Editor’s note: See additional analysis of the first year of Pioneer ACO results from Joseph Damore and Wes Champion.

The Centers for Medicare and Medicaid Services yesterday announced results from the first performance year of the Pioneer Accountable Care Organization Model.  The Pioneer Model should be seen as part of a crucial phase of testing alternative payment and delivery models in an effort to achieve greater value in health care.  One year’s results should not be seen as a definitive outcome or leading to a dispositive conclusion, but rather as a valuable source for learning.

The Pioneers represent 32 of over 425 public and private market ACOs reportedly in operation nationally.  The two-sided risk model they assume is more advanced than the shared-savings-only approach of the Medicare Shared Savings Program Track 1 and most commercial ACO arrangements, but less advanced from a full risk perspective than many capitated provider arrangements in place around the country.  All of these efforts bring greater definitional clarity to accountable care and accountable care organizations.

The participating providers are required to be both financially and clinically integrated.  The ACOs must, among other core attributes, have an effective governance and leadership structure, have the ability to apply evidence-based medicine and care coordination processes, meet quality measures, have a savings distribution formula, develop a robust electronic health record infrastructure, and, importantly, be able to effectively engage patients in their care and their health.

ACO skeptics clearly remain.  Among the doubts and cautions that have been raised are that ACOs will drive insufficient change in physician behavior and patient engagement, result in insufficient savings, create a specialist (and thus patient) backlash, suffer from lack of agreement over measures and metrics, and drive up prices due to consolidation.  These are legitimate concerns.  But such concerns should not mean that the testing should slow down.  Rather, in my view, it should accelerate, given some early positive results and the ongoing cost and quality challenges faced by our health care system.

CMS reports overall quality improvements and some measurable savings in the first year of the Pioneer Model.  According to CMS, all 32 Pioneer ACOs successfully reported quality measures and achieved the maximum reporting rate for the first performance year, with all earning incentive payments for their quality reporting efforts.  Thirteen of the 32 produced shared savings with CMS.  Two had shared losses.  Commentators likely will debate the degree and significance of those results.  MSSP first year results should be available later this year.

The broader ACO picture.  Meanwhile Cigna, Aetna and United, along with other insurers, are stating publicly that they expect to put hundreds more ACOs in place over the next several years.  And early reported results from various non-Medicare ACOs currently in operation are promising:

  • NewHealth Collaborative (Summa Health System) lowered its costs by 8.4 percent in its first year as a Medicare Advantage ACO, largely because of reduced hospital use, including a 10 percent reduction in readmissions.
  • Growth in health care costs for North Shore-LIJ employees under a full risk, self-insured plan dropped to less than 2 percent in 2011; they anticipate similar slow growth for 2012.
  • The Accountable Care Alliance and Nebraska Medical Center found that costs for enrollees in their population management program rose just 4.2 percent over the past five years, compared with 27.4 percent nationally.
  • University Hospital Case Medical Center reported a drop in ER use and length of hospital stays its first year, as well as more attention to wellness.
  • Kelsey-Seybold has reduced costs by as much as 20 percent compared with other organizations.
  • Under the Advocate-Blue Cross agreement in Illinois, hospital admissions are down 6 percent; days spent in the hospital are down nearly 9 percent; the average length of a stay has declined; Advocate has achieved a small but significant savings of about 2 percent below projected costs.
  • Mission Point is up to 50,000 members in its ACO and has cut medical costs each year for its 15,000 original members by 12 percent.
  • NovaHealth ACO experienced a 50 percent reduction in their inpatient hospital days, 45 percent lower hospital admissions, and 56 percent fewer readmissions in their Medicare ACO product in 2012.

Pioneer Model ACOs served 669,000 beneficiaries in their first year according to CMS.  Estimates are that a total of 4 million beneficiaries in 47 states are currently being served through Medicare ACO programs (250+ ACOs), mainly through the MSSP.  Another 8 to 14 million beneficiaries reportedly are in private sector ACOs.  At least 14 states are developing accountable care models in Medicaid and CHIP.  Thus, while the testing in the first year of the Pioneer Model is instructive, it is only a small part of the learning system.  When bundled payment programs and the various value-based and other performance-based payment arrangements in place are added, the field of learning can be seen to be quite broad.

Moving toward transformation.  Much more experience will need to be gained and research done to reach more firm conclusions about ACO results.  The MSSP program, based on CMS learning over its first three years, likely will undergo some changes in 2014.  Many of those who have been early ACO adopters indicate anecdotally that the early years mostly allow for the capture of the proverbial “low hanging fruit,” and that the real challenge will be to transform care permanently and materially over the next several years.  Nevertheless, a growing number of payers and providers, and some employers, have committed to the transformational change that is signified by the “volume to value” dynamic.

Last September, the Institute of Medicine released an important report entitled Better Care at Lower Cost:  The Path to Continuously Learning Health Care for America.  As the IOM concluded in the report: “It is only through shared commitments, in alignment with a supportive policy environment, that the opportunities offered by science and information technology can be captured.  The nation’s health and economic futures—best care at lower cost—depend on the ability to steward the evolution of a continuously learning health care system.”

The Pioneer Model is an important part of that learning system.  The fact that not all participants achieved savings in the first year should not be seen as a weakness or a failure.  Rather, it supports the view that this is a legitimate test, that not all Pioneers will do equally well across quality and cost measures at any given time.  The fact that 7 Pioneers are moving to the MSSP and that 2 are dropping out also shows that testing results in learning.  The question should be: Does the program help us to learn and therefore to move in the right direction?   I believe the answer is yes.