Limitations of the MBI-HSS in Colombian health personnel psychometric: Evaluation after modification of its response options

Abstract

Objective: To establish the construct validity, internal consistency, and difficulty and discrimination of the items in the Maslach Burnout Inventory-Human Services Survey, with modified response options, for measuring burnout in healthcare workers in Colombia. Methods: A descriptive study of validation of measurement instruments was conducted. With the support of a panel of experts, the response options were modified, and the new version was administered to 377 healthcare workers. Internal consistency was evaluated, and confirmatory factor analysis was conducted to assess the structural construct validity. The structural Equation Modeling model was calculated using the diagonally weighted least squares estimation method. Finally, item discrimination and difficulty thresholds of the response options were calculated using the generalized partial credit model of Item Response Theory. Results: The emotional exhaustion and personal accomplishment subscales showed good internal consistency (α= 0.87 and 0.77), while the depersonalization subscale was low (0.45). The confirmatory factor analysis indicated an acceptable but not ideal fit (NFI and RFI <0.95; RMSEA >0.06). Inadequate functioning of the response options and measurement bias were identified. Conclusions: The previously validated version of the Maslach Burnout Inventory-Human Services Survey and the adapted version in this study have limitations in Colombian healthcare workers, especially in the depersonalization subscale. A thorough revision is suggested to appropriately measure burnout in healthcare personnel in our context.