Episode-Based Performance Measures: A reality?

Pay-for-performance has become very fashionable of late. One way to measure physician performance is with episode groupers. This software groups together some or all of the services related to the care of a patient’s chronic or acute medical conditions. Policymakers can then use the episode as the unit of observation for: feedback on physician performance, public reporting, pay-for-performance, or ‘bundled payments for groups of services. These pay-for-performance measures could offer a significant increase in the scope of quality measures, which currently only evaluate process (e.g., did a heart attack patient receive a beta-blocker).

A paper by Hussey et al. (Health Affairs 2009) looks at the current state of the episode groupers. They note that these groupers are currently being tested in the following settings: (1) Geisinger Health System’s payment for cardiac care episodes; (2) the Medicare Acute Care Episodes (ACE) demonstration,(3) the Medicare Physician Hospital Collaboration demonstration; and (4) PROMETHEUS Payment’s pilot-testing of episode-based payments for several acute and chronic conditions.

One problem with assigning an episode to any individual physician is that Medicare patients are typically treated for many physicians, even within an episode. “Medicare beneficiaries receive care from a median of seven physicians, and the typical primary care physician must coordinate with 229 other physicians working in 117 practices.”

Secondly, it is unclear whether the physician should be the unit of analysis. Poor hospital hygine may be the cause of certain hospital-born infections. Any single physician likely has little input in the overall sanitation level of the hospital. Thus, it is unclear if one should assign an episode to a physician, a physician group, a facility (e.g., hospital) or a larger health care institution.

Finally, many Medicare beneficiaries have co-occurring health events. For instance, beneficiaries had an average of “…eight or more episodes of care during a year, some of which were for interrelated conditions. For example, many beneficiaries who had an AMI also had hypertension (63 percent), CHF (54 percent), or diabetes (35 percent) episodes.”

While episode-based performance measures offer much promise, there are many significant obstacles that need to be overcome before these measures will be able to significantly improve the efficiency of how medical care is delivered today.


  1. In other words, an “episode” is a collectivism; no-one individual owns it — except the patient, who can’t act as an owner under our present system.

    So, it behaves like the “commons” in “tragedy of the commons.”

Leave a Reply

Your email address will not be published. Required fields are marked *