Pitfalls of PPACA – Accountable Care Organizations

In addition to Medicare Advantage payment cuts and potential reductions in fee-for-service payment updates, PPACA includes various provisions intended to facilitate ongoing Medicare cost containment, notably creation of the Independent Payment Advisory Board and the Center for Medicare and Medicaid Innovation. In addition to CMI’s broad scope, PPACA requires specific pilot projects, including (in Section 3022) demonstration of accountable care organizations (ACOs).

What does PPACA mean by an ACO? Dr. Elliott Fisher of Dartmouth Medical School, a primary originator of the concept, defined it as “a provider-led organization whose mission is to manage the full continuum of care and be accountable for the overall costs and quality of care for a defined population” and listed several provider groupings that could form ACOs. PPACA provides additional criteria, including having a formal legal structure and administrative systems, meeting CMS requirements for quality assurance and reporting, and serving at least 5000 Medicare beneficiaries. PPACA also specifies a deadline for the ACO pilot: “Not later than January 1, 2012, the Secretary shall establish…a program…”

The goal of an ACO is to reduce costs and improve quality of care through cooperation and coordination among providers, similar to that achieved by integrated delivery systems like Geisinger, HealthPartners, and Intermountain Health Care, but within what may be essentially a virtual organization superimposed on a loose network of providers and covering only a subset of patients.

The ACO concept has been enthusiastically supported by an impressive list of health care experts, plus Dartmouth and the Brookings Institute, but not by Jeff Goldsmith, the author of a critical piece posted on the Health Affairs blog. Goldsmith is particularly skeptical about the difficulty of getting providers to work together, noting “a thundering absence of collegiality – in my view, the central precondition of assuming risk and managing care…” Goldsmith is also doubtful about economic aspects of ACOs: “40% of physicians no longer have any Medicare hospital-related fee income. So squashing hospitals and physicians back together into economic interdependence in a joint hospital/physician economic pool makes no real-world sense.” In a rebuttal piece, Brookings’ Dr. Mark McClellan and others defended ACOs as a critical step away from volume-based health care payment and toward better care at lower cost, but provided no examples of successful implementations.

And that’s the problem. Like some mythical medieval creature, the ACO has not been sighted, other than within existing formal organizational structures in which providers are subject to centralized management—and payment.

This doesn’t mean that we won’t be reading about some ACO successes.  The Brookings-Dartmouth ACO Learning Network has attracted an impressive list of interested provider organizations and is assisting in pilot implementations. However AultCare and Carilion Clinic—the Network’s first pilot sites—both have existing health plans and have much in common with integrated delivery systems. Even so, efforts to revamp IT and financial systems have been substantial, with key details—like how to reward providers for cost savings—still to be worked out, according to a recent article in Modern Healthcare.

Prospects for the virtual forms of ACOs seem much less promising, given the need to create, more or less from scratch, the support systems necessary to make the concept feasible, plus the extreme difficulties of changing the mindset of providers who may demonstrate Goldsmith’s “thundering lack of collegiality.” Vermont—where the ACO concept has been studied since 2008, and where the schedule for pilot project start-up has already slipped a year, to 2011—provides a measure of the time and effort involved, with a recent Commonwealth Fund report providing a lengthy list of “lessons learned” (so far, that is).

An even greater obstacle than lack of provider collegiality may be provider fear of loss of income.  Physicians used to having control over their own fee-for-service world may well hesitate to sign up for a  “share of savings” that is dependent on reduced billings by ACO providers, especially if the share will be net of the administrative costs necessary to support the ACO.

None of this means that Dr. Fisher and his colleagues are wrong in their objectives. If the health care “waste” identified by the Dartmouth Atlas project and other studies is to be reduced, providers must work cooperatively.  It may even be that successful implementation of ACO concepts, first by formal integrated delivery systems, and then by less tightly organized systems with existing central management, will start to put enough pressure on other providers they too begin to look beyond billings to better integrated patient care.

Or more likely not, at least in the near term, given the practical and behavioral problems involved. ACO principles of a continuum of care, resource planning, and performance measurement represent huge challenges to all but centrally managed systems. Unless the CMS ACO pilot is to be based on such a system, its deadline seems unlikely to be met, and yet its objective—to use another medieval analogy—of finding a way to turn the leaden dross of health care waste into the gold of high performance care depends on making the ACO concept work in much looser networks —and CMS has shown very little talent for alchemy.

Roger Collier was formerly CEO of a national health care consulting firm. His experience includes the design and implementation of innovative health care programs for HMOs, health insurers, and state and federal agencies.  He is editor of Health Care REFORM UPDATE [reformupdate.blogspot.com].

Categories: Uncategorized

Tagged as: , ,

12 replies »

  1. Today, I was approached with a proposition to develope a local ACO. It sounds like a new name for an old plan like PPOs and others. Is our memory so short that we can’t recall the core reason for their failure.
    This ACO scheme is being fostered by our government as opposed to the old schemes which were by the insurance companies.
    The scheme goes like this.
    1. Dazzle the pigeons, with increased visions of revenue increases, cost savings, economies of scale, simplified billing, all for the benefit of the patient.
    2. Down play the transfer of risk from the insurance company or government to the ACO, making them the defacto insurance company.
    3. The ACO does not have the deep pockets of the insurance companies or government to cover the risk.
    4. The physicians of the ACO can not be both patient avocate and the insurance denyer/delayer of claims.
    5. The ACO becomes financially unviable.
    6. The socialist government rides in on a white horse to take over and save the day.
    The people who are pushing this plan need to be in close proximity to Bernie Madoff. They could teach him a thing or two.

  2. What does a position profile begin to look like for the head of an ACO. It certainly seems that as ACO’s grow across the country they will present leadership role opportunities, but what does the CEO role for an ACO begin to look like

  3. I am intrigued by this model and the world of possibilities it might open up. Given Medicare’s solvency issues and the need to reduce costs in healthcare – and the tighter, more efficient delivery system portended by an elusive (but promising) reality of ubiquitous HIT interoperability within healthcare communities – could community-wide ACOs (say, headed up by local medical societies) be a solution germane to our times? If the Medicare crisis really is as severe as our medical societies tout – such that a mass exodus of Medicare coverage could pose a crisis for providers and patients both – why could there not be little Mayo Clinics pop up around an ACO model nationwide? That approach might even be what provides a more realistic incentive toward widespread EMR interoperability than the possibly anemic Meaningful Use model or NHIN direct approach. Forgive my ignorance for those more familiar than I with the details; but as a family physician, I’m interested in exploring this.

  4. I don’t know the specifics of the “California” Model. But I am very experienced in the “Florida” model – outside of Miami. Payment and cost levels for MA plans outside of Miami are pretty much national average levels.
    After benefit adjustment down to std Medicare .vs MA reduced cost consumer cost sharing AND using Medicare FFS per capita costs as the “target” the current crop of “at risk provider groups” (PCP based) “get it done” for something like 75%+/- 5% of the current Medicare per capita cost.
    Most of the hospital contracts w/ MA plans are above Medicare allowable, I would trade 100% for hospital for “network leakage” avery single day of the week.
    HEDIS, etc scores are good, so why aren’t these results supportive of a network or physician group structered ACO?

  5. Three issues:
    1) economics…assuming 25% medicare revenue mix, 50% gain sharing, and 10% savings (year-to-year), that means a practice will realize less than 2 points improvement in income. Probably not enough to overcome “thunderous absence of collegiality.”
    2) leadership…where are the leaders that would lead these ACOs coming from? And the management teams they need? Any capable leader would already have secured pre-paid, capitated contracts and built a practice…they would have figured out that they need not give away 20-50% if they went and did it anyway.
    3) what payer is going to invest in the systems work to support the ACO’s patient continuity of care integration with others?
    ACOs are the end objective sure. Question is, what will it take to spawn them besides a yawner gain sharing scheme.

  6. OK, so what options could a major payer (like a medicaid program) look at to spur medical care that met the improved outcomes/decrease cost goals of an ACO model?

  7. My gut and experience in developing IPA model HMOs, their PPO, and EPO derivatives since the mid 80s in California, tells me Jeff is right on in his skepticism. As usual, the ‘devil is in the details.’
    Much of Ellwood’s ‘Supermed derailment’ (one reason we we are where we are), which envisioned regional provider asset concentrations that managed global risk and effectively migrated provider communities from solo, fee for service cultures, and forged them into enlightened clusters of group practices, or physician networks (whether virtual or real) that prudently managed limited medical resources, was due to a mis-guided but VC backed national corporate roll-up of first generation risk bearing IPAs absent both group culture and essential management infrastructure.
    The intentional flexibility written into the PPACA as it relates to ACOs is a good thing. Yet to leverage the freedom, one need ask, what did we learn from the failure of capitation 1.0, and how might we build on those lessons via ‘managed care 2.0, the sequel?’
    At the end of the day, the transformational ‘secret sauce’ may be much more about the intangibles of physician culture, and a patients’ first mindset, than the technical superiority of open vs. proprietary IT solutions, let alone the unwarranted optimism of belt way bandits or ‘out of touch’ academic false profits.
    For a related discourse on medical aggregators, another piece in the failed roll-up and taming of dis-organized medicine story, and the demise of the PPMC industry, see: http://2healthguru.wordpress.com/2010/06/01/the-medical-aggregators-are-we-entering-round-deux/
    Thanks for your piece Roger!

  8. And yet, I look out and see multiple examples where the model seems to be working and sense some keen physician interest out there. The blog response: kvetching about why is shoulnd’t work, couldn’t work, and really can’t be expanded.
    How about some discussion of the take-away from the apparent successes?

  9. Jeff and Roger if you would induldge my lazyness does anything in these regualtions supercede state licensing laws for insurance companies and accepting capitation? It would seem for an ACO to work they would need to be paid a set amopunt then rewarded with a bonus if total cost falls under set goals. Most states I work in this is illegal unless you are a licenses HMO or the payment is comming from a licenses HMO. Requiring the particiaption of HMOs seems counter productive.
    In my opinion if people where serious about trying these out they would open them up to the self funded market. There is no better way to test ideas then the tens of thousands of self funded plans. They are cutting edge and willing to try things like this. Far smarter and knowledgeanble then the politicians and academics that “study” such ideas, each plan is small, if it fails your not crashing an entire system or government program. They are also far quicker to react and progress. Sadly every time we try things like this some DOI issues a cease and disist.
    For example in Ohio I think it would be great for employers to offer their employees a choice of UH or CC and let them choose. We pay each system an agreed upon amount with some small adjustments and a bonus if they outperform expectations. This solves a number of problems we have with PPOs and holding people accoutnable, both the member and provider. We couldn’t get away with this under current state law though.

  10. My problem with ACOs is that I don’t understand the math.
    Assuming for a moment that virtual organizations can be created, the primary care docs will find themselves in the old familiar position of gatekeepers, since it will be up to them to throttle the access to the real money guzzling services of specialists and hospitals. However, even if they do a good job at that, hospitals can still fail at their piece of the puzzle, in which case PCPs will also suffer the consequences.
    And speaking of consequences, what are they? If an ACO manages to show savings, they will only receive a percentage back. So is it better to forgo the savings and get everything to start with? If there are penalties for overspending, will those be allocated based on the biggest spenders in the virtual ACO, or across the board? I guess that depends on who is controlling the payments from CMS and I have a hard time imagining that PCPs will be controlling hospital payments.
    I think it’s more than flippant lack of collegiality. I just don’t see how PCPs in a virtual ACO benefit financially from this arrangement.
    Of course the story is very different for financially integrated entities. Is this what the ultimate goal really is – do away with the independent small practices?

  11. As Roger notes, most of the success stories, including Mansfield, relied upon provider sponsored health plans, where physicians participated voluntarily as part of a medical group or IPA. There are a lot of these examples.
    Rather than trying to encourage more of these formal risk sharing partnerships, the ACO “movement” is trying to be all things to all people, skirting the difficult challenges of peer review, formal fee negotiations, medical policy development, and grafting a gain sharing formula onto a fundamenally inflationary payment model. 5% bonsuses back at the end of a year in which you make most of your money “doing more” doesn’t change anything.
    Virtual organizations, virtual patients, virtually no impact likely.
    Patients and employers need to choose to participate in the ACO, which they could easily do by selecting a provider sponsored health plan. Because health plans are no longer PC, however, we’re trying to have things both ways. Leave dysfunctional structures and cultures alone and pretend the constraints which make managed care so difficult just don’t exist.
    The ACO is a huge consulting franchise, a hobby horse for well meaning, thoughtful academics light years removed from the actual day-to-day operational realities of a troubled health system. It’s not a silver bullet, more like a gumball. . .

  12. What about the Marshfield Clinic? It appears to have worked very well as a demonstration project.

Leave a Reply

Your email address will not be published. Required fields are marked *