RESUMO
The aims of this study were to evaluate adherence of rheumatoid arthritis (RA) patients to biological disease-modifying antirheumatic drugs (bDMARDs), identify potential risk factors, and analyze the discriminative ability of the Morisky-Green test (MGT) to detect bDMARD nonadherence. One hundred and seventy-eight adult RA patients treated with bDMARDs were included. Adherence was measured using the medication possession ratio (MPR) of the previous 6 months. An MPR >80% was considered good adherence. Patient demographics, clinical characteristics, and MGT scores were assessed through a standardized clinical interview at the cross-sectional date. One-hundred and twelve patients (63%) were taking subcutaneous bDMARDs, while 66 (37%) were taking intravenous drugs. One-hundred fifty-eight (88.8%) showed good adherence to bDMARDs, while 79 (61.2%) also correctly took concomitant conventional synthetic DMARDs (csDMARDs). In logistic regression models, nonadherence to bDMARDs was associated with higher disease activity [odds ratio (OR) 1.45; 95% CI, 1.03-2.03; p = 0.032] and subcutaneous route (OR 3.70; 95% CI 1.02-13.48; p = 0.040). MGT accurately identified an MPR >80% of bDMARDs in 76.9% of the patients. A sensitivity of 78%, specificity of 70%, positive predictive value of 95.3%, negative predictive value of 28.5%, positive likelihood ratio (LR) of 2.6, and negative LR of 0.3% were obtained. Adherence may be good for bDMARDs but is low for csDMARDs. Low adherence for bDMARDs is associated with poorer disease control during the past 6 months and use of subcutaneous route. These findings should alert doctors to consider possible low adherence before declaring treatment failure.