The FDA Fails to Protect Americans from the Medical Device Industry

Americans cannot rely on the Food & Drug Administration (FDA) to guarantee that medical devices are safe, and a recent Government Accountability Office (GAO) report only strengthens that conclusion.

Two years ago, Senators Dick Durbin (D-IL) and Richard Blumenthal (D-CT) requested that the GAO investigate the state of the FDA’s recall process for medical devices. This request came after reporting revealed that device maker Philips Respironics had received thousands of adverse event reports from 2010-2021. Internal components in its breathing machines were degrading, releasing toxic materials into people’s lungs and causing significant harm such as cancer.

The company not only hid thousands of these reports from the public and the FDA, but it conducted its own internal evaluation, concluding that the issue could cause “permanent impairment,” and yet it still did nothing. While Philips Respironics hid many reports, the company did send hundreds to the FDA. Nonetheless, the agency took no action even though it knew that Philips Respironics had withheld adverse event reporting in the past.

Ultimately, Philips Respironics itself initiated the recall of over 15 million devices in 2021. Even still, many patients did not learn of the recall for years after, with the FDA receiving 561 reports of deaths linked to the devices post-recall as of January 2024.

While a federal court ordered Philips Respironics to improve safety in several ways, neither the company nor its employees have faced a fine or incarceration as of today.

New Report Documents Widespread Problems  

The new GAO report – which the agency published on December 12, 2025 – continues to paint a devastating picture. It is companies rather than the FDA that initiate recalls in nearly every case; the agency has only mandated that a company recall a medical device four times since 1990 (when Congress gave the FDA that authority).

Device recalls have been on the rise — 3,934 from 2020 to 2024 — but they are still all voluntarily recalled by the manufacturers. During that four year period, there have been more voluntary recalls of medical devices than recalls of drugs and biologics combined.

The adverse event reporting system that leads to recalls is also passive and significantly controlled by manufacturers. Adverse event reporting comes from patients and physicians identifying issues, making the connection between those issues and medical devices, and then filing reports. Such passive reporting leads to significant undercounting, as a 2012 GAO report found that only an estimated 14 percent of adverse events experienced by sampled Medicare beneficiaries were reported in hospitals incident reporting systems in October 2008.

Manufacturers are responsible for gathering such information and conducting investigations that ultimately lead to reporting adverse events to the FDA. This system thus relies on reliably good behavior from device makers. Yet, Philips Respironics is not the only company that has hidden the dangers of their products from the public and regulators, as other companies have done the same. For example, juries have ordered Johnson & Johnson to pay over a billion dollars for hiding the dangers of poisoning from its metal-on-metal hip implants from doctors, patients, and regulators alike.

In 2011, the Institute of Medicine (IOM, now the National Academy of Medicine) concluded that postmarketing surveillance of medical devices is so inadequate that it is “impossible to confidently draw broad conclusions about the safety and effectiveness of products that are on the market.” A 2018 study – which the GAO has cited – estimated that medical devices potentially caused 1.7 million injuries and 83,000 deaths from 2008-2017.

Moreover, the recent December GAO report concludes that insufficient staffing has hampered the FDA’s ability to adequately oversee the medical device recall process in a timely manner. For example, the FDA was late 73.9 percent of the time from 2020-2024 in terminating device recalls after manufacturers let the agency know that it had completed recalling those devices. Thus – even for instances where manufacturers successfully recall a device – the FDA is so compromised that it fails to let patients and physicians know for more than 90 business days.

Thus, the new GAO report and the Philips Respironics story behind it show that medical devices face significant safety issues. Yet, the report alone massively undercounts the dangers of medical devices by not discussing how the FDA has approved or cleared the vast majority of medical devices without evidence that they are safe and effective.

How The Medical Device Approval Process Works 

Before the FDA approves a drug, the pharmaceutical company must provide evidence that the product is safe and effective. They are expected to do this through the gold-standard method: a randomized control trial. While there are many weaknesses in the current drug approval system with the weakening of evidence standards, it is significantly stronger than how the FDA approves medical devices for sale on the market.

From 1976-2020, around 99 percent of new medical devices that the FDA approved or cleared to go on the market went through the 510(k) pathway. This mechanism allows device makers to get their products to patients without proving that they are safe and effective.

All the companies have to do is show that the device is substantially equivalent to devices already on the market, otherwise known as a predicate device. Unless the FDA notices significant technological differences, the device maker does not need to show that a new device produces the same results or is as safe as the predicate device.

This mechanism allows the FDA to allow several generations of medical devices on the market without any trials proving safety and effectiveness. For example, the FDA cleared device manufacturer Penumbra’s JET 7 Reperfusion Catheter in 2019 as part of a lineage of 510(k) clearances that lacked data proving safety and effectiveness. Penumbra voluntarily recalled the JET 7 – which the FDA later classified as Class I, the most severe type of recall – following a likely underestimate of 200 adverse events and 18 deaths.

Lack of testing has led to tens of thousands of deaths. A 2023 study in JAMA(formerly known as the Journal of the American Medical Association) found that more than one in ten of 35,176 medical devices that the FDA cleared via the 510(k) pathway from 2003-2018 were recalled, and these devices were linked to nearly 25,000 deaths from 2003-2020.

The FDA has also cleared eventually recalled devices based on predicates devices that were also recalled. Of the 127 out of 156 medical devices subject to Class I recalls from 2017-2021 where researchers had access to predicate data, 44.1 percent got FDA clearance based on predicate devices that also were subject to Class I recalls.

Of the devices that do undergo the premarket approval (PMA) process where a company has to demonstrate safety and effectiveness, multiple analyses show that most trials have significantly weak designs. For example, a 2009 JAMA study found that among the 123 studies backing approval for 78 high-risk cardiovascular medical devices from 2000-2007, the significant majority did not have essential characteristics of a reliable study: randomization (73.2 percent) and blinding (86.2 percent). The FDA made 65.4 percent of the 78 approvals based on a single study.

The Big Picture

Looking at the state of medical device regulation as a whole, the FDA does not reliably ensure medical devices are safe and effective. The agency clears the vast majority of devices without manufacturers having to prove they are safe and effective. The proof of safety and effectiveness for the minority of devices that undergo the PMA process is very poor because of weak trial designs.

Thus, protecting Americans from harmful medical devices relies on the surveillance system that oversees devices already on the market and in use by patients. Such postmarket surveillance is passive, undercounts harms, and ultimately relies on manufacturers to sufficiently and honestly investigate and report claims of harm to the FDA.

While the FDA has the ability to force manufactures to recall dangerous medical devices, it has only exercised this power four times since 1990. Nearly all device recalls have occurred due to voluntary action by device manufacturers, the same companies with a history of hiding evidence and lying to the FDA.

The FDA is not just a weak actor in an industry-dominated system; agency officials strongly collaborate with device manufacturers. In December 2015, prior to the passage of the 21st Century Cures Act, top FDA officials – including two-time FDA commissioner Dr. Robert Califf – met with top industry leaders to work “together on proposed language for most of the device provisions in 21st Century Cures.” When the news source Inside Health Policy reached out to both the FDA and industry for comment, the response was that such collaboration was standard protocol.

The present administration claims to want to Make America Healthy Again and get rid of the corruption of the federal health agencies. Reforming weak medical device regulation and ending the FDA’s surrender to industry is a necessary step.

This first appeared on CEPR.

The post The FDA Fails to Protect Americans from the Medical Device Industry appeared first on CounterPunch.org.