Objective Fluctuations in weight have been linked to cardiovascular (CV) outcomes in the general population. The present study was undertaken to evaluate whether weight fluctuation was independently predictive of CV events in patients with rheumatoid arthritis (RA). Methods We studied patients with RA from the Corrona registry. Weight change was categorized as loss of ≥10%, loss of 5–10%, stable, gain of 5–10%, and gain of ≥10%. We also categorized patients by quintile of variability in weight in prior observation periods. Cox proportional hazards models explored independent associations between time‐varying weight change and weight variability and risk of CV events before and after adjusting for CV risk factors, RA disease features, and disability. Results Among 31,381 participants, those who lost or gained 10% of their weight had greater disease activity and worse physical function, and they were more likely to smoke, have diabetes mellitus, receive corticosteroids, and be disabled. In adjusted models, a greater risk of CV events was observed in those who experienced 10% weight loss (hazard ratio HR 1.18 95% confidence interval (95% CI) 1.03–1.36, P = 0.02) or weight gain (HR 1.20 95% CI 1.04–1.38, P = 0.01). The association between weight change and CV events was stronger among participants with body mass index <25 kg/m2 for 10% weight loss (HR 1.34 95% CI 1.08–1.66, P = 0.001 and 10% weight gain (HR 1.74 95% CI 1.41–2.24, P < 0.001). Patients with greater variability in weight had a higher risk of CV events. Conclusion Recent changes and high variability in weight predict CV events in RA, particularly among thin patients. Further study is necessary to determine if weight fluctuation has adverse cardiometabolic consequences that are independent of other risk factors.