Most of the time, when I think about interoperability, I visualize communication between various database-driven applications, such as EMRs, laboratory information systems and claims records. The truth is, however, that this is a rather narrow definition of interoperability. It’s time we take medical device data into account, the FDA reminds us.
In early September, the FDA released its final guidance on how healthcare organizations can share data between medical devices and other information systems. In the guidance, the agency asserts that the time has come to foster data sharing between medical devices, as well as data exchange between devices and information systems like the ones I’ve listed above.
Specifically, the agency is offering guidelines to medical device manufacturers, recommending that they:
- Design devices with interoperability in mind
- Conduct appropriate verification, validation and risk management to ensure interoperability
- Make sure users clearly understand the device’s relevant functional, performance and interface characteristics
Though these recommendations are interesting, I don’t have much context on their importance. Luckily, Bakul Patel has come to the rescue. Patel, who is associate director for digital health the FDA‘s Center for Devices and Radiological Health, offered more background on medical device interoperability in a recent blog entry.
As the article points out, the stakes here are high. “Errors and inadequate interoperability, such as differences in units of measure (e.g., pounds vs. kilograms) can occur in devices connected to a data exchange system,” Patel writes. Put another way, in non-agency-speak, incompatibilities between devices and information systems can hurt or even kill patients.
Unfortunately, device-makers seem to be doing their own thing when it comes to data sharing. While some consensus standards exist to support interoperability, specifying things like data formats and interoperability architecture design, manufacturers aren’t obligated to choose any particular standard, Patel notes.
Honestly, the idea of varied medical devices using multiple data formats sounds alarming to me. But Patel seems comfortable with the idea. He contends that if device manufacturers explain carefully how the standards work and what the interface requires, all will be well.
All told, If I’m understanding all this correctly, the FDA is fairly optimistic that the healthcare industry can network medical devices on the IoT with traditional information systems.
I’m glad that the agency believes we can work this out, but I’d argue that such optimism may be premature. Patel’s assurances raise a bunch of questions for me, including:
- Do we really need another set of competing data exchange standards to resolve, this time for medical device interoperability?
- If so, how do we lend the consensus medical device standards with consensus information system standards?
- Do we need to insist that manufacturers provide more-consistent software upgrades for the devices before interoperability efforts make sense?
Hey, I’m sure medical device manufacturers want to make device-to-device and device-to-database data sharing as simple and efficient as possible. That’s what their customers want, after all.
Unfortunately, though, the industry doesn’t have a great track record even for maintaining their devices’ operating systems or patching industrial-grade security holes. Designing devices that handle interoperability skillfully may be possible, but will device-makers step up and get it done anytime soon?