Can machine-learning improve cardiovascular risk prediction using routine clinical data?

Summary

Background Current approaches to predict cardiovascular risk fail to identify many people who would benefit from preventive treatment, while others receive unnecessary intervention. Machine-learning offers opportunity to improve accuracy by exploiting complex interactions between risk factors. We assessed whether machine-learning can improve cardiovascular risk prediction. Methods Prospective cohort study using routine clinical data of 378,256 patients from UK family practices, free from cardiovascular disease at outset. Four machine-learning algorithms (random forest, logistic regression, gradient boosting machines, neural networks) were compared to an established algorithm (American College of Cardiology guidelines) to predict first cardiovascular event over 10-years. Predictive accuracy was assessed by area under the ‘receiver operating curve’ (AUC); and sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) to predict 7.5% cardiovascular risk (threshold for initiating statins). Findings 24,970 incident cardiovascular events (6.6%) occurred. Compared to the established risk prediction algorithm (AUC 0.728, 95% CI 0.723–0.735), machine-learning algorithms improved prediction: random forest +1.7% (AUC 0.745, 95% CI 0.739–0.750), logistic regression +3.2% (AUC 0.760, 95% CI 0.755–0.766), gradient boosting +3.3% (AUC 0.761, 95% CI 0.755–0.766), neural networks +3.6% (AUC 0.764, 95% CI 0.759–0.769). The highest achieving (neural networks) algorithm predicted 4,998/7,404 cases (sensitivity 67.5%, PPV 18.4%) and 53,458/75,585 non-cases (specificity 70.7%, NPV 95.7%), correctly predicting 355 (+7.6%) more patients who developed cardiovascular disease compared to the established algorithm. Conclusions Machine-learning significantly improves accuracy of cardiovascular risk prediction, increasing the number of patients identified who could benefit from preventive treatment, while avoiding unnecessary treatment of others.

Related articles

• Dosage Calculations in Nursing

  This document provides a comprehensive overview of dosage calculations in nursing, emphasizing the importance of precision in medication administration using various measurement systems, and highlighting common errors and guidelines for safe practice.

• The Dynamic Nurse-Patient Relationship, Function, Process and Principles

  This document discusses Ida Jean Pelletier's Deliberative Nursing Process Theory, emphasizing the interactive nature of nursing and the essential components involved in nurse-patient interactions. It outlines how critical thinking and reflective practice enhance patient care.

• The Dynamic Nurse-Patient Relationship: Function, Process and Principles

  This document discusses the Deliberative Nursing Process Theory conceptualized by Ida Jean Pelletier, highlighting its core principles and applications in nursing care. It emphasizes the dynamic interaction between nurses and patients and the importance of critical thinking in addressing patient needs.

• The Roy Adaptation Model: A Comprehensive Nursing Framework

  The Roy Adaptation Model is a nursing framework developed by Sister Callista Roy that emphasizes the person as a holistic adaptive system interacting with a changing environment, focusing on promoting adaptation to enhance health and quality of life.

• The Life of Martha Rogers

  Martha Rogers' Theory of Unitary Human Beings provides a revolutionary framework for nursing that emphasizes understanding humans as irreducible energy fields rather than merely treating symptoms. This theory integrates concepts from quantum physics and general system theory.