Heart attack, a complex health problem in which the electrical activity of the heart becomes chaotic due to extreme heart failure conditions, has been ranked the deadliest human diseases. Recent studies have reported that remote monitoring of patients with heart failure disease could help quantify their level of risks and provide useful information for efficient therapy. Additionally, such platforms could help increase accessibility to health care delivery at a relatively lower cost. Therefore this study proposed a context aware clinical decision support model using support vector machine (SVM) for heart failure risk prediction. The proposed model's performance was evaluated using dataset of potential heart failure patients with metrics including prediction accuracy, sensitive, specificity, and receiving operating characteristic (ROC). An average prediction accuracy of 87.9% and 82.0%, were respectively achieved for the training and testing sessions of the built Radial basis function (RBF) based SVM classifier with a sensitivity value of 76.9%. The results obtained from this study might aid the development of an efficient context aware clinical decision support systems in smart cities and the society at large. © 2017 Elsevier Ltd