By Sharon Begley and M.B. Pell
NEW YORK, April 9 (Reuters) - Medicare billing data released by the federal Centers for Medicare and Medicaid Services (CMS) gives the public an unprecedented look at how the program pays medical providers. But the data can lead to incorrect inferences about how much doctors and others are really making.
Below are some of the caveats:
1. The data are organized according to a "National Provider Identifier" (NPI). That 10-digit number was designed to be unique to a physician, physical therapist, clinical lab, or other provider. In practice, some medical practices and labs file Medicare claims under a single NPI assigned to one of their members, said Karthik Balakrishnan, senior vice president of fraud solutions and analytics of Verisk Health, which specializes in healthcare data: doctors can bill under a group NPI, a hospital NPI, or a clinic NPI, he said.
As a result, while it may seem that one individual received outsized payments, they may have been spread among many.
That was the case with a pathologist at the Mayo Clinic in Rochester, Minnesota, who is listed as receiving $11 million from Medicare for 23 million lab tests in 2012, the fourth-highest amount in the country. In fact, that represents total billings for blood work and other tests at a clinical lab where he serves as director, said Mayo spokesman Bryan Anderson. Dr Franklin Cockerill, he said, "is a salaried employee of Mayo Clinic and is not making big money off Medicare."
"More than one person can file under one NPI, including nurses, physician assistants, and other team members," AMA president Dr Ardis Dee Hoven said in an interview. "That's one of the challenges in the way this data has been released."
In reply to a question at a news conference on Wednesday, CMS's Niall Brennan said it is also "not outside the realm of possibility" that a provider's Medicare identification number is being used by someone else.
2. Physicians and other providers did not check the information about themselves, so it may contain errors, argued the American Medical Association, which fought for decades to keep the information from the public.
Asked about physicians who denied to reporters that they administered certain expensive drugs, as shown in the Medicare data, CMS principal deputy administrator Jonathan Blum pointed out that the data show actual payments to medical providers.
"If a doctor says they never used some drug, they have a legal obligation to pay that back to the (Medicare) program," he said.
3. Some of the sky-high payments, especially to ophthalmologists and oncologists, reflect how Medicare reimburses for medications given in a doctor's office. Those include cancer drugs and eye drugs, whose prices often run into the four figures. Medicare is required by law to reimburse physicians 106 percent of a drug's average sales price.
The reported reimbursements in the database therefore reflect what physicians receive before accounting for what they paid for medicines, and thus do not directly describe what physicians make.
On the other hand, physicians can capitalize on the spread between their actual costs and Medicare's reimbursement rate. Those who buy a drug for much less than its average national sales price can pocket the difference.
4. Consumers trying to find physicians who perform a difficult procedure frequently, so as to choose someone with extensive experience, can be misled. Because the newly released data cover only Medicare, they do not show what procedures physicians performed on patients under 65. The data also excludes information from the nearly one-third of elderly patients who are members of Medicare Advantage health plans administered by private insurers. (Reporting by Sharon Begley and M.B. Pell. Editing by Andre Grenon)