Article questions Consumer Reports analysis of device registration process

Consumer Reports has long been a trusted source people turn to when they want to find the best washers and mattresses (among a host of other products). But it has recently taken aim at the FDA 510(k) clearance process, maintaining that many devices are not tested for safety before they go on the market.

But at least one device expert is taking with issue with this claim, labeling the magazine's reporting as "dangerous" and not quite correct in its summation of the device regulatory process.

"It is unfortunate that in its quest to make the medical device system understandable to a layperson, Consumer Reports misrepresented some key details in the regulatory process, thereby causing additional confusion and creating a dangerous precedent in media coverage of the medical device industry," writes Heather Thompson, editor-in-chief at Medical Device and Diagnostic Industry.

Floyd Larson, president of PaxMed, agrees, and particularly questions its contentions about the testing of devices. Consumers Union, the publisher of Consumer Reports, didn't take other testing outside clinical trials into account, he maintains. "You wouldn't believe the amount of testing that is done on devices," he explains, adding that much of it is done on a pro-bono basis.

Thompson also takes issue with the Consumer Reports' assertion that manufacturers only pay a user fee of roughly $4,000 and file some paperwork before their products come to market. She points to a 2010 report led by Stanford University researcher Josh Makower and sponsored by the Medical Device Manufacturers Association that put the cost of getting a device through the 510(k) process at about $24 million.

So, what does she make of the magazine's attempts at making suggestions for improving the device clearance system? She maintains they are flawed, too. For example, the magazine asked that it be required that implants and other life-sustaining devices be tested at least as rigorously as drugs. Well, there's a problem with that idea, as Eucomed's John Brennan points out. "When it comes to requiring the same type of clinical data for devices as for drugs, it is worth highlighting that for pharmaceuticals randomized control trials are relatively straight-forward to perform, and efficacy and relative safety can be statistically demonstrated in support of obtaining a marketing authorization provided that the clinical trials are sufficiently powered (depends on the number of patients)," he says. Furthermore, "you can't implant a placebo hip," as he sagely points out.

Thompson has other issues with Consumer Reports' plan. (Throwing more money at the system to "improve" it won't help the system, for example.) Whether the authors of the Consumer Reports article are paying attention remains to be seen, but Thompson raises some excellent points that readers of the magazine should be made aware of.

- see Thompson's take