Abstract
BACKGROUND: Burnout affects nearly half of all U.S. nurses and physicians, and has been linked to poor outcomes such as worse patient safety. The most common measure of burnout is the well-validated Maslach Burnout Inventory (MBI). However, the MBI is proprietary and carries licensing fees, posing challenges to routine or repeated assessment. OBJECTIVE: To compare a non-proprietary, single-item burnout measure to a single item from the MBI Emotional Exhaustion (MBI:EE) subscale that has been validated as a standalone burnout measure. DESIGN: Cross-sectional online survey. PARTICIPANTS: A sample of primary care providers (PCPs), registered nurses, clinical associates (e.g., licensed practical nurses (LPNs), medical technicians), and administrative clerks in the Veterans Health Administration surveyed in 2012. MAIN METHODS: We compared a validated one-item version of the MBI:EE and a non-proprietary single-item burnout measure used in the Physician Work Life Study. We calculated kappa statistics, sensitivity and specificity, positive predictive (PPV) and negative predictive values (NPV), and area under the receiver operator curve (AUC). We conducted analyses stratified by occupation to determine the stability of the correlation between the two measures. KEY RESULTS: We analyzed responses from 5,404 participants, including 1,769 providers and 1,380 registered nurses. The prevalence of burnout was 36.7% as measured on the single MBI:EE item and 38.5% as measured on the non-proprietary single-item measure. Relative to the MBI:EE, the non-proprietary single-item measure had a correlation of 0.79, sensitivity of 83.2%, specificity of 87.4%, and AUC of 0.93 (se = 0.004). Results were similar when stratified by respondent occupation. CONCLUSIONS: A non-proprietary single-item measure served as a reliable substitute for the MBI:EE across occupations. Because it is non-proprietary and easy to interpret, it has logistical advantages over the one-item MBI.