Reliability and validity study of the Thai adaptation of the Copenhagen Burnout Inventory-Student Survey (CBI-SS) among preclinical medical students at the Faculty of Medicine Siriraj Hospital, Mahidol University, Thailand

Abstract

Burnout syndrome is a syndrome of emotional exhaustion, professional efficacy and cynicism. A significant proportion of medical students reported having burnout syndrome during their training in medical education. Several tools including the Copenhagen Burnout Inventory-Student Survey (CBI-SS) are considered to be a valid measurement of burnout syndrome in medical students. This study aimed to translate, culturally adapt, and validate the CBI-SS for assessing burnout syndrome among preclinical medical students in Thailand. This study was conducted during February to March 2019 at the Faculty of Medicine Siriraj Hospital, Mahidol University (Bangkok, Thailand), which is Thailand’s largest and oldest medical school, and Thailand’s largest national tertiary referral center. After receiving formal permission to do so from the copyright owner, the original English language version of the CBI-SS was translated to Thai language using an internationally recommended and accepted forward-backward translation protocol. The Thai version of the CBI-SS (Thai CBI-SS) comprises 25 items, including 6 items for personal burnout, 7 items for study-related burnout, 6 items for colleague-related burnout, and 6 items for teacher-related burnout. Standardized Cronbach’s alpha coefficient was calculated to evaluate internal consistency reliability, and correlation coefficient was computed to determine test-retest reliability. A total of 414 preclinical medical students participated in this study. Due to sub-optimal factor weights (<0.50), items 6, 10 and 17 were excluded. The Cronbach’s alpha coefficients of the 22-item Thai CBI-SS for personal, study-related, colleague-related, and teacher-related burnout were 0.898, 0.896, 0.910 and 0.900 respectively. The correlation coefficients for test-retest reliability after three weeks were 0.820, 0.870, 0.821, and 0.787 for personal, study-related, colleague-related, and teacher-related burnout, respectively. Maximum likelihood analysis with oblimin rotation indicated four main components, and confirmatory factor analysis revealed good fit indices of the Thai CBI-SS. Confirmatory factor analysis showed good fit indices of CBI-SS domains (χ2/df = 2.39; CFI = 0.957; GFI = 0.909; RMSEA = 0.058; TLI = 0.949; and NFI = 0.928). The convergent validity analysis using the Average Variance Extracted (AVE) and the Composite Reliability (CR) was adequate for all dimensions (personal: AVE = 0.626, CR = 0.893; study-related: AVE = 0.601, CR = 0.899; colleague-related: AVE = 0.677, CR = 0.913; teacher-related: AVE = 0.606, CR = 0.900). The HTMT values for all variables are in the range from 0.315 to 0.833, confirming the discriminant validity. The Thai CBI-SS was found to be a valid and reliable tool for evaluating burnout syndrome in preclinical medical students in Thailand.