Reliability and validity of the Brief Pain Inventory-Short Form in individuals with rotator cuff-related shoulder pain

Abstract

Purpose To investigate the test–retest reliability and construct validity of the Brief Pain Inventory-Short Form (BPI-SF) in individuals with rotator cuff-related shoulder pain (RCRSP). Methods Sixty-one participants with RCRSP completed the BPI-SF twice with an interval of two to seven days and Shoulder Pain and Disability Index (SPADI) at the initial visit. The BPI-SF pain severity subscale, pain interference subscale, and stand-alone pain severity items were analysed using intraclass correlation coefficients (ICCs) and minimal detectable change at the 95% confidence interval (MDC95). The construct validity of BPI-SF was assessed against SPADI using Pearson’s correlation. Results The BPI-SF pain severity and pain interference subscales presented moderate test–retest reliability (ICC = 0.73, 0.53) and MDC95 were 2.05 and 2.36. All stand-alone BPI-SF pain severity items presented a moderate reliability (ICC = 0.62, 0.70). BPI-SF interference items presented poor to moderate reliability (ICC = 0.39, 0.68). The correlation coefficients between the BPI-SF and SPADI subscales or total scores were large (r = 0.61, 0.75). Conclusions BPI-SF pain severity and pain interference subscales have a moderate reliability in individuals with RCRSP. BPI-SF pain severity and interference subscales showed high construct validity in individuals with RCRSP. MDC95 values are useful metrics for interpreting a true change in BPI-SF scores following interventions in individuals with RCRSP.