Measuring physicians' productivity: a three-year study to evaluate a new remuneration system.

Purpose: To evaluate a new assessment tool measuring physicians' academic productivity and its use in a performance-based remuneration system.

Method: The authors developed an assessment tool based on existing tools to measure productivity. Yearly, from 2008 to 2011, physicians at the University of Western Ontario received a score of up to three points for each of four components (impact, application, scholarly activity, mentorship) in each of four domains (clinical practice, education, research, administration). Scores were weighted by the percentage of time physicians spent on tasks in each domain. Year 1 scores were a baseline. In Years 2 and 3, scores were tied to remuneration. The authors compared scores and associations, accounting for age and academic rank, across the three years.

Results: The 37 participating physicians included 11 assistant, 23 associate, and 4 full professors. The mean weighted total baseline score across all four domains was 7.44. Years 2 and 3 scores were highly correlated with Year 1 scores (r = 0.85, Years 1 and 2; r = 0.89, Years 1 and 3). Year 2 mean weighted scores did not differ significantly from Year 1 scores. Assistant professors' scores improved significantly between Years 1 and 2 (+1.08, P < .001). Lower Year 1 scores were correlated with a greater improvement in scores between Years 1 and 2, and age was negatively correlated with score changes between Years 2 and 3.

Conclusions: Although the tool may be a robust measurement of physicians' productivity, performance-based remuneration had no effect on physicians' overall performance.