Biologic treatment for rheumatic disease: real-world big data analysis from the Greek country-wide prescription database.

Objectives: To directly assess the prevalence of inflammatory rheumatic disease under treatment with biologic disease modifying anti-rheumatic drugs (b-DMARDs) and compare treatment patterns between rheumatoid arthritis (RA) and spondyloarthropathy (SpA), including psoriatic arthritis.

Methods: The obligatory country-wide prescription electronic database covering 10.223.000 Greek citizens (95.1% of the population, 99.5% Caucasian), all of whom with fully reinbursed access to b-DMARDs, was used to retrospectively capture all patients under b-DMARDs for RA/SpA between June 2014-May 2015. Age, gender and medications for RA/SpA and co-morbid classical cardiovascular risk factors (hypertension, dyslipidaemia, diabetes) were retrieved and analysed.

Results: A total of 9.824 RA (61.2±14.0 years, 79% women) and 9.279 SpA patients (51.4±13.1 years, 41% women) using pharmacy-dispensed prescriptions for b-DMARDs were identified (overall prevalence 0.19%). Tumour necrosis factor inhibitors were used in 73% and 99% of RA and SpA patients, respectively. B-DMARD monotherapy (RA: 18.71%, SpA: 52.11%), b-DMARD switching during 12 months (RA: 7.73%, SpA: 6.26%), and use of methotrexate (RA: 50.25%, SpA: 27.35%) and corticosteroids (RA: 55.8%, SpA: 23.63%) differed between the two patient subgroups. In both subgroups, women received more often than men methotrexate, leflunomide, hydroxychloroquine and corticosteroids, and less often b-DMARD monotherapy. After adjustments for age, gender and concomitant drugs, the probability for anti-hypertensive and lipid-lowering drug prescription was higher in SpA than RA [OR=1.41 (95%CI: 1.29-1.54) and 1.24 (1.14-1.36), respectively, p<0.001], whereas for anti-diabetics it was similar.

Conclusions: In the first country-wide study that examines the characteristics of rheumatic disease patients under b-DMARD we show that their exact prevalence is 0.19%, with RA patients being older by 10 years, only slightly more numerous, and less likely to receive treatment for hypertension and dyslipidaemia than their demographically matched SpA counterparts. Longitudinal studies should assess the implications of these novel findings on the differential financial burden of rheumatic diseases, as well as on cardiovascular morbidity and mortality of these patients.