Of Rising Risk and ACO Success in the Medicare Shared Savings Program: Community Health Network and Pharos Innovations

To each according to their need, from each according to their ability.

While that famous (and paraphrased) adage was originally used to attack capitalism, it's also not a bad way to think about effective care management programs.  Ill persons have varying needs, and care providers have varying resources.  In a resource-constrained environment, the health system that matches the right needs with the right resources will win.

A case in point may be one of the few Medicare Shared Savings Program (MSSP) ACOs that met CMS' savings threshold.

"Community Health Network" is a joint venture between HealthEast Care System and Entira Family Clinics that serves Medicare beneficiaries in Minnesota's Twin Cities.  According to this Pharos Innovations case study (which can be found here), the organization analyzed their population's past utilization and future risk and realized that a relatively small group of beneficiaries were high need.

High need was defined as those individuals with "rising risk" or what the Population Health Blog has termed the "sweet middle." These are individuals who are not only burdened by chronic medical conditions (such as heart failure), but have a constellation of issues (such as a recent discharge from a hospital) that also can be mitigated or impacted.

Once these patients were identified, Pharos Innovations provided the "transition coaches" and "engagement specialists" for the high need/rising risk/sweet middle patients who were most likely to have an admission and/or readmission.

By pinpointing state-of-the-art heart failure treatment protocols, care management, self-care coaching and discharge-care interventions, avoidable admissions and avoidable readmissions - compared to a control group of patients - dropped in significant manner. That was enough to skew the ACO's overall admission and readmission costs downward and reach CMS' savings threshold.

Population Health Blog lessons:

1) If a healthcare organization is willing to take on the financial consequences of health insurance risk, it will have to array its covered population's risks from high to low, and deploy interventions that can address the needs of patients at high - yet modifiable - risk.

In other words, within any high risk subpopulation are sub-populations with specific health care needs that can be addressed with smart population-based care management interventions. 

2) The PHB looks forward to these data appearing in the peer-reviewed literature.  In the meantime, kudos to Pharos Innovations for achieving credible outcomes based on a comparison to a valid control group.

3) Patients who don't have to go into a hospital are better off for it.  Despite Karl Marx's antipathy, aligned economic incentives can be win-win for everyone.  Community Health Network deserves to be financially rewarded by CMS.

4) There is a good reason to believe that Community Health was destined to do well versus the other MSSP participants. As pointed out here, physician leadership may be one ingredient to the MSSP ACO secret sauce. According to the case study linked above, Community Health Network expended considerable resources to convince their physicians that this approach was a good idea.  What's more, their board is majority physician controlled.  In addition, Pharos Innovations was started by a doc and they have additional physician representation on their board.

The Majority of Medicare ACO Participants Appear to Have Lost Money in 2014

There's no other way to put it.

Like many wonks, the Population Health Blog glommed onto this recent CMS report report on the 2014 performance of the Pioneer and Medicare Shared Savings Accountable Care Organizations (ACOs). 

While there's some quality reporting data, the PHB decided to focus on the economics.

It ain't pretty.

Briefly, as the PHB understands it:

The 20 Pioneer and 333 Medicare Shared Savings Accountable Care Organizations generated a total of $411 million in savings.

Among the Pioneer participants:

15 out of 20 generated savings.  Only 11 of the 15 earned enough savings to trigger a payment from CMS that totaled $82 million.  The PHB calculates that's an average payment of approximately $7.5 million for each ACO. 

Three of the Pioneer ACOs had to provide clawbacks to CMS of $9 million, or an average $3 million each.

Of the 333 Medicare Shared Savings participants:

92 out of 333 saved $806 million in health care costs.  They received checks totaling $341 million.  The PHB calculates that's a payment of $3.7 million per ACO. 

Another 89 of the Medicare Shared Savings reduced costs, but not enough to trigger a payment from CMS.  That also means that the rest of these ACOs didn't even reduce costs.

The PHB's conclusions:

ACOs in the Pioneer program have about a 50% chance of getting some money back.  Assuming that there are from $2 million to $7 million per year in program support costs - in addition to the all of the foregone billable services - it's not clear to the PHB that the business model is sustainable (for example) for many of the Pioneer participants.  To add downside-risk insult to injury, there's a 15% chance a Pioneer ACO would have to pay Medicare.

ACOs in the Shared Savings program have a 75% chance that they won't be able to generate enough savings to cover the lost of income from fewer billable service or their program costs.

That's a majority of the participating ACOs.

Admittedly, there are several advantages to ACOs.  They 1) are an answer to the threat of rising health care costs, 2) are a laboratory for bundled payments, 3) promote care coordination and 4) are linked to medical homes.

But that's all for naught if the majority of the program participants are losing money in a massive exercise in risk transfer involving hundreds of millions of Medicare dollars.

This is health reform?

++++++++++++++++++++++

Coda: The PHB can't help noting that the title of the CMS report is "Medicare ACOs Provide Improved Care While Slowing Cost Growth in 2014."   That may be technically true, but that title is more about spin than about the science. The findings haven't been submitted to the scrutiny of peer-review, and until it is, the PHB won't really know what to believe.

Of Medicare Shared Savings Program (MSSP) ACOs, Start-Up Costs, Preliminary Financials and Data Support

The Disease Management Care Blog didn't know there was a "National Association of ACOs" either, but they've just released results from a "web" survey of the organization member ACOs that are participating in the Medicare Shared Savings Program.  You can read more about the Program here. 

Of the total number of 123 MSSP participants, 35 anonymously participated in the survey. Their covered beneficiary numbers ranged from 5,100 to 78,000.

Among the findings:

Start up costs in the first year of operation averaged $2 million, with a range from $300,000 to $6.7 million.  With continued operations, the average cost over two years was $3.5 million. Total capital needs averaged $4 million.

While Medicare has yet to release any formal financial results, the ACOs' estimated results showed that 13 guessed they would break even. Nine will gain an average of $1.3 million and six will lose $1.3 million.  Six had no estimates, while other gains and losses ranged from positive $9 million to negative $10 million, respectively.

The biggest problem? "CMS data and learning to access it and process it."  This required pricey information technology with an average of $413,000 internal and $443,000 external costs.

The DMCB's take:

Running an integrated delivery system or physician-hospital organization as part of an ACO is a very expensive and capital intense enterprise.  Given the additional costs of new technology, electronic records and personnel, some of the ACOs may not be able to afford the loss of millions of dollars.  It remains to be seen how Medicare will handle the downside of hospital lay offs or clinical program discontinuations among some of the MSSP participants.  Will members of Congress have to get involved on behalf of their local constituents?  Stay tuned!

As the disease management industry learned, it's one thing to "save money," it's another to save money in excess of fees plus program costs.  $3.5 million over two years is a lot of money to make up before you break even, making the DMCB wonder if cheaper programs (such as population health management) with a more modest scope (such as reducing readmissions) may have a better long term value proposition. Once again, stay tuned.

And the Disease Management Care Blog predicted there'd be problems with the data feeds here.  Recall that one of the problems with the Medicare Health Support program were "execution" problems with the timely provision of utilization data from CMS.  ACOs - and the Medicare beneficiaries they're taking care of - deserve better.