An interpretable machine learning model for predicting 1-year major adverse cardiovascular events in patients with type 2 diabetes and hypertension.
A logistic regression model predicted 1-year major cardiovascular events with good accuracy.
An interpretable machine learning model for predicting 1-year major adverse cardiovascular events in patients with type 2 diabetes and hypertension.
Evidence for prediction models developed specifically in established T2DM-hypertension comorbidity population remains limited.
To methodologically explore and preliminarily evaluate an interpretable machine learning framework for 1-year major adverse cardiovascular events prediction in hospitalized patients with coexisting T2DM and hypertension using routine clinical data.
This retrospective study included 1,054 hospitalized patients with T2DM and hypertension, of whom 249 (23.6%) experienced major adverse cardiovascular events during 1-year follow-up.
LASSO identified six stable predictors: HbA1c, age, hypertension duration, cystatin C (cystatin C), T2DM duration, and carotid intima-media thickness (carotid intima-media thickness).
Sex was additionally incorporated based on clinical relevance.
Multivariable logistic regression showed that HbA1c, age, hypertension duration, T2DM duration, cystatin C, and carotid intima-media thickness were associated with 1-year major adverse cardiovascular events risk, whereas sex was not statistically significant.
An interpretable logistic regression model based on seven routine clinical variables showed relatively good internal performance for predicting 1-year composite major adverse cardiovascular events risk in hospitalized patients with coexisting T2DM and hypertension.
cystatin C provided additional prognostic information beyond its conventional role as a renal filtration marker, although this association should be interpreted as prognostic rather than causal.
External validation is required before the model can be considered for clinical decision support.