Evaluation of RABBIT risk score in Japanese patients with rheumatoid arthritis starting first bDMARDs: a validation study using the IORRA cohort data.

Abstract

To examine the ability of the Rheumatoid Arthritis Observation of Biologic Therapy (RABBIT) risk score to predict the occurrence of serious infections in Japanese patients with rheumatoid arthritis (RA), after initiating their first biologic disease-modifying antirheumatic drug (bDMARD). We used data from the Institute of Rheumatology, Rheumatoid Arthritis (IORRA) cohort from 2008 to 2020. Patients with RA who were started on their first bDMARDs were included. Those with missing data required to calculate the score were excluded. A receiver operating characteristic (ROC) curve was used to evaluate the discriminatory ability of the RABBIT score. A total of 1,081 patients were enrolled. During the one-year observational period, 23 (1.7%) patients had serious infections; the most frequent one was bacterial pneumonia (n=11, 44%). The median RABBIT score in the serious infection group was significantly higher than that in the non-serious infection group (2.3 [1.5-5.4] vs 1.6 [1.2-2.5], p<0.001). The area under the ROC curve for the occurrence of serious infections was 0.67 (95% confidence interval 0.52-0.79), suggesting that the score had low accuracy. Our present study revealed that the RABBIT risk score did not have sufficient discriminatory ability for predicting the development of severe infections in Japanese patients with rheumatoid arthritis after initiating their first bDMARD.