(889): Development and validation of the modified Brief Pain Inventory short form (mBPI-sf) in CABG and general surgery patients

Abstract

A modified version of the Brief Pain Inventory short form (mBPI-sf) was developed to measure severity and interference of acute post-operative pain. The mBPI-sf contains 12 items measuring pain intensity (4 original BPI items) and pain interference (general activity, mood, walking ability, relations with others, sleep, coughing, deep breathing, and concentration). The psychometric qualities of the mBPI-sf were evaluated. Data were obtained from two randomized clinical trials: non-CABG general surgery trial (N=1050) and CABG surgery trial (N=1636). Data on the mBPI-sf and pain intensity were collected for 10 days. Physician and patient global evaluations of medication were collected at transition from parenteral to oral study medication, and at end of study. Factor analysis and item response theory (IRT) analysis were employed. Internal consistency and test-retest reliability, construct validity and responsiveness were all evaluated. Analyses were conducted for all patients and subgroups of CABG and GI patients. Three factors were identified in the “all patients” analysis: pain severity, pain interference, and coughing and breathing. In GI and CABG patients, two factors were identified: pain severity and pain interference. Confirmatory factor analysis found acceptable model fit with the two-factor structure in the “all patients” analysis. IRT analyses showed all items exhibited good characteristics supporting their inclusion in the mBPI-sf. All analyses of reliability (internal consistency 0.85; 0.87; test-retest 0.75; 0.81) and validity showed good psychometric properties for the two mBPI-sf subscales in the combined sample. There were no substantive differences in reliability or validity for different sub-samples of patients. The mBPI-sf pain severity and pain interference scales demonstrated good reliability and validity in CABG and general surgery patients. Coughing and Breathing are more relevant to GI and CABG patients than to other general surgery patients, but their inclusion does not affect the measurement properties of mBPI-sf for other general surgery patients.