Validation of the Amharic Version of the Brief Pain Inventory for Utility in Ethiopian Cancer Patients

BACKGROUND: Pain is a highly prevalent and distressing symptom in patients with cancer but at the same time inadequately treated in many of these patients. The major reason for such imbalance is inappropriate pain assessment. To overcome such a barrier, the use of a valid and reliable pain assessment tool in the language the patient understands is invaluable. Unfortunately, until now, no such standardized tool has been utilized in Ethiopian patients with cancer for both clinical and research purposes. OBJECTIVES: This study tried to validate the Amharic language version of the Brief Pain Inventory (BPI-Am) and evaluate the adequacy of cancer pain management using the Pain Management Index (PMI). METHODS: A standard forward/backward translation technique was used for translating the BPI from its original English version into Amharic (BPI-Am). The BPI-Am was administered to 291 Ethiopian patients with cancer who fulfilled the inclusion criteria. Factor analysis was used to identify the underlying constructs of the tool. Mean comparison was used to confirm the sensitivity of the BPI-Am to known patient groups that differed based on their performance status. Cronbach's α and intraclass correlation coefficients, respectively, were used to assure internal consistency and test-retest reliability of the BPI-Am. The PMI was calculated to identify the level of inadequate pain management in the current sample. Stepwise logistic regression was used to identify potential predictors of inadequate pain management. RESULTS: Factor analysis yielded 3 factors-pain severity, physical activity interference, and psychosocial interference-with Cronbach's α coefficients of 0.85, 0.87, and 0.77, respectively. The BPI-Am showed the capacity to detect higher mean pain severity and mean pain interference scores in patients with poor performance status as compared with those having a good performance status (P < 0.001). Intraclass correlation coefficients for test-retest reliability were 0.75 and 0.78 for the pain severity and pain interference composite scores, respectively. Sixty-seven percent of patients in the current sample were inadequately treated for their pain according to the PMI. Good performance status (odds ratio [OR] = 2.9, P < 0.01), absence of cancer-related complications (OR = 2.1, P < 0.05), and being unemployed (OR = 2.6, P < 0.01) were identified as predictors of inadequate pain management. CONCLUSION: The BPI-Am is a valid and reliable tool for use in Ethiopian patients with cancer. The inadequacy of cancer pain management in these groups of patients is high, which calls for needed attention.