The Trouble With Accountable Care Organizations
Dr. Ashish Jha did us all a favor recently by pulling back the curtain on the Obama administration’s recent press release touting the supposed success of the Accountable Care Organization (ACO) effort.
The administration claimed that ACOs operating under the Medicare Shared Savings Program (MSSP), as opposed to the Pioneer ACO or “Next Gen” ACO demonstration programs, reduced Medicare’s costs in 2015 by $429 million. But that figure excludes the payments made by the federal government to those ACOs having savings that exceeded a certain threshold which made them eligible for bonus payments. Include these bonus payments in the calculation, and the MSSP ACO program actually increased Medicare spending by $216 million in 2015 — a rather different bottom line from the one implied by the press release.
Furthermore, as Dr. Jha notes, nearly as many MSSP ACOs lost money in 2015 as saved money. And the ones that saved enough money to be eligible for bonuses were concentrated in markets with high benchmarks, raising the possibility that only ACOs in excessively costly regions are able to reduce costs in any significant way.
The continuing underperformance of the MSSP ACO program is complicating the Obama administration’s preferred narrative of recent health-care history. The administration has gone to great lengths to suggest to the media that cost escalation is slowing down throughout the health sector, that the Affordable Care Act's (ACA) “delivery system reforms” are an important reason for this development, and that the ACO initiative is the most important of the ACA’s delivery system reforms.
Unfortunately for the administration, its explanation of the cost story doesn’t stand up to the slightest scrutiny. For starters, to the extent that there’s been a slowdown in cost growth over the past decade, it predates the enactment of the ACA by several years. And the Congressional Budget Office has estimated that the much-discussed “delivery system reforms” of the ACA are minor events at best, even if they work as planned. The biggest cuts in spending in the ACA aren’t from these provisions but from blunt, across-the-board cuts in Medicare that almost no one believes can be sustained over the long run.
As for the ACOs, if they’ve produced any savings at all — which is questionable — the total of the previous four years is less than a rounding error in the nation’s massive $3 trillion per year health system.
ACOs were conceived as an alternative to insurance-driven managed care. Prior to the ACA, Medicare beneficiaries already had the option to enroll in private insurance plans through Medicare Advantage (MA), including scores of HMOs with decades of experience in managing care. The authors of the ACA wanted to give beneficiaries another option: provider-driven managed care. ACOs must provide the full spectrum of Medicare-covered services, so that means hospital and physician groups must work together to provide patients with the full spectrum of care. But there’s no requirement for ACOs to accept a capitated payment or operate like an insurance plan.
The fundamental problem with the MSSP ACO effort is the method of beneficiary enrollment. The ACA stipulates that a beneficiary is to be assigned to an ACO if the beneficiary’s primary physician has joined the ACO. Beneficiaries don’t really have a say in the matter and are never really informed in a clear way about their assignment to an ACO. They are under no obligation to get care from the providers within the ACO’s network and can see any physician they want to under the usual rules of traditional fee-for-service Medicare. (The Next Gen ACO demonstration is testing the payment of incentives to beneficiaries for staying within the ACO network for care.)
This assignment of beneficiaries to ACOs has undermined the ability of the MSSP ACOs to operate like genuine managed-care entities. The patients have no incentive for, or interest in, complying with the plan’s effort to control costs, and many times the physicians don’t have any real idea who among their patients is in the ACO.
The recently enacted “doc fix” legislation, called the Medicare Access and CHIP Reauthorization Act, upped the ante on ACO coercion. In future years, physicians will get paid more by Medicare only if they join an alternative payment model, which effectively means that they will have to join an ACO to get any kind of reasonable increase in their fees. And when physicians join an ACO, their patients will automatically come with them.
The administration is hoping eventually to herd all of the nation’s physicians — and thus the vast majority of Medicare beneficiaries — into ACOs by effectively giving them no other choice. But this won’t lead to “delivery system reform” or a more efficient health system. Rather, it will lead to widespread resentment among physicians and beneficiaries alike because neither will have fully consented to participate in the ACO model. The result will be a care delivery system that is indistinguishable in reality from unmanaged and inefficient fee-for-service, albeit with lower payments from the government.
A better approach would be to trust the Medicare beneficiaries to make choices for themselves. ACOs should be converted and rebranded into genuine, provider-driven integrated delivery networks (IDNs), with less regulation by the government but stronger incentives to cut costs to attract enrollment. Beneficiaries would be given the option to enroll in competing IDNs, Medicare Advantage plans, or the traditional fee-for-service program and would pay higher premiums for enrolling in the more expensive options. Competition based on price and quality would push the IDNs and the MA plans to improve their performance each year.
The MSSP ACO program has been in place now for four years, which is long enough to see that it’s not going to deliver what was promised. The problem is fundamental: Managed care plans that are formed based on assignment of beneficiaries, rather than consumer choice, will never have the legitimacy that comes from a patient’s genuine consent to submit to far-reaching changes in the care-delivery process. A very different approach to provider-driven managed care is required for that — one that relies less on government regulation and more on strong competition in the marketplace to deliver higher-value care for patients.