Background In the general population, polypharmacy (PP) is associated with increased risk of adverse events. The relationship between adverse outcomes and PP in Rheumatoid Arthritis (RA) has not been studied in depth. The mantra of treatment in RA encourages PP through combination Disease Modifying Anti-Rheumatic Drugs (DMARD).
Objectives To study the relationship between PP and serious adverse events in RA, including the influence of DMARDs within the PP count.
Methods Data from the British Society for Rheumatology Biologics Register were analysed. PP was defined as number of drugs co-prescribed at baseline, with two models: (1) including DMARDs (2) excluding DMARDs from the medication count. PP was stratified by 0–5, 6–9 and >10. Patients were studied from initiation of 1st biologic until 1st serious adverse event (SAE), 3 years of follow up, or last available visit, whichever came first. A Cox-proportional hazard model was used, with adjustment for a priori selected cofounders.
Results This study included 15,004 patients commencing biologics. The demographics are shown in table 1. Excluding DMARDs from the PP cohort, 7,115 (47%) of the patients were taking up to 5 drugs; 6,010 (40%) were taking 6 to 9 drugs; 1,870 (12%) were taking 10 or more medications. Higher levels of PP associated with older age, more severe disease, and longer disease duration. PP predictably associated with comorbidities; the relationship was not linear: comorbidity count appeared to show a ceiling effect. The overall incidence of SAEs was 25.5/100 person years (95% CI 24.7–26.3). The rate of SAEs increased across the PP counts (See Table 1). The relationship remained significant after adjusting for comorbidities. Including DMARDs within the PP count attenuated the association.
Conclusions PP is common in patients with RA and is associated with adverse outcomes especially when patients are on >10 drugs. Including or excluding DMARDs from the PP model had negligible impact on findings. The relationship between PP and comorbidity is worthy of further research, as PP represents a potentially simple but valuable predictor of adverse outcomes, and a suitable surrogate for comorbidity in epidemiological analyses.
Disclosure of Interest None declared