The abstract is read at the publisher; the summary is JClub's.
A machine learning model predicting rheumatoid arthritis remission between 6 and 24 months in 1,807 Swiss treatment courses achieved an external AUC of 0.797.
- Why it matters: Clinicians need reliable, simplified tools to estimate remission after starting or switching RA therapies, but prior prediction approaches using routine registry data are limited.
- What they did: Using 21,675 courses for training and 5,418 for internal validation, an XGBoost model with 63 baseline variables was externally tested; remission was CDAI ≤2.8 within 6–24 months.
- The result: External performance showed sensitivity 0.804, specificity 0.650, PPV 0.454, and NPV 0.902, with similar results for a 10-variable model (AUC 0.802) and logistic regression (AUC 0.809).
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.