Psychometric evaluation of the Bengali brief pain inventory in a sample of Bangladeshi cancer patients

Abstract

Pain is a common complain among oncology patients, and success of its management requires accurate assessment. However, the status of pain treatment in Bangladesh is largely undocumented. The purpose of this study was to evaluate the psychometric properties of the Bengali version of Brief Pain Inventory (BPI-Bengali) in a Bangladeshi sample of cancer patients. BPI-Bengali and the Questionnaire on Stress in Cancer Patients Revised (QSC-R 23) were administered to a convenience sample of 60 adult oncology patients receiving pain treatment. Cronbach alpha coefficients were 0.89 and 0.94 for the severity and interference items respectively. The Pearson correlation coefficients for the test-retest stability were 0.65 (p<0.01) for the pain intensity scale and 0.88 (p<0.01) for the pain interference scale. Exploratory factor analysis (EFA) yielded two components, indicating pain severity and interference clusters by explaining 75% of the variance. The confirmatory factor analysis indicated the model as moderate fit to the data (CFI=0.89, RMSEA=0.16). Convergent validity of the pain severity and interference was demonstrated by significant correlations with stress level (r=0.67, p<0.01), (r=0.66, p<0.01). The findings of this study support the psychometric properties of Bengali BPI among cancer population in terms of validity, reliability and factor structure.