FDA Medical Devices Approval

A recent meta-analysis suggested that FDA-approved high risk medical devices often have lacking support data. Many devices' support data had no randomization, no blinding, no controls, and often based on a single study. Though I have not read the entire article, it should be important to note that "soft" evidence does not mean no evidence. Furthermore, many drugs, such as generics, also go through a relatively simple approval process. For instance, generic drugs' ANDA often only needs BA/BE data, with no efficacy trials. Moreover, the BE data only looks at AUC, no cmax, tmax, or overall PK profile. Despite all this, generics often work as well as the brands! Also, many devices are probably "followup" or "me-too" devices and would not need extensive data for efficacy/safety support. Finally, if regulatory stringency is increased, the cost of development will increase, stifling innovation, driving up healthcare costs.

JAMA. 2009;302(24):2679-2685. Strength of Study Evidence Examined by the FDA in Premarket Approval of Cardiovascular Devices