John HarrisNew payment models offer an unprecedented opportunity to reward physicians for their efforts to improve quality and cost performance. That’s the good news. But when you accept the role of distributing funds to physicians, you had better set it up right. 

The funds used to reward physicians often come in the form of incentives or shared savings from payers. Ideally, you will be able to decide how to distribute them to physicians to reward performance. 

Here are a few tips for setting up an incentive distribution model.

Carole Graham Agree on key principles. An important characteristic of a successful incentive distribution model is that it feels fair and appropriate to physicians. One way to cultivate this feeling in physicians is to establish principles jointly with physician leadership, and then design the incentive distribution model according to those principles. Possible principles include: 

  • Supporting quality improvement
  • Aligning with payer reward calculation
  • Allocating rewards based on impact
  • Promoting Simplicity
  • Ensuring Transparency

Identify metrics. Metrics should be selected to support initiatives that will lead to success in the payer contract. Ideally, those metrics will already be important to your hospital and physicians and will link to your goals (e.g., core measures for hospitals and Medicare PQRS measures for physicians). 

Organizations pursuing an accountable care organization (ACO) strategy should consider that initial measures typically center on primary care physicians. Initial measures for organizations pursuing bundled payment models tend to be more specialty-focused. Metrics can also be used to create incentives for other behaviors you are seeking, such as attendance at certain staff meetings or educational programs.

Assuming that you are successful and that there are incentive funds to be distributed, these metrics can become the building blocks for the physician incentive pool distribution model. 

Define the incentive distribution methodology. Allocating points to each metric creates a simple distribution methodology, and one that is easily explained to physicians. For example, a basic model could have 10 metrics, and each metric could have a value of one point. 

At the end of the measurement period, you divide the total points earned by all physicians into the total dollars in the incentive pool, thereby establishing a value per point. Physicians then receive an incentive equal to their points earned, multiplied by the value per point. 

If overall performance results in more dollars in the total incentive pool, the value of each point will go up. Over time, the distribution of points can be adjusted to focus on specific initiatives. For example, if an organization’s goal is to reduce readmissions for congestive heart failure, then the point model can be adjusted to either include more CHF-related measures or to reallocate the points so that CHF-related measures are more heavily weighted. 

Other Considerations. In some ways, principles, metrics, and methodology are the easy things. There are many other considerations that make these efforts “interesting". Securing physician buy-in is likely to be challenging; the logistics of data collection will surely be more difficult than identifying the measures. Incentives will need to be designed to drive both individual and group practice performance, and to affect both independent and hospital-employed physicians appropriately. Lastly, all of this has to be done in a way that complies with regulatory requirements. 


John M. Harris is a principal, DGA Partners, Bala Cynwyd, Pa., and a member of HFMA’s Metropolitan Philadelphia Chapter.

Carole J. Graham, RN, is a manager, DGA Partners, Bala Cynwyd, Pa.

Publication Date: Thursday, June 06, 2013