Evaluating prediction of short‐term tolerability of five type 2 diabetes drug classes using routine clinical features: UK population‐based study
Authors' conclusion
A model-based approach to predict drug-specific discontinuation for individual patients with T2D has low clinical utility. Instead of likely tolerability, prescribing decisions in T2D should focus on drug-specific side-effect risks and differences in the glycaemic and cardiometabolic benefits of available medication classes.
Study context
- Population
- People with T2D initiating one of five drug classes in UK primary care, Clinical Practice Research Datalink 2014–2020
- Participants
- 182,194 treatment initiations
- Exposure
- Short-term discontinuation across five T2D drug classes (GLP-1 RA, DPP-4i, SGLT2i, thiazolidinediones (TZD), sulfonylureas (SU)) by routine clinical features
- Outcome
- Short-term discontinuation (proxy of overall tolerability) and model prediction of drug class-specific discontinuation
- Follow-up
- 3-month
Methodological detailsUK routine clinical data (Clinical Practice Research Datalink, 2014-2020), machine learning (Bayesian Additive Regression Tree)
- UK routine clinical data (Clinical Practice Research Datalink, 2014-2020)
- Machine learning (Bayesian Additive Regression Tree)
Main finding
As reported in the abstract
Model combining all features had low ability to predict discontinuation (AUC)
0.61
Observational data: associations, not effects.
Original publication
Evaluating prediction of short‐term tolerability of five type 2 diabetes drug classes using routine clinical features: UK population‐based study. Diabetes Obes Metab. Published 16 May 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →