With 5% Error Rate Gone As the Extrapolation Trigger, OIG Emphasizes Medicare 60-Day Rule

A 5% error rate—the point at which many providers believe they should extrapolate an overpayment instead of repaying it dollar-for-dollar—has moved to a more subjective judgment under the Medicare 60-day overpayment refund rule. The HHS Office of Inspector General sees the shift away from the 5% standard in corporate integrity agreements (CIAs) as a way to give health care organizations more flexibility and align with the 60-day rule, but it’s causing anxiety because 5% has been widely used in internal audits and is considered the only precise statement of when to apply extrapolation, experts say. With the 60-day rule applying pressure on health care organizations, there is concern about the lack of clarity in when to pull the extrapolation trigger.

Five percent has been kind of sacrosanct in the compliance world for many years. “I think everyone has been pointing to the 5% error rate as a guide in their internal overpayment calculations, at least for the first line of scrutiny, but now that it’s gone, there isn’t much to suggest what the government’s expectations are,” says San Francisco attorney Judy Waltz, with Foley & Lardner LLP. This has bubbled to the surface because attorneys have noticed that “the 5% threshold has gone away” from CIAs, notes Kelly Sauders, a partner with Deloitte & Touche in New York City.

The 5% guideline in CIAs changed almost two years ago, says Nicole Caucci, deputy branch chief of OIG’s Administrative and Civil Remedies Branch in the Office of Counsel. “We transitioned from claims-review language we previously had, which was an initial discovery sample of 50 claims. If the error rate was 5% or greater, that would trigger a full sample and systems review. You had to pay back an extrapolated overpayment.” OIG got rid of that structure and now requires a sample of 100 paid claims, with the independent review organization (IRO) doing the initial sample, she says.

This document is only available to subscribers. Please log in or purchase access


    Would you like to read this entire article?

    If you already subscribe to this publication, just log in. If not, let us send you an email with a link that will allow you to read the entire article for free. Just complete the following form.

    * required field