In the Literature
Jul 2015

Accountable Care Organizations: The First Two Years’ Performance and Directions for the Future

Kieran Holzhauer
AMA J Ethics. 2015;17(7):630-636. doi: 10.1001/journalofethics.2015.17.7.nlit1-1507.


McClellan M, Kocot SL, White R. Early evidence on Medicare ACOs and next steps for the Medicare ACO program. Health Affairs Blog. January 22, 2015.

In December 2014, the Centers for Medicare and Medicaid Services (CMS) released a notice of proposed rulemaking that detailed proposed changes to its initial policies governing accountable care organizations (ACOs). The notice prompted health care economists to appraise the performance of ACOs to date. The evaluation posted by McClellan, Kocot, and White on the Health Affairs Blog [1] goes a long way in helping us understand the goals set and obstacles faced by this innovative model for care delivery and physician payment.

Background: Accountable Care Organizations

One provision of the ACA established an avenue for the creation of accountable care organizations (ACOs)—voluntarily formed groups of doctors, hospitals, and other health care organizations that work together to minimize the cost of providing high-quality care [2]—for patients covered by Medicare. Proponents of the ACO model argue that providing a unified health care “home” can improve the quality and lower the cost of health care [3]. Advocates also assert that the role of physicians as leaders in ACOs gives the patient-physician relationship a more central role within the health care system [4].

The Centers for Medicare and Medicaid Services (CMS) has set up several incentive options for ACOs: the Medicare Shared Savings Program (MSSP), which is subdivided into two arms, the Pioneer ACO Model and the Next Generation ACO Model. ACOs that are part of the MSSP can earn “shared savings,” a percentage of the money they have saved CMS, by meeting minimum savings and quality standards. In the “two-sided risk” arm, ACOs can earn up to 60 percent of the savings they generate by reducing the cost of care; if, on the other hand, they fail to meet savings and quality benchmarks, they will face penalties. In the “one-sided risk” arm, ACOs face no penalties for failing to meet these benchmarks but can earn only up to 50 percent of savings realized by CMS. The Pioneer ACO Model operates similarly for the first two performance years, but with higher levels of risk and potential earnings. In subsequent years, Pioneer ACOs can shift to a population-based payment system, in which reimbursements are based on the number of beneficiaries as opposed to individual services provided [5]. The Next Generation ACO Model provides an avenue for experienced ACOs that have adopted the Pioneer or MSSP model to assume higher levels of risk and reward [6].

There are two additional possibilities for physician-owned and rural ACOs. The ACO Investment Model and the Advance Payment Model both provide “prepaid savings,” i.e., advance payments, to encourage new ACOs (particularly those owned by physicians) to form in rural and underserved areas and existing ACOs to take on greater financial risk.

Data: The First Two Years, 2012 and 2013

Early data published by CMS revealed mixed financial results [7]. While CMS-released quality data is still sparse, early quality performance looks positive. However, in neither the MSSP nor the Pioneer ACO Model has a direct correlation appeared between quality and savings [1, 8].

Cost savings. Savings are determined by comparing expenditures to a historical benchmark determined by an approximation of the organization’s per-beneficiary spending over the prior three years [7].

Image removed.

Figure 1. Cost savings in the first two years of the MSSP and the Pioneer Model, 2012 and 2013 [1, 9].

Of the 220 ACOs in the MSSP, 48 percent reduced costs. Just over half of that group met the requirements to earn shared savings [1]. In the first year, 41 percent of the 32 ACOS in the Pioneer Model saved enough to qualify for shared savings, 3 percent owed money, and the remaining 56 percent performed within the minimum range to neither earn nor owe money. In the second year, 48 percent of the 23 returning ACOs earned shared savings, 13 percent owed money, 26 percent performed within the minimum loss or savings range, and 13 percent chose to postpone reconciliation until the third year [9].

Quality improvement. CMS measures an ACO’s care quality using 33 metrics divided among 4 domains: patient/caregiver experience, care coordination/patient safety, preventive health, and at-risk population [10]. The measures are collected through patient surveys, claims, the Medicare and Medicaid Electronic Health Record Incentives Program data, and the Medicare and Medicaid Group Practice Reporting Option Web Interface [10]. In performance year one, pay depends on reporting of quality measures. In performance years two and three, pay for performance standards gradually replaces pay for reporting [11]. Points earned based on the percentile an ACO achieves in these measures factor into the determination of the percentage of shared savings it receives.

Image removed.

Figure 2. Quality improvement in the first two years of the MSSP and the Pioneer Model, 2012 and 2013 [1, 9, 12].

In the Pioneer model, all 23 participants that went on to the second year of the program saw overall quality improvements between years one and two [9]. In the MSSP year 1 data, 209 of the 220 ACOs successfully reported quality measures [12]. Of those 209, 52 also achieved shared savings [1]. Of the 11 ACOs that did not report quality measures, 6 met the financial requirements for shared savings.

Overall, results on cost savings and adherence to program requirements are mixed, but data on quality improvement seems promising.

The Authors’ Analyses

A second article discussing this early data, by Scott Heiser, Carrie Colla, and Elliott Fisher, points to geography as one potential predictor of ACO success [8]. An ACO located in a higher-cost area has a higher baseline cost per beneficiary, which allows for greater comparative savings. McClellan, Kocot, and White note, however, that the minimum cost reduction rates required for particular organizations to earn shared savings are region-specific, based on past performance in the hospital’s referral region. While being in a higher-cost region, such as Florida and Texas, may make it easier to reduce costs quickly, successful and unsuccessful ACOs exist throughout the country; geography does not solely account for varying performance. They

note that the beginning financial benchmark explains less than 10 percent of the variation in early financial performance, and note that there are many successful ACOs in lower-cost areas as well…. Many other factors matter…. Based on our work in the ACO Learning Network, as well as the findings of many case studies and other anecdotal reports, more important factors likely include clinical interventions, analytic capabilities, leadership and culture, and other up-front investments. This suggests that Medicare ACOs will continue to benefit from experience and learning before most are willing to accept downside financial performance risk, a critical objective for the program…. The results also suggest that allowing flexibility and focusing on results…could help more organizations succeed [1].

Potential Ethical Issues in ACOs

Ethical concerns could arise in Medicare ACO programs if the pressure to increase savings and improve quality scores overwhelms the desire to act in the best interest of the health and well-being of patients. Some have expressed concern about the loss of patient autonomy given the ACOs’ financial interest in referring patients to physicians within the ACO [13]. A second concern is the preferential acceptance of patients with low expected long-term health care costs to the organization, which would be an unethical means of reducing costs. If ACOs become the standard model of health care provision, patients who have predictably high costs of care will not have adequate health care options. Over time ACOs could also find loopholes to decrease costs by maintaining quality scores but reducing the actual quality of care by preferentially treating lower-cost patients or cutting corners in areas not directly measured.

The Medicare ACO quality assessment includes metrics that examine physicians’ clinical, communication, and interpersonal skills, but physicians are not asked about their feelings towards the ACO and the care that it provides. Although conflict of interest precludes physician-completed assessments from factoring into shared savings, regular reflection and self-evaluation by physicians could improve the quality of care and physician well-being.The transition to an ACO also changes physicians’ incomes, work environment, autonomy, and routine [4]. The level of upheaval associated with ACOs warrants examination of how physicians are responding to the transition and the changes that they see in quality of life for themselves and quality of care for their patients.

ACOs’ Future

According to McClellan, Kocot, and Wright, a better understanding of the determinants of ACO success will improve the sustainability and growth of the program by reducing participant uncertainty [1]. There is uncertainty, for example, about whether joining a CMS ACO provides an achievable route to lowered costs and improved quality; this deters entry to and triggers withdrawal from ACO programs. They posit that further honing of MSSP policies, particularly surrounding benchmark calculations, will reduce some of the financial uncertainty. A successful ACO program, they argue, rests upon improving current programs and providing more clearly delineated steps for experienced ACOs to continue to improve savings and quality while accepting more risk.

Physicians solely or in conjunction with hospitals lead more than 80 percent of ACOs [4]. Physician leadership has been identified as one of the potential advantages of ACOs because of physicians’ ability to maintain patients’ interests and uphold strong physician-patient relationships [4]. Smaller physician-led ACOs have also been associated with a greater degree of financial success [1], perhaps due to the expedience with which change can be designed and implemented. CMS’s standards require strong clinical and managerial leadership to meet financial and quality goals, suggesting a need for leadership training to be incorporated into medical education. Although the Institute of Medicine recommended that medical schools support the changing organization of patient care, education, and research by contributing to the development of leaders, few medical schools have introduced leadership topics into their curricula [14, 15]. If ACOs are to continue to expand their share of the health care market and if physicians are to continue to take on leadership roles in ACOs, leadership training must be integrated into medical education.

Moreover, having both quality and financial requirements prevents ACOs from being rewarded for reducing cost at the expense of quality or vice versa—the reason the managed-care model became unethical. But these dual requirements diminish the sustainability of these programs; a time will come when continuing to reduce costs while maintaining high quality care is impossible and improvement and innovation can no longer be measured by increased quality and decreased price. This stage will require more creativity, and even then the model may need to be replaced or rethought.


  1. McClellan M, Kocot SL, White R. Early evidence on Medicare ACOs and next steps for the Medicare ACO program. Health Affairs Blog.January 22, 2015. Accessed April 2, 2015.

  2. 2018;

    Centers for Medicare and Medicare Services. Accountable care organizations. Accessed April 10, 2015.

  3. Fisher ES, Shortell SM. Accountable care organizations: accountable for what, to whom, and how. JAMA. 2010;304(15):1715-1716.
  4. Colla CH, Lewis VA, Shortell SM, Fisher ES. First national survey of ACOs finds that physicians are playing strong leadership and ownership roles. Health Aff (Millwood). 2014;33(6):964-971.
  5. Centers for Medicare and Medicaid Services. Pioneer ACO model: frequently asked questions. Accessed April 7, 2015.

  6. Centers for Medicare and Medicaid Services. Next generation accountable care organization fact sheet. March 10, 2015. Accessed April 7, 2015.

  7. Centers for Medicare and Medicaid Services. Medicare Shared Savings Program: shared savings and losses and assignment methodology: specifications. December 2014. Accessed April 15, 2015.

  8. Heiser S, Colla C, Fisher E. Unpacking the Medicare shared savings proposed rule: geography and policy. Health Affairs Blog. January 22, 2015. Accessed April 4, 2015.

  9. US Government Accountability Office. Medicare: results from the first two years of the Pioneer Accountable Care Organization Model. April 2015. Accessed May 26, 2015.

  10. Centers for Medicare and Medicaid Services. Guide to quality performance standards for accountable care organizations starting in 2012: pay for reporting and pay for performance. Accessed April 5, 2015.

  11. In performance year four and beyond, the measures of quality will be compared against quality performance standards that CMS determines each year. CMS enumerates quality data by allotting points for each quality metric for which an ACO performed above the “minimum attainment” level, which is the thirtieth percentile compared to national Medicare fee-for-service claims data and Medicare Advanced Quality data. ACOs must meet minimum attainment for at least 70 measures in each domain in order to remain in good standing in the program. Centers for Medicare and Medicaid Services. Guide to quality performance scoring methods for accountable care organizations. Accessed April 4, 2015.

  12. Centers for Medicare and Medicaid Services. Shared savings program: news and updates. Updated March 3, 2015. Accessed May 26, 2015.

  13. DeCamp M. Ethical issues in accountable care organizations. Virtual Mentor. 2013;15(2):156-161. Accessed April 21, 2015.

  14. Kohn LT; Institute of Medicine. Academic Health Centers: Leading Change in the 21st Century. Washington, DC: National Academies Press; 2004.

  15. Varkey P, Peloquin J, Reed D, Lindor K, Harris I. Leadership curriculum in undergraduate medical education: a study of student and faculty perspectives. Med Teach. 2009;31(3):244-250.


AMA J Ethics. 2015;17(7):630-636.



The viewpoints expressed in this article are those of the author(s) and do not necessarily reflect the views and policies of the AMA.