Friday, May 30, 2014

ACO Results: What We Know So Far

May 30th, 2014 

  • by Matthew Petersen and 

  • David Muhlestein

Editor’s note: For more on this topic, stay tuned for the upcoming June issue of Health Affairs, which features a series of articles on accountable care organizations. 
Accountable care is a relatively recent addition to the health care vernacular, but its roots can be traced to the decades-long effort to coordinate medical care. In the United States, health care has evolved into a fragmented pay-for-volume system which has both driven up cost and decreased quality. Coordination of care is meant to reverse this trend.
Through such solutions as Health Management Organizations (HMOs), Integrated Delivery Networks (IDNs) and now Accountable Care Organizations (ACOs), policymakers, providers and payers have sought to consolidate and coordinate patient care. Contemporary care coordination efforts focus on accountable care which increases provider accountability for the cost and quality of care.
The driving principle behind the formation of ACOs is the Institute for Healthcare Improvement’s triple aim: improving the patient experience of care, improving the health of populations, and reducing the per capita cost of health care. One of the broadest applications of this concept is the creation of Medicare ACOs under the Patient Protection and Affordable Care Act. This includes the Pioneer ACO Program and the Medicare Shared Savings Program.
More recently, states have also pursued ACO contracts to cover Medicaid populations. In the private sector, providers have forged ACO contracts with commercial payers. At the close of 2010, only 41 preliminary Accountable Care Organizations existed. The number of ACOs more than tripled to 138 a year after the passage of the PPACA. By 2012 the number nearly tripled again, and by the end of 2013 more than 600 ACOs were operating across the U.S.
In the past year, CMS has begun releasing both financial and quality results from Pioneer and Medicare Shared Savings Program (MSSP) ACOs. Some commercial ACOs have released selected results as well. While results are preliminary and incomplete, both CMS and commercial ACO results warrant a cautious but optimistic outlook on ACOs and their ability to accomplish the triple aim.

Sample Group

The Leavitt Partners Center for Accountable Care Intelligence conducted an analysis of ACO results to determine the cost and quality implications of the ACO model on the U.S. health care system. Information was gleaned from primary and secondary research, including the Leavitt Partners ACO Database of over 620 ACOs. Information about Pioneer and MSSP ACO results was gathered from CMS, and includes press releases, announcements, and data sets.
Data was supplemented with information gathered through interviews and surveys carried out with the leadership of more than a hundred ACOs nationwide. Commercial ACO results were gathered primarily through publically available data such as press releases by affiliated providers or payers and supplemented by interviews with ACO leadership. A breakdown of how many ACOs were represented in our study can be found in Table 1.


Although ACOs share common goals, they vary widely in terms of organization and level of development. Results will be discussed separately for Pioneer, MSSP, Medicaid and Commercial ACOs. Where available, both financial and quality results will be discussed and analyzed.
Pioneer ACOs
Thirty-two organizations began the Pioneer ACO program in 2012. Of these organizations, 23 remain in the ACO Pioneer program. Nine ACOs left the pioneer program, with seven of those transitioning to the MSSP ACO program and two leaving completely.
“We really did learn a lot as a Pioneer ACO,” said the VP of one of the departing ACOs. “However, we’d be better off putting our energy into the health plan we already have… We didn’t have the confidence, based on historical trends, that we could beat the trend. We would have been in a loss position and writing a check to Medicare.”
The Pioneer program generated $147 million in total savings with approximately $76 million in savings returned to ACOs. Of the original 32 Pioneer ACOs, 12 shared in savings while 19 did not share in savings or losses. Only one ACO shared in losses. Addressing these mixed results, the CEO of one Pioneer ACO that neither shared savings nor losses stated, “Our objectives were not to do well in a particular financial cycle. We believe the payoff is going to be accumulated clinical transformation.”
Figure 1

Pioneer ACOs were held to a set of 33 ACO quality metrics, which are also common to the MSSP program. These metrics span four quality domains: patient experience, care coordination, patient safety, preventive health and at-risk populations. ACOs were held responsible only for the reporting of these metrics, not for any quality improvement.
All Pioneer ACOs successfully reported quality metrics to CMS and showed improvement where comparable data was available. In interviews with Leavitt Partners, Pioneer ACO leaders outlined a few tools they used to improve the quality of clinical care including best practices, evidence-based medicine, and electronic health records.
The MSSP ACO program is broader than the Pioneer program with less stringent rules for participation. CMS has released preliminary results on the first two cohorts of MSSP ACOs, which include 114 ACOs that started in 2012. Of the 114 MSSP ACOs, 54 kept costs below budget benchmarks and 29 of those saved more than 2 percent, thus qualifying for shared savings (see figure 2). These 29 ACOs received $126 million in savings and generated $128 million in total CMS trust fund savings. The other 60 MSSP ACOs experienced spending above their set benchmark.
Figure 2

One of the principle differences in the MSSP program is the ability to choose between an upside-risk-only contract (sharing in savings; no risk for losses) or an upside/downside-risk contract (sharing in savings while being at risk for losses). ACOs accepting both upside and downside risk would receive a larger share of any shared savings due to their willingness to risk shared losses. Only four ACOs elected to take downside risk and two of those shared in losses.
The CEO of one ACO that incurred shared losses remained positive when reporting to MedPAC stating, “I’m actually quite optimistic about ACOs as a real catalyst to change the paradigm of care delivery… I’d like to wait and give these ACOs a chance to perform. You know, we haven’t gotten a lot of negative feedback from the marketplace or from our members.”
MSSP ACOs were held to the same aforementioned set of 33 ACO quality metrics. Again, MSSP ACOs were required only to report quality metrics. Failure to do so resulted in forfeiting a portion potential shared savings. All but five MSSP ACOs successfully reported their quality metrics.
Medicaid ACOs
Medicaid ACOs are still in their infancy and have only been adopted by a few states, including Oregon, Iowa, Vermont and Colorado. The maturity of these programs varies widely and little information is available in the way of results. Perhaps the best test case can be found in Oregon where Medicaid ACOs have been designed to cover the entire geography of the state. Detailed financial results released by the Oregon Health Authority (OHA) show that Medicaid ACOs were able to decrease cost of care for 19 out of the 21 financial measures tracked. Areas of cost increases were focused around outpatient primary care. While the overall savings were marginal, the OHA is, “encouraged by the first nine months of progress data.”
In their February 2014 report, OHA highlighted results of their 17 quality metrics. A focus on utilization resulted in a 13 percent decrease in emergency department visits and an 8 percent decrease in all-cause readmission while hospitalization for chronic conditions was cut by a third. Other areas of improvement include technology (EHR adoption has doubled in Oregon), primary care, and preventive care. Colorado’s Medicaid ACO program has also highlighted positive preliminary results including $44 million in gross savings in its second year. Few other state programs have publically released their quality or financial metrics. It remains to be seen if shared savings will offset investment costs.
Commercial ACOs
Perhaps the most diverse group of ACOs are those with commercial contracts. Like Medicare ACOs, commercial payers with ACO contracts strive for the “triple aim” goals of improved patient experience, improved quality of care, and decreased cost of care. However, they are not necessarily held to the same financial requirements, quality metrics, or reporting timeline used by the Center for Medicare and Medicaid Services (CMS). Publically available commercial results tend to highlight mostly positive aspects of a particular ACO.
Results are more difficult to compare than Medicaid ACOs due to their lack of uniformity in measurement and reporting. According to the Leavitt Partners ACO Database, there are 287 ACOs with commercial contracts, only 12 of which have reported financial results of some sort. Eleven of the 12 commercial ACOs report having saved money. Very few of these have reported a dollar figure for savings, but costs were reported to have decreased by between 2 and 12 percent.
Successes include one New England ACO that reported a medical cost trend 1.2 percentage points better than its market overall, as well as a large Northeast ACO which shared approximately $2 million in their contract with United Healthcare. Savings aside, the cost of ACO investment was made clear by one Northwestern ACO that reports spending about $1 million on infrastructure and only earning $125,000 in savings in the first year.

In addition to negotiating their own financial arrangements with providers, commercial payers with ACO contracts also determine their own quality metrics. Some metrics are similar to those set by CMS while others are unique to a specific payer.
Table 2 provides insight into the quality metrics of some of the leading players in ACO commercial contracts. Commercial ACOs have been tight lipped about their quality metrics; quality metrics found in table 2 were garnered from publically available sources and are not a comprehensive list. Commercial contracts focus on preventive care management of chronic illnesses and access to care. Fifteen commercial ACOs reported quality results, although only about 50 percent of those provided quantifiable data.

Winners and Losers

More important than providing a pulse on the ACO movement, the true value of these results lies in their ability to influence those organizations considering entering into the world of accountable care. These results represent a variety of sources including large health care systems, smaller physician groups, private payers, government contracts, etc. This makes them applicable to a wide variety of providers cautiously considering accountable care.
The results here go beyond answering the question “is it working?” They show winners and losers in the ACO game and highlight successful strategies as well as potential pitfalls.