Originally published on Law360.com in the Health, Insurance, and Competition sections on April 28, 2010.

In spite of the continued public and political division over the Patient Protection and Affordable Care Act (the “Act”), there are elements of the Act that policymakers, regardless of politics, might actually agree upon.

Hidden away behind the most contentious provisions of the Act1 and found in the Republican health care alternative proposals2 is a policy change that could assist in delivering cost-effective care to the marketplace. The change involves the creation of Medicare pilot programs for “accountable care organizations” (“ACOs”).

While many aspects of ACOs are not new to healthcare, focusing the federal government on reimbursing integrated providers for providing a broad base of care to patients is new. Each of the proposals identifies general, non-specific features that an ACO must implement in order to participate in the pilot. However, providers looking for both an early advantage in the creation of qualifying ACOs and a path to avoid potential antitrust liability should consider the valuable guidance that the Federal Trade Commission (“FTC”) already has provided in very similar contexts.

The Shared Savings Pilot Program and ACOs

The primary objective behind the ACO provisions in the Act is to provide financial incentives to promote the delivery of coordinated care across the health care delivery spectrum (i.e., from primary care physician to tertiary hospital care) to a defined patient population of Medicare beneficiaries. No later than January 1, 2012, the Secretary of Health and Human Services must establish a “shared savings program” which utilizes the ACO model that “promotes accountability for a patient population and coordinates [Medicare] items and services . . . , and encourages investment in infrastructure and redesigned care processes for high quality and efficient service delivery.”

The Act makes a variety of integrated providers eligible to be an ACO, including physician group practices, networks of physicians, physician-hospital organizations or joint ventures, and hospitals with employed physicians. The ACO model requires provider integration and cooperation in order for the providers to coordinate the care of patients effectively.

Under the Act, among other things, an ACO must have:

(1) a sufficient number of primary care providers to care for the patients for whom the ACO is accountable;

(2) leadership, management and access to specialists and processes (such as the use of telemedicine technologies and use of websites to promote best-practices within the ACO) necessary to promote evidence-based medicine and to coordinate care; and

(3) a legal structure able to distribute financial incentives related to meeting quality performance measures.

The manner in which Medicare will reimburse the ACO providers will help drive such coordinated care. The Act combines existing fee-for-service payments with a shared savings incentive payments for achieving certain quality performance measures and for reducing Medicare expenditures within the ACO by a certain percentage below a predetermined benchmark.

The Secretary also has the flexibility of paying ACOs other than through the traditional fee-for-service and shared savings bonus. The Act specifies that one of the alternative payment models could be a partial capitation model in which the ACO would be at financial risk for some, but not all, of the Part A and B services provided. However, only “highly integrated” ACOs or ACOs otherwise capable of bearing risk, as determined by the Secretary, would be eligible to participate in a capitation or other risk-based reimbursement model.

In some respects, many features of the ACO proposals are not new. Payers and providers have experience, both good and bad, with capitation and global payments along with more recent pay-for-performance mechanisms. While experience with capitation and other risk-based reimbursement may be mixed, the key difference is that the ACO concept relies heavily on greater integration among the providers within the ACO and greater focus on such integrated providers using shared learning to improve care, which allows providers to reap the benefits of available financial rewards, based on both improved care and reduced expenditures.3

Why Antitrust Law Matters for ACOs

There are many legal issues an ACO will have to manage. For example, contractual integration among hospitals and physicians can implicate Stark and anti-kickback laws. Depending upon the ACO model, if an ACO assumes financial risk, it can trigger certain state insurance laws.

However, because integration among providers is a critical feature of an ACO, the antitrust laws, and what the antitrust regulators have said about integration, also matter. Integrated systems based on a physician employment model, for example, have always had to consider the antitrust ramifications of their actions and expansion, because growth can lead to untenable market share and market power, from a regulator’s, competitor’s, or payer’s perspective.

Integrated systems, such as the Mayo Clinic, Cleveland Clinic and the Marshfield Clinic, might already have a number of the features necessary to be successful ACOs. In some respects, such fully integrated systems can better manage their antitrust risk.

First, integrated systems are in a better position to control the types of activities that could result in antitrust exposure. Unlike a network comprised of competitors, an integrated system need primarily be concerned with its own actions.

Second, most of their antitrust risk is size dependent. They know what their market share is and are cognizant of what they can and cannot do to expand their share and how far they can go.

The ACO models with greater antitrust risk or uncertainty are those involving provider networks, such as independent practice associations (“IPAs”) and physician-hospital organizations (“PHOs”). A core antitrust principle prohibits agreements among competitors to fix prices. A network of independent providers who share price or contract information and collectively negotiate contracts with payers is at serious risk of violating this core principle. Understanding that provider networks can provide benefits to the marketplace, the FTC, along with the U.S. Department of Justice, provided guidance to providers on how to create a network which brings efficiencies to the market benefiting consumers.4 From the first statements on the matter in 1993 and in subsequent revisions in 1994 and 1996, along with numerous advisory opinions, the FTC informed providers that networks can avoid per se violations of antitrust laws by providing efficiencies to the market through either sharing substantial financial risk in the form of risk-based contracting or being significantly clinically integrated.

Under the antitrust laws, unless a network-based ACO, such as an IPA or a PHO, relies substantially on risk-based reimbursement, e.g., capitated reimbursement, a network-based ACO must be clinically integrated to avoid a per se antitrust violation. Of course, this is what the Act expects. An ACO cannot be accountable for a patient population unless the ACO can implement performance measures and have processes in place to promote evidence-based medicine throughout the course of a patient’s treatment.

This is precisely in line with how the FTC views clinical integration. In a May 2008 workshop on clinical integration held by the FTC, Markus H. Meier, Assistant Director, Health Care Services and Products Division, stated that “Clinical integration is an active and ongoing program to evaluate and modify the practice pattern of physicians and create cooperation to control costs and ensure quality.” This statement is consistent with the FTC’s position in two recent positive advisory opinions it granted to Greater Rochester IPA5 and TriState Health Partners, Inc. (a PHO).6

Ultimately, if a network-based ACO qualifies to receive partial capitation or other risk-based reimbursement from CMS, then it might avoid a per se violation of antitrust law by satisfying the sharing of substantial financial risk test. In certain markets, such as California, with a strong history of provider networks, many IPAs might meet the financial risk test as they exclusively operate under capitation arrangements.

However, because the Act only permits the Secretary to reimburse “highly integrated” ACOs or other ACOs capable of bearing risk under a capitation or other risk-based model, all other network-based ACOs (and those that choose to avoid risk-based reimbursement) will have to meet the clinical integration test.

The good news for any network desiring to qualify as an ACO, whether it meets the financial risk test or not, is that the FTC has identified a number of key elements for a network to be clinically integrated.

First, networks must create and implement a broad spectrum of best practice protocols and clinical practice guidelines based on local or national standards or through their own internal development in order to focus the organization’s activities on improving care.

Second, clinically integrated networks depend upon significant investment in the form of time commitment and financial resources. Providers must be willing to participate actively in governance and network training efforts to implement best practice protocols effectively and to implement systems to improve compliance and education efforts, such as web-based systems or utilization of shared electronic health record tools.

Third, an organization must measure success in implementing the guidelines and whether the guidelines are achieving measurable quality or cost improvements.

Finally, successfully integrated organizations have in place compliance, education and enforcement mechanisms. These mechanisms are necessary to educate and facilitate shared learning among providers on clinical guidelines and programs; to educate patients to encourage compliance with treatment plans; and to remove repeat non-compliant providers from the network.

Not only would implementing this guidance help network-based ACOs limit their antitrust exposure, but also it would more likely than not, set networks (even risk-bearing ones) down the path of developing a successful ACO model.


Even though the Act’s ACO provisions only apply to Medicare, there are three reasons why commercial payers might embrace the ACO concept. First, employers, payers and providers desire a means of aligning reimbursement with outcomes. After falling out of favor, capitation payment arrangements have increased, along with performance based reimbursement.7 Second, the expanded use of and financial support for electronic health record systems will facilitate greater information sharing among providers. Real-time information access is integral to the success of an ACO as patients move between ACO providers.

Finally, over the past few years, in a number of markets, there has been greater integration among providers, from direct hospital acquisition of practice groups and employment of physicians to the creation of and expansion of provider networks, including IPAs and PHOs.8 Even those who are highly skeptical of the ACO model, recognize that integration, in addition to significant investment, is required for ACOs to have any chance of success.9

If the environment is ripe for the ACO concept, then providers must start planning now on how to successfully integrate and how to manage the risks of integration.