Abstract
OBJECTIVE: We aimed to identify predictors of short- and long-term sustained remission in early rheumatoid arthritis (eRA) and very early rheumatoid arthritis (veRA), and to assess their relevance within the joint accumulation model paradigm. METHODS: This retrospective real-world cohort study included adult Colombian RA patients. Based on symptom onset, patients were classified as eRA (≤12 months) or veRA (≤3 months). Sustained remission was defined using Clinical Disease Activity Index (CDAI) thresholds maintained for ≥6 months (short-term) and ≥24 months (long-term). Predictors were identified using robust Poisson regression. RESULTS: In eRA, short-term remission was more likely in antinuclear antibodies (ANA)-positive patients (risk ratio [RR] 3.29, 95% confidence intervals [CI] 1.38~7.83) and less likely in those with higher baseline Health Assessment Questionnaire (HAQ) (RR 0.48, 95% CI 0.29~0.78). Long-term remission was more frequent in males (RR 2.67, 95% CI 1.11~6.46) and in patients with lower baseline Simple Disease Activity Index (SDAI) (RR 0.92, 95% CI 0.85~0.99). In veRA, short-term remission was negatively associated with anti-citrullinated protein antibodies (ACPA) (RR 0.74, 95% CI 0.58~0.93), swollen joint count (RR 0.52, 95% CI 0.27~1.00), and patient global assessment (RR 0.61, 95% CI 0.40~0.93). Long-term remission was associated with higher rheumatoid factor levels (RR 1.76, 95% CI 1.20~2.56), lower tender joint counts (RR 0.71, 95% CI 0.55~0.92), and slower joint accumulation rates (RR 0.51, 95% CI 0.29~0.90). These findings were consistent among disease-modifying anti-rheumatic drug (DMARD)-naïve patients. CONCLUSION: Distinct predictors of remission in eRA and veRA underscore the need for stage-specific treatment strategies. Our findings provide real-world evidence supporting the joint accumulation model and highlight the potential to optimize outcomes through tailored time-sensitive interventions.