Utilization patterns and healthcare accesses of JAKi used in rheumatoid arthritis patients in Tuscany: The LEONARDO study

Background: The first JAK inhibitor (JAKi) was approved for rheumatoid arthritis (RA) in late 2018. Objectives: This study is aimed at characterizing JAKi users by evaluating their prescription history and the healthcare utilization patterns using administrative healthcare databases. Methods: This is a descriptive, retrospective cohort study on Tuscan administrative healthcare databases. We identified two groups of Tuscan JAKi new users with the first dispensation: 1) from January 1st, 2018 to December 31st, 2019 (cohort entry 1), 2) from January 1st, 2018 to June 30th, 2019 (cohort entry 2). We included adult patients with at least 10 years of look back period and with at least six months of follow-up. In the first analysis, we calculated the distribution of first-ever JAKi users, the DMARD dispensations in the lookback period, and the mean time and standard deviation (SD) from the firstever DMARD to the JAKi. In the second analysis, we estimated the number of emergency department (ED) accesses and hospitalizations (H) for any causes and the number of RA visits (RV) in the follow-up, the mean time, and SD to the first event of interest. Results: The first analysis included 363 new JAKi users. The mean age was 61.5 (SD 13.6) and 80.7% were females. 78.5% of patients used baricitinib, while 21.5% tofacitinib. The 79% of patients had conventional synthetic DMARD before the cohort entry, with hydroxychloroquine (44%), methotrexate (42%), and leflunomide (33%) the most frequently observed. The 60% had a history of biologic DMARD with a high distribution of etanercept (40%), abatacept (30%) and adalimumab (27%). The JAKi were similarly administered as second (31.1%), third (33.1%), or fourth line (27.8%) treatment. The mean time to JAKi was 7.2 years (SD 3.3). The second analysis included 221 new JAKi users. We observed 109 accesses to ED, 39 H, and 64 RV. All RV were recorded in patients receiving baricitinib. The mean time in days to the first ED admission was 73.5 (SD 54.1), to the first H was 89.6 (SD 54.8) and to the first RV was 60.9 (SD 51.3). The most frequently reported causes related to ED accesses were classified as injury and poisoning (18.3%), skin (13.8%), and circulatory disorders (11.0%), while those associated with H were cardiovascular (69.2%), musculoskeletal (64.1%) and skin manifestations (38.5%). Conclusions: Our findings showed that JAKi are used in accordance with the current RA clinical guidelines. The utilization of healthcare services was similar between the two drugs and the events reported were in line with the known safety profile or the underline disease.