Keith D. Moore
Dean C. Coddington

Whether or not ACOs eventually become a major force in health care, healthcare provider organizations should not stand by waiting to see what will happen. With this journey, the greatest mistake is to be left behind.

At a Glance  

  • In healthcare reform legislation, the concept of accountable care organizations (ACOs) has been deliberately broad.  
  • Providers' efforts to develop ACOs are likely to proceed on different paths and at different paces in various parts of the country, and some organizations will move faster than others.  
  • In deciding which approach to use, a provider should choose a target market, a service area, and a reimbursement methodology; identify the provider structure; design, develop, enhance, and modify core support elements; and identify patient-related and organizational strategies for improvement.  

Accountable care organizations (ACOs) are widely seen as being among the most promising vehicles for increasing value in the nation's healthcare system through improved quality of care and reduced cost. It is for this reason that the concept of ACOs was afforded a prominent place in this year's landmark healthcare reform legislation.

But what is the true potential of ACOs? What does an ACO mean for you and your organization? What are the most promising strategies for developing an ACO? What transition steps are needed to get there? These are some of the questions we asked a mix of policy, health system, and physician leaders this past spring.a The answers suggest a journey that is just beginning, with many twists and turns ahead.

The Rationale for ACOs

As noted previously, ACOs figure prominently in policymakers' ideas for "bending the cost curve" in health care and in hopes for improved quality and patient service. As envisioned by policymakers and set forth in the 2010 Patient Protection and Affordable Care Act (HR 3590, as amended), the concept of an ACO is deliberately broad-enough so to encompass many types of healthcare provider organizations working in different types of communities under a variety of reimbursement structures. Ultimately, most observers believe ACOs will be applicable not only for Medicare, but also for most forms of contracting. Consider the following proposed working definitions:

"An ACO would be a group of providers held responsible for the quality and cost of healthcare for a population of Medicare beneficiaries. An ACO would be a combination of one or more hospitals, primary care physicians, and possibly specialists, and would be accountable for total Medicare spending and quality for the Medicare patients served."
-Commonwealth Fund, April 13, 2009.

"The concept of ACOs is an entirely new paradigm-giving healthcare providers the responsibility and appropriate incentives to improve outcomes and giving them the flexibility to design the most efficient and effective way to do so."
-Harold Miller, "How to Create Accountable Care Organizations," Center for Healthcare Quality & Payment Reform, September 2009

"[Under the ACO concept,] payers identify the primary care patients of a physician-hospital network that is willing to take responsibility for the full continuum of care. A spending target is set for these patients, and if the ACO meets quality benchmarks and reduces per-beneficiary spending below the target, providers receive a share of savings."
-Jonathan Skinner and colleagues, "Looking Back, Moving Forward," New England Journal of Medicine, Feb. 17, 2010

ACOs are a key concept in many existing and proposed reimbursement structures. A wide variety of reimbursement pilots are under way now, and still more are in the planning stage. One way to think of these reimbursement alternatives is as a continuum based on who bears the risk of care. At one end of the continuum-the fee-for-service end-the risks associated with the costs of care are born primarily by the patient and/or the payer. As one moves along the continuum, a higher and higher proportion of this risk is born by the providers of care.

 View Exhibit 1  


In this context, ACOs are a means for transferring risk associated with the costs of care, managing this risk, making payments associated with care, and then distributing associated rewards (or performance payments) to providers.

Some see ambitious goals for ACOs, and high potential. One physician leader within the Centers for Medicare & Medicaid Services (CMS) said, "People haven't focused on the true implications of this act-it's a Trojan Horse. The real issue is that it sets in motion a revolution as to how care is delivered in this country."

Others have far more jaded views. One leader of an integrated system (including a 250-physician multispecialty group, plus a hospital and a health plan) said, "ACOs are a nice way to avoid using the 'capitation' word. The sooner we get over that, the better. You can write about them, but in the end, they will not change the cost curve."

Similarly, the Deloitte Center for Health Care Solutions characterizes ACOs as "a compilation of integration tactics that have been tried at different times and in different systems." Deloitte does not see ACOs as representing a significant paradigm shift (Accountable Care Organizations: A New Model for Sustainable Innovation, April 2010, p. 18).

Group discussions about ACOs reflect the lows and highs of emotions (and the mass confusion) around the concept of healthcare reform. Discussions focus one moment on the threats of "rationing care" and "death panels," and the next moment on how "20 percent of the cost of care could be cut without harming, indeed while improving, quality and service" and "ACOs have to be an improvement, because the current system makes no sense."

Preparing for the Journey

Despite reservations, most healthcare organizations are moving forward to develop an ACO strategy (or, in many cases, multiple strategies). This is the smart thing to do. Even if a healthcare leader believes personally that ACOs are going to fall on their face and be terminated, the risks of being unprepared for a future that includes ACOs are unacceptable. Indeed, the potential consequences of a "false negative" (proceeding on the belief that ACOs will not be an important part of the healthcare landscape and then finding you were wrong) are much more severe than the consequences of a "false positive" (proceeding on the belief that ACOs will be important and then finding they are not).

At least as important, the process of developing an ACO entails building, adding, or enhancing certain capabilities. This effort will entail, for example:

  • Developing the ability to measure quality and the organizational processes to enhance quality
  • Developing the ability to identify and analyze options for cost reduction and the organizational processes to implement cost reduction
  • Establishing closer physician-physician and physician-health system working relationships, with the ability to share more operating data
  • Establishing more timely and relevant clinical information and extending IT linkages

The reality is that these are initiatives that should be undertaken regardless of the outcomes of pilot projects on reimbursement structures.

Martin Arrick of the rating agency Standard & Poors expresses a similar perspective: "Will these pilots show that bundling is the way to go? It depends … depends … depends." On the other hand, Arrick says, "Even with the uncertainties, it's encouraging to see that providers are not waiting to move in the direction of payment reform and ACO development."

As the journey begins, many key questions must be answered:

  • Who should the ACO include, and what organizational form should it have?
  • How should it be developed?
  • How should it be governed?
  • How should the IT be integrated and the necessary information flows developed?
  • How should it be financed?
  • What should its strategy be?
  • How should the transition be managed?
  • How should the organization develop the culture and day-to-day decision making required to be successful?
  • How can risks be minimized?

These questions can be addressed as part of a seven-step decision-making process (see also Exhibit 2):

  • Choose a target market
  • Choose a service area
  • Choose a reimbursement methodology
  • Identify the provider structure
  • Design, develop, enhance, and modify core support elements
  • Identify patient-related strategies for improvement
  • Identify organizational strategies for improvement

View Exhibit 2


Different Points of Departure

The Patient Protection and Affordable Care Act requires that ACOs include primary care physicians, but it makes few other requirements as to organizational composition. A clinically integrated network (CIN) of primary care physicians therefore can be an ACO, as long as it is willing to accept and manage risk.b Ruth Benton, CEO of New West Physicians, a 60-physician primary care group in Denver, points out, "We've functioned as an ACO for 15 years now." New West contracts with United Healthcare to manage a panel of Medicare Advantage patients. Members of the group use the same electronic medical record, follow the same clinical care pathways for key chronic conditions, and share in the performance payments. The group contracts with specialists and hospitals for the care it needs from them, and is willing to make changes in these arrangements should performance or cost problems develop.

Certainly, an integrated delivery system (IDS) can be an ACO. One could even argue that IDSs that include a health plan-e.g., Kaiser Permanente based in Oakland, Calif.; Scott & White in Temple, Texas; Geisinger Health System in Danville, Pa.; and the Carle Foundation Association in Urbana, Ill.-have functioned like ACOs for some time. Recognizing their advantages, IDSs are already taking significant steps to move their entire organizations, including culture, still further in this direction. "We assessed a number of options, and this was clearly the direction our organization (and our community) needed to go," says Edward Murphy, MD, CEO of the Roanoke, Va.-based Carilion Clinic, which is an ACO pilot site for Medicare.

One common form of ACO is expected to be a CIN of private practice physicians that is organized by a health system and that also includes that health system's employed or integrated physician practices. Advocate Physician Partners (in the Chicago area), Memorial Hermann Health Network Providers (in the Houston area), and Catholic IPA (in Buffalo, N.Y.) are examples of CINs that have been operating for several years and appear well positioned to serve as ACOs.

The examples above represent organizational forms that exist today and are somewhat "ACO-ready." Unfortunately, these represent a minority of the existing situations in the United States. Many others will begin this journey further back, and will have more twists and turns ahead. Should a hospital lead an ACO that includes the independent physician practices that are represented on its medical staff? Hospitals are a "natural" to lead ACO development in many cases. They can come up with the required capital investment, and they have or can acquire the required management expertise. In addition, most hospitals are community-oriented, have built reputation and trust, and include community representation on their boards-all sources of strength for an ACO. In addition to primary care groups and health systems, other potential "integrators of care" (and of ACO leadership) include ad hoc groups of physicians, health plans, and venture capitalists.

For some hospitals, providing strong leadership in an ACO will require a big change in culture and expectations. As one hospital executive put it, "We are pretty much accustomed to listening to physicians, and giving them anything they want. If we start to say 'no' to some equipment requests, and to call some physicians on the carpet for not following protocols, we will be operating in a very different world." Still, the alternative-not to lead the ACO and to have some other interest group lead it-will often simply be untenable.

Some medical centers may find a need for more than one ACO. For example, a major medical center with substantial outreach may need one closely coordinated ACO that is capable of accepting and managing significant financial risk. It may need a second ACO that is less tightly coordinated, but operates over a much larger geographic area. And it may need a third ACO for children's services. In this case, the three ACOs may use the same core services platform (e.g., IT, legal, and human resources) but include different physicians, have different quality and patient service initiatives, and use different branding. If the medical center is part of a larger health system, it may participate in a fourth, systemwide ACO, and it may use the system's core services platform.

Compounding Factors

It seems clear that the journey to develop needed capabilities will proceed on different paths, and at different paces in different parts of the country, and some organizations will progress more quickly than others. Jeff Selberg, former CEO of Denver-based Exempla Health (which has a bundled payments pilot under way), points out that to be successful, an ACO will "need to develop the culture, leadership, infrastructure, and technique to engineer more effective care processes"-and that the more integrated a system is at the beginning, the bigger head start it has. Other differences will also come into play; for example, some leadership teams are able to move much faster than others.

Another compounding factor is that even if health systems were neatly organized to manage care in ACOs, patients are not neatly organized and ready. Even patients who visit highly integrated systems usually do not confine their care choices to these systems. John Koster, MD, CEO of Providence Health and Service, based in Renton, Wash., points out that even in Portland (where Providence provides its most highly integrated approach to care), "Only 12 percent of our patients receive all of their care from the Providence system." Koster observes, "The patients of our clinics and hospitals come from all over the place, and not necessarily through our primary care network or our health plan."

Still another compounding factor is the mix of physicians within systems. One concept of the integrated system is that it includes a balance of primary care physicians, specialists, and inpatient capacity, a so-called "right sizing" of care resources to match the needs of a given population. However, this is simply not the case in most of today's integrated systems. Nicholas Wolter, MD, CEO of Billings Clinic in Billings, Mont., indicates the clinic's specialists are reliant on outside referrals (as opposed to referrals from the organization's primary care physicians) for more than half of their cases. The numbers are roughly the same for The Carilion Clinic. This type of situation is common, particularly in rural areas where cases need to be aggregated from a large geographic area to provide the required volume for specialty services.

Different Journeys

How aggressively should ACOs pursue new forms of contracting? There is not a "one-size-fits-all" answer to this question. Organizations that are ACO-ready need to build on this advantage by adding market share, physician affiliates, and/or net revenues. Organizations that are not ACO-ready need to get ready.

For early-stage situations, the easiest first step may be to form an organization and initiate a pilot. No organization that is not ACO-ready wants to receive a large portion of its revenues from a new performance-based contract next year. However, there is a trade-off here. It is difficult to form this type of organization without having someone put some dollars on the table. The logical sources for this seed capital include government payers, commercial payers, and health systems. To be credible to a health plan and to participating physicians, the number of patients and amount of dollars involved must be significant.

Government-sponsored pilots are one option. However, pilots with one or more health plans are also an alternative. Often, pilots with health plans can get started faster.

What type of contract is best to start with? For most organizations, it may be best to start with a reimbursement structure that is closer to the fee-for-service end of the spectrum. For example, a medical-home agreement often simply grafts a primary care payment on top of a fee-for-service schedule. Similarly, some pay-for-performance arrangements graft a success fee on top of a fee-for-service schedule. A key early goal is getting the right information flows working, so that the new ACO can begin to take stock of its performance and where it needs to go.

Managing Risk

Risk management is likely to be a critical success factor in the development of ACOs. The early risks come in many forms. The following are just a few examples.

Timely information. Many early pilots have begun to distribute performance payments before they are able to generate timely information. The payments are therefore based on information that arrives too late to correct behavior. The information flows need to be in place before the performance pay process is initiated.

Effective assignment of risk. Health systems are working hard not to assume more risk earlier than they are prepared for. They also are engaged in ongoing efforts to refine the approaches used to differentiate between actuarial risk (the risk that the organization's group of patients is statistically likely to have higher costs) and performance risk (the risk that controllable costs will not be controlled).

Unanticipated start-up costs. Most early clinically integrated networks, which are precursors to ACOs, took longer than was anticipated to put in place and had greater than expected start-up-cost and staff requirements.

Inertia. Start-up efforts for some organizations in the first year have had the quality of a "book club," focusing largely on literature reviews and lengthy discussions. Although this experience can be valuable, and has resulted in significant education, it often does not move the organization forward in achieving its longer term objectives.

Mix of participating physicians. Some efforts have begun with too few physicians or with the wrong physicians. In some cases, for example, the initiative might begin with the wrong mix of specialties or without the participation of the natural leaders in the medical community.

Specialist flight/alienation. Some early efforts have offended key specialists whose support is necessary to keep the existing fee-for-service revenues flowing in the right direction.

No Small Trek

Initiating an ACO is demanding work that requires balancing several strategic considerations: It must be closely integrated with other priorities and investments, requires careful attention to finances, should move neither too slow nor too fast, and demands the best efforts of organizational leaders and a high-priority focus.

Let the journeys begin …

Web Extra:See a presentation of payment reform and ACOs published with the permission of McManis Consulting.

Keith D. Moore is CEO and chairman, McManis Consulting, Greenwood Village, Colo. (

Dean C. Coddington is a senior consultant, McManis Consulting, Greenwood Village, Colo., and a member of HFMA's Colorado Chapter


a. Discussions were held with 15 healthcare leaders-including health system CEOs, physician group leaders, health plan leaders, association leaders and staff, and leaders of public sector healthcare entities-between March 1 and May 30, 2010.

b. The Federal Trade Commission has established guidelines whereby a clinically integrated network (CIN)-a group of otherwise independent practices that share clinical information, develop and adhere to common clinical protocols-can negotiate for contracts as a single entity. Many CINs include a hospital or health system as well.

Publication Date: Monday, August 02, 2010

Login Required

If you are an existing member, please log in below. Username and password are required.



Forgot User Name?
Forgot Password?

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:


   Become an HFMA member instead