<h3>Background</h3> Adherence in the treatment of rheumatoid arthritis (RA) ranges 20% to 70% in worldwide population.¹ In Colombia there are no studies comparing adherence and persistence to conventional and biological treatment. <h3>Objectives</h3> To determine adherence and persistence and associated factors to the treatment of conventional disease-modifying anti-rheumatic drugs (cDMARD) and biological DMARD (bDMARD) in patients with RA under real world data. <h3>Methods</h3> We conducted an observational, analytical retrospective cohort study from January 2015 to December 2016. The study population was 552 RA patients older than 18 years who received any cDMARD treatment in monotherapy or combined with two or more cDMARD and bDMARD (Rituximab and Tofacitinib were excluded). Clinical information was obtained from electronic clinical records and Morisky-Green test was performed during the follow-up. Univariate analysis (proportions and medians), bivariate analysis [relative risk (RR)] and multivariate analysis (logistic regression and Kaplan-Meier survival curve) were developed. <h3>Results</h3> Eighty nine percent were women, the median age was 59 years, 50% have ≥12 years of duration of disease. Adherence for cDMARD was 61% and for bDMARD 56%. There was an association between adherence and high degree education level (adjusted RR=2.1; <b>CI: 1.09–4.14</b>) and non-adherence with clinical factors such as high disease activity (adjusted RR=0.41; <b>CI: 0.22–0.75</b>); in the cDMARD group the greatest persistence was leflunomide (mean: 631 days) followed by methotrexate (mean: 526 days) and in the bDMARD group was etanercept (mean 1577 days) and tocilizumab (mean 1064 days). Patients with social support had greater persistence in the treatment with cDMARD (adjusted HR=2.1; <b>CI: 1.11–4.28</b>). <h3>Conclusions</h3> In real world data, education level and disease activity significantly impact adherence level in RA patients. Social support positively impacts the persistence of the treatment of RA patients, which suggest the implementation of care programs taking this aspect into consideration in order to improve outcomes. <h3>Reference</h3> [1] De Klerk E, van der Heijde D, Landewé R, van der Tempel H, Urquhart J, van der Linden S. Patient compliance in rheumatoid arthritis, polymyalgia rheumatica, and gout. J Rheumatol2003;(1):44–54. <h3>Disclosure of Interest</h3> None declared