Development of a predictive model for gastrointestinal side effects of metformin treatment in Chinese individuals with type 2 diabetes based on four randomised clinical trials
Authors' conclusion
To predict gastrointestinal side effects in diabetic patients for initial use of metformin, a few easily obtained features are needed to establish the model. The model can be applied to the Chinese population in clinical practice.
Study context
- Population
- Chinese patients with T2D on initial metformin, from four randomised clinical trial cohorts
- Participants
- 1,736 patients4 RCT cohorts
- Comparison
- Extra Tree model vs Simplified model
- Outcome
- Prediction of gastrointestinal side effects after initial metformin
Methodological detailsmachine learning, pooled data from four randomised clinical trials, training, internal validation and test sets, Shapley values used to report feature contribution, Online web-based tool established, Training set 1216 (70%), internal validation 260 (15%), test 260 (15%)
- Machine learning
- Pooled data from four randomised clinical trials
- Training, internal validation and test sets
- Shapley values used to report feature contribution
- Online web-based tool established
- Training set 1216 (70%), internal validation 260 (15%), test 260 (15%)
Performance
As reported in the abstract
Prediction of gastrointestinal side effects
| Test or model | AUC |
|---|---|
| Extra Tree model | 0.87 |
| Simplified model | 0.76 |
AUC
Extra Tree model0.87
Simplified model0.76
- Extra Tree had the highest AUC of ten machine learning models in validation and test set
- 0.87
Original publication
Development of a predictive model for gastrointestinal side effects of metformin treatment in Chinese individuals with type 2 diabetes based on four randomised clinical trials. Diabetes Obes Metab. Published 28 November 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →