Importance of specialist engagement in accountable care organizations.

W e are in the midst of a tumultuous time in health policy. As new payment models gradually replace fee-for-service systems, new quality metrics proliferate, and costs of care become a bona fide health outcome, expensive imaging and procedures face increasing scrutiny. Responsible for >$300 billion in medical spending in 2010, 1 more than any other diagnostic group, cardiovascular disease is understandably at the forefront of many new efforts to improve quality and reduce costs of care. One of the most visible such efforts is the movement toward accountable care organizations (ACOs). As designed within the Medicare program, ACOs are groups of physicians and other providers who agree, as a unit, to provide high-quality care at costs that are lower than projected; if they do so, they are eligible to share in the savings. Currently, there are 368 Medicare ACOs up and running and 154 more in the private sector. 2 The hope, of course, is that providers participating in ACOs cut costs by cutting back on discretionary care where safe and appropriate while maintaining current levels of provision of nondiscretionary, high-value care. In this issue of Circulation, Colla et al use evidence from the pre-ACO demonstration project on which much of the ACO program is modeled, the Physician Group Practice Demonstration (PGPD), to determine whether this in fact took place. 3 Interestingly, the authors find that, despite the investment of millions of dollars into infrastructure, there was no difference in trends in utilization of either discretionary or non-discretionary cardiovascular imaging or procedures between the PGPD groups and local controls. In fact, the groups were remarkably similar in both the pre-PGPD period and the post-PGPD period on all metrics of utilization that the authors examined. 3 Of course, every story has both good news and bad news. First, the good news: There was no drop in the use of non-discretionary cardiovascular imaging or procedures over the study period for PGPD providers relative to controls. This can be interpreted as a safety signal or, in this case, a lack thereof. The authors found no evidence that care was inappropriately withheld from patients who needed it simply because it was expensive. This is highly reassuring, particularly given concerns that have been raised about whether the ACO model and spending targets will lead to rationing or withholding of high-value, potentially lifesaving care. Now, the bad news: There was also no reduction in the …