Public health implications when heart valves, hip replacements and other medical devices are not tested in clinical trials

The Food and Drug Administration (FDA) requires companies making new prescription drugs to submit clinical trials to support an application for approval, but there is no similar requirement for more than 90% of medical devices. As a result, most medical devices that are sold in the U.S. are not proven safe in clinical trials, and no scientific clinical data are available to patients or physicians on effectiveness or safety. By law, clinical trials are required for FDA approval of high-risk medical devices, such as heart valves, defibrillators, HIV tests, and other devices that can sustain, save, or harm a life. However, the FDA has been criticized for not always following the law. Using the FDA's own data, we analyzed the devices that were subject to “high-risk recalls” from 2005 through 2009. We found that the FDA recalls of devices that were considered potentially fatal or seriously harmful were usually of devices that had not undergone clinical trials prior to FDA approval, or afterwards. Only 20% of these “high-risk recalls” had undergone clinical trials prior to sale. When our findings were published online in the Archives of Internal Medicine in February 2011 and reported on national TV news programs and national newspapers, the response from medical device manufacturers and several key Members of Congress was immediate. They stated that clinical trials were not necessary for most medical devices, that the current FDA system for devices was too burdensome rather than too relaxed, and that the changes we proposed would cause job losses as medical device businesses closed. In my proposed presentation I will provide new data on the public health costs of defective medical devices that are later recalled, focusing on implanted medical devices. I will provide data on deaths and serious injuries that were reported to the FDA, the risks of undiagnosed diseases from inaccurate diagnostic tests, as well as medical costs to patients and the medical care system. I will also discuss whether the FDA is appropriately classifying devices and recalls as “high risk;” for example, glucose test strips were classified as moderate risk by the FDA, but when they were found to be inaccurate, they were recalled as high risk.