Pragmatic test-retest and concurrent reliability of three PROMIS® measures in breast cancer rehabilitation

Research Objectives To evaluate test-retest and concurrent reliability of three Patient Reported Outcome Measurement System (PROMIS®) measures in outpatient breast cancer rehabilitation. Design A retrospective, secondary analysis of practice-based data. To evaluate test-retest reliability we grouped breast cancer survivors (BCS) into cohorts by year of initial evaluation (2020, 2021 or 2022), then used ANOVA to compare baseline and change scores between cohort. To evaluate concurrent reliability, we examined correlation (r) between PROMIS and QuickDASH. Setting National, outpatient community-based rehabilitation clinics. Participants BCS (N=2,706) attended physical or occupational therapy (PT/OT) in 2020 (n=217), 2021 (n=1,400), or 2022 (n=747). BCS completed selected PROMIS measures and the Quick-DASH. Interventions Individualized PT/OT provided by oncology-trained therapists. Main Outcome Measures PROMIS-Global health (10-item, scored into Global Physical Health [GPH] and Global Mental Health [GMH]), PROMIS-Physical function (4-item, PF), PROMIS-Ability to participate in Social Roles and Activities (4-item, SRA) and the QuickDASH. Results At initial evaluation, there was no between-group difference in PROMIS T-scores: GPH [F(2, 2,361)=.569, p=.566], GMH (F[2, 2,361]=.610, p=.543), PF (F[2, 2,361]=1.68, p=.186) and SRA (F[2, 2,361]=2.17, p=.114). From initial evaluation to discharge, there was no between-group difference in in PROMIS T-score change: GPH (F[2, 2,417]=.223, p=.800), GMH (F[2, 2,417]=.052, p=.949), PF (F[2, 2,417]=1.79, p=.168), and SRA (F[2, 2,417]=.939, p=.391). Correlation between PROMIS-T-scores and QuickDASH at initial evaluation (GPH [r= -0.614, p<.001], GMH [r= -0.275, p<.001], PF [r= -0.618, p<.001] and SRA [r= -0.671, p<.001]) and change over the course of rehabilitation were small-to-moderate and significant (GPH [r=0.473, p<.001], GMH [r=0.257, p<.001], PF [r=0.548, p<.001], and SRA [r=0.497, p<.001]). Conclusions PROMIS-GPH, -GMH, -PF, and -SRA outcomes demonstrated appropriate test-retest and concurrent reliability in our real-world sample, indicating these outcomes are appropriate and reliable for use in breast cancer rehabilitation. Author(s) Disclosures Drs. Pergolotti, Wood and Kendig receive a salary from Select Medical (ReVital Cancer Rehabilitation).