Radiographs present several attractive features for the assessment and monitoring of patients with rheumatoid arthritis (RA). Radiographic erosions are the closest to a pathognomonic sign in RA. Radiographs provide a permanent record of permanent damage. Excellent quantitative scoring systems have been developed by Larsen, Sharp, van der Heijde, Genant, Rau, and others. However, quantitative radiographic scoring is used only in research studies and is not included in usual treatment. Furthermore, magnetic resonance imaging and ultrasonography may be more sensitive than radiography in detecting abnormalities. Moreover, treatment of patients with RA should be initiated before evidence of damage. Reports that biologic therapy is superior to methotrexate in preventing radiographic progression are accurate for groups of patients, although methotrexate and other disease-modifying antirheumatic drugs control inflammation in 70% to 80% of patients and most patients present no radiographic progression with methotrexate. Radiographic findings are also much less significant and functional measures are far more significant in the prediction of severe outcomes of RA, including costs and mortality. Whereas prevention of radiographic progression is certainly desirable, it appears that prevention of functional disability is far more important for successful patient outcomes.