RESUMEN
BACKGROUND: Studies on the psychometric properties of the Copenhagen Burnout Inventory (CBI) in medical residents are scarce despite their susceptibility to burnout. Moreover, none of these studies were conducted in Spanish. AIM OF THE STUDY: To analyze the psychometric properties of the Spanish version of the CBI among Mexican medical residents. METHODS: This cross-sectional study was conducted online on medical residents from a public medical institution (n = 525). The English version of the CBI (19 items organized into three domains: personal-related burnout, work-related burnout, and patient-related burnout) was translated into Spanish and again into English. Content, convergent, discriminant, and concurrent validity were assessed, along with reliability. RESULTS: The CBI Spanish version showed acceptable content, convergent and concurrent validity. Exploratory factor analysis showed two factors, but confirmatory factor analysis showed three factors with adequate fit (Root Mean Square Error of Approximation = 0.08, Comparative Fit Index = 0.95, Tucker-Lewis Index = 0.94, and Standardized Root Mean Square Residual = 0.04). There was no good discrimination between personal-related and work-related burnout. Cronbach's alpha coefficients for the personal-related, work-related, and patient-related burnout domains were 0.94, 0.95, and 0.93, respectively. CONCLUSIONS: The Spanish version of the CBI in Mexican medical residents is reliable, and it meets adequate content, convergent and concurrent validity. The construct validity was not consistent. This should not diminish the importance of the CBI.