OP0222 TRAJECTORIES OF ANTIMALARIAL ADHERENCE AMONG NEWLY DIAGNOSED RHEUMATOID ARTHRITIS AND SYSTEMIC LUPUS ERYTHEMATOSUS PATIENTS: A POPULATION-BASED COHORT STUDY

Abstract

BackgroundAdherence to antimalarial regimens are suboptimal in rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) patients. Also, adherence is dynamic in nature, and varies over the time. Computing a single adherence level over a period may not explain the adherence trajectories of antimalarial in RA and SLE patients over time.ObjectivesTo identify the groups of patients with similar patterns or trajectories of antimalarial adherence over time and evaluate the baseline determinants of the group membership of adherence trajectories.MethodsAll patients with incident RA/SLE and incident antimalarial use in British Columbia, Canada, between January 1997 and March 2021, were identified using previously published definitions and administrative health data. Patients were followed up for 12 months from the index date, the time when subjects met RA/SLE criteria and were on antimalarials. We calculated a measure of adherence, the proportion of days covered (PDC) for all patients each month. Then, we used group-based trajectory model (GBTM) analysis on monthly PDC values to identify the latent groups of antimalarial adherence trajectories. The number of groups was selected using the AIC and minimum percentage criterion. Finally, we used an ordered logistic regression to evaluate the baseline determinants of the group membership of adherence trajectories. Baseline determinants of adherence for assessment included sociodemographic factors (neighborhood income quartile, region, age, sex), disease-related factors (disease type (RA vs. SLE), disease duration at index date, hypertension, angina, COPD, modified Charlson comorbidity index), healthcare system factors (hospital visits, physician and specialist visits), and medication use factors (glucocorticoids, immunosuppressives, biologics, Cox-2 selective NSAIDs).ResultsWe identified 27,510 patients with incident antimalarial use (23,997 RA and 3,513 SLE patients, mean ± SD age 56.8 ± 15.5 years, 74.8% female). Using GBTM analysis, we identified four groups for antimalarial medication adherence trajectories, representing an ordered pattern of antimalarial adherence from worse to better. Those trajectory groups were - Group 1: quick deterioration (19%), Group 2: moderate deterioration (15.7%), Group 3: slow deterioration (18.4%), and Group 4: consistent high adherence (46.8%) (Figure 1). Significant determinants of the group membership of adherence trajectories from the ordinal logistic regression model are shown in Table 1. The odds of better adherence were higher for those who, at baseline, were older, had higher income, had SLE compared with RA, had hypertension, had rheumatologist visits, and used glucocorticoids, immunosuppressives or Cox-2 selective NSAIDs.Table 1.Determinants of belonging into better adherence trajectory groups: results from an ordinal logistic regression model including only significant factors.FactorsAdjusted odds ratio (95% CI)Age1.009 (1.008-1.011)Neighborhood income quartile(Ref: 3)10.914 (0.854-0.978)20.963 (0.899-1.032)41.075 (1.002-1.153)51.041 (0.969-1.117)SLE vs RA1.551 (1.444-1.665)Have hypertension1.068 (1.002-1.138)Have angina0.878 (0.776-0.995)Have COPD0.861 (0.761-0.975)Rheumatologist visits1.035 (1.022-1.048)Glucocorticoids use1.082 (1.033-1.134)Immunosuppressives use1.177 (1.117-1.239)Cox-2 selective NSAIDs use1.080 (1.017-1.147)Biologics use0.559 (0.426-0.734)ConclusionAmong incident RA/SLE incident antimalarial users from a population-based cohort, 53.2% did not continuously adhere to the antimalarial regimen in the first year of treatment. We identified four distinct antimalarial adherence trajectory groups in this study. Sociodemographic, disease-related, healthcare system, and medication use factors associated with better adherence trajectories could help inform strategies to improve antimalarial adherence among RA and SLE patients.REFERENCES:NIL.Acknowledgements:NIL.Disclosure of InterestsNone Declared.