Why the study?
Patients with coexisting T2DM and hypertension face synergistically elevated MACE risk, but evidence for prediction models developed specifically in this population remains limited.
Does an interpretable machine learning model based on routine clinical variables predict 1-year MACE in hospitalized patients with coexisting T2DM and HTN?
Population
1,054 hospitalized patients with coexisting T2DM and HTN
Comparison
Four algorithms (logistic regression, random forest, support vector machine, and XGBoost)
Design
Retrospective study
Follow-up
1-year
Key result
An interpretable logistic regression model based on seven routine clinical variables predicted 1-year major adverse cardiovascular events with an ROC-AUC of 0.828 in hospitalized patients with coexisting type 2 diabetes and hypertension.
Authors
Loading...
May inform MACE risk stratification in T2DM-HTN; leaves open prospective validation before clinical use.
Cohort (n=1,054)
No
Does an interpretable machine learning model based on routine clinical variables predict 1-year MACE in hospitalized patients with coexisting T2DM and HTN?
Effect estimate: ROC-AUC 0.828 (95% CI 0.749-0.895)
An interpretable logistic regression model using seven routine clinical variables demonstrated good internal performance for predicting 1-year MACE risk in patients with coexisting T2DM and hypertension.
Lv et al. (2026) conducted a cohort in Type 2 diabetes mellitus and hypertension (n=1,054). Logistic regression prediction model was evaluated on 1-year major adverse cardiovascular events (MACE) (ROC-AUC 0.828, 95% CI 0.749-0.895). An interpretable logistic regression model based on seven routine clinical variables predicted 1-year major adverse cardiovascular events with an ROC-AUC of 0.828 in hospitalized patients with coexisting type 2 diabetes and hypertension.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: