Driving UDI adoption: notes from the 2016 UDI Conference

Today, there are approximately 550,000 unique device identifier (UDI) records in the Food and Drug Administration’s (FDA) Global UDI Database (GUDID). In a little over three onths, that number will have soared past 3 million, as manufacturers meet the third deadline for submission, for all Class II devices. Class II devices represent the largest regulated class of medical-surgical products sold in the United States, about 43 percent of products. Combine that with products already in the database, including the highest risk group (Class III) and all implantables regardless of class, and the majority of products will be assigned a UDI and listed in the GUDID by September 24, 2016.

This represents an important shift in the UDI regulation, from something that manufacturers must do to keep selling their products in the U.S. to a tool that the healthcare system can start putting to good use. Recognition of this shift was apparent by the near standing room only crowd at the “Realizing the Value of UDI” track on the second day of the UDI Conference held in Baltimore in April. Presented by the Association for Healthcare Resource and Materials Management (AHRMM), the track featured presentations by providers, suppliers and technology companies working together to achieve value and improve healthcare efficiency from more accurate and unambiguous identification of medical devices. Representatives from the FDA and the National Library of Medicine also provided an update on AccessGUDID, the online portal that allows the public to search and download data from the GUDID.

One of the most tangible arguments for the value of UDI came on the first day when the FDA’s Senior Advisor for UDI Adoption, Terrie Reed, compared how recalls are handled for automobiles and medical devices, respectively. As the chart below depicts, the numbers of certain makes and models of new cars sold in 2010 were similar in magnitude to specific types of procedures involving implantable devices performed that same year. But there is considerable difference in how the data linking people and products is captured and how that impacts the recall process. For example, many of us have likely received notices from automobile manufacturers about recalls or repairs that need to be completed on our cars. Those recall notices find us, even if we have moved out of state. That’s because the automobile industry works in concert with state and local governments to match specific cars by Vehicle Identification Numbers (VINs) with the individuals who own them via automobile registration data.

Now compare that to medical devices. In a world without UDI, a patient undergoing a procedure in which a device is implanted would typically receive a paper implant card containing the manufacturer’s proprietary code and the serial number for the device. The patient would then need to store that card in a safe place and would likely not have it with them in the case of a medical emergency. Now, with the availability of UDIs for all implantable devices, we have the opportunity to capture the UDI (including lot and serial number) for that device in a patient’s electronic medical record. Barring the interoperability issues we still face with electronic health records, healthcare providers would know, even if the patient did not or was not able to tell them, that he or she has a specific medical device implanted. There still remains the challenge of an easy and swift way to identify which patients have which devices and where those patients are in the event of a recall, but capturing the data electronically in a patient’s record is a foundational step in the process.

Now consider one other critical difference. When was the last time you saw someone in the automobile industry try to capture the VIN manually? Barcode scanners are routinely used to capture VINs, but in healthcare we still capture product data manually, especially in the OR and other procedural suites where the patient-product connection is most critical. Perhaps this is one use case where a shorter proprietary product code is easier, but it is certainly not safer, given that the same product code could stand for very different products sold by different manufacturers. Yes, the UDI is longer, but we have technology (barcode scanners, RFID readers) that can overcome that challenge. If other industries can do it, so can healthcare. UDI takes us one step closer.