The pathway to approving many new medical devices is broken beyond the point of repair, according to a new report from the Institute of Medicine (IOM).

Medical equipment that comes with moderate risk—more than bandages but less than pacemakers—rarely goes through clinical testing for safety or efficacy before reaching patients, the report warned. As long as these moderate-risk items, which include artificial hips and blood-glucose monitors, are "substantially equivalent" to devices already on the market, they do not need to undergo testing before getting clearance. But such ambiguous stipulations might be putting patients in harm's way—and at the very least not ensuring the devices will do their job well. It is time to develop a new way of evaluating devices before they hit the market, the report authors concluded.

The report, which was commissioned by the U.S. Food and Drug Administration (FDA), came under fire from the medical device industry even before its July 29 release. Stephen Ubl, president of the industry group Advanced Medical Technology Association (AdvaMed), called the current process "overwhelmingly safe" in a prepared statement. Citing already "escalating review times, lack of clarity in data requirements and inconsistency in clearance decisions," he recommended that the report not receive "consideration from the Congress or the administration."

But the 12-person committee that produced the report, an expert array of physicians, inventors, lawyers and others, is not convinced by the industry's arguments. And, says committee chair David Challoner, vice president for health affairs emeritus at the University of Florida, "we surprised ourselves, that after we looked at all of this data that we came together with a unanimous recommendation."

Device precedent
The current federal law regulating medical devices was passed in 1976. Challoner notes that after studying the law's history, implementation and follow-up since its initial passage, he and the others on the committee found that "there's nothing in there about safety, efficacy or the public health." Instead, device-makers apply for clearance through a process known as 510(k), or "pre-market notification," by showing that the devices meet the "substantially equivalent" criteria and do not need to be considered for more formal pre-market approval and testing. Devices that were already on the market before the first law went into place—and by extension, newer ones similar to them—were grandfathered in without approval. "It has just allowed a daisy chain of devices back to '76 to be cleared for the marketplace," Challoner says.

The rules, although updated twice in the 1990s, also fall short when it comes to assessing software, which is becoming increasingly important to medical devices; in the case of some mobile applications the software is a medical device in its own right. "There is little specificity about the contribution of software failures that lead to recalls," the committee noted. With software updates being an ever-present factor in how machines and devices run, there should be clear ways that software will be assessed for safety and efficacy, the report noted.

The committee's qualms with the current clearance process did not stop with the pathway to the market. The report also found that the FDA's systems for keeping tabs on device performance once they are in use "often do not provide sufficient information about potential harm or lack of effectiveness." Major incidences, such as the widespread failure of metal-on-metal artificial hips, have garnered widespread attention, but subtler or more sporadic mishaps are not always assessed in a systematic fashion.

Without a solid system for gathering and comparing this post-market data, the report stated, it is "impossible to draw confident conclusions about the performance of medical devices now on the market." That, in turn, could make comparing new and old devices—a crucial facet of more evidence- and value-based health care—virtually impossible.

One of the industry's repeated complaints about the current process is that it is often variable and unpredictable, with poorly defined criteria and occasionally shifting requirements. And Challoner agrees, noting that the FDA has a big task ahead in instituting more consistency and transparency in its assessments. "That alone would make the device-makers happier," he says.

But refining these processes—even to make them more efficient in the long run—takes an investment of capital. And the report's authors acknowledge that the FDA "is persistently hindered" from regularizing the approval process by shortages of "human, fiscal and technologic resources." The agency already receives some 4,000 applications for 510(k) clearance annually. Other countries and regions have dealt with the device issue by allowing companies to compile their own safety and efficacy data. A recent report of medical devices in the European Union, for example, found that even high-risk devices were not subject to regulatory pre-market approval.

Comments