Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement
Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.
Clinical Utility and Validity of Self Assessed Joint Involvement in Rheumatoid Arthritis (RA) Patients with Low Active Disease.
Radner4, Helga, Stamm3, Tanja Alexandra, Grisar2, Johannes, Smolen1, Josef S., Aletaha3, Daniel
Joint swelling and tenderness are essential clinical manifestations of RA disease activity. Especially in patients in low disease activity states without significant joint involvement regular performance routine visit are usually less frequent. In these patients, self assessment of joints, and potentially, the use of a purely patient derived disease activity score might be an acceptable alternative. As a first step, we evaluated the validity of patient self assessed joint counts and a composite disease activity scores calculated from patient derived scores.
212 patients with established RA were instructed to perform self-assessment of joints regarding swelling and tenderness on the 28 joint count (SJC28, TJC28) before an independent routine assessment by a trained evaluator. A randomly chosen subgroup of patients (n=79) received short training of joint assessment by a physician.
Based on patient derived joint counts a composite index of disease activity was calculated (patient clinical disease activity index CDAIp = TJC28 patient + SJC28 patient + patient global assessment (PGA) in cm) and correlated with the standard CDAI (CDAIe) via Spearman Correlation. Furthermore we calculated the concordance of CDAI states and differences in swollen and tender joint counts between the two different assessors using Wilcoxon test. Subsequently patients were divided into their level of disease activity defined by CDAIe and agreement was analyzed in the 4 subgroups. Last, differences of agreement for trained and untrained subgroups were calculated separately to investigate the effect of training.
Table 1. Wilcoxon test comparing patient and evaluator derived swollen and tender joint counts (SJC28p, TJC28p; SJC28e, TJC28e) as well as the clinical disease activity index calculated by patient derived values (CDAIp) compared to the standard one (CDAIe) in the total cohort and subgroups defined by CDAI
We found a significant correlation (p<0.01) of patient and assessor derived SJC28 (r=0.42), TJC28 (r=0.71) and CDAI (r=0.79). Wilcoxon test showed good accordance of CDAI level between CDAIp and CDAIe (ties=129 of 212) and slightly lower agreement of SJC28 (ties= 61) and TJC28 (ties=81). In different subgroups of CDAI-level we found a high concordance of patient and evaluator derived values using Wilcoxon test (see table) and significant (p<0.01) correlation of CDAI in all subgroups of disease activity except HDA (REM r=0.61; LDA r=0.50; MDA r=0.38; HDA r=0.36 p=0.13).
Comparing trained and untrained patients we found no significant differences between these two subgroups regarding Spearman correlation of CDAIp and CDAIs (trained r=0.74; untrained r=0.83) or Wilcoxon test (ties CDAI level trained =48 of 79, untrained =81 of 133).
Patients self assessment of joints show good correlation with those performed by specialists, regardless of training. To improve effective follow-up, especially in patients with REM or LDA, a purely patient derived disease activity index seems valid and might be a helpful tool in daily clinical practice.
To cite this abstract, please use the following information:
Radner, Helga, Stamm, Tanja Alexandra, Grisar, Johannes, Smolen, Josef S., Aletaha, Daniel; Clinical Utility and Validity of Self Assessed Joint Involvement in Rheumatoid Arthritis (RA) Patients with Low Active Disease. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1745