Background The relevant elements to evaluate health states in RA should include [ref]: a) Remission according to the new criteria of ACR/EULAR b) A criterion representing good functional ability: HAQ≤0.5 and HAQ deterioration≤0.25 c) A criterion representing structural damage: absence of progression of the Sharp score over 1 year. The concept of favorable outcome could then be defined by the association of these 3 elements. As remission is highly related to therapeutic decision-making, we may propose an alternative definition of favorable outcome keeping only the last 2 criteria (function and structure).

Objectives 1) To describe patients with favorable outcome according to the two definitions (ie using 2 and 3 criteria) at 1 year in early arthritis (EA) in daily clinical practice 2) To assess which variables at baseline are associated with favorable outcome at 1 year.

Methods Patients: from the French cohort of EA ESPOIR (at least 2 swollen joints for less than 6 months and suspicion of RA) and treated for at least 3 months with synthetic DMARDs (methotrexate, leflunomide or sulfazalasine) without biotherapy during the first year.

Data collected: at baseline, patients characteristics; every 6 months, bio-clinical variables and yearly, X-rays

Analysis: The characteristics and health status of patients with favorable outcome at 1 year (according to the two definitions, ie using 2 and 3 criteria) were analyzed. Multivariate logistic regression was used to determine which variables at baseline were independently associated with favorable outcome at 1 year according to the two definitions.

Results Within the first year of follow-up of 813 EA patients, 457 received at least 3 months of synthetic DMARDs. The mean age of patients was 49±12 years; 74% were women. At 1 year, 65 (15%) were in remission according to the new Boolean ACR/EULAR definition, 267 (61%) had favorable function criterion and 260 (65%) had no structural progression. In all, 157 (39%) presented favorable outcome using the 2 criteria (function+structure) and 44 (11%) presented favorable outcome using the 3 criteria (clinical remission+function+structure). In the logistic regression model, lower symptom duration, lower tender joint count, better perceived health state (EQ5D), and lower Sharp score were associated with favorable outcome using the 2 criteria (function+structure). In the logistic regression model, lower age, being married/in a common-law relationship and lower tender joint count (TJC) were associated with favorable outcome using the 3 criteria (Table).

Conclusions Favorable outcome appears to be achievable outcome in early RA: its predictors are close to the predictors of other validated outcomes, but favorable outcome may be a more relevant outcome in particular in terms of long-term prediction. Favorable outcome should be further assessed.

1. Felson DT et al. Ann Rheum Dis 2011; 70(3):404-13

Disclosure of Interest None Declared