AB0223 PHYSICIAN’S GLOBAL ASSESSMENT OF DISEASE ACTIVITY IN RHEUMATOID ARTHRITIS IS A RELIABLE AND RESPONSIVE TOOL IN CLINICAL Practice

Background: Physician’s global assessment of disease activity (PhGA) is highly influential upon treatment decisions taken by rheumatologists, surpassing the impact of DAS28. [1, 2]. However, data regarding its psychometric properties are scarce. Objectives: To evaluate the reliability and responsiveness of PhGA. Methods: We included two consecutive visits of RA patients followed in a Tertiary Rheumatology Department. Socio-demographic (age and gender) and clinical data were collected including tender (TJ28) and swollen (SJC28) joints in 28 count, C-Reactive Protein (CRP), Erythrocyte Sedimentation Rate (ESR), Disease activity Score (DAS28-3v-CRP, DAS28-3v-ESR, DAS28-4v-CRP, DAS28-4v-ESR), PhGA and Patient Global Assessment of disease Activity (PGA) through a Visual Analogic Scale (VAS) 0-100mm. Changes (Δ) between the two visits were calculated. Only patients without missing data were included. Correlations between ΔPhGA and change of other variables were assessed using Pearson’s correlations. Reliability was evaluated through Intraclass Correlation Coefficient (ICC) between two consecutive appointments in a subgroup of patients with stable disease activity (Δ DAS28-4vESR [-0.6 to 0.6]. An ICC above 0.8 was considered indicative of excellent reliability. Sensitivity to change was assessed in the subgroup of patients who improved their disease activity at least 0.6 on DAS28-4V-ESR, through Standardized Response Mean (SRM). The respective intervals of confidence were obtained through bootstrapping procedures. SRM above 0.8 were considered large. Independent factors associated with ΔPhGA were identified through multivariate linear regression analysis. p<0.05 was considered statistically significant Results: 121 RA patients (84.3% female and 64.0±12.6 years) were included. Δ PhGA was weakly correlated with ΔCRP (r=0.23), Δ PGA (r=0.31) and Δ pain (r=0.37). Moderate to strong correlations were observed with Δ DAS28-3V-ESR (r=0.55), Δ SJC28 (r=0.56), Δ DAS28-3V-CRP (r=0.58), Δ DAS28-3V-CRP (r=0.60), Δ TJ28 (r=0.62) and Δ DAS28-4V-CRP (r=0.63). ICC between two consecutive visits was 0.7, [95%CI:0.47-0.83] and SRM was -1.01 [95%CI:-1.26-(-0.73)]. In the multivariate regression analysis, ΔSJC28 (β=4.01; 95% CI:3.07 to 4.96) and Δ Pain (β=0.18; 95%CI: 0.07 to 0.28) remained as independent factors associated with ΔPhGA (R2:0.49, p<0.01) Conclusion: In this study, PhGA showed a high reliability and sensitivity to change regarding disease activity, in clinical practice. Changes in SJC had the strongest association with change in PhGA scoring, but Δ Pain was also significantly correlated (graph 1). Figure 1. Graph 1 – Explicative model to variations on PhGA References: [1]Choy T et al. Rheum (Oxford, England). 2014;53(3):482-90. [2]Rohekar G et al. Jour Rheum. 2009;36(10):2178. Disclosure of Interests: LILIANA SARAIVA: None declared, Luisa Brites: None declared, Ana Rita Cunha: None declared, Helena Assunção: None declared, Ana Rita Prata: None declared, Mariana Luis: None declared, Flavio Costa: None declared, Pedro Freitas: None declared, Marlene Sousa: None declared, José Antonio P. da Silva Grant/research support from: Pfizer, Abbvie, Consultant of: Pfizer, AbbVie, Roche, Lilly, Novartis, Catia Duarte: None declared