Abstract. [Purpose] This study was designed to evaluate the clinical utility of the Brief Pain Inventory-12, 23 (BPI-12, 23) for predicting shoulder pain in stroke patients and to compare the functional ability of the BPI-12, 23 test to discriminate between subgroups with and without shoulder pain.[Subjects] The subjects were 62 persons who were chronic stroke survivors with shoulder pain.[Methods] The patients answered questions on shoulder pain using the BPI question 12 (BP1-12) and Pain-related Quality of life (BPI-23). Therapists measured the performance of combined upper-limb movements including the hand behind the neck (HBN) maneuver, and added passive painfree shoulder external rotation range of motion, and the Modified Ashworth Scale (MAS) score of the elbow flexors. [Conclusion] The BPI-12 and 23 both have potential as screening tools for risk factors of shoulder pain. They can be performed during evaluation to increase the likelihood of determining those who complain of hemiplegic shoulder pain after stroke. However they first require confirmation in a prospective study.