Friday, June 25, 2010

Putting the Cart Before the Horse on ACOs

Unfortunately, too many discussions about Accountable Care Organizations these days are putting the cart before the horse – defining an ACO based on its ability to participate in a particular payment model, rather than defining the payment system that will best enable an organization to become accountable for the costs and quality of care it delivers.

This wrong-headed approach has been encouraged by the way the federal Patient Protection and Affordable Care Act (PPACA) was written. Contrary to popular belief, there isn’t a section of the law titled “Accountable Care Organizations.” The provisions about ACOs are in a new section of the Social Security Act entitled “Medicare Shared Savings Program”. In other words, the law defines a payment method, and then defines ACOs as entities that can accept that payment method.

What we SHOULD be doing instead is identifying what kinds of changes care delivery systems can make to get better outcomes and lower costs, and then defining the payment changes needed to support those changes in care.

For example, a clear focus for ACOs should be reducing the rate at which people with chronic diseases are admitted to the hospital. We know that can be done. Study after study has shown that things like having nurses make home visits, encouraging patients to call their doctor early, and improving access to primary care practices on evenings and weekends can dramatically reduce ER visits and hospitalizations. But current fee for service payment systems don’t pay for the nurse care managers, they don’t pay doctors to talk to patients on the phone, and the payers don’t even tell PCPs how often their patients are being admitted to the hospital. So the obvious solution is to pay for those things, in return for PCPs accepting accountability for reducing admission rates.

Another example is hospital-acquired infections. Tens of thousands of people still get infections in hospitals, and some infection rates are going up, even though we know infections can be eliminated with appropriate protocols. But under current payment systems, hospitals lose money – a lot of money – when they prevent infections. Again, the solution is obvious – pay for care that has a warranty, the same way we pay for products and services in every other industry.

Those care delivery changes should be central to an Accountable Care Organization, and the priority should be creating payment changes that support those delivery changes. But most people in Washington aren’t talking about those changes. Instead, they’re mostly talking about “shared savings.” Unfortunately, shared savings does nothing to enable the PCP to afford to hire a nurse care manager; it provides no upfront money and only the uncertain prospect of future payments long after the PCP pays the nurse’s salary. Shared savings does little to change the fact that hospitals lose money by preventing infections or readmissions – losing less money is still losing money.

In fact, shared savings does nothing to fix any of the problems with the current fee-for-service system. In other words, shared savings isn’t really fundamental payment reform. It’s just another form of pay-for-performance (P4P), and a pretty weak form at that. And we know that pay for performance has shown very modest success in overcoming the powerful negative incentives built into the underlying payment system.

Shared savings can be useful if it’s paired with appropriate changes in the underlying fee-for-service structure. However, too many people are assuming that the “right” payment reform is merely shared savings on top of the current fee for service system, and nothing else.

Indeed, too many discussions about payment reform today are based on the erroneous belief that we have to give financial incentives to doctors and nurses to provide better care. The reality is that most doctors and nurses WANT to provide better care, they don’t need a financial incentive to do that. The problem is that the current payment system penalizes them for doing the right thing. So the key thing is to start paying for the right things and stop paying for the wrong things, i.e., remove the disincentives.

If we aren’t explicit about how we think care is going to improve and how payment changes will support that, what will consumers think when they hear that ACOs are fueled by “savings?” They’re likely going to assume the worst – that rationing is going to occur. And if we don’t ensure the payment system supports genuine improvements in care, some providers are going to use rationing as the way they create those savings.  If they do that, it will taint the ACO concept for everyone, the same way that bad versions of managed care in the 1990s led to the demise of good versions.

It’s also unrealistic to expect that providers are going to suddenly be able to manage the total cost of care when they haven’t been expected to manage any costs of care for decades. So if a small physician practice or an IPA comes forward and says “we’re willing to take accountability for some things – hospital admissions, ER visits, use of high-tech imaging,” we shouldn’t say “No, you can only be an accountable care organization if you’re able to take accountability for total costs right away.” In fact, we should do exactly the opposite: we should provide payment systems that enable them to tackle the costs that they can, and not penalize them for the costs they can’t control. Shared savings doesn’t do that, because it’s based on total costs: for example, even if a provider successfully reduces ER visits or preventable readmissions, if costs go up somewhere else – in a part of the system they have no control over – they may get no shared savings payment at all to cover the upfront costs they had to incur to achieve the successes they had.  That’s just another disincentive to change.

To make matters worse, even if a provider comes forward and says “I’m willing to take accountability for the total costs of care for a subset of my patients,” they won’t be able to do that, either. Why? Because under the shared savings model, Medicare will determine which patients the provider is accountable for through retrospective statistical attribution, i.e., the provider won’t know who its patients are until after the care has already been delivered.

Think about it – is it better to have only 10% of the providers in the country taking accountability for 100% of their costs while the other 90% are accountable for nothing, or is it better to get 80% of the providers taking accountability for 50% of costs (or whatever portion of costs they can control)? Simple math shows the latter is far better, but a pure shared savings is likely to result only in the former.

Fortunately, although the new Section 1899 of the Social Security Act is titled “Shared Savings,” a separate part of PPACA (Section 10307 of the bill) added language to Section 1899 giving the Secretary of HHS the “Option to Use Other Payment Models.” PPACA says that this can include partial capitation or “any payment model that the Secretary determines will improve the quality and efficiency of … services…” This provision deserves much more attention than it has received to date, because it provides the flexibility that the Medicare program needs to support the specific kinds of care changes that ACOs should be implementing.



  1. Congressional stalemate on the SGA “Doctor Fix” leaves many physicians in organized medicine (AMA) as well as independents quite skeptical about the rank order of the “Healthcare Quality and Payment Reform” monniker. If indeed quality is a primary focus and not a secondary after thought, bureaucrats and economists must realize that a “buy in” in necessary at several levels. Policy makers who have a separate but equal personal healthcare provider option menu, surrender all credibility (think Animal Farm). When our economist and system’s analysts, as this website/organization so adroitly attempts, clothe cost savings in a vestiges of quality improvement, the buy in of the Hippocratic Oath “Primum Non-Nocere…First Do No Harm” clearly relegates patient outcomes to a back seat. The first iteration of the 2 year medical homes initiative was a failure by the metric of patients perceiving no additional value added. Unlike other industries where providers are anonymous production sources, healthcare delivery success depends on a ethical/moral buy by buyers and sellers alike, which so often escapes the MBA curriculum.

  2. Comment by Texas Doctor — June 27, 2010 @ 11:02 am

  3. A lack of governmental buy in is evident when primary provider shareholder/caregivers are so clearly excluded from guideline construction (you will know what is in the bill after we pass it). Breast cancer screening guidelines by the U.S. Preventive Services Task Force in the Nov. 17, 2009, Annals of Internal Medicine. They caused a firestorm then and continue to stir controversy. The line of doctors wanting to comment on the issue when it came up at a committee session stretched out the door.

    The task force statement recommended against routine screening mammography in women age 40 to 49 and called for women age 50 to 74 to receive the procedure every two years. The recommendations were an update of the 2002 guidelines, which said women age 40 and older should consider having a screening mammography every one to two years.

    The task force is sponsored by the U.S. Agency for Healthcare Research and Quality and develops guidelines for preventive services in the primary care setting. Members are experts on prevention in the primary care setting. But many AMA delegates felt that physicians who treat women with breast cancer, such as radiation oncologists, should have been represented. Systematic exclusion in guideline construction is a very effective barrier against caregiver buy in and fosters unneccessary animousity between groups who should be naturally at the same table. Quality, research, and reform will not advance via bureaucratic fiat.

  4. Comment by Texas Doctor — June 28, 2010 @ 7:42 am

  5. I have not been following the implementation of HC Reform (as relates to Medicare beneficiaries) as much as I wish, in order to know whether any regulations to penalize hospitals for certain readmissions (per sets of patient-specific criteria) and to incentivise PCPs and other docs./providers to prevent those very readmissions (ideally characterized by those same pt.-spec. criteria).

    Your piece implies that such regulations have not been made or if made, not yet implemented. And, I know that Dr. D. Berwick’s appointment to finally fill the CMS post with a permanent director (not permanent, but, you know what I mean – someone other than interim) is being delayed.

    When and if such regulations are made and being enforced, I would be so curious to see how might local home care agencies/companies be enlisted (via some arrangement/partnershi0) to possibly take on the care management function you speak of. Similarly, have any ACO pilots/studies utilized this model (vs. the physician practice employing the Nurse Care Mgr. directly) already and if so, is there any data about process and effectiveness?

  6. Comment by Jean Sale-Shaw — July 8, 2010 @ 11:24 am

RSS feed for comments on this post.

TrackBack URL

Leave a Comment

©2008-2011 Center for Healthcare Quality and Payment Reform. All rights reserved.
320 Ft. Duquesne Blvd., Suite 20-J - Pittsburgh, PA 15222 - (412) 803-3650 -