Psychometric properties of a custom Patient-Reported Outcomes Measurement Information System (PROMIS) physical function short form and worst stiffness numeric rating scale in tenosynovial giant cell tumors

PURPOSE: The purpose of this study was to evaluate the psychometric properties of the PROMIS-Physical Function (PF) and Worst Stiffness Numeric Rating Scale (NRS) among patients with tenosynovial giant cell tumors (TGCT). METHODS: Measurement properties of the customized lower extremity (LE) and upper extremity (UE) PROMIS-PF scales and Worst Stiffness NRS were assessed using data from the Phase 3 ENLIVEN trial (n = 120). Anchor- and distribution-based analyses were utilized to derive a responder threshold for meaningful change over time. The Patient Global Rating of Concept (PGRC)-Physical Functioning and Patient Global Impression of Change (PGIC)-Stiffness served as anchors. Responsiveness and responder threshold analyses were from baseline to week 25. RESULTS: Cronbach’s alpha values for internal consistency reliability were 0.93 and 0.91 for the PROMIS-PF LE and UE, respectively. Test-retest reliability intra-class correlation coefficients were > 0.75 for both instruments. Convergent validity for both instruments was supported by moderate to strong correlations (≥0.30) with the Brief Pain Inventory and EQ-5D. Known-groups validity was established between subgroups stratified by pain level (p < 0.05). Responsiveness was supported by evaluating change scores among different levels of change in PGRC-Physical Functioning and PGIC-Stiffness (overall F values < 0.001). Triangulation of responder definition analyses resulted in a threshold of ≥3 for the PROMIS-PF and ≥ 1 for the Worst Stiffness NRS. CONCLUSION: This study is the first to establish the psychometric properties of patient-reported outcome measures in TGCT. The evidence demonstrates that the PROMIS-PF and Worst Stiffness NRS have good reliability, validity, and responsiveness, and provides guidance for the interpretation of meaningful change.