• diagnosis
• health care quality
• access
• and evaluation
• policy
• public health
• therapeutics

Postmarket surveillance represents a critical component of the US Food and Drug Administration’s (FDA) lifecycle assessment of medical devices, as many devices—including high-risk implanted devices—are often approved on the basis of limited premarket clinical efficacy and safety data.1 The FDA collects information on real-world adverse events through passive surveillance and shares it with the public using the agency’s Manufacturer and User Facility Device Experience (MAUDE) database, a national database of device-associated deaths, injuries and malfunctions. Reporting to the FDA is mandatory for certain entities, including manufacturers and healthcare facilities, and voluntary for others, such as physicians and patients. This system is similar to the one in place under the European Union Medical Device Regulation, except that only manufacturers are required to report, and other stakeholders, including healthcare facilities and physicians, are encouraged to report to manufacturers and national competent authorities.2 With greater than 70 000 adverse event reports submitted every month, the MAUDE database currently represents the FDA’s primary source of safety signals triggering medical device safety communications, and has the capacity to provide valuable data to inform risk–benefit assessments and support evidence-based clinical care involving medical devices.3

Despite its potential value, there are increasing concerns regarding the usefulness of the MAUDE database in its current form for detecting and characterising device-related adverse events. Recent reports have highlighted high rates of misclassification for major clinical outcomes, …