But he and CMS Principal Deputy Administrator Jonathan Blum, in response to a follow-up question, conceded that in some states, physician assistants and others who administer treatment under the technical authority of a doctor are required to use that physician's NPI number in billing Medicare.
However, they also said that the data released Wednesday indicate that the practice of using someone else's NPI number occurs at a higher rate than can be explained by the rules of some states.
Read MoreHere's what really fueled the Obamacare surge
Lawrence Vernaglia, a health-care lawyer at the firm Foley & Lardner, said he was worried that the raw data release by CMS could be misconstrued by the general public.
For example, Vernaglia said, "Just because somebody got paid a lot doesn't mean they took home a lot."
But some patients may look at the data online "and think their doctor makes too much money," he said
The American Medical Association, which in 1979 had obtained a federal injunction that remained in effect for nearly 35 years that prevented Medicare reimbursement data from being disclosed, raised similar concerns about the accuracy and usefulness of the data to the public.
"We believe that the broad data dump [Wednesday] by CMS has significant short-comings regarding the accuracy and value of the medical services rendered by physicians. Releasing the data without context will likely lead to inaccuracies, misinterpretations, false conclusions and other unintended consequences," the AMA said in a statement.
"Thoughtful observers concluded long ago that payments or costs were not the only metric to evaluate medical care. Quality, value and outcomes are critical yardsticks for patients. The information released by CMS, will not allow patients or payers to draw meaningful conclusions about the value or quality of care."
"The AMA is disappointed that CMS did not include reasonable safeguards that would help the public understand the limitations of this data."
But Brennan and Blum, in their remarks to the media Wednesday, said having the data become public was not only the right thing to do because Medicare is paid for with taxpayer money, but also because having researchers and reporters review the data could lead to exposure of fraudulent billing and of variations in treatment between geographical regions that might warrant changes in policy.
"For too long, this information was not made public, it was protected, which raised many questions of what the program was spending, how it was spending," Blum said. "We want the press and outside researchers to mine this data to find outliers, to identify spending that does not appear to be in the beneficiaries interest and the taxpayers' interest."
McGinnis, the New Jersey pathologist, said, "I don't have any fear of data as long as it's correct."
—By CNBC's Dan Mangan