Objective: To develop a composite score for measurement of disease activity in polymyalgia rheumatica (PMR) and assess its internal and external validity.
Methods: A PMR activity score (AS) was designed and assessed for internal and external validity in two patient cohorts: 57 international patients evaluated primarily for development of the PMR-AS at baseline, weeks 4 and 24; and for validation, 24 Austrian patients assessed at baseline, week 4, and at a mean (SD) point of week 33.6 (24.5). The PMR-AS was calculated as: CRP (mg/dl)+VAS p (0-10)+VAS ph (0-10)+(MST (min)x0.1)+EUL (3-0); Cronbach's alpha was calculated. Factor analysis by linear regression was applied, and responses calculated on the basis of the PMR response criteria and the PMR-AS applied. PMR-AS values at different times were compared by paired t tests.
Results: Cronbach's alpha for the composite score was 0.91 and 0.88 in the two cohorts. Factor analysis showed that each single item contributed significantly to the total score and the relative weight of each item in both cohorts was equally distributed. Mean PMR-AS at baseline was 27.54 and 28.72, respectively, at week 4, 5.99 and 8.99, and at the final visit 5.35 and 5.92 (NS). PMR-AS values at baseline and at later visits were significantly different (p<0.0001). PMR-AS values <7 indicated low disease activity, 7-17 medium disease activity, and >17 high PMR activity. In a third control cohort the PMR-AS correlated highly with patient's global assessment, patient satisfaction, and ESR (p<0.001).
Conclusion: The PMR-AS provides an easily applicable and valid tool for monitoring disease activity, and in combination with the PMR response criteria provides a better description of response.