Identifying Risk Factors for Female Cardiovascular Disease Patients in Malaysia: A Bayesian Approach


  • Nurliyana Juhan Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia, 81310 Johor Bahru, Johor, Malaysia
  • Yong Zulina Zubairi Centre for Foundation Studies in Science, University of Malaya, Kuala Lumpur, Malaysia
  • Zarina Mohd Khalid Department of Mathematical Sciences, Faculty of Science, Universiti Teknologi Malaysia, 81310 Johor Bahru, Malaysia
  • Ahmad Syadi Mahmood Zuhdi Cardiology Unit, University Malaya Medical Centre, 50603 Kuala Lumpur, Malaysia



Cardiovascular disease (CVD) includes coronary heart disease, cerebrovascular disease (stroke), peripheral artery disease, and atherosclerosis of the aorta. All females face the threat of CVD. But becoming aware of symptoms and signs is a great challenge since most adults at increased risk of cardiovascular disease (CVD) have no symptoms or obvious signs especially in females. The symptoms may be identified by the assessment of their risk factors. The Bayesian approach is a specific way in dealing with this kind of problem by formalizing a priori beliefs and of combining them with the available observations. This study aimed to identify associated risk factors in CVD among female patients presenting with ST Elevation Myocardial Infarction (STEMI) using Bayesian logistic regression and obtain a feasible model to describe the data. A total of 874 STEMI female patients in the National Cardiovascular Disease Database-Acute Coronary Syndrome (NCVD-ACS) registry year 2006-2013 were analysed. Bayesian Markov Chain Monte Carlo (MCMC) simulation approach was applied in the univariate and multivariate analysis. Model performance was assessed through the model calibration and discrimination. The final multivariate model of STEMI female patients consisted of six significant variables namely smoking, dyslipidaemia, myocardial infarction (MI), renal disease, Killip class and age group. Females aged 65 years and above have higher incidence of CVD and mortality is high among female patients with Killip class IV. Also, renal disease was a strong predictor of CVD mortality. Besides, performance measures for the model was considered good. Bayesian logistic regression model provided a better understanding on the associated risk factors of CVD for female patients which may help tailor prevention or treatment plans more effectively.