Physician Fraud Is a Big Problem. Big Data May Have Big Solutions.
Healthcare is big business in the United States, accounting for roughly 20% of its overall GDP. And big businesses inevitably become vulnerable to fraud in a big way.
That’s why experts at the University of Maryland’s Robert H. Smith School of Business and the College of Behavioral and Social Sciences — under a grant from the National Institute of Justice — are looking to closely examine how state-of-the-art data science might create a better way to predict and detect fraud within the system.
NIJ, the research, development and evaluation agency of the U.S. Department of Justice, recently awarded a grant of roughly $843,000 to Maryland Smith’s Center for Health Information & Decision Systems (CHIDS) and its Center for the Study of Business Ethics, Regulation, & Crime (C-BERC). The grant will examine physician behavioral big data for high precision fraud prediction and detection.
The study will harness the latest data science to assess the value of incorporating non-clinical physician behavioral data into fraud-assessment models and will develop a machine learning algorithm that will estimate the likelihood that a physician will engage in fraud.
The Maryland Smith and criminology researchers working under the grant will explore whether models using big data on non-clinical physician behaviors — such as illegal behavior, consumer complaints and malpractice, conspicuous spending, other life stressors — can successfully predict whether a physician will engage in fraud in the next year or two.
They will also dive deeper into those behavioral factors, to explore which of those represents the greatest risk for fraud engagement.
And they’ll seek to determine which machine-learning algorithm is most accurate in foreseeing the medical professional’s risk of committing a fraud.
On the team are Sally Simpson, Distinguished University Professor of Criminology and Criminal Justice; Guodong "Gordon" Gao, Maryland Smith professor of Decisions, Operations and Information Technology; Ritu Agarwal, Maryland Smith Interim Dean, Distinguished University Professor; Kenyon Crowley, Managing Director, CHIDS; Michelle Dugas, Senior Research Scientist, CHIDS; and Weiguang Wang, Graduate Research Fellow, CHIDS.