Concerns about medicines predict medication adherence in rheumatoid arthritis after 3 and after 12 months

Abstract

Background: According to the “necessity-concerns frameworkâ€, beliefs about medicines influence the degree of medication adherence. This has been empirically corroborated in various populations. However, evidence from longitudinal studies is scarce. This study investigated whether beliefs about medicines predict adherence in people with rheumatoid arthritis (RA) three and twelve months later. Methods: 361 patients with physician-diagnosed RA (30.5% male, mean age 60.2 years (SD=13.4)) completed the “Beliefs about Medicines-Questionnaire†(BMQ), the “Medication Adherence Report-Scale†(MARS) and sociodemographic items at three points in time: baseline, 3-months follow-up (T1), 12-months follow-up (T2). The BMQ-subscales “necessity†and “concerns†at baseline were used as explanatory variables, dichotomized MARS scores (full vs. suboptimal adherence) at follow-ups as dependent variables. Multivariate logistic regression analyses adjusted to covariates were computed. Findings: 80.9% of initial 361 patients provided information at T1, 67.9% at T2. Patients lost to follow-up did not differ from those remaining in the study. At both follow-ups, about two thirds of patients were non-adherent (T1: 68.0%, T2: 66.3%). Concerns at baseline predicted medication adherence each at 3-months (OR=1.50, 95%-CI: 1.02-2.22, p=.041) and at 12-months follow-up (OR=1.79, 95%-CI: 1.18-2.72, p=.006), with stronger concerns increasing the chance of being non-adherent. However, beliefs about necessity of medicines were not significantly associated with adherence at neither follow-up. Discussion: Employing a longitudinal study design, we demonstrated that even if RA patients hold strong beliefs about the necessity of their medicines, concerns may hinder them to fully adhere to their medication. These concerns should therefore be specifically addressed during doctor-patient consultations.