Physician global assessment at 3 months is strongly predictive of remission at 12 months in early rheumatoid arthritis: results from the CATCH cohort.

OBJECTIVE The objective of this study was to determine predictors of 1-year remission in early RA (ERA) using baseline and 3-month data. METHODS The Canadian Early Arthritis Cohort (CATCH) patients were included if baseline, 3- and 12-month data were available. Regression analyses for four different definitions of remission at 12 months were done to determine baseline and 3-month predictors of remission. RESULTS Five hundred and sixty-two patients had complete data at 12 months (mean age 53.4 years, disease duration 6.2 years, 73% female). The factors at baseline associated with all four remission outcomes at 12 months were age, gender, income, education, tender joint count (TJC), patient global assessment (PtGA), HAQ and pain. Baseline ESR was associated with the 28-joint DAS (DAS28) remission only. At 3 months, all four remission definitions were associated with TJC, swollen joint count, physician global assessment (PGA), PtGA, HAQ, pain, ESR and CRP in univariate analyses. In the regression model, variables associated with Simple Disease Activity Index remission were PGA [odds ratio (OR) 0.77, P < 0.001), pain (OR 0.85, P = 0.004), age (OR 0.98, P = 0.006) and HAQ (OR 0.49, P = 0.011); Clinical Disease Activity Index remission was associated with PGA (OR 0.77, P < 0.001), pain (OR 0.85, P = 0.003), age (OR 0.98, P = 0.015) and CRP (OR 0.80, P = 0.031). DAS28 remission was predicted by ESR (OR 0.95, P < 0.001), PGA (OR 0.76, P < 0.001), age (OR 0.98, P = 0.001), HAQ (OR 0.57, P = 0.006) and male gender (OR 2.01, P = 0.005), whereas Boolean remission was associated with pain (OR 0.79, P = 0.009), age (OR 0.98, P = 0.016), PtGA (OR 0.83, P = 0.025) and PGA (OR 0.86, P = 0.038). CONCLUSION A low PGA at 3 months was consistently associated with 1-year remission in ERA.