Measurement properties of the Polymyalgia Rheumatica Activity Score (PMR-AS): reliability, measurement error and discriminative validity

OBJECTIVES: The Polymyalgia Rheumatica Activity Score (PMR-AS) is the best validated composite disease activity measure in PMR to date, but its reliability, measurement error and discriminative validity have not been studied. METHODS: Inter-rater reliability, measurement error and discriminative validity were assessed in a cross-sectional cohort of PMR patients in routine care (April-August 2025). For inter-rater reliability and measurement error, five physicians scored the PMR-AS, with two independently evaluating each patient at the same visit. Test-retest reliability and measurement error were assessed using data from two randomized trials in newly diagnosed PMR (2019-2024, EULAR/ACR criteria positive). Patients with two visits 2-4 weeks apart and no reported change in disease state were included. Analyses included intraclass correlation coefficients (ICCs), standard error of measurement (SEM), smallest detectable change (SDC/SDD) and ROC analyses. Agreement between disease activity categories (PMR-AS <10 and ≥10) was evaluated using Cohen's kappa (ᴋ). RESULTS: A total of 124 patients contributed to the inter-rater analyses and 38 for test-retest analyses. For inter-rater reliability, the PMR-AS showed excellent agreement (ICC 0.93, 95% CI 0.90-0.95). Test-retest reliability was lower (ICC 0.66, 95% CI 0.38-0.82), reflecting lack of variability in disease activity (median PMR-AS 1.6). Measurement error was notable (inter-rater: SEM 2.46, SDD 6.83; test-retest: SEM 2.07, SDC 5.75), but discriminative validity was good (AUC 0.93, ᴋ 0.69). CONCLUSION: The PMR-AS demonstrated excellent inter-rater reliability and discriminative validity, supporting its use for classification of disease activity. However, its measurement error may limit interpretation of small intra-individual changes.