Pioneer accountable care organizations: traversing rough country.

Abstract

In their article, Nyweide and colleagues1 present results from the first 2 years of the Pioneer accountable care organization (ACO) program. Like the Medicare Shared Savings ACO Program (MSSP), the Pioneer program rewards health care organizations that accept accountability for a population of beneficiaries and score well on measures of cost, quality, and patient experience.2 The Pioneer program gives ACOs the opportunity to gain a greater share of any cost savings they produce but also gives them more risk if costs for their population exceed targets. ACOs are one of the centerpieces of the Affordable Care Act, and understanding how they have performed is critical in the United States. Using a difference-in-differences approach, Nyweide et al compared the cost of care for Medicare beneficiaries for whom the 32 Pioneer ACOs were responsible with cost for other beneficiaries in their areas. The increase in cost per beneficiary was $36 per beneficiary per month less for Pioneer beneficiaries in 2012 and $11 less in 2013. Smaller increases in the cost of hospital inpatient care accounted for the largest share of the difference (nearly 50%), whereas physician services accounted for nearly 25%. Pioneer beneficiaries also had smaller increases in outpatient procedures, imaging, tests, and emergency department visits, although the differences between Pioneer beneficiaries and and those in the comparison group were small. Despite these decreases in utilization, Pioneer beneficiaries’ reported experience of care, including timeliness and ease of obtaining care, access to specialists, and clinician communication, was at least as high as for beneficiaries in the fee-for-service Medicare and Medicare Advantage programs. Nyweide et al provided fewdata relevant to the quality of care. They did show that the proportion of beneficiaries who saw a physicianwithin 7 days of hospital discharge increased significantly more for Pioneer than non-Pioneer beneficiaries. However, therewas no significant difference in readmission rates, which declined for both groups. Limited information available todate fromother studies suggests that Pioneer ACOsare improvingtheirperformanceonqualitymeasuresand perform at least as well as comparison groups.2-4 The relatively smaller increases in costs foundbyNyweide et al are compatible with, but larger than, an earlier Centers for Medicare &Medicaid Services (CMS) estimate2,4 and a recent estimate for the first year of the Pioneer program.3 The differences result frommultiple subtledifferences in themethodologies used in these studies, but Nyweide et al had access to the most accurate data (eg, the physicians and beneficiaries actually included in the Pioneer ACOs, rather than estimates of these physicians and patients) and conducted multiple sensitivityanalyses,whichsupported their estimates, and suggest these are the best data to date regarding cost savings associated with the Pioneer ACO program. Nyweide et al estimated that Pioneer ACOs achieved savings for CMSof $280million in their first year. This represents a savings of approximately 4%. This amountmay seemsmall, but if this rate of savings could be sustained, and achieved throughout a large part of the US health care system, it would be more than enough to “bend the cost curve” so that health care expenditures do not continue to increase as a percentage of the gross domestic product and the federal budget.5 Can this rate of savings be sustained? The Pioneer ACOs produced savings in year 2 that were one-third of year 1 savings. It is possible that during the first year these ACOs were able to“grasp the low-hanging fruit”—toaddress relativelyeasy ways to control costs—and that the savings they generatewill be much smaller, at best, in subsequent years. Alternatively, itmay be that it will take time for ACOs to develop better processes to improve the care of their patients and that they will be able continue to lower costs for years to come.6 Savings generated by ACOs will have little effect on US health care unless a large number of ACOs can do so. The 32 Pioneer ACOs were selected because they are sophisticated organizations thought to be capable of succeeding. But many organizations that have developed reputations for successful “population health management”—organizations like Kaiser and Geisinger—elected not to participate. Of the 32 original Pioneer ACOs, 13 have left the program. There are more than 400 ACOs in the MSSP program, but very few selected the higher potential reward/higher risk track available in MSSP.7 In theory, ACOs should be attractive to physicians. They provide an opportunity to proactively improve care for patients. They are an alternative to other methods of controlling costs, such as cuts in payment rates and extensive use of priorauthorization.8But forACOstobebroadlysuccessful, they will needstronger incentives, closerongoingconnectionswith patients, better logistical support fromMedicare, and regulatory relief. For ACOprograms to growandbe sustainable, physicians andhospitalsmust believe that theywill be at least aswell off financially if they become a high-functioning ACO as they wouldbe if they continuedwithbusiness asusual. Thismeans that theremust be substantial rewards for ACOs that perform Related article page 2152 Opinion