TY - JOUR
T1 - Inferring Rheumatoid Arthritis Disease Activity Status From the Electronic Health Records Across Health Systems
AU - Cheng, David
AU - Wang, Xuan
AU - McDermott, Gregory C.
AU - Hanberg, Jennifer S.
AU - Love, Zoe
AU - Zhong, Katherine
AU - Jeffway, Mary
AU - Hou, Jue
AU - Panickan, Vidul
AU - Sangar, Rahul
AU - Qi, Ying
AU - Melley, Connor
AU - Costa, Lauren
AU - Feil, Dakota
AU - Matty, Rachael
AU - Weisenfeld, Dana
AU - Animashaun, Abisayo
AU - Schreiner, Aimee
AU - Morini, Sara
AU - Rusnak, Lauren
AU - Cagan, Andrew
AU - Paudel, Misti
AU - Gaziano, J. Michael
AU - Sauer, Brian
AU - Weinblatt, Michael
AU - Baker, Joshua
AU - England, Bryant
AU - Ho, Yuk Lam
AU - Cho, Kelly
AU - Monach, Paul
AU - Cannon, Grant W.
AU - Shadick, Nancy
AU - Mikuls, Ted R.
AU - Cai, Tianxi
AU - Liao, Katherine P.
N1 - Publisher Copyright:
© 2026 The Author(s). Arthritis & Rheumatology published by Wiley Periodicals LLC on behalf of American College of Rheumatology.
PY - 2026
Y1 - 2026
N2 - Objective: Disease activity plays a central role in rheumatoid arthritis (RA) clinical studies. The inconsistent availability of data on disease activity in real-world electronic health records (EHRs) data has limited the ability to generate real-world evidence (RWE). This study aimed to develop and validate scalable machine learning (ML) models to infer RA disease activity from EHR data. Methods: We used EHR data from Mass General Brigham (MGB) and the Department of Veterans Affairs (VA) linked with RA registries that prospectively collected the Disease Activity Score with 28-joint counts (DAS28). Features for the algorithm were extracted from the EHRs including structured data (eg, diagnosis codes and narrative data using natural language processing [NLP]). ML models were trained on the registry-collected DAS28. The performance of models trained within the same institution and across institutions was evaluated. To assess face validity, we estimated the association between inferred disease activity and major adverse cardiovascular events (MACEs) with stratified Cox models. Results: We studied 1,105 MGB and 2,631 VA patients with RA. Models with structured data achieved an area under the receiver operating curve (AUC) of 0.68 to 0.70; models incorporating structured and NLP achieved higher performance (MGB, AUC = 0.843; VA, AUC = 0.833). Cross-institution validation demonstrated limited transportability of algorithms across sites (MGB→VA, AUC = 0.679; VA→MGB, AUC = 0.718). Within the same institution, inferred disease activity was significantly associated with increased risk for incident MACEs (MGB, hazard ratio [HR] = 1.12; VA, HR = 1.14). Conclusion: RA disease activity can be inferred at scale from within-institution EHR data, though cross-institution performance is limited. The inferred disease activity replicated known associations with MACEs, and the results support its use in future studies to generate RWE.
AB - Objective: Disease activity plays a central role in rheumatoid arthritis (RA) clinical studies. The inconsistent availability of data on disease activity in real-world electronic health records (EHRs) data has limited the ability to generate real-world evidence (RWE). This study aimed to develop and validate scalable machine learning (ML) models to infer RA disease activity from EHR data. Methods: We used EHR data from Mass General Brigham (MGB) and the Department of Veterans Affairs (VA) linked with RA registries that prospectively collected the Disease Activity Score with 28-joint counts (DAS28). Features for the algorithm were extracted from the EHRs including structured data (eg, diagnosis codes and narrative data using natural language processing [NLP]). ML models were trained on the registry-collected DAS28. The performance of models trained within the same institution and across institutions was evaluated. To assess face validity, we estimated the association between inferred disease activity and major adverse cardiovascular events (MACEs) with stratified Cox models. Results: We studied 1,105 MGB and 2,631 VA patients with RA. Models with structured data achieved an area under the receiver operating curve (AUC) of 0.68 to 0.70; models incorporating structured and NLP achieved higher performance (MGB, AUC = 0.843; VA, AUC = 0.833). Cross-institution validation demonstrated limited transportability of algorithms across sites (MGB→VA, AUC = 0.679; VA→MGB, AUC = 0.718). Within the same institution, inferred disease activity was significantly associated with increased risk for incident MACEs (MGB, hazard ratio [HR] = 1.12; VA, HR = 1.14). Conclusion: RA disease activity can be inferred at scale from within-institution EHR data, though cross-institution performance is limited. The inferred disease activity replicated known associations with MACEs, and the results support its use in future studies to generate RWE.
UR - https://www.scopus.com/pages/publications/105038878649
UR - https://www.scopus.com/pages/publications/105038878649#tab=citedBy
U2 - 10.1002/art.70181
DO - 10.1002/art.70181
M3 - Article
C2 - 41972843
AN - SCOPUS:105038878649
SN - 2326-5191
JO - Arthritis and Rheumatology
JF - Arthritis and Rheumatology
ER -