This week the Centers for Medicare & Medicaid Services (CMS) Innovation Center published an update on the Pioneer ACO Model experience. This is definitely a PR piece and it comes off as a desperate salvage attempt. This CMS sponsored program started in 2012 with 32 hand-picked, high-functioning, EMR savvy, medical care organizations. The purpose was to explore the cost-saving and quality enhancing potential of a model Accountable Care Organization (ACO) for the Medicare population. Financial incentives are offered to those organizations who beat certain cost reduction goals. Starting last year penalties will be applied to those ACO’s whose performance falls below established levels. This happened to 6 of the 23 remaining ACO.s. Eleven of them earned shared savings with an average of $4.2 million. The usual bag of quality of care metrics (very few are outcomes) is monitored. In 2013 9 of these organizations pulled out of Pioneer and recently Sharp Health Care in San Diego left. CMS reported $87 million in savings for 2012 and $96 million in 2013. These are not very impressive numbers for 669,135 Medicare beneficiaries. It’s obvious that the program leaves much to be desired especially when one considers that all of these organizations are atypically sophisticated. Nevertheless one of the departing organizations was the University of Michigan. They stood to gain some shared savings but apparently they quit because of administrative complexity and lack of computer interoperability. Interestingly CMS doesn’t reveal what the start-up and maintenance costs have been for the participants. “Innovation pods” have been developed to help generate that information but I have not been able to find any published data.. Apparently some of last year’s cost-saving came from a waiver CMS gave to Pioneer for the 3-day hospitalization rule for admission to a skilled nursing facility. Of course that waiver and those savings could be applied to all Medicare recipients without involving an ACO. So in this instance the savings figure has been “doctored.” Assignment Instability and Alignment are huge problems in any grouped outpatient plan. In a recent study only 66% of beneficiaries were consistently assigned to one ACO over a two year period. To make matters worse, 66.7% of office visits with specialists were outside of the ACO. And, of course, that presents a problem with applying quality metrics. CMS uses claims analyses to assign patients to an ACO. They are considering allowing patients to identify their primary physician and determine assignment based on that. If that worked it still wouldn’t solve the problem of leakage where a patient receives services outside of the organization. The last paragraph of this published viewpoint is interesting. Whereas the piece started out with beautifying words such as “collaborative”, “rich”, “shared”, “sophisticated”, etc., we then see “recognition of the need for more tools.” And finally, “challenging”, “continues to mature”, “fueled by”, “as ACO’s become…”, “as CMS becomes more effective…”, “apply lessons learned”, “development of new models” and ”CMS will evaluate whether these Pioneer results warrant expansion nationally.” And the final throw-away pitch, “Early success in the Pioneer model suggests that in the long term, accountable care will offer patients the improved outcomes they deserve and ACOs the sustainable business model they need to stay focused on delivering high-value care.” In other words, it isn’t working.