JISAR

Journal of Information Systems Applied Research

Volume 12

V12 N2 Pages 4-17

August 2019


A Machine Learning Approach to Optimizing Diabetes Healthcare Management Using SAS Analytic Suite


Taiwo Ajani
Ferrum College
Ferrum, VA 24088, USA

George Habek
North Carolina State University
Raleigh, NC 27695, USA


Abstract: A typical healthcare business challenge was explored to demonstrate the effectiveness of data mining and machine learning techniques on large-scale medical and pharmacy claims data for about 70,000 patients newly diagnosed with type II diabetes. The business challenge was to move uncontrolled diabetic patient (H1AC > 7) to a controlled state (H1AC < 7). Two algorithms were explored for this purpose and the regression was observed to perform slightly better than decision tree. Regression model was subsequently used to score “new” data. Analyses revealed the drivers and probabilities of a patient being diagnosed as controlled. Obtained results can provide incentives for the business decision maker to explore interventional programs that could enhance the quality of treatment for the uncontrolled diabetic. The article provides an added value to business and the analytic literature by exploring and explaining predictive analytics and associated techniques from the perspective of the business.

Keywords: Data mining, Machine Learning, healthcare analytics, Regression,, SAS, Decision Tree

Download this article: JISAR - V12 N2 Page 4.pdf


Recommended Citation: Ajani, T., Habek, G. (2019). A Machine Learning Approach to Optimizing Diabetes Healthcare Management Using SAS Analytic Suite. Journal of Information Systems Applied Research, 12(2) pp 4-17. http://jisar.org/2019-12/ ISSN: 1946-1836. (A preliminary version appears in The Proceedings of CONISAR 2018)