Pay-for-performance in health economics

This application is most appropriate for a health economics class. It may also be appropriate in a principles of economics or microeconomics theory course, if those courses touch on incentive and health economics. Before approaching this application, students should know what pay-for-performance is, and should have a basic knowledge of the kinds of things that can be measured.

Overview

Students will identify the most important outcome measure for a doctor pay-for-performance scheme.

Expected Student Learning Outcomes

Students will be able to articulate the advantages and disadvantages of different systems for measuring the productivity/quality of doctors.

Information Given to Students

If you were an insurer implementing a pay-for-performance program for doctors, which of the following would be most important in your doctor performance measure?
A. Patient experience (How did your patient rate their experience with you?)
B. Outcomes (What is your surgery death rate? Were your patients readmitted to the hospital? What was your patients' insulin at the next appointment?)
C. Process (Did you give aspirin when needed?, Did you ask about family history?, etc)
D. Cost of care and Utilization rates (How many MRIs did you order over what is considered "normal"?)

Pay_for_performance_application (Microsoft Word 2007 (.docx) 31kB Jun23 18)

Prefatory remarks: The professor should make sure that students know what pay-for-performance is and how it works before the application begins. They should also know what kind of processes might be measured (% of doctors giving chest pain patients aspirin, % of doctors giving 50-year-old women a mammogram, etc).

In the discussion, one of the most important concepts to highlight relates to a new vocabulary term that you can introduce after students have brought up the idea on their own..
Negative defensive medicine: Doctors avoiding sick or complex patients because the doctor may be penalized in a pay-for-performance or doctor ranking system if the patient has negative outcomes due to the patient's condition or behavior.
This contrasts with...
Defensive medicine: Ordering too many tests (or otherwise providing too much care) to protect ones' self from lawsuits. (Students are often familiar with this one)

The patient's behavior, genetics and social conditions have a big impact on their outcomes, so a P4P scheme that rewards for outcomes will penalize doctors who take on the sickest and most complex patients.

The problem with processes as a measure in P4P is that they are difficult and expensive to measure, and often don't take into account nuanced judgement. If something can be measured by a general rule, then it is likely that a nurse's aid could perform the task, rather than an MD.

The downsides of rewarding for patient experience should include the fact that there is imperfect information, so that patients cannot properly evaluate quality care. Bias against doctors of color and female doctors has also been shown to impact quality ratings that these doctors receive from patients.

Basing pay on utilization or cost containment, of course, risks under-provision of care. Students almost never choose this option.

After finishing the discussion, students are often interested to see which of the four is most common. Process measures are currently the most commonly used metric. About half of P4P systems use process measures.

Process measures: 50%

Outcome measures: 19%

Utilization (11%) and cost of care (3%) = 14%

Patient experience: 6%

If you look up the article, you get a nice pie graph of these...

Source: Higgins, A., Veselovskiy, G., & McKown, L. (2013). Provider performance measures in private and public programs: achieving meaningful alignment with flexibility to innovate. Health Affairs, 32(8), 1453-1461.

If students can articulate both the advantages and disadvantages of each way of measuring doctor performance, then they have fulfilled the goal of the application.