Feeling the burn in the era of COVID-19: cross-cultural adaptation and validation of the Arabic version of the Copenhagen Burnout Inventory among community pharmacists

Abstract

Background: Community pharmacists (CPs) are one of the frontline healthcare workers (HCWs) working diligently to provide much-needed services during the COVID-19 pandemic. Burnout was one of the detrimental outcomes of the pandemic on the mental health of Lebanese CPs. To assess the extent of this syndrome among Lebanese CPs, a psychometrically reliable and valid tool is needed. Objectives: This study aimed to validate the Arabic version of the Copenhagen Burnout Inventory (CBI-A) for use in the assessment of burnout among CPs. Methods: A web-based cross-sectional study was conducted among Lebanese CPs over February 2021. Data were collected using an anonymous Arabic self-administered questionnaire that includes information on socio-demographic characteristics, work-related variables, in addition to the measurements: the CBI which includes personal, work-related, and patient-related dimensions of burnout, and the hospital anxiety and depression scale. Data were analyzed using SPSS and Amos software. Exploratory factor analysis and confirmatory factor analysis were performed to explore the factorial structure and to measure model fit. Cronbach’s alpha was used to assess internal consistency. The criterion validity of the CBI was assessed. Multivariable linear regression analyses were used to explore the association between different aspects of burnout and mental health outcomes such as depression and anxiety. Results: The CBI-A showed high internal consistency with Cronbach’s alphas varied from 0.774 to 0.902 and a low floor and ceiling effect (1–9%). As for the CBI-A construct validity, the exploratory factor analysis showed three factors with good factor loadings and explained 72.17% of the variance. The confirmatory analysis supported the three-factorial structure of the CBI that presented a good overall fit based on the goodness-of-fit indices. Ad hoc modifications to the model were introduced based on the modification indices to achieve a satisfactory fit by allowing one covariate error between one pair of items within the personal burnout domain. All of the 19 items were kept in the construct since they showed a good factorial weight. The CBI-A is associated with burnout-related factors in expected directions, including extensive working hours, sleeping hours, and job satisfaction, indicating, therefore, the criterion validity of the tool. CBI subscales were also found positively associated with mental health outcomes such as depression and anxiety demonstrating, in turn, a predictive validity. Conclusion: This study provides evidence for the validity and reliability of the Arabic version of CBI as an adequate tool for assessing burnout among CPs. Such an instrument could be useful for assessing such syndrome among other healthcare workers.