Psychometric Properties of Frailty Instruments in Adults With Rheumatoid Arthritis: A Systematic Review

Abstract

Objective Examine psychometric properties of frailty instruments used in adults with rheumatoid arthritis (RA) to inform selection of frailty instruments for clinical and research use. Methods A systematic review was registered in PROSPERO. Studies measuring frailty in adults with RA published before May 25, 2022, were searched in six electronic databases. Level of evidence of psychometric properties were synthesized and graded for each frailty instrument using Consensus‐Based Standards for the Selection of Health Measurement Instruments methodology. Results There were 22 articles included in the review, and psychometric properties of 16 frailty instruments were examined. RA cohorts were predominantly female with moderate RA disease activity, mean age was 60.1 years, and frailty prevalence ranged widely from 10% to 85%. Construct validity was the only psychometric property routinely examined for frailty instruments in RA, and nearly all (14/16) performed favorably in this domain. Frailty correlated most frequently with older age, higher RA disease activity, and worse physical function. Internal consistency, measurement error, and content validity were examined infrequently. Reliability and responsiveness data were not reported. Six frailty instruments were rated highest in adults with RA: three adaptations of Fried's Criteria, 32‐Item and 45‐Item Frailty Indexes, and the Comprehensive Rheumatologic Assessment of Frailty. Conclusion Six frailty instruments possessed the highest‐rated psychometric properties in RA. These instruments demonstrated construct validity of frailty with important outcomes in RA. Frailty assessment shows promise to inform risk stratification in RA, but studies are needed to evaluate reliability, responsiveness, and validity to support accuracy of frailty measurement in adults with RA who may have disease‐related features that differentially impact outcomes.