Association Between Physician Burnout and Self-reported Errors: Meta-analysis

Abstract

Objectives: Burnout among physicians is an increasingly recognized phenomenon affecting different aspects of patient care and safety. This meta-analysis quantifies association of burnout and its subscales with self-reported medical errors among physicians.

Methods: This meta-analysis followed the principles formulated in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses and Meta-analyses of Observational Studies. The MEDLINE, PubMed, Web of Science, PsycInfo, and Eric databases were searched until February 5, 2019, using various combinations of key terms without any language restrictions: burnout, physicians, error, safety, and quality. Reference lists of selected studies were hand searched. Data were extracted from published reports. All quantitative studies reporting prevalence of burnout and its association with self-reported errors among physicians were considered. The analyses of heterogeneity (Cochran Q, I2), publication bias (Begg-Mazumdar and Egger), three subgroups, and sensitivity were performed. The effect of overall burnout and Maslach Burnout Inventory subscales on self-reported errors was calculated as odds ratios with 95% confidence interval.

Results: Thirteen studies on 20,643 physicians and residents were included. The overall burnout among participants was associated with a significantly increased risk of self-reported errors (odds ratio = 2.72, 95% confidence interval = 2.19-3.37). Emotional exhaustion, depersonalization, and personal accomplishment were all independently predicting factors of self-reported errors. Cochran Q test and inconsistency index I2 were as follows: Q = 27.2; P = 0.0013, I2 = 67% (36%-83%).

Conclusions: The results provide evidence that not only overall burnout but also its subscales independently are to be associated with a significantly increased risk of self-reported errors among physicians. As self-reported errors may translate into different types of adverse events, this strong and unequivocal association should be of major concern to healthcare organizations.