Editor’s Note: There is considerable overlap in the nature and goals of the “alphabet soup” of network structures—Accountable Care Organizations (ACOs), Independent Practice Associations (IPAs), Clinically Integrated Networks (CINs) and Physician Hospital Organizations (PHOs). Often, there is little practical difference among the network types, regardless of the latest popular acronym chosen as the label. They all present similar legal challenges to establishing, structuring and operating a successful provider network. In a recent webinar, Manatt decoded the acronyms—and explained, for both lawyers and nonlawyers, how to navigate the legal issues involved in launching and maintaining a provider network.
In part 1 of our article summarizing the webinar, we explained critical terminology, as well as reviewed entity formation issues. In part 2 of our summary, below, we focus on antitrust issues and risk management. Click here to view the webinar free, on demand, and earn CLE. Click here to download a free copy of the webinar presentation.
The key antitrust statute is Section 1 of the Sherman Act, which prohibits two types of agreements that unreasonably restrain competition:
Antitrust Issues and Provider Collaborations
All provider collaborations are agreements between independent parties that are subject to Section 1. The agreement that we are talking about here is the participating provider agreement. All collaborations between competing or potentially competing providers—including the documents providers sign to be part of a network entity—are potentially subject to per se liability.
Because provider networks typically want to engage in price negotiations with payers and enter into a single agreement covering the services of the entire group, the key antitrust concern is price fixing. If the agreement between competing providers is just to negotiate jointly, it could be considered a per se price-fixing agreement.
To identify if there is an antitrust issue, it is critical to determine if there are competing providers in the network. Not all providers compete. Some offer different services. For example, orthopedists don’t compete with obstetricians. Others practice in different regions. Two primary care physicians (PCPs) don’t compete with each other if their practices are 150 miles apart. The bottom line is, however, that most networks are going to have competing providers in them.
It is important to remember that competition isn’t always obvious. There may be overlaps between different specialties. For example, some PCPs can provide some of the same services as a cardiologist. Multispecialty practices and federally qualified health centers can cloud the picture even more, because they can include a variety of providers and specialists. Telemedicine further complicates the issue, because it can create competition between providers, even if they practice in distant locations. The fact that certain providers are friendly and may even refer patients to each other does not mean that they are not competitors. In addition, hospitals may be part of the physician competitive set if they employ PCPs or specialists who compete with private practice physicians in the surrounding community.
Managing Antitrust Risk: Integration Models
Integration programs seek to achieve efficiencies, such as cost savings and quality improvements. There are three types of integration models—financial, clinical and hybrid (which includes elements of both financial and clinical integration). A network will avoid the per se price-fixing concern if the joint contracting is ancillary and necessary to achieve the efficiency goals of the integration. Joint contracting can’t be the primary purpose of the integration.
Over time, the Federal Trade Commission (FTC) and Department of Justice (DOJ) have issued guidance on when an integrated network can engage in joint contracting, including the 1996 Healthcare Statements and the 2011 Medicare Shared Saving Program (MSSP) ACO Guidance. Nothing has been released from the agencies, however, since 2013.
The 1996 Healthcare Statements define financial integration as a network of otherwise independent providers sharing financial risk in such a way that each member has an economic incentive to ensure that the network as a whole generates efficiencies that benefit consumers. Financial integration can’t be the goal. The goal has to be to creating meaningful prospects for improving efficiency, controlling costs, managing utilization and/or improving quality of care.
Substantial financial integration includes the concept of both upside and downside risks. There is no clear guidance on the exact level of risk that is required, although 15% is the common benchmark.
The 1996 guidance also sets out safety zones. If a collaboration falls within the safety zone, the agency is presuming that it will not have an anticompetitive impact. For exclusive physician networks, participants can constitute only 20% or less of each physician specialty in the relevant geographic area. If a network is nonexclusive, the number goes up to 30% or less. For a network to be considered nonexclusive, there must be evidence that its physicians can participate in and earn substantial revenue from other networks. Physicians also must be able to contract individually.
Networks outside the safety zone are judged under the rule of reason. The integration must be likely to produce significant efficiencies that benefit customers, and price agreements must be reasonably necessary to achieve those efficiencies. Where there is high market share, collaborations will face greater scrutiny to ensure they are producing the efficiencies, cost savings and quality improvements that would justify joint contracting.
Three Examples of Financial Integration
Following are examples illustrating three types of financial integration.
The first example is based on a 1980s-era approach, with fee-for-service payments being made by the health plan. The plan pays 80% of the claim but holds back 20% in a collective pool. If year-end performance metrics are achieved (or exceeded), the withheld funds (and possibly an additional bonus) are paid to the billing providers. This approach demonstrates the “skin in the game” required to be viewed as financially integrated.
The second example demonstrates full-risk capitation, sometimes referred to as population-based payments. Under full-risk capitation, the payer pays $100 per member per month to the network. The network entity pays participants’ fee-for-service claims or has the health plan pay on its behalf. At the end of the year, all the risk is held by the network entity, not by the health plan. The availability of funds for the network entity to pay all claims is based on the collective performance of the entire network. Therefore, the network is considered financially integrated.
In the third example, the payer pays the standard fee from its existing fee-for-service schedule. If aggregate claims payments for the year are less than the payer projects, the payer shares a percentage of the savings with the network entity in a year-end lump-sum payment. The network entity decides how the lump sum will be allocated to downstream providers.
The 1996 Healthcare Statements define clinical integration as an “active and ongoing program to evaluate and modify the practice patterns by the network’s physicians and create a high degree of interdependence and cooperation among the physicians to control costs and ensure quality.” The definition is deliberately open-ended. The analysis focuses on the substance of what’s being done, not the exact form. The 1996 Healthcare Statements provide some examples—such as mechanisms to monitor and control utilization, the selective choice of network participants, and significant investment of monetary and human capital. But they don’t prescribe.
As with financial integration, the overall goal of a clinical integration program must be to create a meaningful prospect of jointly improving efficiency, controlling costs, better managing utilization and enhancing care quality. Any agreements on price must be “reasonably necessary” to realize these improvements.
In discussing clinical integration, it is important to be aware of alternative meanings when the term is used outside the antitrust context. For example, many use the term “clinical integration” when discussing healthcare reform to describe the tools needed to attain the Triple Aim.
To add to the confusion, some states have their own definitions of clinical integration that may or may not be the same as the antitrust definition. It is important not to assume that the state regulators’ standard for clinical integration is sufficient to meet the DOJ’s and FTC’s definitions of the same term.
In this complex environment, how is it possible to determine if there is an antitrust problem with a clinically integrated entity? There are a number of possible approaches:
Clinical Integration: Medicare ACO Policy Statement
The Medicare ACO Policy Statement applies to ACOs that intend to be or have been approved to participate in the Medicare Shared Savings Program (MSSP). It also applies to other ACO initiatives from the CMS Innovation Center (such as Pioneer ACO) and to Medicare ACOs that participate in commercial markets. It does not formally apply to non-Medicare ACOs. An ACO with the same features, however, is likely to satisfy clinical integration standards.
The primary point of the Medicare ACO Policy Statement is that the FTC and DOJ will apply rule of reason analysis to Medicare ACOs. The CMS eligibility criteria are broadly consistent with the indicia of clinical integration that the FTC and DOJ have defined. The belief is that MSSP ACOs are likely to be genuine arrangements to reduce the cost of providing healthcare. CMS is charged with monitoring the MSSP ACO results. The Medicare ACO Policy Statement is focused on clinical integration and does not discuss financial risk sharing in any detail.
The Policy Statement sets out safety zones, requiring combined shares of not more than 30% for each common service. The Statement defines markets and services based on physician specialties and CMS definitions of major diagnostic categories—which are not necessarily reflective of market realities. Other requirements include:
Not many ACOs have invoked the safety zone in their CMS applications, because it’s a complex and data-intensive process, particularly for physicians.
ACOs outside of the safety zone will be analyzed under the rule of reason. The FTC is unlikely to challenge unless the ACO has a very high market share and engages in conduct the guidance has stated should be avoided, including:
The 2011 Policy Statement offers an expedited voluntary antitrust review for Medicare ACOs. No one has gone through that process yet, so there are no examples of the FTC or DOJ formally looking at a Medicare ACO to see whether it meets the standards for clinical integration. Therefore, ACOs that are outside the safety zone have to fall back on the principles in FTC advisory opinions, including:
Hybrid Models—Belt Plus Suspenders
Some networks qualify as both clinically integrated and financially integrated—described as a belt-and-suspenders approach. Since the basic aims of both financial and clinical integration are the same, it makes sense that networks would develop both types of integration. A clinical integration program will provide the data and coordination needed for an ACO to take on financial risk. A financially integrated ACO usually needs some sort of clinical integration to achieve the systems improvements that enable financial risk-taking.
Managing Antitrust Risk: Single-Entity Models
Single-entity models consist solely of providers who are affiliated with a single health system. The affiliation consists of ownership or control by a common parent entity of all the participating providers. The providers may be employees or exclusive to an entity owned or controlled by the parent. The providers cannot compete with each other, because they are all effectively part of the same entity. There is no basis for liability under Section 1 of the Sherman Act, because there is no agreement among independent contractors.
Managing Antitrust Risk: Messenger Models
If a network entity is neither clinically nor financially integrated, the remaining option is the messenger model. In the messenger model, there is no joint venture among the competitors. Each competitor must make its own decisions in a silo without knowing what its competitors are offering or accepting.
By definition, under the messenger model, the network entity does not negotiate on behalf of numerous providers. Instead, the network entity relays offers and acceptances between the parties—the providers and the payer. Participation in the network is tentative and subject to final acceptance by each provider of each specific payer or contract.
Many ACOs do not negotiate payer fee schedules. For example, under the MSSP and NextGen ACO models, CMS pays all providers at the standard Medicare fee schedules, which exist regardless of the ACO arrangement. Instead, these ACOs often negotiate only supplemental provisions regarding:
While the same antitrust rules apply, antitrust officials are less likely to be concerned when fee schedules are not being altered or renegotiated.