Monday, November 5, 2012

Pay for Performance vs Intrinsic Motivation

Among the many stories about health care heroes during Hurricane Sandy, this was my favorite:

Allison Chisholm, 46, who works for the Visiting Nurse Service, lives with a frail mother in Park Slope, Brooklyn. When the lights started flickering during the storm on Monday, she had images of her mother falling in the dark. But she also had patients who needed her, including one receiving hospice care in a 12th floor apartment in Chinatown, and one needing an intravenous round of antibiotics in the West Village.

“It was treacherous driving during the hurricane,” said Ms. Chisholm, fitting an intravenous line into the arm of Jill Gerson, 71, who teaches social work at Lehman College in the Bronx. “But it’s just something you have to do as a nurse. That continuity of care helps the healing. I don’t see this as being heroic. I have a conscience. I need to get to sleep at night.”
Ms. Chisholm was responding to intrinsic motivation - her values as a nurse, embodying the tradition associated with Florence Nightingale and Mother Theresa. She wasn't being "incentivized" (one of my least favorite words) by pay-for-performance, unless we regard the threat from her conscience that - like Lady Macbeth - she would "sleep no more" if she failed to put her values into action as a performance management system, as an "incentivizing" force!

Pay-for-performance has considerable face validity. Extrinsic motivation clearly works in vast swathes of the economy. But as my friend Dr. Steffie Woolhandler's recent post on the Health Affairs blog shows, it's  not at all clear that pay-for-performance is effective in domains that have historically rested on intrinsic motivators such as idealism, altruism, and care. Pay-for-performance can increase the behaviors that are being measured, but evidence that these systems enhance patient outcomes is weak or absent. And there is substantial evidence from the behavioral economics literature that monetary rewards can actually decrease motivation for tasks that are intrinsically rewarding.

My own reaction when I hear of programs to "incentivize physicians to do [XYZ desirable clinical behavior]" is decidedly negative. When I began my own fee-for-service practice in the 1970s I took pleasure in including Medicaid beneficiaries, but after a time the burdensome paperwork and inefficient reimbursement process, combined with microscopic fees, acted as a disincentive for doing what I wanted to do, and I limited the number of Medicaid beneficiaries I took on. I didn't need to be "incentivized" but I would have responded well to a reduction of disincentives.

Rats in a Skinner box are "incentivized" by food pellets. But as the interview with Ms. Chisholm reflects, the kinds of caretaking we want to encourage in medicine flows from values, not P4P pellets. Program managers will do better by recognizing, respecting, and supporting intrinsic motivation. This is best done by removing impediments, not by the condescending view of doctors and nurses as reluctant laborers.

No comments: