AAMI recently convened a two-day meeting to follow up on last fall’s Wireless Workshop, which outlined several broad themes, including:
- Clarify roles and responsibilities.
- Manage the spectrum to improve safety and security.
- Design the wireless infrastructure for high reliability.
- Learn from other industries.
- Manage risk and prevent failure: Patient safety comes first.
Under the expert leadership of Bob Stiefel, the group of roughly 20 experts from industry, health delivery organizations, research groups, and other organizations discussed how to best address these themes. While the discussion was very lively and animated at times, there seemed to be strong agreement among the attendees concerning the various approaches that could be followed.
Over time, the mission evolved into increasing awareness of the issues related to wireless medical devices and systems, and supporting the people who may not know a lot about risk management but need to manage the hospital IT network so that patient safety is not compromised. Some of the key action items included:
- Increase the visibility of the issue and the availability of established processes such as ANSI/AAMI/IEC 80001-1 to manage wireless networks. We need to raise awareness among the management of health delivery organizations and the accreditation agencies to pay attention to this topic, especially given the almost exponential growth of wireless devices and systems.
- Increase awareness of ANSI/AAMI/IEC 80001-1, the companion TIRs, as well as supporting documentation and webinars.
- Build on the existing material to support the efforts of hospital network risk management and provide more specific checklists, data concerning expected data loss under certain conditions, and examples of risk management files based on actual use cases.
- Investigate pathways to obtain “protected” spectrum, which would allow the use of off-the-shelf components
One of the key areas of discussion related to interference and how to deal with it. Risk management (in our context) is about identifying risks and understanding the possible impact on patient safety. What is useful is understanding the probability of interference and the possible impact on the data stream (lose a packet a day, a packet a minute, a packet a second, or some other metric). Then the risk manager must map that into the possible impact on patient safety based on the use case/intended use, and, most importantly, what he or she will do to mitigate the risk to an acceptable level — if not there already.
To give an easy example, we know that microwave ovens can interfere with certain wireless transmissions. If the use case is such that there is no risk to patient safety if packets are lost due to microwave emissions, then the staff can continue to warm up their lunches. If the use case is such that the emissions may disrupt critical patient alarm information every time it is turned on, then they may have to take away the microwave and replace it with a toaster oven.
Traditional spectrum management, largely coming out of a Department of Defense background, has essentially answered the question, “Tell me when I have no risk.” There is an easy answer to that: Don’t use any wireless and you will have no risk of wireless interference. The risk manager wants to know when he really has something to worry about. To give him that, we have to connect bit and packet loss into patient or user impacting events. There is some work to do to connect the two. But it will be rewarding to make that contribution.
Senior Principal Architect
Mindray North America