Observational study: Adherence to Disease Modifying Anti Rheumatic Drugs in Tunisian patients with Juvenile Idiopathic Arthritis

Abstract

Abstract Introduction: Non-or poor adherence to medication appears to be particularly common in chronic inflammatory rheumatic diseases, especially in juvenile idiopathic arthritis (JIA). It leads to a reduction of the therapeutic potential, aggravation and progression of the disease and therefore represents a real health and economic issue. The main objective of our study was to evaluate the compliance of Tunisian patients with JIA with disease modifying drugs and the potential factors that may influence this compliance. Methods This was a cross-sectional study including patients with JIA (ILAR criteria), currently taking csDMARDs and/or bDMARDs for at least 3 months. Socio-demographic, clinical, biological, radiological and therapeutic data were collected. Compliance was assessed using two methods: self-reported compliance by the patient and the parent and compliance measured by two self-administered questionnaires: the PARQ (Parent Adherence Report Questionnaire) and the CARQ (Child Adherence Report Questionnaire). Results Thirty patients, 16 girls and 14 boys, with a mean age of 24.8 ± 11 years [8–47] were included. A cs DMARD was prescribed in76.7% of patients and a bDMARD in 26.7%. Eighty percent of the parents and 76.7% of the patients reported taking their treatment as prescribed. Adherence according to PARQ and CARQ had a mean of 74.58 ± 36 [0-100] and 74 ± 34 [0-100] respectively. In the univariate analysis, PARQ compliance was positively correlated with ESR (p = 0.001 ; r = 0.643) and CRP (p = 0.008 ; r = 0.561) and negatively correlated with the mother's age (p = 0.005 ; r=-0.572), the difficulty of taking medication according to the parent (p < 0.0001; r=-0.698) and negative reactions to medication according to the patient (p = 0.012; r=-0.506) and the parent (p = 0.001; r= -0.651). Adherence to medication according to the CARQ was significantly related to the "indigent card" type of social coverage (p = 0.019). In the multivariate analysis, the predictive factors for non-adherence according to the PARQ were old age of the mother (p = 0.004), low ESR (p = 0.029) and negative reactions to medication (p < 0.0001). For the CARQ, the only predictive factor for non-compliance was the difficulty following treatment (p = 0.042) reported by the parent. Conclusion Our study showed good compliance in Tunisian patients with JIA. High maternal age, low disease activity, negative reactions to medication and difficulty in following treatment were predictive of non-adherence. Appropriate educational strategies could improve patient compliance and thus the prognosis of JIA patients.