Peer Feedback in Clinical High-Volume Activities: How Ranking Designs and Ability Affect Physician Effort

Abstract
Although relative performance feedback generally appears to be effective in improving performance in healthcare, there is limited understanding of how the design of rankings and individual abilities affect physician effort. Using a controlled lab-in-the-field experiment with practicing and future physicians (N=352), we systematically analyze effort choices under different ranking as a peer-feedback mechanism to improve performance in small teams. Ranking designs are exogenously varied by the number and position of thresholds in the outcome distribution. We observe that an additional threshold, which increases the number of ranks, leads to a higher effort only among individuals capable of exceeding that threshold. For the others, effort remains unaffected or even may even decrease. A ranking with thresholds spanning the entire outcome distribution maximizes overall physician effort. Individuals are affected in heterogeneous ways. Setting a threshold at the upper end of the outcome distribution significantly motivates individuals capable of achieving the highest rank. Thresholds in the middle and lower ranges prevent demotivating effects on those who cannot reach the highest rank. Our behavioral results suggest that in order to effectively motivate teams of physicians with peer feedback, clinical leaders should set thresholds such that all physicians can improve their rank through increased effort, while also preserving exclusivity for both the top and the bottom rank.