:: Volume 14, Issue 6 (3-2013) ::
2013, 14(6): 538-547 Back to browse issues page
Prediction of Myocardial Infarction Using Weibull Accelerated Failure Time Model: A Population Based Study
Kourosh Sayehmiri , Sajjad Alizadeh , Morteza Motedayen , Fereidoun Azizi
Zanjan university of medical sciences , mor.mot@yahoo.com
Abstract:   (9453 Views)
Introduction: Cardiovascular diseases rank first among causes of death in Iran and most other countries. Determination of the risk factors of these diseases and their importance in the incidence of myocardial infarction is very important in formulating preventive programs for these diseases. This study was designed to evaluate the relationship between the risk factors of heart diseases and the time to event for myocardial Infarction using the Weibull model. Materials and Methods: In the TGLS study, 5183 men and women, aged over 30 years, without cardiovascular disease from the East of Tehran were studied. Data were analyzed using Weibull distribution and Cox proportional hazard models. Results: During 6.5 years of follow up, a total of 83 new cases of MI (53 men and 30 women) accured. Cumulative incidence of MI was 0.0025. The Weibull multivariate regression model, showed significant relationships between FBS, sex, SBP and smoking and MI Wiebull regression models, showed that FBS levels over 126 mg/dl, adjusted for age, sex, LDL, SBP, DBP, HDL and TG, decrease time to MI by 2.87 compared to normal persons. Conclusion: Variables of FBS, age, sex, LDL, SBP, DBP, HDL, TG, and smoking are the most important predictive factors of Myocardial Infarction.
Keywords: Prediction Time, Myocardial Infarction, Risk Factors, Weibull distribution
Full-Text [PDF 330 kb]   (2714 Downloads)    
Type of Study: Original | Subject: Cardiology
Received: 2012/03/12 | Accepted: 2013/03/6 | Published: 2013/03/6

XML   Persian Abstract   Print

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Volume 14, Issue 6 (3-2013) Back to browse issues page