Objectives: (1) To determine 6-month follow-up adherence and persistence with disease-modifying antirheumatic drugs in patients with early rheumatoid arthritis with disease under control. (2) To compare disease flares across adherent, nonadherent, persistent and nonpersistent patients. (3) To identify differences in adherent and persistent rates among therapeutic regimens. (4) To identify baseline prognosticators of poor compliance.
Methods: Ninety-three patients (86% female) had 4 consecutive 2-month apart evaluations during which the 28-joint disease activity score and the Health Assessment Questionnaire were scored, comorbidities and treatment recorded and a compliance questionnaire and a drug record registry applied. Descriptive statistics, Student t and χ tests and logistic regression analysis were used.
Results: At the study entry, patients had mean ± standard deviation age of 40.8 ± 13.9 years, the 28-joint disease activity score of 2.1 ± 1.1, the Health Assessment Questionnaire of 0.09 ± 0.2, and 68 of them (73.1%) had remission. During follow-up, 47 patients (50.5%) were adherent and 51 (54.8%) persistent; 14 of 68 patients (20.6%) who achieved remission had a disease flare. Incidence rate and individual risk of a disease flare were significantly greater in nonadherent and nonpersistent patients. Compared with methotrexate monotherapy, therapeutic regimens with >3 disease-modifying antirheumatic drugs had increased risk of nonadherence and nonpersistence (P ≤ 0.02). Higher previous serial erythrocyte sedimentation rate was associated to nonadherence (as was a shorter follow-up at the Clinic) and to nonpersistence (odds ratio: 1.03; 95% confidence interval: 1.01-1.05 for both, P = 0.05 and P = 0.001, respectively).
Conclusions: Therapy behavior of patients with rheumatoid arthritis with mild/no disease activity and disability was poor and translated into disease flares. Higher serologic activity was associated to poor compliance with therapy.