Title: Predictive risk modelling for forecasting high-cost patients: a real-world application using Medicaid data
Authors: Sai T. Moturu, William G. Johnson, Huan Liu
Addresses: Department of Computer Science and Engineering, School of Computing and Informatics, Arizona State University, Tempe, AZ 85287-8809, USA. ' Center for Health Information and Research, Department of Biomedical Informatics, ASU Biomedicine, Phoenix, AZ 85004-2430, Mail Code 8120, USA. ' Department of Computer Science and Engineering, School of Computing and Informatics, Arizona State University, Tempe, AZ 85287-8809, USA
Abstract: Approximately two–thirds of healthcare costs are accounted for by 10% of the patients. Identifying such high-cost patients early can help improve their health and reduce costs. Data from the Arizona Health Care Cost Containment System provides a unique opportunity to exploit state-of-the-art data analysis algorithms to mine data and provide actionable findings that can aid cost containment. A novel data mining approach is proposed for this challenging healthcare problem of predicting patients who are likely to be high-risk in the future. This study indicates that the proposed approach is highly effective and can benefit further research on cost containment.
Keywords: predictive risk modelling; healthcare expenditure; Medicaid; high-cost patients; data mining; non-random sampling; risk adjustment; skewed data; imbalanced data classification; healthcare costs; care cost containment; high-risk patients; risk prediction.
International Journal of Biomedical Engineering and Technology, 2010 Vol.3 No.1/2, pp.114 - 132
Published online: 30 Nov 2009 *Full-text access for editors Access for subscribers Purchase this article Comment on this article