Background Suboptimal adherence is a persistent problem with far-reaching consequences in chronic diseases. In patients suffering from rheumatoid arthritis (RA), a weekly intake of methotrexate (MTX) is the basic drug treatment in most patients.
Objectives This observational study aims to investigate whether MTX is taken correctly by RA patients and to identify determinants of non-adherence.
Methods Eligible patients with RA according to ACR criteria and treated with MTX were included by the rheumatologist during outpatient visits between September to November 2013. Intake of MTX was recorded with an electronic monitoring system (MEMS, Aardex, Switserland) for 16 weeks. In addition, two compliance questionnaires (the Medication Adherence Report Scale (MARS-5) and Compliance Questionnaire Rheumatology (CRQ)) as well as a visual analogue scale (VAS) measuring MTX compliance were used as self-reports of adherence.
As possible determinants of adherence data on demographics, disease and treatment characteristics i.e. disease activity (DAS28), functional limitations (Health Assessment Questionnaire) and somatic symptoms (Patient Health Questionnaire physical symptoms PHQ-15) were collected. Also data on depression (Patient health questionnaire depression PHQ-9), illness cognitions (Illness Perception Questionnaire IPQ-K), motivation (Treatment Self-Regulation Questionnaire) and social support were collected.
Adherence was defined as the correct intake of at least 90% of the prescribed medication. A multivariate logistic regression model was used to identify the factors explaining the variance in adherence.
Results Of 146 eligible patients, 140 agreed to participate. 129 patients recorded their MTX intake successfully with the electronic monitoring system and completed all questionnaires. Over the 16 weeks, a high average adherence rate of 87% was found. Three quarters of the patients met the 90% adherence criterion. In the multivariate model living alone was associated with non-adherence (OR:0.18, 95% CI: 0.06-0.55) and a slightly better adherence was noticed in older patients (OR: 1.05, 95% CI:1.02-1.09). Other factors negatively related to adherence were nausea as a side effect of MTX treatment (non-adh 7%, adh 28%, p=.002), incapacity of work (non-adh 7% adh 21%, p=.022), somatic symptoms (non-adh 20% adh 22%, p=.022) and depression (non-adh 4%, adh 6%, p=.042).
Of the three self-reported adherence scales, the VAS correlated best with the results of the electronic monitoring system (r=0.552). The correlation between the MARS-5 questionnaire and the electronic monitoring system was much weaker (r=0.356). There was no significant correlation of the CQR with the electronic monitoring system.
Conclusions A high adherence was observed in RA patients treated with MTX. Social factors (living alone, incapacity of work) and side effects of the medication showing a significant association with non-adherence should be taken into account by the rheumatologist and the rheumatology nurse. In daily practice a simple VAS scale seems to be the most reliable measure of adherence.
Acknowledgements L. De Clerck Grant/resarch support from: MSD, E. De Cuyper: None Declared, S. Maes: None Declared, V. De Gucht: None Declared, Y. Van Camp: None Declared, M. Elseviers: None Declared
Disclosure of Interest None declared