditioned on Better Calibrated Payments

One could draw two lessons from these trends. Perhaps policymakers should be cautious about granting overly broad antitrust exemptions to ACO’s in a field where competition law’s prerogatives have already been whittled away.

Or one could call health care antitrust a largely failed project, and start regulating dominant ACO’s as veritable health care utilities, as critical to regional infrastructure as roads, electricity, or water. The logic of concentration seems inevitable in the field: insurers and providers have long been in an arms race for bargaining power. Joe White has explained the dynamic:

One might wonder why consolidation among insurers [in the US over the past 20 years] did not allow them to resist the providers’ demand for increased payments. The simple answer is that there were two concentrated parts of the market and one fragmented part. The insurers had to choose between fighting a full-pitched battle with the providers or exploiting their own market power vis-à-vis the employers. Raising premiums to employers was a lot easier.

In theory, employers could have demanded restrictive networks (at lower prices). But since everyone had agreed that employees did not like restrictive networks, and providers (especially hospitals) were not willing to discount much to get into such networks, there were not many available for purchase. Individual employers could not invent such a product; they could only shop around and find the relatively best deal by customizing other contract terms, such as cost sharing. The system left substantial room for entrepreneurship, but this entrepreneurship did not serve to improve health care values.

What can be done in a health care marketplace that’s increasingly looking like the “clash of the titans?” Perhaps inspired by the utility model, Maryland has implemented a hospital payment system where “all payers—public and private—pay the same rates.” If ACO’s deliver a coup de grace to insurers’ efforts to control provider prices, it’s only fair that the same governmental authorities behind the ACO movement condition its rewards on the responsibility to provide affordable care.

ACA stands for Affordable Care Act, not Accountable Care Organizations Above All Else. ACOs may work, but only if policymakers can replace classic instruments of health care regulation with calibrated financing decisions that reflect new industry realities.

Accountable Care Organizations: Landmark Report from California

Earlier this week the Integrated Healthcare Association (IHA) released an important whitepaper entitled “Accountable Care Organizations in California — Lessons for the National Debate on Delivery System Reform.” (Click here for a pdf of the whitepaper). The IHA describes itself as: “a statewide multi-stakeholder leadership group that promotes quality improvement, accountability and affordability of health care in California. “

ACOs have been discussed before at HRW in the context of reforming the health care delivery system. Unfortunately, ACOs are relatively new and untested. This is what makes California — a tried-and-true state policy laboratory — a valuable laboratory for health reform, particularly for models like the ACO that the reform law attempts to leverage. The tables provided in the whitepaper make it easy to appreciate California’s experience with ACOs:

The upshot of IHA’s report are the ten lessons that they have distilled from the 285+ physician organizations that together display many of the characteristics of ACOs. These lessons are as follows:

Lesson One: A variety of organizational structures are effective at delivering high quality coordinated care; at least as important to success as structure are an organization’s capabilities, culture and infrastructure, as well as the alignment of goals between the organization and its individual physicians.

Lesson Two: In California, a range of relationships exist between physician organizations and hospitals. Alignment of incentives between physician organizations and hospitals offer important opportunities for performance improvements across the entire continuum of care.

Lesson Three: As a method of payment, capitation can be effective at encouraging coordinated care, but payment methods should vary across ACOs depending on an organization’s ability to assume risk.

Lesson Four: Health plans acting in concert on payment methods and performance measurement helped facilitate the growth of California’s provider organizations, and should also play an integral part in fostering ACO development nationally.

Lesson Five: ACOs are not a panacea for healthcare spending control.

Lesson Six: ACOs must be agnostic to insurance type; most provider organizations in California have focused on commercial, Medicare and Medicaid HMO plans for their patients, but for ACOs to be viable across the country, mechanisms must be found to encourage PPO and traditional Medicare and Medicaid patients to use their services.

Lesson Seven: Balancing patient choice with the desire to decrease costs and effectively coordinate care is difficult. California’s experience underscores the challenge of promoting care coordination in an environment of unrestricted provider choice.

Lesson Eight: Regulation of the financial solvency of provider organizations is important to ensure market stability.

Lesson Nine: Consumer protections from capitated provider organizations need to be balanced, not overburdening.

Lesson Ten: Special attention must be given to establishing ACOs in areas with social and economic challenges.

There is a common thread that runs through many of these lessons: the tension between new organizational models and the payment models that currently exist:

In California, provider organizations have developed hand-in-hand with HMO products, and have been largely unsuccessful in their attempts to diversify into serving PPO patients. This has been driven in part by regulatory restrictions at both the State and  Federal  level  surrounding  providers  accepting capitation  and  FFS  payments. Downward trends in HMO enrollment in California have meant that this failure to diversify has limited the impact of the state’s physician organizations. In order for ACOs to flourish, laws and policies must allow for innovative provider payment arrangements, regardless of insurance type, and internal organizational changes will be needed to adapt to different payment methods.

As the report goes on to describe:

The greatest challenge and greatest opportunity facing ACOs in California and elsewhere is the potential for integrating the coordinated care programs developed originally for HMO and other narrow network insurance products into PPO and other broad network products.

The managed-care backlash of the 1990s is a reminder that past is prologue.  Increased consumer choice through products like PPOs is likely here to stay for quite some time. What remains to be seen is how the future regulations governing ACOs will promote greater collaboration in the delivery of care while embracing the payment models that consumers have come to appreciate and expect.

A Guide to Accountable Care Organizations, and Their Role in the Senate’s Health Reform Bill

[Ed. Note, April 24, 2011: This article by Jordan Cohen was written in March, 2010. It has been cited and linked by everyone from the New York Times to the Texas Tribune and all points in between-- for good reason, it is both thorough and informative. But since the Accountable Care Organization (ACO) model  has continued to gain coverage and currency, we believe it worthwhile to point out recent HRW pieces that will help you get caught up on ACO implementation and impact after you finish the overview of ACOs discussed below. These include:

• It’s Not the ‘Shared Savings’, Stupid: Why ACOs Under the Proposed Rule Will Change Medicine As We Know It - 4/17/2011
• The Normative Meets the Practical: Who Should Can Lead ACOs - 4/12/2011
• ACOs: OIG Guidance, CMS regulations, and Interpretive Tasks - 4/5/2011
• Summary of CMS Proposed Rule on Accountable Care Organizations - 4/4/2011
• Accountable Care Organizations and Antitrust: A New PSA Test - 4/1/2011]

The accountable care organization has been a model for health care reform, yet its modest success has been limited to a handful of health care systems across the country. However, the accountable care organization model has recently taken on far greater significance since being introduced as one of Medicare’s pilot programs in the Senate’s health reform bill.

The phrase is attributed to Dr. Elliot Fisher of Dartmouth Medical School.  Dr. Fisher has led the Dartmouth Atlas Project — a project that has, for the last 30 years, painstakingly documented the variation in care across the United States. (Click here for an interactive map of some of the Dartmouth Atlas results). The Dartmouth Atlas has focused on both the quality of health care as well as its cost. More importantly, they have reported on the relationship between the two, and their findings are nothing short of an indictment of our current paradigm.

Specifically, their findings illustrate that there exists wide variations in the cost of care across the country, and profoundly, that the regions that spend more per patient do not necessarily obtain better outcomes. So what to do? Dr. Fisher believes he has found at least part of the answer: the Accountable Care Organization, known as an “ACO”.

What is an ACO, and How Does it Differ from Other Payment Reforms?

In his paper “Creating Accountable Care Organizations: The Extended Hospital Medical Staff,” Dr. Fisher acknowledges that the term ACO “grew out of an exchange between he and Dr. Glenn Hackbarth at a MedPAC meeting in November of 2006″. (Fisher, 2006 n. 7). Dr. Fisher’s purpose in writing the aforementioned paper was to help identify the proper “locus for shared accountability” for a patient’s health care. HMO’s and other health insurers are obvious candidates, but as Dr. Fisher notes, HMOs only comprise a small percentage of the current market, and health plans in general have focused on negotiating favorable prices within relatively open networks of providers. (Fisher, 2006, p. 45).  The “medical home” (also referred to as a Patient Centered Medical Home or PCMH) is another candidate, but is taken out of the running by Dr. Fisher because of the untested nature of medical homes, and their requirement of new payment mechanisms. (Id.).

Dr. Fisher notes that a better option already exists: “virtual” organizations consisting of the various physicians that are associated with local acute care hospitals. As Dr. Fisher notes, these physicians are either directly affiliated with such hospitals through their inpatient work, or through the care patterns of the patients they serve.  Dr. Fisher refers to these multi-speciality group practices that are bunched around local hospitals as an “extended hospital medical staff.” He argues that improving quality and lowering cost should be realized by fostering greater accountability on the part of this “extended medical staff.”

In a recent Urban Institute paper on ACOs by Kelly Devers and Robert Berenson (pdf summary here, full pdf here), the authors point to three essential characteristics of ACOs:

1. The ability to provide, and manage with patients, the continuum of care across different institutional settings, including at least ambulatory and inpatient hospital care and possibly post acute care;
2. The capability of prospectively planning budgets and resource needs; and
3. Sufficient size to support comprehensive, valid, and reliable performance measurement. (Berenson, p. 2.).

In exchange for investing in this reformed health care provider structure, the ACO members will  share in the savings that results from their cooperation and coordination. Thus, ACOs can–theoretically–act as a reform tool by incentivizing more efficient and effective care. This would help to combat the current perverse incentives of overutilization and overbuilding of health care facilities and technology.

In 2007, Dartmouth’s Institute for Health Policy and Clinical Practice headed by Dr. Fisher and Dr. James Weinstein, teamed up with the Brookings Institution’s Mark McClellan to create The Brookings-Dartmouth ACO Learning Network. The ACO Learning Network will serve as a support tool for providers looking to transition to the ACO framework. In the “Overview” section of their site (available as a pdf here), the Brookings-Dartmouth team provide a useful chart comparing the ACO model to other payment reform models such as “bundled payments,” “medical homes” and capitation. Click the image below to enlarge.

Various Extant Structures Utilized

Since Dr. Fisher’s introduction of the ACO concept, the idea has continued to be refined. In their 2007 paper “Accountable Care Systems For Comprehensive Health,” Dr. Stephen Shortell and Dr. Lawrence Casalino envision a broad range of ACOs in addition to the “extended medical staff” originally described by Dr. Fisher. Drs. Shortell and Casalino identify extant organizational structures that could be leveraged to create ACOs, including the Multi-speciality Group Practice (MSGP), the Hospital Medical Staff Organization (HMSO), the Physician-Hospital Organization (PHO), the Interdependent Physician Organization (IPO), and the Health Plan Provider Organization or Network (HPPO/HPPN). (Shortell et al., 2007, p. 10). Below is a table from their paper that organizes the different ACO models while comparing their capabilities. Click the image below to Enlarge.

ACOs In Practice

Building on the Physician Group Practice (PGP) demonstration project that rewarded the provision of quality care with a share of the savings, the Brookings-Dartmouth group propose a “voluntary and incremental” ACO program. (Fisher et al., 2009, p. 2).  The ACO would have to be a legal organization that can receive shared savings, and would have to incorporate primary care physicians who solely practice under the ACO. (Id.). Furthermore, the Brookings-Dartmouth group believes there would have to be at least 5,000 beneficiaries in the ACO for it to be viable. The ACO would provide CMS with a list of their providers willing to participate in the ACO. As discussed above, the beneficiaries would be determined by, among other things, the patterns of patient referrals in the region. However, beneficiaries would not be “locked in” to a given provider. (Fisher et al., 2009, p. 4). The ACO would receive savings if their risk-adjusted, per beneficiary spending levels were below their benchmark. Id.

An Ultra-Simplified Example

A hypothetical independent practice association (IPA) teams up with a community hospital to create an ACO. Medicare determines a benchmark, that is, what it will cost to treat the average beneficiary in that geographic area per year–let’s say $10,000. The physicians submit their traditional claims to Medicare under the RBRVS system while the hospital submits its typical DRG-base claim. Thus, the traditional fee-for-service system remains in place. At the end of the year, Medicare determines if the ACO has provided care for less than $10,000. If they have, the ACO is entitled to share in the cost savings, and the savings are divided among the providers and hospital. Though simple in theory, ACOs become more difficult when attempting to construct payment models that will distribute the savings of the ACO to the individual providers. Shortell provides another helpful chart that lays out some of the options; Click on the image to enlarge.

Criticism of the ACO Model

The strongest criticism that I am aware of is from Dr. Jeff Goldsmith PhD, president of Public Health Services at the University of Virginia. In his Health Affairs article entitled “The Accountable Care Organization: Not Ready for Prime Time,” Dr. Goldsmith recalls previous attempts to at implementing payment reform models based on shared risk:

The problem with this movie is that we’ve actually seen it before, and it was a colossal and expensive failure. During the 1990s, many hospitals and physicians believed that the Clinton health reforms would force them into capitated contracts with health plans. . . . Risk-bearing physician/hospital organizations and hospital-sponsored preferred provider organizations (PPOs) sprang up all over the country. . . . Some of these hospital/physician efforts actually succeeded and survive today. . . . However, these were outliers in an expensive failure.  Employers and patients preferred open panels managed by health insurers to closed panels managed by providers. Billions of dollars were lost.. . . Many of the practice acquisitions were reversed, as hospital systems sought to rein in their expenses and adjust to an open-panel world dominated by point-of-service style health plans

However, the 1990s left behind an expensive legacy:  highly concentrated local provider markets….There were numerous reasons for the 1990s collapse of at-risk hospital/physician partnerships, besides the failure to find willing buyers of their services. These efforts lacked infrastructure, experienced management, as well as reliable and timely cost information to support cost management. They assumed global risk but paid for care on a fee basis, just as Fisher and colleagues propose.  But these hospital-sponsored organizations could neither redistribute income nor exclude their high-cost providers (who inconveniently generate most hospital profits).

Some things have clearly changed in the ensuing decade. . . . A rapidly increasing percentage of physicians, particularly primary care physicians, are now hospital employees.  A larger percentage of the physician community receives hospital subsidies for call coverage. Many of these subsidies are, in fact, extorted from the hospital by specialists in scarce supply, destined to become scarcer.  An entire generation of 80-hour-a-week baby-boomer physicians are retiring and being replaced by younger physicians who want to work 30 hours a week.  You are not going to see a lot of these younger physicians in utilization review committee meetings after hours; they are going to be at their kids’ soccer practices.

What hasn’t changed is the fragmentation of care, the huge disparities in income and political power inside physician communities, and also the level of suspicion that physicians have of their now much more powerful local hospitals.  There is also, sadly, a thundering absence of collegiality – in my view, the central precondition of assuming risk and managing care. This absence is palpable in suburbs and even more pronounced in many “lifestyle dominated” resort communities in the sunbelt.. . .They are “collections” of physicians, not communities.

The hospitals in these areas appear formidable:  they have beautiful campuses,  prestigious boards, and deep financial reserves. . . . But these hospitals have been picked clean of vital outpatient services by their medical “communities.”

….Entire disciplines have disappeared from hospitals: ophthalmology, cosmetic surgery, gastroenterology, urology. Even community-based internists and family practitioners have stopped coming to the hospital; their patients are cared for by hospitalists who work full time inside the hospital.

The result of our previous attempts at ACO-like integrated care, Dr. Goldsmith points out, is that…

. . . .while the hospital has become more involved in subsidizing physician practice, physician communities have drawn away from the hospital and function increasingly independently on a day-to-day basis. Wennberg’s own data show that something like 40% of physicians no longer have any Medicare hospital-related fee income. So squashing hospitals and physicians back together into economic interdependence in a joint hospital/physician economic pool makes no real-world sense.

Dr. Goldsmith goes on to note that there have been some successful ACOs, but that they haven’t been “virtual” in the sense that Dr. Fisher points out, rather, they are

. . . real organizations with P+Ls, medical directors, and management infrastructure. Prominent examples in my home region include Carilion Health System in Roanoke and the Bon Secours Health System in Richmond. Voluntary ACO arrangements, with Medicare and with private insurers, may find enthusiastic partnerships with many of these hospital-sponsored physician groups. . . .

The Senate Bill

In defense of the Brookings-Dartmouth model, the group has gone on record in favor of voluntary ACOs. To Dr. Goldsmith’s relief, the Senate’s health reform plan incorporates ACOs on a voluntary pilot program basis. You can read their rebuttal to Dr. Goldsmith here.  Section 3022 of the Senate bill — which amends Title XVIII of the Social Security Act (42 U.S.C. 1395) — introduces ACOs under the name “Medicare Shared Savings Program. (View a pdf of the extracted ACO part of Senate bill here).

The Senate’s plan is remarkably similar to the Brookings-Dartmouth model. Under the Senate’s plan, ACOs will be eligible to receive a percentage of the cost savings that they have realized under the traditional fee-for-service Medicare system. The requirements are set forth in section (B)(2). Furthermore, the ACO shall enter into a 3 year agreement with HHS whereby the ACO must agree to contain at least 5,000 Medicare beneficiaries, while being prevented from engaging in risk selection. The ACO must define processes to promote evidence-based medicine and patient engagement, report on quality and cost measures, and coordinate care, such as through the use of telehealth or other remote patient monitoring tools. The ACO must also demonstrate to HHS that it meets the yet-to-be defined criteria for “patient-centeredness”.

Whether ACOs will succeed is impossible to determine with certainty. The panopticon that would be ACO management looking over the shoulders of physicians may be enough to turn off many physicians. Nevertheless, as even Dr. Goldsmith acknowledges, some ACOs have thrived. Moreover, the voluntary ACOs in the Senate’s bill represent a measured approach towards reforming our system without a wholesale transformation. As Dr. Atul Gawande describes in a lesser-cited pre-”Cost Conundrum” article, the most sound approach is often “path-dependent,”  that is, it builds on what already exists. As Dr. Gawande notes:

. . .accepting the path-dependent nature of our health-care system—recognizing that we had better build on what we’ve got—doesn’t mean that we have to curtail our ambitions. The overarching goal of health-care reform is to establish a system that has three basic attributes. It should leave no one uncovered—medical debt must disappear as a cause of personal bankruptcy in America. It should no longer be an economic catastrophe for employers. And it should hold doctors, nurses, hospitals, drug and device companies, and insurers collectively responsible for making care better, safer, and less costly. . .

Whether the shared savings will entice physicians on a large scale is uncertain.  What is certain is that our current fragmented system incentivizes providers to offer neither cost-effective nor coordinated care. Though it is unlikely that physicians and hospitals will flock to ACOs from the start, the vision of ACOs conceived of by the Dartmouth-Group and described in the Senate bill may nevertheless prove itself a useful tool in a larger arsenal of approaches meant to salvage our unsustainable health care system. In other words, the Senate’s approach could provide a path-dependent solution toward the collective responsibility and better outcomes that Dr. Gawande mentions. And as described in the Senate bill, physicians and hospitals will not be offered a new path, but rather a resurfaced path that would retain fee-for-service, while providing a safer and smoother ride for the patient.

« Previous Page