An Accounting of ACOs

Where they are, what they are, and how many there are
Peter Wehrwein

Certainty is hard to find when it comes to accountable care organizations (ACOs). What they are, what they do — it’s a bit elusive and hard to squeeze into a nutshell.

There are, of course, publicly available rules and regulations for the Center for Medicare & Medicaid Service’s (CMS) Pioneer and Medicare Shared Savings programs. A menagerie of organizations — integrated delivery systems, independent practice associations, hospital groups, hybrid organizations run jointly by hospitals and physicians — are participating.

Meanwhile, detailed information about the ACO contracts that organizations have negotiated with commercial payers is hard to come by. Adding to the confusion is that the word accountable is now very much in vogue because of ACOs and is being used to describe almost any payment arrangement or incentive system that’s not strictly fee for service.

So it’s a relief to talk with David Muhlestein and experience a little bit of clarity. Muhlestein is director of research at Leavitt Partners, the Salt Lake City health consulting company founded by Michael Leavitt, the former Utah governor and health and human services secretary in the George W. Bush administration, and Rich McKeown, Leavitt’s chief of staff at HHS. Muhlestein’s employer has both feet in the ACO consulting business, so he’s hardly a neutral bystander. But Leavitt has been an important source of information about ACOs since early 2011 when the company started keeping a running ACO tally and making it publicly available. After the passage of the Affordable Care Act, Muhlestein says, the company leaders decided that providing ACO advice would be a fruitful line of business, along with health insurance exchanges and Medicaid expansion.

‘We define an ACO as the group of providers’ with accountable care contracts, and often a private and public ACO is the same entity, says David Muhlestein, of Leavitt Partners.

What counts as an ACO?

‘We started tracking ACOs out of necessity because no one else was doing it,’ he says. To conduct their research, Muhlestein and his colleagues scrounge online for ACO news and press releases and survey the organizations they identify.

It’s a business school adage that you can’t manage what you can’t measure. A corollary might be that you can’t count what you can’t define. Muhlestein deploys more elaborate definitions (Figure 1), but for quick, handy comprehension he offers this three-part, flashcard-sized definition of ACOs: a provider-led organization that takes on the financial risk for the health care of a defined population.

Figure 1 What is accountable care?

Provider-led and defined population: Those are pretty straightforward and easy boxes to check. It’s the financial risk aspect where things are muddy. Some provider-led organizations get pay-for-performance programs and care management fees, ‘but they are not really at financial risk for the population,’ notes Muhlestein.

The scope of the risk also matters. Primary care physicians organized into a patient-centered medical home (PCMH) may be at financial risk for the care they provide but not for the care they don’t: ‘If one of their patients is admitted to the hospital, it is not going to affect their per-member, per-month care management,’ says Muhlestein.

In his — and Leavitt Partners’ — book, what separates ACO claimants from the real deal is that the ACO shares financial responsibility for all of the health care of a population and ‘not just your own little slice of it.’ The mechanisms for sharing that financial risk vary and include old-fashioned capitated payments as well as the elaborate shared savings and losses schemes used by CMS and others.

Contracts and covered lives

Muhlestein’s most recent running total shows that the pace of the creation of new ACOs has leveled off this year after a burst of activity last year (Figure 2). The dynamics of early adoption is one reason for the trend line plateau.

Figure 2 Overall trajectory

After the initial wave, more cautious organizations are inclined to wait and see what works for ACOs — and that is far from clear yet, says Muhlestein: ‘All the followers — the organizations coming second — want to see a path forward.’

Meanwhile, though, existing ACOs are negotiating additional contracts, so the number of covered lives has climbed faster than the number of ACOs (Figure 3). Adding contracts means organizations can take advantage of the IT, management, and other investments they made to get the ACO off the ground. A second or third bite of the ACO apple may also allow for some tinkering with quality metrics. Ideally, these metrics are both convenient and inexpensive and truly improve outcomes and reduce costs. The ACO with several contracts may also speed the weaning of physicians and other providers off of the volume-driven fee-for-service practice of medicine. ‘You no longer are living in two worlds,’ says Paul Gardner, a senior analyst at Leavitt Partners.

Figure 3 Covered lives

Muhlestein says he has counted more than 900 ACO contracts and found that roughly a third of the more than 300 organizations participating in the CMS ACO programs also have commercial ACO contracts. Notably, Muhlestein avoids using the term ‘commercial ACOs’ because it suggests the organizations with commercial contracts are different from those in the CMS program when, in fact, they’re often the same. ‘We define an ACO as the group of providers that enters into accountable care contracts and we don’t differentiate with names based on which contracts they have.

‘We try to refer to CMS’s ACOs as ‘ACOs participating in the MSSP [Medicare Shared Savings Program]’ or ‘ACOs in the Pioneer program’; for commercial contracts we call them ‘ACOs with commercial contracts.’ There are some people who refer to the latter as ‘commercial ACOs’.’

ACO map shaped by local factors

Muhlestein has several charts showing the geographic distribution of ACOs (Figure 4). The hotspots vary with the measure (the number of ACOs versus number of lives covered).

Figure 4 Percentage of lives covered by ACOs, by hospital referral region

The reasons that ACOs sprout — or don’t — vary from state to state and market to market, says Muhlestein. Some markets are dominated by brand-name provider organizations that do well with fee-for-service payment and see no reason to join or establish an ACO. In others, when the major provider sets up shop as an ACO, others are tugged along. State policies and programs can be a major factor.

‘To be an ACO is a competitive edge [in Florida]. And if your competitors are doing it, then you’d better do it, too.’

In Oregon, for example, the state’s Medicaid program has encouraged providers taking care of its beneficiaries to set up ACO-like coordinated care organizations (CCOs). According to the program’s Web site, 17 of these CCOs are now in operation, serving roughly 90% of the state’s Medicaid population. In Massachusetts, several pieces of legislation and Blue Cross Blue Shield of Massachusetts, with its Alternative Quality Contracts program, are pushing global payment, and global payment means that providers have to come up with ways to control costs and shoulder risk.

Enter the ACO. It’s certainly not a coincidence that 5 of the 19 ACOs remaining in CMS’s Pioneer ACO program are in Massachusetts. In central Florida, a crowded marketplace has spurred small physician groups to become ACOs, according to Muhlestein. ‘To be an ACO is a competitive edge there. And if your competitors are doing it, then you’d better do it, too.’

Species in the ACO genus

David Muhlestein, director of research at Leavitt Partners, sorts ACOs by the type of sponsoring organization (Figure 5). One source of apprehension about ACOs has been that they would accelerate the consolidation of the health care system into anticompetitive, hospital-led behemoths. But by Muhlestein’s count there are a few more physician group-sponsored ACOs than hospital-sponsored ones (273 versus 256), although the hospital-sponsored ACOs have more covered lives than their physician group-led counterparts (10.5 million versus 7.8 million).

Figure 5 ACO sponsoring entities

Muhlestein has also come up with an ACO taxonomy (Figure 6) that includes six different types of organizations, ranging from what he has dubbed full-spectrum integrated systems — think Kaiser Permanente — to expanded physician groups. Others are also assembling ACO taxonomies. Elliott S. Fisher, MD, MPH, the Dartmouth professor credited with inventing ACOs, recently published a cluster analysis that resulted in a not-dissimilar ACO taxonomy of three groups: large, integrated systems; smaller, physician-led practices; and a hybrid set of groups run jointly by hospitals and physicians.

Figure 6 ACO types

Predictions that ACOs would spell the end of physician groups are premature, in Muhlestein’s view, because groups are figuring out ways to manage care — and attendant financial risk — that they don’t directly provide. Some groups are employing hospitalists to help manage inpatient care and oversee the discharge to outpatient, according to Muhlestein. Others have a common EMR with nearby hospitals, so inpatient care is charted and documented for the physician group. There’s no one right way to do it, he says: ‘There are a bunch of different models out there where the physician group does not directly provide the care but is involved in knowing what’s happening on the inpatient side.’

But if there is one common theme to ACOs, it’s that estimates about how long it takes to set them up with the necessary IT and physician buy-in are overly optimistic.

‘It’s a like a homeowner’s do-it-yourself project. It always takes three times as long as you thought it would,’ says Muhlestein. ‘ACOs are finding the exact same thing.’


Medical Devices Summit Boston, MA February 19–20, 2015
7th Health Plan Innovation Summit on Consumerism Orlando, FL February 25–26, 2015
3rd Annual Summit to Improve Adherence and Enhance Patient Engagement Philadelphia, PA March 9–10, 2015
Value-Added Solutions for Enhanced Customer Experience Philadelphia, PA March 9, 2015

Our most popular topics on