Inferring rheumatoid arthritis disease activity status from the electronic health records across health systems

Key Points

• The aim is to assess how well rheumatoid arthritis disease activity can be inferred from electronic health records across different health systems.
• Analyzed electronic health record data from multiple health institutions.
• Evaluated within-institution and cross-institution performance of disease activity inference.
• Investigated associations between inferred disease activity and major adverse cardiovascular events (MACE).
• Inferred disease activity was consistent with known associations with MACE.
• Cross-institutional performance showed limitations compared to within-institution evaluations.
• Findings support the use of inferred disease activity in generating real-world evidence.