Moving Beyond Fee For Service (Part Two of Three)

In the last post, I reviewed some of the perverse incentives of fee for service payment for medical care.   In this post, I will talk about some of the key elements to a transition away from fee for service.   In the final post in this series, I will examine four efforts to provide a platform to pay something other than straight fee for service for health care.


 Even if fee for service payment doesn’t encourage the most integrated, high value care for our patients, there are good reasons that fee for service endures (and indeed made a substantial rebound after many capitation efforts foundered in the late 1990s).


1)      The current level of organization of ambulatory providers is not well suited to accepting capitation or bundled payment.  Small groups should not take bundled payment or capitation, because their numbers are small enough that impressive success or ignominious failure might be due to chance, rather than actual clinical performance.

2)      Risk adjustment has historically been inadequate, encouraging providers to be reluctant to recruit or retain the sickest and most needy patients.  It’s less expensive to take care of healthy patients than very sick patients, even with very impressive care coordinatioin.

3)      Many capitated health plans, especially those in California, paid provider organizations a percent of total heath care premium.  Over the years the health insurance plans were willing to sell insurance for too low a premium to offer sustainable capitation rates. 

4)      While fee for service might be suboptimal as the major payment mechanism, there are good reasons to pay fee for service for evidence-based services that are historically underutilized.  For instance, we would not want to see the cost of human papilloma virus (HPV) vaccine ($360 per course, indicated for all young women) built into a capitation, as this would discourage aggressive provider efforts to deliver more of this vaccine.    

5)      The physicians who are big winners in capitation are relatively quiet about this, for fear their capitation payments would be cut. The physicians who are big losers can’t shut their mouths – giving the impression that physicians all hate capitation.


How could we move toward making capitation politically acceptable, possible to operationalize, and make it more likely we will improve care rather than make care worse in the transition?


1)      Encourage consolidation of ambulatory practices, so that there are the critical mass of physicians more likely to be able to accept bundled payments.  Current efforts to encourage or even force adoption of electronic medical records has the impact of discouraging solo or small independent practices, for which these systems are usually unaffordable.   Government can do this through differential payment to those providers who have greater infrastructure (which are more likely to be larger practices). 

2)      Not offer “cost of living” increases in fee for service reimbursement, making bundled payment more attractive to providers.

3)      Continue to invest in robust risk adjustment – so that the sickest patients are not harmed by a transition away from fee for service

4)      Fee for service generally increases access, so efforts to move to a bundled system will need to measure impact on access and incorporate incentives to maintain or improve access.

5)      For bundled payment to have a substantial impact on the delivery of care, bundled payment must represent a critical mass of total reimbursement. That’s a challenge, since for many adult physicians Medicare is the largest payer, and there are structural impediments to Medicare offering payment other than fee for service.

6)      To protect patient access, I believe that it’s optimal for bundled payments or capitation to represent less than the entire practice revenue stream. My experience in a multi-specialty group with about half capitation and about half fee for service was excellent – the group did not try to “churn” or increase rate of service because of the large capitated base, but strived to increase access to benefit from fee for service patients.  Patients individually went back and forth between capitated and fee for service plans, so there was no practical way to discriminate.  I don’t know what the optimal mix is, but policy-makers shouldn’t fret about a portion of ambulatory revenue remaining in fee for service.

7)      Any type of bundled payment will need to be a work in progress.  Standardization will be required, but we will need flexibility for bundled payments to change as we gain experience with them.  We will also need to change payments with the advance of medical knowledge.  We wouldn’t want to retain elements that encouraged delivery of certain services by carving them out of a capitation, for instance, if the medical literature revealed that those services did not benefit patients.  

8)      It will be critical to avoid imposing large income losses on physicians immediately.  This will constrain the cost savings possible, but will make it les likely that providers who would be net losers would sabotage the transition.

9)      The new payment system will need to be simple enough that it can be explained in a paragraph or less.  Payment methodologies that require hundreds of pages of rules are not likely to be transparent.  There is a tradeoff here;  a simple payment system will not include all the different adjustments that many would like.