Background Rheumatoid Arthritis (RA) is a systemic autoimmune disease affecting up to 1% of the general population. Continuous measurement of RA activity facilitates clinical decision-making and treatment to target strategy. The CDAI measurement provides a sensitive and accurate reflection of disease activity.
Objectives We aimed to validate the accuracy of the assessment of RA disease activity by a healthcare professional (nurse) trained in joint examination using an electronic version of CDAI as compared to rheumatologist assessments.
Methods Thirty-eight consecutive patients with RA attending a biologic therapy infusion centre were included in this study. The CDAI scores were obtained for the same patients by rheumatology nurses trained in joint counting and a rheumatologist. Both the nurse and rheumatologist were blinded to the other's assessment results. The order of the assessments by examiners was changed every second patient. The total score, Swollen and Tender Joint Counts (SJC & TJC) and the Global Assessment (GA) of disease activity were compared between evaluators using student t-test and Mann-Whitney U test. All scores were assessed based on CDAI Disease Activity State levels: Remission-0 (<2.8); Low-1 (2.8-9.9); Moderate-2 (10-22); and High-3 (>22). The Pearson's correlation (r) and the level of agreement between evaluators were also investigated using Kappa (Chi-square) statistics. The HAQdas iPad app V4.0 (NLRT) was used in all cases.
Results Thirty-eight RA patients (84.2% females) with mean (SD) age of 43.1 (19.2) years and mean (SD) duration of disease 14.6 (9.4) were assessed during the study. All measurements taken by nurses were strongly correlated with measurements provided by the physician: SJC (r=0.86, p<0.001); TJC (r=0.97, p<0.001); GA of disease activity (r=0.94, p<0.001); CDAI scores (r=0.98, p<0.001); and Disease Activity State (r=0.95, p<0.001). Nurses tended to count less swollen joints (5.2 (4.7) vs. 5.4 (5.1), p=0.63) and more tender joints (11.6 (9.8) vs. 11.1 (9.5), p=0.78) compared to the physician. The mean CDAI score obtained by nurses and the physician did not differ significantly (24.6±17.8 vs. 24.9±17.6, p=0.579; 95%CI -1.42-0.80). Even though, nurses tended to rate the Disease Activity State significantly lower than rheumatologist (2.1±1.0 vs. 2.2±1.0; p=0.023; 95%CI -0.24-(-0.02)), the agreement between evaluators was high with approximately 80% (p<0.001) of patients concordantly classified by Disease Activity State.
Conclusions There was strong agreement between evaluators of different backgrounds in their assessment of RA patients' disease activity (CDAI). The study results suggest that measuring and following disease activity in RA utilizing the CDAI can be employed effectively by a trained Health Care Professional (HCP), independently and probably in different locations and settings. The electronic versions of the CDAI seem to be useful in treating to target strategy, as a clinical index of RA disease activity and planning adjustments of therapy. This approach would probably provide consistency in patients' follow-ups by nurses in clinics and remote areas where rheumatologists are not available to do the assessments at each visit.
Disclosure of Interest : None declared