OBJECTIVE We aimed to systematically review rheumatoid arthritis (RA) disease severity indices for use in administrative healthcare databases. We also provide an overview of alternative methods to control for RA disease severity in administrative database research. METHODS We conducted a systematic review of studies that developed/validated an index for RA disease severity using variables in administrative databases, and compared the convergent validity/reliability of the index with a standard measure of RA severity. RESULTS After reviewing 539 articles, 2 studies were included. The claims-based index for RA severity (CIRAS) was developed in one study. Components of the CIRAS included tests for inflammatory markers, number of chemistry panels/platelet counts ordered, rheumatoid factor test, number of rehabilitation and rheumatology visits, and Felty's syndrome. The CIRAS correlated moderately well with a previously validated RA medical records-based index of severity. The second study assessed whether current and lifetime treatment with disease-modifying antirheumatic drugs and/or biologics accurately predicted RA severity, as measured by the patient-reported Patient Activity Scale (PAS). Treatment variables did not fully distinguish patients in the highest and lowest quartiles of PAS scores (67.2% correctly classified). CONCLUSIONS Two claims-based indices of RA severity were identified but have some limitations for routine use. A concerted effort from experts in the field is needed to define, develop, and validate a widely applicable measure of RA disease severity for administrative database research.