Monday, April 04, 2011

ACO rules: Where's the beef?

I'm sorry, but I just don't get it. Last week, CMS announced proposed regulations about setting up Accountable Care Organizations. Here's the statutory background and the theory of the case, as set forth in the March 31 Medicare Fact Sheet:

Section 3022 of the Affordable Care Act, added a new section 1899 to the Social Security Act (the Act) that requires the Secretary to establish the Shared Savings Program by January 1, 2012. This program is intended to encourage providers of services and suppliers (e.g., physicians, hospitals and others involved in patient care) to create a new type of health care entity, which the statute calls an “Accountable Care Organization (ACO)” that agrees to be held accountable for improving the health and experience of care for individuals and improving the health of populations while reducing the rate of growth in health care spending. Studies have shown that better care often costs less, because coordinated care helps to ensure that the patient receives the right care at the right time, with the goal of avoiding unnecessary duplication of services and preventing medical errors.

Here's the introductory paragraph from the CMS summary:

ACOs create incentives for health care providers to work together to treat an individual patient across care settings – including doctor’s offices, hospitals, and long-term care facilities. The Medicare Shared Savings Program will reward ACOs that lower growth in health care costs while meeting performance standards on quality of care and putting patients first. Patient and provider participation in an ACO is purely voluntary.

How will this work? And, will it work?

Let's dig in.

The proposed rule would require providers participating in an ACO to notify the beneficiary that they are participating in an ACO, and that the provider will be eligible for additional Medicare payments for improving the quality of care the beneficiary receives while reducing overall costs or may be financially responsible to Medicare for failing to provide efficient, cost-effective care. The beneficiary may then choose to receive services from the provider or seek care from another provider that is not part of the ACO.

. . . Medicare would continue to pay individual providers and suppliers for specific items and services as it currently does under the fee-for-service payment systems. The proposed rule would require CMS to develop a benchmark for savings to be achieved by each ACO if the ACO is to receive shared savings, or be held liable for losses. Additionally, an ACO would be accountable for meeting or exceeding the quality performance standards to be eligible to receive any shared savings.

So, the PPO character of Medicare would not change: "The provider may not require a beneficiary to obtain services from another provider or supplier in the same ACO."

How can you be held accountable, as a provider group, if you cannot control the management of care of your patients? I'm not blaming CMS for this contradiction. The agency is simply implementing what Congress and the President ordered it to do. There is no way Congress will limit choices among the Medicare population.

Real cost savings will not result from ill-conceived government laws and regulations: They will occur when physicians and other health professionals redesign the work that takes place in their offices and hospitals. Attempts to generate that redesign by government regulations, especially self-contradictory ones like this, will fail.


Ileana Balcu said...

I think this happens through referrals: I would go to the doctors/labs/testing centers that my doctor sends me to unless there is a good reason not to do it: either it's faster or cheaper or I am treated better somewhere else. So my doctor has the incentive to make sure that everyone else in the group works in a competitive way, my doc warmly recommends the other group partners to their patients and their patients agree with the recommendation.

Maybe this is just theory and it really won't work, but it doesn't seem that stupid to me.

Anonymous said...

Coordination sounds good. Like patient participation, which it requires. But I don't see either as explicit and accountable goals in ACO contracts. If organizations and providers do not share risk contracts or metrics accountable to coordination (including patient involvement in care), this is no more than hopeful sales by CMS.

I've seen very few metrics of care coordination. And process improvements in most institutions still dance carefully around the actual behavior of providers. Although coordination has come a long way where technologies and payments are salient (pathology, radiology), real coordination of care requires appreciation by hospitals of the costs and constraints of service siloing, and of providers of the costs and constraints of medical decision autonomy to patient health and collegial reciprocity. That is not a discussion I see in the public sphere.

Will a diabetic patient's cardiologist speak to her primary care provider before changing medications? Will the endocrinologist read the depression screen? Will the nurse call to see if the patient was able to find a psychiatrist? When ACOs incentivize these behaviors, we'll know we are really getting somewhere with this model.

Barry Carol said...

Under one CMS payment model, as I understand it, ACO’s will not even be penalized if they exceed cost targets during the first two years yet they will still be eligible for a bonus if they save costs and meet or exceed quality metrics. I don’t see how ACO’s can work unless there is potential financial risk as well as reward.

I think cost savings are more likely if we can develop and implement interoperable electronic medical records and tiered provider networks that subject patients to higher coinsurance if they use more expensive, less cost-effective providers. Hospitals could reduce avoidable harm through more consistent use of checklists, hand washing and the like. Better discharge planning could reduce readmission rates. Paying for palliative care consults and storing patient end of life preferences on a registry could cut futile and often unwanted care at the end life. Substantive tort reform could mitigate the excessive use of imaging. Price and quality transparency tools can help referring doctors to more easily identify just who the most cost-effective providers are. In short, there is a lot we could do to attack the growth of healthcare costs without the dubious ACO structure. Providers, insurers and regulators should pursue these approaches on their own with or without CMS support and political cover

Toni said...

From Facebook: You are so right. This has fatal flaws in the basic design. Right now ACOs seem to be the stimulus act for consulting firms.

Keith Marple said...

Why all the negativity? Yes, the Shared Savings Program is clearly a half-measure. Yes, the lack of closed networks or risk-sharing greatly lessen the potential impact.

But what's the cost? If it fails, it pays out some small amount of savings to organizations that are already performing well and arguably deserve the bonus payment. And if it succeeds, it only pays out a portion of the savings it realizes, becoming a net positive for our government.

This no-risk program allows providers to put a toe in the ACO waters and start internally incentivizing collaborative behavior. Commercial payers are bound to follow suit with similar incentive programs, further aligning incentives. And if it's successful, you can be sure there will be follow-on rounds of legislation to extend the program into partially-capitated models.

Baby steps, people.

Barry Carol said...

Keith – I think there could be considerable costs to establish ACO’s. Management infrastructure needs to be organized and established. IT costs may be considerable. It’s quite likely that ACO’s could lose money during the first two or three years before the upfront investments start to pay off. At the same time, management time and attention may be diverted from other approaches that could have a more immediate and lasting affect on healthcare costs. Finally, under CMS’ approach, it will take several years to determine whether or not the effort was successful.

Patients will be allowed to seek care outside of the ACO which makes it harder for ACO’s to control costs. A tiered network approach that requires patients to pay higher coinsurance to access more expensive providers would make more sense. Also, it would be better if patients benefitted directly from lower premium payments if they get their care from more efficient and cost-effective ACO’s. There is also the potential for hospital led ACO’s to lead to even more concentration of provider market power than exists now resulting in higher costs. The FTC and DOJ need to be mindful of this.

Anonymous said...


Apropos of anti-trust was an article in today's WaPo:

These cases seem few and far between, however, compared to what we can see with our own eyes in places like Boston. I think they must be difficult to prove, like many of the securities fraud cases we have seen.


Barry Carol said...

Nonlocal – I think any discussion between a hospital and an insurer about how much that insurer will pay other hospitals or whether it contracts with them at all should be strictly off limits. At the same time, a so-called Most Favored Nation (MFN) clause whereby an insurer strikes a deal that guarantees it will pay a hospital no more than the lowest amount accepted as full payment from any other commercial insurer is probably OK. Medicaid insists on similar clauses in its prescription drug contracting, I believe.

To illustrate the disparities in payment rates, I recently heard the CFO of a large healthcare system, which included a well known AMC, state that the payment rates it received from the local Blue Cross plan were about half of what it was able to collect from other commercial insurers who lacked the Blue’s local market power. To what extent the local Blue plan insisted that the hospital system charge its competitors such large premiums over what it paid, I have no idea. This is yet another illustration of why I think disclosure of actual contract reimbursement rates would be a good thing.

Paul Levy said...

I agree. Unless you believe that market power should be determinative of rates, there is no reason to hold such matters confidential. But there is no reason why market power should be determinative in such an essential sector of society.

Keith Marple said...

You're right that market power shouldn't determine payment rates. So what's one good way to remove that from the equation? Fully-capitated ACOs, which allow providers to compete on cost and ability to keep a population healthy.

The Shared Savings Program, much like the national health care reform bill, doesn't solve all or most of the issues but is a step in the right direction. It's not tiered, or closed network, or even partially-capitated. Provider consolidation is definitely an issue, as is the risk that providers will start acting like insurers and dropping their sickest patients. It's definitely not perfect; that's the reality of the political process.

But the program allows providers to start getting rewarded for improving patient health, which isn't always the case now. And it serves as a large-scale proof of concept, which addresses the need for additional evidence, and if successful would make a strong case for more aggressive models. While providers may expend some cost and mental energy re-engineering their systems, the ACO model doesn't dictate how providers achieve low costs and high quality, so existing initiatives can continue unabated. And don't we all want providers focused on how best to lower costs and keep their patients healthy?

Paul Levy said...

I think you are mixing up concepts. Full capitation does not address market power, unless you are saying that all ACOs should get the same capitated rates from the insurers. How will you ensure that this happens? It either requires direct rate-setting by a state or full disclosure of the rates paid.

I also think you fail to address the point that I made above. You can't get a real proof of concept in a PPO environment.

Keith Marple said...

Thanks Paul, as always you bring up really good points. I was envisioning a flat PMPM rate for participating ACOs, but I guess I hadn't fully considered the mechanisms to create that flat rate. Rate setting probably would be required - maybe patients could choose to pay out of pocket for costlier ACOs, a la HPHC?

One thing I still don't understand is why the program, as currently constituted, can't serve as a large-scale test. Let's say we're three years in, and both the risk-adjusted costs and health of patients served by participating providers improved more (or worsened less) than the control group of patients served by non-participants. Couldn't that prove that the incentives had had a positive effect, PPO model notwithstanding?

Thanks again for the insight and terrific blog.