Wednesday, 28 June 2017 Readmissions bring financial pressure to the health care system, create duplicate workload for medical staff and increase fatality rate of patients. Statistical models were developed to predict the likelihood of patients being readmitted. In other words, high risk patients were provided with additional treatments or extension of hospital stay depending on the model output. In the case study of pneumonia patients, Becker's Hospital Review showed how data mining models helped to identify patients who were likely to be readmitted with 75% accuracy. By keeping these patients longer in the hospitals, readmission rate has improved by 217%. From 1st October 2012, hospitals in the United States will be heavily penalised for high readmission rate. The adoption of predictive modelling provides win-win outcome on multiple fronts - hospitals will avoid potential penalties; healthcare system will run more efficiently; and most importantly, patients will receive a better service and fast track their road to recovery. Becker's Hospital Review outlines the use of data analytics to reduce re-admissions. In addition, a paper in BMJ provides the technical details in predicting re-admissions.