Medication adherence and predictive factors among patients with rheumatoid arthritis: A COM-B model guided structural equation modeling analysis

Objective: To assess medication adherence of patients with rheumatoid arthritis (RA) and explore predictive factors under the guidance of the Capability, Opportunity and Motivation to Behavior (COM-B) model. Methods: A cross-sectional study was conducted among 221 RA patients conveniently recruited from rheumatology outpatient clinics of a university-affiliated hospital in China. Data about patients' sociodemographic and disease characteristics, medication adherence, functional disability, social support, beliefs about medicines, selfefficacy, and depression were collected via self-administered questionnaires. The influence of factors within the COM-B model on medication adherence were analyzed by the structural equation model. Results: The mean score of medication adherence was 63.19 (SD 8.83), and 214 participants (96.8%) were considered non-adherent to their medication regime. Greater functional disability, higher social support, more positive beliefs about medicines, higher self-efficacy, and lower depression were significantly positively associated directly or indirectly with medication adherence, explaining 66% of the total variance. Conclusion: Patients with RA demonstrate poor medication adherence. Essentials in improving medication adherence are delaying the occurrence of disability, promoting social support, shaping beliefs about medicines, enhancing self-efficacy, and relieving depression. Practice Implications: The value of integrated interventions targeting the drivers and barriers to medication adherence identified in this study should be further explored.