Uncategorized

What a Physician-Led ACO Can Teach Us about Getting It Right

Farzad MostashariSeveral of the provisions included within the Affordable Care Act in 2011 designate Accountable Care Organizations (ACOs) as formal, contractual entities.

However, in the real world ACOs come in a variety of shapes and sizes.

When compared to larger, hospital-sponsored ACOs, rural and small physician-led ACOs face a tough challenge, because despite limited resources they need to come up with substantial upfront capital and infrastructure investment to establish a strong ACO foundation.

To help ease this burden, 35 ACOs were selected to participate in the Advanced Payment Model ACO demonstration through a grant program from the Center for Medicare and Medicaid Innovation (CMMI). The grants provided a portion of upfront capital to determine whether or not this financial assistance would help ease the startup burden for smaller ACOs, and increase their success rate.

One of those 35 organizations includes the central Florida-based Physicians Collaborative Trust ACO, LLC (PCT-ACO). They are participants in the January 2013 Medicare Shared Savings Program (MSSP) ACO cohort, along with 106 other ACOs.

Larry Jones, PCT-ACO’s CEO, describes his personal mission as an effort to “preserve and protect the independent practice of medicine.” For over 25 years he has been advocating for physicians through their efforts to organize, negotiate with health plans, and other challenges.

He’s led the formation of 18 multispecialty independent physician associations (IPAs), created a large single specialty IPA, and led several management services organizations.

We recently spoke with Larry about his “freshman year” experience in the Advanced Payment Model demonstration, including PCT-ACO’s challenges, big wins, and even bigger lessons learned.

Change is Good.

Jones painted a broad picture of what it means to be an Advanced Payment ACO, while maintaining the independent practice of medicine. PCT-ACO’s 13,500 Medicare patients are managed by 35 primary care physicians across 18 independent practices, none of which are hospital-based.

However, altering the structure of the primary care practices did not occur overnight, and considerable physician buy-in was necessary to achieve such a dramatic change. Jones made it clear to his physicians that with the health care reforms rolling out reimbursements are moving away from fee-for-service, and toward value-based payments.

Therefore, the need to embrace new concepts of care will only continue to grow.

But rather than feel threatened by change, Jones advocates for “putting physicians back in control of the delivery of care and showing doctors that there is a light at the end of the tunnel.

Jones was able to successfully bring his physicians on board with the structural changes, in part, because of his savings distribution model. The MSSP model allows ACOs to recoup a share of the savings if expenditures for beneficiaries are below target cost benchmarks. Jones likens their unique position to being issued a “CMS debit card with $140 million on the tab.”

The ACO is entitled to half the funds left on the “debit card” at the end of the year with the rest of the savings going back to CMS. The ACO is then allowed to divvy up this chunk of savings as they see fit. After infrastructure spending, 90% of PCT-ACO savings are given back to the providers.

Within that 90% of funds, Jones allocates 70% to primary care physicians and 20% to specialists.

If the ACO can reduce total costs by 8%, which he believes is feasible, primary care physicians would receive nearly $100,000 in extra annual revenue.

High-risk can mean high reward. 

Jones knew very well that changing the financing model alone would not automatically bring about savings for their ACO. They also needed to implement some essential care processes to both improve the quality of care and find areas to reduce costs. Jones asked the physicians to identify the highest-risk patients among the 13,500 in their network.

Physicians identified approximately 300 patients based on high emergency department (ED) use, inpatient admissions, hospital readmissions rates, presence of chronic conditions and other factors. Once the claims data became available, physicians were able to identify the 10% of patients driving 60% of the practice’s entire costs (many of whom were already flagged by the clinicians).

Although an analysis of its incremental value has not been completed, Jones (like many others) believes that a provider’s intuition and relationship with their patients are just as valuable as the algorithms created by sophisticated analytics tools.

With these high cost patients identified, providers and the health coaches were able to begin their more targeted efforts to reduce ED visits and chronic disease management. The Advance Payment funds allowed PCT-ACO to bring in new team members, including three health coaches to meet with patients and serve as an extension of the primary care practice.

The health coaches not only support coordination of patient care – a key characteristic of an ACO – but they also act as a resource for patients when they are in need of immediate medical attention. As the coaches became a more integral part of the care team, patients began initiating direct contact with them, instead of rushing to the ER or hospital.

On Measuring and Improving Quality.

In the first year of the program, the ACOs must report quality measures in order to achieve shared savings, but in subsequent years, the amount of shared savings depends on actual performance determined by standardized ACO measures.

These results are reported to CMS for monitoring purposes and maintaining accountability.

However, assembling the data necessary for reporting these measures can be an incredible challenge, particularly for smaller organizations. Fortunately, PCT-ACO has been able to establish direct interfaces between electronic health records (EHR) and their privatehealth information exchange (HIE) for about half of the practices.

In addition, PCT-ACO administers a patient survey to complement the data for the EHR.

On Creating Referral Networks.

 While PCT-ACO has started working on ways to curb spending of their highest-cost patients, Jones found that a significant portion of PCT-ACO’s costs are tied to specialist care. To find ways to reduce overall costs, PCT-ACO has started creating specialist agreements to create a high-quality referral network that lays out the mutual expectations of coordinating patient care and encourages specialists to cover their costs for health IT infrastructure.

In addition, Jones has asked his primary care providers to identify their top seven to ten referral specialist physicians, which has resulted in a list of roughly 200-250 specialists. Eventually, PCT-ACO hopes to establish specialist scorecards that assess care quality, and provide a means of comparison for utilization patterns and resource use.

It is expected that outpatient specialists that provide high quality visits and procedures without costly hospital facility fees may become more preferred.

Looking Ahead.

Though the MSSP provides a critical boost toward accountable care, redesigning practices and organizational change for a small proportion of patients is incredibly difficult. To enhance the financial model, Jones is busy bringing in commercial plans to the mix.

Blue Cross Blue Shield has agreed to contract with them as their preferred network provider in 6 counties, and negotiations with Cigna for a shared savings contract and other large plans are underway.

Jones was able to leverage the ACO infrastructure to secure the largest BCBS contract in the area, and they will manage all of their Medicare Advantage lives across 6 counties.

While the Advanced Payment model does not hold the answers to all of the challenges ahead for small ACOs face, it certainly opens up an opportunity for new ACOs to “get in the game” and begin the transition toward truly accountable care.

Ultimately, the success of smaller physician-led ACOs are essential for the health care field, and as Jones notes, “We need to fight against the mentality that physicians either have to be owned by a hospital or put out of business.”

Farzad Mostashari, MD, ScM (@Farzad_MD) is a visiting fellow of the Engelberg Center for Health Care Reform at the Brookings Institution. He was previously the National Coordinator for Health Information Technology at the U.S. Department of Health and Human Services.

Anna Marcus is a staff assistant in economic studies at  Engelberg Center for Health Care Reform. 

This post originally appeared in the Brookings Up Front Blog

8 replies »

  1. Vince,
    We’re still finalizing our shared savings distribution model for our physician network-managed Medicare ACO. Our working draft has 50% going to primary care, up to 25% to specialists who have collaborated on care of complex patients w/multiple chronic/acute conditions, and 25% being retained by our physician network for IT investments, care coordinators, etc.. PCPs will earn savings (a) based on number of member months their total patient have in our Medicare ACO, and (b) their quality metric scores, severity-adjusted cost metrics, and participation in care management and cost management activities.

    One flaw in the current Medicare Shared Savings Program is that when a group of physicians has their practice purchased by a hospital and leave the ACO, their ACO Medicare patients remain in the ACO for up to a year. How we’re supposed effectively to manage the care of thousands of Medicare patients whose docs are no longer in our ACO is a question about which CMS does not seem to be concerned, unfortunately.

  2. On first glance the closed panel HMO is doing what the ACO is doing.

    There are some important differences however:

    The patient thinks he/she is a FFS patient–the ACO was established to manage FFS Medicare patients–and is not aware that when he arrives
    in a particular doc’s office, that he is now being managed by an ACO. And that it might be only for a portion of care and that the ACO style may not be permanent in his management.

    Because this group of doctors is trying to save money on the patient and to improve care, these providers have qualitatively different principal-agent relationships to the patient compared with the usual FFS Medicare doc that the patient sees.

    The patient must have this explained to him. Viz., that he is now being taken care of by docs who are working together trying to save money on his care AND trying to improve his care and that there might be a financial bonus for the doctors attached to saving money on his care.

    It’s important for the patient to know what is happening to him so that information about this style of care can return to the HHS.

  3. sr,

    I find that NPs or PAs are MUCH more likely to order imaging tests of limited usefulness than MDs (I am a Radiologist and a financial beneficiary of these ordering patterns)

    Where does your information come from?

  4. As one of the specialists discussed, I agree that a lot of referrals involve issues that could be taken care of by a primary provider. With no grander statement or conclusion intended, I find this to occur much more frequently when the primary provider is a NP or PA, rather than an MD. I assume it’s a matter of training and comfort level, but it makes me wonder if the majority of the providers in this type of a physician led ACO are physicians or ancillary providers?

  5. I key in on the 70/20 payment split — 70% to primary care, 20% to specialists.

    This strikes me as a key success factor in catalyzing a primary care led accountable care initiative.

    It also strikes me that this type of split would be a political impossibility in a hospital-led or multispecialty-led ACO. Would love to see some data on how this split occurs elsewhere.

  6. “On Measuring and Improving Quality.
    “In the first year of the program, the ACOs must report quality measures in order to achieve shared savings…”
    __

    Will we EVER get around to outcomes data as the basis for payment? CQMs are hypothesized, tangential process proxies.

    http://tinyurl.com/lmm8wg3