Bridging Government Offices of Inspectors General and Academic Statistics to Reduce Fraud and Abuse

• Activity Number: 250
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Social Statistics Section
• Author(s): Cathy Furlong*+ and Timothy F. Champney and Richard Kusserow
• Companies: Caucus for Women and Integrity Management Services, LLC and Strategic Management Systems
• Keywords: Fraud, waste, and abuse ; sampling ; estimation ; Bayesian methods ; predictive modeling ; data mining

Abstract

The amount of data in Federal agencies has grown tremendously. Federal agencies have an Office of Inspector General (OIG) charged with policing fraud. Large data stores within federal agencies offer potential for monitoring fraud and applying statistical methods to save government funds. In Congressional testimony, M. Sparrow, distinguished Harvard professor observed that while fraud in government programs runs into hundreds of billions of dollars, "academia has paid almost no serious attention to this critical problem." (http://www.hks.harvard.edu/news-events/news/testimonies/sparrow-senate-testimony)

HHS OIG has been at the forefront of statistics among OIGs. We will examine the role of statistics within the HHS OIG and how it has evolved and the potential to apply statistical methods across the OIGs to help them accomplish their mission. We also examine the adoption of statistical methods in other OIGs. We discuss the application of statistically valid sampling and estimation, predictive modeling, and Bayesian methods. Our goal is to encourage collaboration between government OIGs and academic institutions to improve government's capacity to detect fraud, waste, and abuse.