Health Affairs editor-in-chief Susan Dentzer described the atmosphere at a recent conference on the hot topic of accountable care organizations (ACOs) as follows: “As newly minted ACO ‘experts’ opined on the subject, the audience alternately expressed confusion, excitement, hope, fear, skepticism, and disbelief that the fragmented U.S. health care system could be restructured accordingly. It was a bit like describing a future of flying unicorns to hard-bitten denizens of a thoroughbred racetrack.”¹

Regulatory fatigue aside, the hubbub about ACOs would be worthy of discussion because of its prominent place in the Patient Protection and Affordable Care Act – the health care reform bill. It becomes even more interesting considering that many observers inside and outside of Washington (but particularly inside!) are contrasting ACOs with the patient-centered medical home (PCMH) model as a vehicle for primary care re-empowerment and health system reform. The issue is being positioned by some as “either/or” – with the ACO replacing the PCMH as the model going forward.

We've seen a lot of activity in the blogs and discussion forums on this issue, many viewing the shift in Washington as a blow to family medicine, if not to all of primary care. Frankly, I don't see this as something to be alarmed about, although I do think the AAFP leadership and members should carefully consider the nuances of the argument in favor of ACOs as the principal framework for primary care rejuvenation in this country.

First, a few places to read up on ACOs if you haven't done so already: Probably the best short piece on ACOs is a recent article in Health Affairs that argues that several flavors of ACO should be allowed to flourish.² These would include both tightly integrated organizations paid by capitation and more loosely structured arrangements paid mostly on a fee-for-service basis, the latter designed around primary care IPAs and virtual physician organizations. Another, more comprehensive, source for current thinking on ACOs is Harold Miller's report, published by the Center for Healthcare Quality and Payment Reform, titled “How to Create Accountable Healthcare Organizations.”³ Finally, the Brookings Institute and the Dartmouth Institute for Health Policy and Clinical Practice have teamed up to create the ACO Learning Network, with materials at their new web site.

In each of these sources you will find a strong policy statement that ACOs are to be based on a foundation of primary care, and suggestions that the PCMH is the model on which primary care should operate. At the same time, these writers consider the PCMH just one of several components required for ACO success.

Here is how the Health Affairs authors put it: “Accountable care organizations will be largely based on physician practices that, in turn, may be organized as patient-centered medical homes. Many ACOs will also include hospitals, home health agencies, nursing homes, and perhaps other delivery organizations.”

Miller's support for the PCMH is more nuanced and qualified, but he clearly believes that the PCMH can be important to the success of an ACO: “Efforts to help primary care practices become more effective, such as the tools of Patient-Centered Medical Homes, the Chronic Care Model, etc., are helpful, but not sufficient. In order to create a successful Accountable Care Organization, primary care practices must add the capability to manage both cost and quality outcomes. Moreover, not all of the standards in current Medical Home accreditation programs may be necessary to success as an Accountable Care Organization.”

And from the Brookings-Dartmouth web site: “The ACO approach also builds on current reform efforts that focus on one key group of providers, as in the medical-home model, or on a discrete episode of care, as in bundled payments. … If adopted within a framework of overall accountability for cost and quality as is envisioned in the ACO model, both the medical home and bundled payment reforms would have added incentives to support not only better quality, but also lower overall spending growth.”⁴

Perhaps the bottom-line message about the shift from PCMH to ACO-with-PCMH-as-component is the realization that health care payment reform will involve providers taking on performance risk, and that to do that successfully, we will need organizations that can manage populations of 50,000 patients or more. The idea of small practices, even those organized as PCMHs, being able to assume performance or insurance risk is what people are challenging, not the core ideas of the PCMH.

Concomitant with this worry that small PCMHs may not be able to assume risk is the notion that they may not be able to afford the necessary management tools unless part of a larger organization. Jeff Goldsmith, PhD, captures this well in another Health Affairs article: “For this [PCMH] model to work, primary care must be enabled by information technology that can predict and track identifiable health risks. It must also include nonphysician services, such as patient education, that can help patients who are at risk for expensive medical problems to manage their own health more effectively. There have been numerous public and private experiments with this model. A major concern … is whether the medical home concept is becoming so logistically complex that only large group practices or hospital systems can afford the requisite information technology systems and overhead.”⁵

I would encourage AAFP leadership and members to engage in this conversation about ACOs at every opportunity, as I think it bespeaks a maturation and evolution of the best thinking about “systemness” in health care and the role of family medicine and primary care in general within the framework of care delivery and payment reforms. We all know that there are places in this country where small medical practices will remain important, and in some cases the only, resources for community primary care. On the other hand, we recognize the complexity involved in assuring continuity and coordination of care within a complex and sometimes crowded and fragmented provider ecosystem. It's not an “either/or” situation, but a “both/and” environment that demands our flexibility.