Abstract 21081: Coffee Intake Affects Heart Failure and Stroke Survival and is Significant in Predicting Heart Failure and Stroke Risk
Introduction: Heart Failure (HF) and stroke are complex diseases with multiple phenotypes. While many risk factors for these diseases are well known, it is likely that there are as-yet unidentified risk factors given the complex pathophysiology of each disease. We investigated the diet domain in the Framingham Heart Study (FHS), Cardiovascular Heart Study (CHS), and Atherosclerosis Risk In Communities (ARIC) study to identify potential diet factors associated with HF and stroke.
Methods: We used random forest machine-learning methods to identify risk factors associated with cardiovascular diseases in the Framingham Heart Study. We used Cox proportional hazards analysis to investigate these findings further in the FHS and validated our findings using the CHS and ARIC datasets.
Results: Random forest analysis identified known risk factors such as blood pressure, age, and cholesterol. In addition, coffee and red meat intake were also found to be important in predicting heart failure and stroke. Survival analysis confirmed these results suggesting coffee intake is significantly associated with HF and stroke. Coffee decreased risk with HF and stroke by 7% and 8% respectively with every increased cup consumed per week (p = .001 for both HF and Stroke). Validation of this trend was confirmed in the CHS and ARIC which resulted in similar outcomes. Furthermore, we built a risk model using known risks from the Framingham Risk Score guidelines and predicted CHF and stroke outcomes. By including coffee in the model, the prediction accuracy increased by 4%.
Conclusion: Increased coffee intake is associated with reduced HF and stroke outcomes in FHS, CHS and ARIC.
Author Disclosures: L. Stevens: None. C. Görg: None. D. Kao: None.
This research has received full or partial funding support from the American Heart Association, Yes.
- © 2017 by American Heart Association, Inc.