• Did we already try ACOs?

    Some say ACOs cannot work because they are too much like certain, failed managed care contracts of the 1990s. Some say ACOs cannot work because they are not enough like those managed care contracts of the 1990s. Some people are a little bit right even if everyone is a little bit wrong.

    Rick Mayes and I sort it out in a new paper that appears in Health Affairs today. Below is the abstract. I’m preparing a longer summary of it as part of the remarks I’ll make at a Health Affairs forum held at the National Press Club this Friday. My remarks will also appear around 10AM on Friday on the AcademyHealth blog.

    Abstract: A key issue in the decades-long struggle over US health care spending is how to distribute liability for expenses across all market participants, from insurers to providers. The rise and abandonment in the 1990s of capitation payments—lump-sum, per person payments to health care providers to provide all care for a specified individual or group—offers a stark example of how difficult it is for providers to assume meaningful financial responsibility for patient care. This article chronicles the expansion and decline of the capitation model in the 1990s. We offer lessons learned and assess the extent to which these lessons have been applied in the development of contemporary forms of provider cost sharing, particularly accountable care organizations, which in effect constitute a search for the “sweet spot,” or appropriate place on a spectrum, between providers and payers with respect to the degree of risk they absorb.

    The full paper is gated, but please read it if you have access.


    • In your article you discuss several types of payment options.

      In two of them it is clear the payor takes on volume risk: FFS and bundled payments.

      On the other hand, might one argue that payors also take on volume risk under capitation?

      After all, if providers don’t “hit their expected volume targets” for whatever reason, payors would have been better off under a FFS or bundled payment option.

      • If I understand, you’re thinking of a per-person capitated rate, right? So if the provider serves/enrolls more people, that’s the volume risk? How is the payer better off under FFS in that case?

        If you’re thinking of something else, spell it out for me. I’m having trouble seeing how capitation doesn’t shift all the risk that’s possible to shift to providers. It’s a lump-sum payment.

        • Sorry for the delay.

          A provider will have a per-person capitated rate as you metion. Say based on the number of enrollees they received $1M per month in capitated payments. The provider receives their cap payment each month regardless of the amount of care they provide during the month. If the volume of care as measured under a FFS scheme, i.e. billable CPT codes, they provide falls short of $1M, say $800K, the payor would have been better off under FFS, no? Of course if they were good at estimating accurate capitated payment rates and given the nature of health care this is unlikely to be a problem for payors, but it is still a risk.