ABSTRACT
The Brief Pain Inventory (BPI) is a widely used pain measurement tool. There are 2 versions, the BPI Long Form (BPI-LF) and Short Form (BPI-SF), which share 2 core scales measuring pain severity and pain interference but which use different recall periods (24 hours vs. 1 week). To date, the BPI-SF has not been validated for use in Spain. This study investigated the psychometric properties of the BPI-SF Spanish version and compared results on the core scales between BPI-LF and BPI-SF. The data came from a 3-month observational study of 3,029 nononcologic patients managed in Spanish pain units. The BPI-SF's reliability, validity, and responsiveness were assessed. The effect of different recall periods was investigated by using intraclass correlation coefficients (ICCs) to determine the strength of correlation between BPI-LF and BPI-SF. The BPI-SF showed good reliability, with Cronbach's alphas of 0.931 for the severity and interference scales, which also discriminated well between patients reporting different levels of quality of life on EuroQol-5D dimensions (between group effect sizes [ESs] over 0.8). Substantial improvements were seen on both subscales after 3 months of treatment (ES of 1.76 for pain severity and 1.51 for pain interference). Recall period did not noticeably affect scores; ICCs (95% CI) between the long and short versions were 0.946 (0.938 to 0.954) and 0.929 (0.919 to 0.939) for the severity and interference subscales, respectively. The Spanish version of the BPI-SF is a valid and reliable instrument to measure pain severity and interference.
