SCIENCE

FDA Approval Of Medical Devices Involves Shockingly Little Research

A new report casts doubt on the rigor of even the agency's strictest approval processes.

Are federal regulators playing fast and loose with the health of the tens of millions of Americans who rely on medical devices to see, walk and survive? 

That's the suggestion of a new study that finds most medical devices go to market with little evidence that they work or are safe.

Groups like Consumer Reports and the Government Accountability Office have long criticized the U.S. Food and Drug Administration for approving medical devices that end up killing thousands of people a year. They've been most critical of the fact that the FDA allows 99 percent of new devices to go to market without any new clinical testing whatsoever.

But a team at the Yale School of Medicine found that even the 1 percent of devices that are studied before getting FDA approval likely aren't being studied enough.

This 1 percent consists of so-called "high-risk" devices, which are very different from any others on the market and which are implanted within the body or directly support human life. The Yale researchers found that these devices are subjected to an average of just one major study apiece before being granted FDA approval. This means that doctors are implanting devices like coronary stents, replacement hips and intraocular lenses in thousands of patients after they've been tested on groups of just a couple hundred people each.

Dr. Joseph Ross, a professor at Yale School of Medicine and one of the authors of the paper, said that this state of affairs is the result of federal legislation stipulating that the FDA can only require manufacturers to do the bare minimum amount of research to demonstrate the efficacy and safety of a new medical device.  

"The laws are written in a way to facilitate earlier access to therapy, because that's what most lawmakers think their constituents want," Ross said. "But as a result, we don't have the data we need to make the best decisions for our patients."  

For their study, Ross and his co-authors looked at the 28 high-risk devices approved by the FDA in 2010 and 2011. They found that, on average, each device was the subject of just two studies before getting FDA approval, and that in most cases, one of the two studies was focused more on the basic mechanics of the device than on its long-term clinical effects.

As a condition of its approval, the FDA did stipulate that manufacturers would have to carry out further studies of the effects of these devices after their release. The FDA called for 33 such studies in all, among the 28 devices approved in that period -- a little more than one additional study per device. But the research team found that just six of these 33 studies have been completed so far, five years after some of the devices were released. Moreover, many of the clinical trials that have been completed failed to compare the medical device in question with any other medical device.

Though none of the 28 devices have been implicated in the death of any patients, one of them -- a stent used in the treatment of brain aneurysms -- was voluntarily recalled because of a flaw that the FDA said could have been deadly. And more than 6,000 of the 150,000 patients who have been implanted with another device, DePuy's Pinnacle Complete prosthetic hip system, have sued the manufacturer, claiming that the metal-on-metal device caused pain and leached harmful metal ions into their bloodstreams.

Ross said that there are no protocols in place to identify patients using any given medical device, so it's impossible to say exactly how many of a particular device are currently being used, or to monitor their impact in a rigorous way.

"Right now, medical devices, unlike your computer, or your television, or even a box of pens you would buy -- they don't have a bar code on them," he said. "So we can't track devices once they leave the manufacturer."

This, at least, is changing. A law mandating identification numbers for all medical devices will be phased into effect starting in 2016

But Ross warned that the 21st Century Cures Act, a major scientific research bill passed overwhelmingly by the House of Representatives last month and currently under consideration by the Senate, could make the FDA's already low threshold for approval of new medical devices even lower.

"There are a lot of measures aimed at FDA deregulation in the bill," Ross said. "It's minimizing their ability to impose stricter evidentiary standards prior to approval for lots of products, including medical devices."

UPDATE: 8/18, 6:20 p.m. -- In an email to The Huffington Post Tuesday, the FDA noted that "some FDA mandated post-approval studies are longer than the three to five year observation period in the study," and that "therefore the studies may not be expected to be completed."

The agency also noted that it "uses a variety of other data sources and methods to monitor the safety and effectiveness of marketed medical devices."

"These can include medical device reports (MDRs) of adverse events and product problems, device registries, review of scientific literature, and other approaches depending on the specific device and public health need," the FDA said in its email.

If you or someone you know has suffered adverse consequences as a result of a high-risk medical device, The Huffington Post wants to know. Please contact the author of this story at joe.satran@huffingtonpost.com.

CONVERSATIONS