Psychometric evaluation of the Adherence to Refills and Medications Scale (ARMS) in Australians living with gout

This study aimed to examine psychometric properties of the Adherence to Refills and Medications Scale (ARMS) in people with gout. We conducted exploratory factor analysis (EFA) and tested internal consistency (ordinal and Cronbach's alpha coefficients) and agreement (intraclass correlation coefficient (2,1)) in ARMS scores across three timepoints (baseline, 6, and 12 months) in 487 people with gout. The Kruskal-Wallis test, Spearman's rank, Kendall's tau-b correlations, and logistic regression were used to examine the criterion-related validity of the ARMS and factors associated with the ARMS. EFA suggested a one-factor structure, explaining 43.2% of total variance. High internal consistency (ordinal alpha = 0.902 at baseline) and moderate agreement in ARMS scores over time (ICCs > 0.5; p < 0.001) were observed. Lower ARMS scores (indicating better adherence) predicted achieving target serum urate (OR, 0.89; 95% CI, 0.83-0.95; p < 0.001), but not urate-lowering therapy (ULT) adherence (Proportion of Days Covered (PDC) ≥ 80%) (OR, 0.93; 95% CI, 0.81-1.05; p = 0.261). Negative correlations between ARMS and PDC were not statistically significant (Kendall's tau-b, r =  - 0.126, p = 0.078; Spearman's rho =  - 0.173, p < 0.073). Differences in median ARMS scores (IQR) of 16 (14-20), 13 (12-15), and 17.5 (15-21) in three groups of participants who reported (1) not taking ULT, (2) taking ULT and adherent, and (3) taking ULT but not adherent, respectively, were statistically significant (p < 0.001). Age was the only patient factor independently associated with optimal adherence (ARMS score = 12) (OR, 1.91; 95% CI, 1.50-2.43; p < 0.001). The ARMS is a reliable and valid measure of medication adherence behaviours in people with gout, justifying its use in gout medication adherence research.