Premorbid predictors of death at initial presentation of coronary heart disease in the Women's Health Initiative study.
Ming-Li Chen, Jin Li, Kruthika R Iyer, Catherine Tcheandjieu, Shirin Jimenez, Elias Levy Itshak Salfati, Liana C Del Gobbo, Marcia L Stefanick, Manisha Desai, Xiaonan Xue, Themistocles L Assimes
Author Information
Ming-Li Chen: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Jin Li: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Kruthika R Iyer: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Catherine Tcheandjieu: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Shirin Jimenez: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Elias Levy Itshak Salfati: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Liana C Del Gobbo: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Marcia L Stefanick: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Manisha Desai: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Xiaonan Xue: Division of Biostatistics, Department of Epidemiology & Population Health, Albert Einstein College of Medicine, Bronx, NY, USA.
Themistocles L Assimes: Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA.
Background: Premorbid health traits that increase the risk of dying at the time of initial presentation of coronary heart disease (CHD) remain poorly characterized. Methods: We followed 148,230 post-menopausal participants in the Women's Health Initiative for a median of 13.3 years. We ascertained the first occurrence of CHD and performed a joint Cox multivariate regression to identify premorbid predictors for a fatal rather than a non-fatal incident event. Results: A total of 10,714 incident CHD events including 513 fatal events accrued during follow up. A five-year increase in age, smoking 5 to 34 cigarettes per day, and a standard deviation (SD) increase in the Cornel voltage product, an electrocardiographic measure highly correlated with left ventricular mass index on echocardiography, each independently increased the relative risk (RR) of dying from one's initial presentation of CHD by 46 % (95 % confidence interval [CI], 35 to 58 %), 30 % (8 to 51 %,), and 17 % (7 to 28 %), respectively. A high level of recreational physical activity (>1200 metabolic equivalent (MET) minutes per week) reduced one's relative risk by 32 % (12 to 49 %). A significant dose-response effect was observed for both physical activity and smoking and the reduction in absolute risk of presenting with fatal CHD associated with a healthy lifestyle was roughly equivalent to the difference in risk observed among women separated in age by approximately 10 years. Conclusions: Modifiable factors affect a post-menopausal woman's risk of dying from her initial presentation of CHD. Our findings may reduce case-fatality risk of CHD by motivating individuals at risk to adopt and/or adhere to established primary prevention strategies.
