Preprints.org
Baseline Prediction of a Two-Year Reduction in Estimated Annual Cardiovascular Risk in Type 2 Diabetes: The DIAL-STEP Study
By Villari P and Labate AM.
Abstract
Background
Longitudinal changes in estimated cardiovascular risk vary across indi-viduals with type 2 diabetes. We investigated whether baseline clinical data predict a relative decrease of at least 10% in the DIAL2-derived calibrated one-year cardiovascular risk estimate (P1Y) over two years.
Methods
This single-centre observational prediction analysis used data collected at ASST Mantova. Of 332 source records, 324 had valid P1Y estimates at baseline (T0) and two years (T2). Elastic net logistic regression and random forest were assessed with three repetitions of nested stratified five-fold cross-validation. We assessed discrimination, average precision, Brier score, and calibration of held-out predictions.
