Statistics, Ethics, and Medicare Fraud

• Activity Number: 102
• Type: Invited
• Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Committee on Professional Ethics
• Author(s): George W. Cobb*+
• Companies: Mount Holyoke College
• Keywords: professional ethics ; consulting ; forensic audit ; Medicare fraud

Abstract

In my talk I summarize some of what I have learned from my work on roughly two dozen cases of alleged Medicare/Medicaid fraud. In these cases I have provided analysis, written reports, and testimony, in all cases paid by a corporation that works with trial attorneys retained by health care providers. In short, my statistical work is used to defend doctors accused of receiving payments that, according to the other side, were not justified.

I first present some background about the legal process and statistical issues involved in recovering unjustified payments, then address a handful of ethical issues: (1) Most obviously, what conflicts arise when a scientist is hired and paid by one party to an adversarial proceeding? (2) How should a statistician handle arguments that are technically correct but weak on statistical substance? Does it matter if the Trier of Fact (typically an administrative law judge) is statistically naïve and likely to dismiss other arguments that do have statistical substance? (3) If you are paid by the hour, how do you draw the line between work that you charge for and work that is "off the clock?" (4) What if you are convinced that federal guidelines and precedent ignore a critical statistical issue? A legal challenge based on the unrecognized issue is unlikely to succeed, but without such challenges, the issue will not get traction. What should a statistician do?