Comparison of reliability, validity and responsiveness of the Japanese Orthopaedic Association Shoulder 36 Ver. 1.3 among different diagnoses of shoulder lesions

Background Shoulder 36 (Sh-36) is an original quality of life measure for shoulder lesions with high reliability and validity; however, in some cases, especially in those with a Bankart lesion, we observed no improvement in Sh-36 during the postoperative follow-up. Sh-36 may be less effective for a certain shoulder lesion. This study aimed to compare the reliability, validity, and responsiveness of Sh-36 among different representative diagnoses of shoulder lesions. Methods Sh-36 and the Disability of the Arm, Shoulder and Hand (DASH) were measured in 192 patients with a Bankart lesion (Bankart group), rotator cuff tear (Cuff group), and SLAP lesion (SLAP group) who underwent arthroscopic surgery. Both measures were evaluated before surgery, and at 3, 6, 9, 12, 18, and 24 months postoperatively, and reliability, validity, and responsiveness of Sh-36 and the DASH were compared among the three groups. Results Significant postoperative improvement was observed in the three groups (p < 0.0001) within 9 months. No marked improvement was observed after 9 months in the Bankart and SLAP groups due to the ceiling effect; however, most domains of Sh-36 increased continuously in the Cuff group during the whole follow-up period. Reliability and construct validity were sufficient in all the groups. The longitudinal validity was sufficient in most domains for the three groups; however, the standardized response mean in the Bankart group was lower than that in other two groups, indicating low responsiveness in this group because of the ceiling effect. Conclusions Sh-36 was a valid and reliable instrument in patients who have undergone arthroscopic shoulder surgery, especially for patient with a rotator cuff tear with high responsiveness. However, Sh-36 had lower standard response mean representing lower responsiveness in the Bankart group due to the ceiling effect and may not be ideal for longitudinal follow-up in patients with a Bankart lesion.