Predicting cardiovascular outcomes in Chinese patients with type 2 diabetes by combining risk factor trajectories and machine learning algorithm: a cohort study
Authors' conclusion
The ML-CVD-C model, incorporating dynamic cardiovascular risk trajectories and a machine learning algorithm, significantly improves risk prediction accuracy for Chinese patients with diabetes. This model may serve as a valuable tool for more personalized cardiovascular risk management in type 2 diabetes.
Study context
- Population
- Chinese patients with T2D, Kailuan cohortKailuan cohort
- Participants
- 16,378 patients
- Comparison
- ML-CVD-C score vs ML-CVD-C, China-PAR and PREVENT
- Outcome
- 10-year cardiovascular risk (cardiovascular death, nonfatal MI, stroke); discrimination (C-index)
- Follow-up
- 10 years
Methodological detailsmachine learning, longitudinal risk factor trajectories, training and testing datasets
- Machine learning
- Longitudinal risk factor trajectories
- Training and testing datasets
Performance
As reported in the abstract
C-index
ML-CVD-C
0.80
95% CI 0.78–0.82
- ML-CVD-C (base), China-PAR and PREVENT C-index values
- 0.62-0.65
Further findings
Calibration
- ML-CVD-C provided more accurate risk estimates; all models tended to overestimate high-risk prevalence
Reclassification (net reclassification improvement (NRI), integrated discrimination improvement (IDI)) of ML-CVD-C vs the other three scores
- NRI gains
- 57.7%, 44.1%, and 47.3%
- IDI gains
- 10.1%, 7.9%, and 8.4%
Transition analysis: CV risk vs reclassification to higher risk category · observation period
- Reduction in CV risk, same risk category vs reclassified higher
- 22%
- Reduction in CV risk, reclassified lower vs reclassified higher
- 86%
Original publication
Predicting cardiovascular outcomes in Chinese patients with type 2 diabetes by combining risk factor trajectories and machine learning algorithm: a cohort study. Cardiovasc Diabetol. Published 7 February 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →