Development, validation, and visualization of a novel nomogram to predict stroke risk in patients

Area of Science:

• Cardiovascular Medicine
• Public Health
• Biostatistics

Background:

• Stroke is a leading global cause of death and disability.
• It imposes significant financial and societal burdens.
• Identifying predictors is crucial for risk stratification and prevention.

Purpose of the Study:

• To identify predictors of stroke risk in a large population.
• To develop and validate a nomogram prediction model for stroke.
• To assess the clinical utility of the developed nomogram.

Main Methods:

• Retrospective analysis of 11,435 participants from NHANES 2011-2018.
• Utilized Least Absolute Shrinkage and Selection Operator (LASSO) and logistic regression for variable selection.
• Developed a nomogram and validated its performance using AUC, calibration curves, and Decision Curve Analysis (DCA).

Main Results:

• Older age, lower education/income, hypertension, diabetes, depression, smoking, drinking, sleep issues, CHF, CHD, angina, and MI were significant stroke predictors.
• The nomogram achieved an AUC of 0.843 (development) and 0.826 (validation).
• The model demonstrated good calibration and clinical utility across various risk thresholds.

Conclusions:

• A validated nomogram effectively integrates demographic, vascular, emotional, and lifestyle factors to predict stroke risk.
• This tool can aid in screening high-risk individuals.
• The nomogram supports physicians in making informed treatment decisions to reduce stroke incidence.