One reason why carriers don't always play it safe with overpayments: the standards are lax for the use of statistical sampling, says attorney Craig H. Smith with Hogan & Hartson in Miami.
It used to be that carriers had a strict standard, in an appendix to the Medicare Carriers Manual, for the number of claims they had to audit before they could perform a statistical sample and extrapolate to all other claims, notes Smith. But in Program Memo B-01-01, issued January 2001, Medicare threw out that standard and replaced it with vague instructions saying that the size of the sample "will have a direct bearing" on the accuracy of the overpayment "but it is not the only factor that influences precision."