Friday, November 28th, 2014
Mostly True
Roskam
"Worldwide credit card transactions, the credit card fraud rate is 0.04 percent, compared to almost 8 percent, 9 percent, 10 percent of Medicare fraud."

Peter Roskam on Wednesday, June 12th, 2013 in an interview with Fox News

Medicare fraud rate is 8 to 10 percent, says Roskam of Illinois

Fraud by its very nature tries to hide from view. So when Rep. Peter Roskam, R-Ill., talked about how big the problem is for the Medicare program, he gave himself some wiggle room.

Roskam was talking to Fox News about a bill he has that would borrow tools used in the credit card industry to pare down the tens of billions of dollars that criminal gangs and unscrupulous operators pocket from the federal government’s health care program for the elderly.

"Worldwide credit card transactions, the credit card fraud rate is 0.04 percent," Roskam told Fox News. "Compared to almost 8 percent, 9 percent, 10 percent of Medicare fraud."

When PolitiFact checked the size of Medicare fraud five years ago, we quickly discovered that hard and fast numbers are hard to come by. In this fact-check, we’ll examine Roskam’s estimate of 8 to 10 percent and see what sources he relies on.

Before we dive into Medicare fraud, a review of credit card studies shows that Roskam is pretty much on the mark when he speaks of a 0.04 percent rate. A 2010 report from the Federal Reserve Bank of Kansas City gives a fraud rate of 0.05 percent for U.S. issued cards, both debit and credit. Roskam’s office pointed us to a trade publication, the Nilson Report, that cites an international rate of 0.04 percent rate.

On Medicare, Roskam’s office cited a Government Accountability Office report, a watchdog website created by an executive order from President Barack Obama, and an article from the U.S. Administration on Aging. All three give us rates that range from 7.9 percent to 8.5 percent for the largest Medicare program, and up to 11 percent for a smaller program, Medicare Advantage.

While the fraud rates fall into Roskam’s range, none of them is talking about fraud alone. Rather, they address the much broader category of improper payments. If a doctor orders too many tests, or provides a service but submits the wrong payment code, those come under the umbrella of improper payments.

Out and out fraud is not as large as improper payments, but it can be egregious. The FBI recently charged 25 people in Miami for allegedly bribing Medicare beneficiaries and then using their account numbers to bill for services that were unnecessary or never provided. The government says the scheme netted the conspirators about $44 million.

Malcolm Sparrow, professor of public management and a specialist in corruption control at Harvard’s Kennedy School, told us that fraud and improper payments are far from identical.

"There is a serious problem with conflating these different types of overpayment," Sparrow said. "They are quite different in origin and require very different types of control mechanisms."

When we raised this with Roskam’s office, his staff sent us several examples where Roskam has spoken collectively of fraud, waste and abuse. On Fox News, however, he spoke only of fraud.

PolitiFact also found a study that restores a measure of credibility to Roskam’s estimate. Donald Berwick, a former head of the Centers of Medicare and Medicaid Services or CMS, the agency that runs Medicare, collaborated with an analyst at RAND to produce a landmark paper in 2012 in JAMA, the Journal of the American Medical Association.

That paper offers three estimates of fraud in the Medicare and Medicaid programs: a low of 3 percent, a medium of 6 percent and a high of 10 percent. CMS told us they have no official estimate of fraud but pointed us to this study, and they cited FBI figures that mirror the numbers in this paper.

If it turns out that the high end of the range in the JAMA article is correct, then Roskam is in the right ballpark.

Of course, nobody knows for sure because fraud is a crime, and criminals don’t advertise their work.

Two caveats for policy makers

The JAMA article doesn’t stop at Medicare and Medicaid. It also looks at fraud in the health care sector as a whole, both public and private. The fraud rates don’t change much when the private sector is included.

For Sparrow at Harvard, this is no surprise.

"The systems and structures they use for control are the same, across public, private, and not-for-profit programs," Sparrow said. "They all tend to share the same strengths and weaknesses, and are roughly equally vulnerable."

Put another way, the nature of the American health care system lends itself to a certain level of fraud, and the Medicare program is no more and no less susceptible to this type of crime.

Sparrow says this also suggests the comparison to the credit card industry might be less useful than might appear. In the first place, the transactions are much simpler. When a thief tries a scam with a credit card, the card holder tends to notice by the next bill. In general, this kind of fraud is more easily detected and tracked. But Sparrow says there’s a bigger difference.

"The losses are borne by banks," he said. That creates a powerful incentive, in real time, to control the problem. In health care, the incentives are "more diffuse and ambiguous."

Our ruling

Roskam said the Medicare fraud rate is 8 to 10 percent. His office pointed us to various documents that analyzed the problem of improper payments, an issue that mixes fraud together with nominally legal activities such as referring patients for more tests than are necessary. This suggested Roskam was using an inflated estimate of fraud. However, a recent study tends, in the worst-case analysis, to support Roskam’s figures.

Roskam’s comparison to credit cards overlooks many key differences between the structures of the health care and credit card industries, and it tends to obscure the systematic nature of fraud in health care, whether public or private. But Roskam is right that credit card fraud is a tiny percentage of all transactions.

We rate the statement Mostly True.