Between 2013 and October 2015, the US Food & Drug Administration received over 5,000 adverse event reports related to Essure, a permanent birth control implant manufactured by Bayer. Most news sources have said that the vast majority of these side effect reports, which often involve severe abdominal pain, abnormal bleeding and allergic reactions, were submitted voluntarily by patients, not health care professionals. But one former FDA employee is casting doubt on that official story.

Essure Reports Riddled With Errors, Device Expert Says

Madris Tomes, who managed the FDA’s Adverse Event Reporting System from 2012 to 2014 according to her LinkedIn profile, has been analyzing the adverse event reports for months now. In February 2016, she delivered the results of an initial review to Pennsylvania State Representative Mike Fitzpatrick, who has thus far led the Congressional charge to have Essure recalled entirely. Fitzpatrick in turn disclosed Tomes’ findings in a well-attended press call on February 17: Essure had been linked to 303 fetal deaths since its approval, not the five cases publicly reported by the FDA.

Now, Tomes has dropped another bombshell. Someone, she says, has been mislabeling adverse event reports. In her new analysis, Tomes identified more than 300 reports submitted by physicians and nurses since August 2015. But for most, the data field listing the reporter’s occupation in FDA’s system doesn’t read “physician” or “healthcare professional.” It says only “other.” Reports on the issue from NBC New York suggest that most of these mislabeled adverse events were submitted via Bayer, Essure’s manufacturer.

Does FDA Trust Doctors More Than Patients?

Adverse event reports can be submitted by either health care professionals or patients. In both cases, reporting is voluntary and thus the true number of side effects actually experienced is necessarily higher than the number reported to FDA. Complicating the picture, some health professionals and consumers report adverse events directly to a product’s manufacturer. In many cases, the company is then required by federal law to pass the report along to the FDA.

Hospital Patient On Gurney

But in all of these cases, the FDA says it treats every report equally, and doesn’t regard physician-submitted reports as more trustworthy than ones sent in by a patient. In comments to NBC, Tomes suggested that such claims of impartiality are far-fetched. She told reporters the mislabeled reports would amount to “concealment” if Bayer, in fact, submitted them incorrectly. What would there be to conceal if the FDA didn’t care one way or the other who submitted a report?

Common logic would appear to be on Tomes’ side. As Elena Mendez, who had to have her implants removed after they perforated her endometrial cavity, says, “Doctors know better. Medical people know better.”

Companies are only required to submit those reports that suggest that a device “caused or contributed to a death or serious injury,” according to FDA. This distinction is more than semantic. As just one example, a federal investigator named Timothy Grome discovered evidence that Essure’s first manufacturer, Conceptus Inc., had withheld several documents reporting instances in which an Essure implant had perforated a patient’s bowels from the FDA. Conceptus disputed Gromes’ findings, saying their reported perforations were the result of physician’s error, not failure of the Essure device.

Computer Glitch?

But the mislabeled reports aren’t Bayer’s fault, the FDA claims. Instead, blame should be placed on a computer glitch. In an official statement, FDA spokesperson Deborah Kotz said the federal agency was taking responsibility for the inaccuracies:

“the FDA became aware of a coding error that occurred during the upgrade of our adverse event reporting database in August of 2015. It affected the ‘reporter occupation’ field in the database and resulted in the data field not correctly specifying whether the person filing the report was a doctor or nurse.”

That’s a very different story from the one NBC heard just days earlier.

In an email sent to the news team five days prior to her public statement, Kotz suggested that the FDA hadn’t made any mistakes, saying that the agency just enters the exact codes submitted by a device manufacturer. The implication seemed clear: anything wrong with the reports was already wrong before it got to us. But now Kotz has backtracked, placing responsibility in the aforementioned “coding error,” a problem she says has been fixed.

Fitzpatrick Calls For Internal Investigation

For his part, Mike Fitzpatrick is questioning the FDA’s timeline. While Kotz claims that a glitch in August of 2015 mislabeled the adverse event reports, Fitzpatrick says problems began even earlier, in July of that year.

Before that time, reports on Essure were almost never coded as “other” or “patient,” but nearly always submitted by physicians or nurses. “This coding changed dramatically in July of 2015,” however, Fitzpatrick wrote in a letter to the FDA, highlighting his “extreme concern” over the issue, since “the FDA announced it was reviewing the safety and effectiveness of the Essure device” on July 17, 2015. That’s a strange coincidence, Fitzpatrick argues, one that warrants internal investigation.