• ACOs: one- vs two-sided risk

    I understand that the “shared-savings” financial incentives for accountable care organizations (ACOs) are one-sided. That is, ACOs can only benefit from meeting quality and cost targets. At worst, if they fail to meet them, they’d still get all their fee-for-service payment they’d receive outside of the shared-savings program. Only Medicare is taking risk for high costs, not providers in this bonus-only arrangement.

    A November 22nd letter from MedPAC to Donald Berwick (brought to my attention by HealthReformGPS) suggests that at least that organization is thinking about a two-sided risk model.

    In a bonus-only model there is some incentive to control spending for services from providers outside the ACO. But that incentive is weaker for services the ACO provides directly because the ACO still receives FFS payments for those services and those payments are certain. Receiving a bonus is uncertain because of random variation and even more uncertain if there is a high threshold to overcome. Thus, the incentives in a bonus-only model for controlling spending are relatively weak. […]
    To increase the strength of the underlying incentives Medicare should consider implementing a two-sided risk model in addition to a bonus-only model. In a two-sided risk model the ACO would share in some portion of savings (and be at risk for the same portion of spending over the target). [Emphasis mine.]
    If Medicare really implements a two-sided risk model, ACOs stand a far better chance at making a dent in program costs. Putting some cost risk on providers is exactly what is lacking from our current system, in which that risk is borne by tax payers and program beneficiaries.
    Share
    Comments closed
     
    • Dear Austin,

      It seems the ACO model is just a step along a journey for transformative delivery system redesign and alternative payments.

      I understand the hope from the delivery system for a one-sided risk but the reality is that we must have a two sided risk. I fully expect that in the opening phase of an ACO, we will see one-sided risks. But if the program to continue, I suspect we will see payment tied to various aspects of delivery system redesign which will tied to some verification or measurement processes. In such a program, I envision CMS will pay a lower percentage of the volume based fees for those not meeting compliance. We will see graduated payments for those with larger aspects of compliance. And, we can see shared savings for those who bent the cost curve.

      Tow other points – first, as long as solutions are zero sum, I doubt we will see substantive delivery system redesign. We must seek non-zero sum solutions.

      Secondly, there are several real dangers in the lemming-like activity around ACOs. The delivery system should participate in two-sided risks – but it should only encompass clinical risks. The delivery system must be vigilant that they do not accept insurance risk. We can hold a system accountable for clinical care but not as accountable for whether or not a population is overwhelmed with the frail elderly subjected to a nasty flu season.

    • Two-sided risk isn’t new.

      In the good (bad) old days of physician groups accepting partial cap, many of those groups paid their docs FFS with a 5% withhold for a bonus pool, with a matching contribution from the plan.

    • New terminology , “two sided risk”, for what I have been doing for ten years in the Humana HMO.
      I like the distinction between clinical risk and insurance risk. It seems to me I take both. Please elaborate on this.
      In your example I certainly have to take care of everyone on my defined roster if they all get the flu. If they all need bypasses then they all need bypasses.
      I am forbidden from extracting high risk patients.
      The reward comes after stabilizing the patients from bypass surgery or the flu .
      The risk adjusted reward continues after the patients are stabilized.
      I think I am running a mini insurance company . What do you think I should be doing?

    • @Owen Under the federal PSO rules and many state insurance rules, physicians can accept risk for clinical services they are able to provide themselves. If things go south, the physician might have to work longer hours, but patients still get cared for.

      Insurance risk would include a physician accepting the financial risk for services they don’t provide, like inpatient hospital costs. Regulators worry about solvency in that situation, and patients being left uncovered.

      Bonus pools funded with withholds are different because the downside is generally capped. So you could put a physician bonus pool at risk for inpatient hospital services and not need an insurance license.

      Of course, if this was legal advice, I’d have to send you a bill!

    • a response for the record. This exchange being on the internet it is for the sake of truth some explanation of the “worry of regulators for financial risk for sevices physicians don’t provide.”

      Bankrupcy was a potential eventuality twenty years ago.
      The system now has safeguards and innovative methods.
      One safeguard is reinsurance.
      Another is membership in a pool. We have 190 members.
      A third is the pooling of risk in MSOs covering many lives. We ahve over 25,000 lives.
      A fourth is the requirement for risk taking MSOs to carry a letter of credit.
      I understand it has to start at a million dollars , but I don’t know how much per physician.
      A fifth protection is ability of MSO to select in good effective physicians.
      The sixth protection is the elimination of hopeless losers
      Not every physician can do this. In fact most can’t.
      But there is no penalty for catastrophic cases. that is the reinsurance of the risk.
      Some physicians can not get their heads around using generic drugs, group bargaining power for adjunctive services. and other manner of regular and routine cost saving.
      Effective risk management denies no-one needed care but saves 50 % over fee for service.
      A seventh and most important protection for any system is system auditing for truth and foundation for the decision. Liars do get ejected
      Fraudsters are outed.
      The latter is enhanced by the future eighth protection verifiable information by health information technology.