Pos0467 discordance in patient and evaluator global assessment of disease activity over time and across disease activity levels in recent-onset rheumatoid arthritis

Background Treatment of rheumatoid arthritis (RA) aims to achieve remission or at least low disease activity. Composite measures (e.g. SDAI) to measure RA disease activity usually incorporate patient (PGA) and evaluator (EGA) global assessments. EGA and PGA are often discordant, reflecting divergences in evaluation of disease impact and potentially towards therapeutic goals. Objectives Patient vs Physician (PGA – EGA) discordance was evaluated according to disease duration over 5 years and to concomitant SDAI scores. Methods Demographics, clinical and serological variables, radiographic damage (Sharp van der Heijde scores), treatments, comorbidities, and Patient-Reported Outcomes (function, depression and coping with disease) were collected at baseline (median symptom duration 3.7 months) and at 12, 18, 30, 42 and 60 months from symptom onset. EGA and PGA were reported using a 0-10 cm visual analog scale (VAS). PGA-EGA discordance was divided in four subgroups[1]: negative discordance group (<-1 cm), concordance group (≥-1cm and ≤+1 cm), positive discordance group I (>1 cm and ≤3 cm) and group II (>3 cm). Results We included 822 patients from the prospective longitudinal Early Undifferentiated PolyArthritis (EUPA) cohort of recent-onset inflammatory arthritis. At baseline, concordance was present in 27.3 %, 24.9% in negative discordance and 47.8% in positive discordance (25.9 % group I and 21.9 % group II). Over 60 months, concordance increased to 45.7%, negative discordance decreased to 7.7%, while positive discordant groups remained stable (46.4 % at 5 years). Although prevalence was stable at a group level, only 40% of patients remained in positive discordance Group II over time. In positive discordant groups, patients were younger (median 58,6 years group I and 59,8 years group II, p <0.003), had higher functional impact (M-HAQ ≥1, 46.3 % to 50.2 % p <0.001), fewer tender and swollen joints (p <0.0001), but rated higher for pain, fatigue, sleep (p <0.0001) and depression (CES-D, median score 18 for both groups, p <0.005). Concordance was most frequent in remission, negative discordance in high disease activity, while group I positive discordance remained stable independent of disease activity. Group II positive discordance was highest (35-38%) with low or moderate disease activity. Conclusion In this longitudinal prospective study of patients with early arthritis, almost half evaluated their global disease activity worse than the evaluator (positive discordance) and this proportion remained stable up to 60 months. Patients in positive discordant group were younger and reported higher functional impact, pain, fatigue, poor sleep and more symptoms of depression but had lower tender and swollen joint counts. Concordance was best when in remission. Negative discordance increased with disease activity. Mild positive discordance (Group I) was stable at 25%, independent of disease activity and disease duration. Marked positive discordance (Group II) was stable over time and more prevalent with low/moderate disease activity. References [1]Desthieux C et al. Arthritis Care Res. 2016;68:1767-73 Acknowledgements We thank staff rheumatologists who recruited and followed recent-onset polyarthritis patients in EUPA: Artur deBrum Fernandes; Ariel Masetto; Lyne Bissonnette; Alessandra Bruns; Guylaine Arsenault; Pierre Dagenais; Javier Marrugo. Disclosure of Interests Audrey-Anne Couture: None declared, Nathalie Carrier: None declared, Sophie Roux: None declared, Hugues Allard-Chamard: None declared, Patrick Liang: None declared, Gilles Boire Consultant of: Abbvie Canada, Janssen Canada, Lilly Canada, Mylan Canada, Novartis Canada, Samsung Bioepis, Sanofi Canada, Teva, Grant/research support from: Unrestricted financial support for investigator-initiated initiatives: Lilly Canada, Pfizer Canada.