Importance: Although leadership behavior of physician supervisors is associated with the occupational well-being of the physicians they supervise, the factors associated with leadership behaviors are poorly understood.
Objective: To evaluate the associations between burnout, professional fulfillment, and self-care practices of physician leaders and their independently assessed leadership behavior scores.
Design, setting, and participants: This survey study of physicians and physician leaders at Stanford University School of Medicine (n = 1924) was conducted from April 1 to May 13, 2019. The survey included assessments of professional fulfillment, self-valuation, sleep-related impairment, and burnout. Physicians also rated the leadership behaviors of their immediate physician supervisors using a standardized assessment. Leaders' personal well-being metrics were paired with their leadership behavior scores as rated by the physicians they supervised. All assessment scores were converted to a standardized scale (range, 0-10). Data were analyzed from October 20, 2019, to March 10, 2020.
Main outcomes and measures: Association between leaders' own well-being scores and their independently assessed leadership behavior.
Results: Of 1924 physicians invited to participate, 1285 (66.8%) returned surveys, including 67 of 117 physician leaders (57.3%). Among these respondents, 651 (50.7%) were women and 729 (56.7%) were 40 years or older. Among the 67 leaders, 57 (85.1%) had their leadership behaviors evaluated by at least 5 physicians (median, 11 [interquartile range, 9-15]) they supervised. Overall, 9.8% of the variation in leaders' aggregate leadership behavior scores was associated with their own degree of burnout. In models adjusted for age and sex, each 1-point increase in burnout score of the leaders was associated with a 0.19-point decrement in leadership behavior score (β = -0.19; 95% CI, -0.35 to -0.03; P = .02), whereas each 1-point increase in their professional fulfillment and self-valuation scores was associated with a 0.13-point (β = 0.13; 95% CI, 0.01-0.26; P = .03) and 0.15-point (β = 0.15; 95% CI, 0.02-0.29; P = .03) increase in leadership behavior score, respectively. Each 1-point increase in leaders' sleep-related impairment was associated with a 0.15-point increment in sleep-related impairment among those they supervised (β = 0.15; 95% CI, 0.02-0.29; P = .03). The associations between leaders' well-being scores in other dimensions and the corresponding well-being measures of those they supervised were not significant.
Conclusions and relevance: In this survey study, burnout, professional fulfillment, and self-care practices of physician leaders were associated with their independently assessed leadership effectiveness. Training, skill building, and support to improve leader well-being should be considered a dimension of leadership development rather than simply a dimension of self-care.