Weight change and longitudinal disease activity in rheumatoid arthritis: a real-world cohort study.

OBJECTIVE: To assess whether weight change during routine clinical care is associated with DAS28-CRP trajectories in rheumatoid arthritis (RA) with obesity. METHODS: Retrospective cohort study of data from Montpellier University Hospital. Adults with RA and obesity (BMI ≥30 kg/m²) were included. Weight change (percentage from baseline) was summarised for each patient as a regression-based slope using all available weight measurements. Outcomes were repeated DAS28-CRP, DAS28-CRP components, remission (DAS28-CRP <2.6) and low disease activity (≤3.2). Mixed-effects models estimated weight change-by-time interactions with prespecified adjustment. Missing data were handled using multiple imputation. Sensitivity analyses included categorical weight-change modelling, a 6-month landmark design, and censoring at DMARD change. RESULTS: Among 106 participants with estimable weight change (24 with ≥5% loss, 64 stable, 18 with ≥5% gain), mean BMI was 33.7 kg/m² and obesity was class I in 76 (71.7%). The weight-change-by-time interaction for DAS28-CRP was null (-0.004 units/year per 5% weight change, 95% CI -0.026 to 0.018). Weight change was not associated with remission (OR 1.01 per 5%, 95% CI 0.96 to 1.06) or low disease activity (OR 1.00 per 5%, 95% CI 0.94 to 1.07), and no signal was observed for DAS28-CRP components. Sensitivity analyses were consistent (weight loss ≥5% vs stable: +0.021 units/year, 95% CI -0.076 to 0.119; weight gain ≥5% vs stable: -0.003 units/year, 95% CI -0.158 to 0.153). CONCLUSIONS: In RA with predominantly class I obesity, we did not observe a clear association between routine clinical care weight change and improved inflammatory disease activity. These findings should be interpreted cautiously given the modest sample size and limited precision, and small clinically meaningful effects cannot be excluded.