Wednesday, June 25, 2008

Fearmongering or medical rigor? RFI induces hazards in critical care medical equipment

In "Hacking an ICD", Roy Poses reported on a research study about the susceptibility of implantable defibrillators to possible radio-based tampering.

The posting elicted responses from some quarters that the authors of the research study and of the blog posting itself were anti-industry alarmists carrying out fear-mongering.

I expressed my thoughts on the issue from the perspective of a smart person who actually knows something (i.e., just the attributes that in today's intellectually and ideologically debased world cause advice to be ignored) in "Hacking an ICD - A Dual Medical Informatics/Ham Radio Perspective."

It's time for a bit more "fear mongering." [explanatory note: the preceding sentence is satirical! -ed.]

In today's papers and newswires there's a report on the June 25 JAMA article "Electromagnetic Interference From Radio Frequency Identification Inducing Potentially Hazardous Incidents in Critical Care Medical Equipment ." Notably:

In 123 EMI tests (3 per medical device), RFID induced 34 EMI incidents: 22 were classified as hazardous, 2 as significant, and 10 as light. The passive 868-MHz RFID signal induced a higher number of incidents (26 incidents in 41 EMI tests; 63%) compared with the active 125-kHz RFID signal (8 incidents in 41 EMI tests; 20%) ...

As a ham radio enthusiast Extra class, and someone who did a clerkship in biomedical engineering in medical school, I find the outcomes of the research on the effects of RFID and other wireless systems on critical medical equipment simpy appalling.

It is reminiscent of the problems with HIT: a huge gap in resilience engineering, a rigor that would ensure the devices function well in context (i.e, in actual medical environments).

In my local paper the following is reported via AP ("Microchips may be hazards in hospitals"):

Nearly 20 percent of the cases involved hazardous malfunctions that would probably harm patients. These included breathing machines that switched off ; mechanical syringe pumps that stopped delivering medications; and external pacemakers that malfunctioned .

The frequencies that caused interference were not "special" Star Trek subspace faster-than-light communication frequencies, they are in the very conventional longwave and UHF spectrums.

The susceptibility to RFI (radio frequency interference) of any piece of electronic equipment is not a random matter. Equipment needs to be deliberately and purposefully engineered to be resistant to RF. Such RF resistance should not be a hit-or-miss proposition.

If the manufacturers had been doing their jobs properly, the results of the experiments would have been negative. However, I believe they are not doing their jobs in terms of RF susceptibility engineering, and the failure is a failure of both imagination and of dedicated concern for the "customer" (ultimately, patients) free of economic concerns (i.e., profits).

The malfunctioning of external pacemakers is particularly alarming. Internal pacemakers and ICD's came under scrutiny some months ago. In " Hacking an ICD - A Dual Medical Informatics/Ham Radio Perspective " I wrote:

The bottom line is: manufacturers need to work a little harder when they deploy wireless devices ... In biomedicine, the most meticulous resilience engineering is never a bad idea.

I believe the same issues will prove relevant regarding the devices found susceptible to common RF in the current JAMA study.

Finally, in my local paper, again via the AP, Harvard Medical School CIO John Halamka, MD was quoted as saying:

"It is absolutely an issue, but you have to manage around it," said John Halamka, chief information officer at Beth Israel Deaconess Medical Center in Boston.

While I agree with John that hospitals must out of practicality "manage around" the weaknesses of their equipment, that is a very partial solution. Just as in HIT, workarounds should be temporary and a source of very strong feedback to vendors.

The long term solution is for physicians and others who are genuinely dedicated to patient care to take the lead in demanding the tools critical to patient care be properly engineered.

-- SS

No comments: