Exploring Health Information Technology Events from FDA MAUDE Database

To facilitate preventing patient safety events, researchers have been using the Federal Drug Administration (FDA) Manufacturer and User Device Experience (MAUDE) database as a publicly accessible data source for retrieving reports of medical devices including Health Information Technology (HIT) devices. A combined search strategy built with keywords in several fields such as Generic Name, Brand Name and Manufacturer is commonly used to extract reports of specific classifications of devices in the MAUDE. However, another structured field, Classification Product Code, is rarely visited. To improve the secondary usage of the MAUDE database in retrieval of HIT events, we extracted HIT reports from a nine-year MAUDE dataset by combining keywords filter and expert review, and explored Classification Product Codes. The distributions and relationships between Product Codes and keywords in Generic Name/Manufacturer in HIT reports were visualized to provide a view of the landscape. This study presents a new perspective for improving the search strategy of HIT events in MAUDEwhich would facilitate the understanding of HIT events for improving patient safety.