SAT0691 Adherence to disease-modifying drugs in chronic inflammatory rheumatic diseases: several questionnaires, diverse patient characteristics and some efficacious interventions – a systematic literature review

Background In chronic inflammatory rheumatic diseases (CIRDs), adherence to disease-modifying drugs (DMD) is only moderate. Non-adherence may lead to complications, unnecessary treatment switches and heightened costs. Physicians are often at a loss when faced with non-adherence. Objectives To obtain an overview of how to measure adherence, who to screen particularly (risk factors of non adherence) and interventions to enhance medication adherence to DMD in patients with CIRDs. Methods A systematic literature review was performed in Pubmed, Cochrane, Embase and websites in 2017. All english and french studies related to methods to measure non-adherence, risk factors regrouped in 5 domains according to World Health Organisation (patient characteristics, health status, treatments, socio-economic conditions and relations with caregivers and the health system) and interventions for non-adherence regrouped in 5 modalities (educational, behavioural, cognitive behavioural, multicomponent interventions or others) were selected. The scope was limited to CIRDs (i.e., rheumatoid arthritis (RA), spondyloarthritis (SpA), psoriatic arthritis (PsA), cristal-induced arthritis (CIA), connective tissue diseases (CTD), vasculitis and auto-inflammatory diseases), and to DMD (i.e., mainly conventional DMARDs, biologics and targeted synthetic DMARDs). Results After screening 1131 publications and 194 other documents, 231 relevant papers were analysed for measuring adherence (60% in RA, 8% in SpA, 6% in PsA, 11% in CIA and 15% in CTD), 117 for predicting non-adherence (55% in RA, 9% in SpA, 14% in gout and 22% in CTD) and 22 for improving adherence (72% in RA, 8% in SpA, 16% in systemic lupus and 4% in gout). Objective measurements of non-adherence included: delivery data (total number of use: n=92), pill counts (n=8), medication event monitoring system (n=9), blood level assessments (n=7). Subjective measurements included: patient global assessment (n=57), and 4 questionnaires. The most used questionnaire was the Morisky Medication Adherence Scale and the most widely validated in rheumatology were the Compliance Questionnaire on Rheumatology and the Medication Adherence Self-report Inventory. Around 100 predictive factors were identified. Polymedication, mood disorders, lack of information and poor physician-patient relationship were associated with lower adherence. Regarding management options for non-adherence, 13/22 studies were randomised controlled trials (1535 patients) and only 5 (38%) were positive (774 patients). Educational interventions were the most represented with the highest level of evidence: 8/13 trials (1017 patients) with 4/8 yielding positive results. Conclusions Despite the importance of medication adherence in CIRDs, this review revealed limitations in methods to measure non-adherence, a multiplicity of non-adherence risk factors and a relative lack of evidence on interventions to improve medication adherence. It’s important to improve the assessment and optimisation of adherence in CIRDs. Acknowledgements AbbVie France funded this initiative but played no role in the points to consider. Disclosure of Interest None declared