To be fair, you almost have to ask the question as to “what area has software not touched” with either glitches or recalls, it’s everywhere and probably the last place we want to see that issue is with medicalimage devices.  Some are much more critical than others of course, especially one that is implanted as we have probably all read the stories about the potential to be hacked.  Sometimes the software is only as good as the next hacker that comes along too.

Medical Device Recalls Involve Software Flaws in Almost 1/4 Cases–Why the FDA Needs Engineers Just the Rest of the World

A few years ago I wrote this magazine article about FDA recalls and myself I wouldn’t mind having “over the counter” drugs having a bar code that I could scan to check for recalls, and of course this would be a massive project.  The FDA has done quite a bit of work on the project, but Jay Crowley who headed the project left not too long ago for a job in private industry.  Here’s a condensed version of links that I have written over the years on the topic. 

FDA Recalls Never End

In addition there’s different classes of devices too with everything from very basic consumer devices all the way up to devices that now algorithmically enter data into electronic medical records.  The mobile movement of mHealth is growing too and you see everything out there with all the ideas floating around too, so in that area with more integration with data and records I think it’s a good idea to put these two divisions of HHS together so we can all benefit.  It’s only going to keep growing and I see a much brighter future of the two collaboration versus dueling over who’s going to regulate. 

mHealth Why Battle Over Where the Responsibility Lies, Make the ONC A Division of the FDA, Better Collaboration and Software Engineering Exposure Both Ways Around

I wish there was an simpler answer to reducing recalls as does everyone I’m sure but the reality of the complex software world we live in today, I don’t think that is going away any time soon.  It’s everywhere and was very public with the release of with software issues so the reality here is that hopefully government and consumers wake up to this fact and stop believing in “miracle algorithms fairies” to come in and fix things in a hurry, it takes time with writing complex code today.  The more devices you have out there, the more recalls you are going to have and the best way to address this is to use analytics to identify big issues of course as well as small one too.  Again I go back to my original idea to make this information available to consumers and clinical personnel as soon as possible.  Bar codes could be it someday. 

I have to admit too the consumer side of so many different devices has tendency to make me nuts as we are at the point now to where it’s like “ho hum” one more device that can do a heart rate or function as a pedometer with the addition of accelerators in cell phone.  You also have to keep in mind that there’s a commitment from the end user to interact with the data and sometimes with busy life styles today, we don’t have time and thus the model created by the software designers doesn’t work as “people don’t work that way”.  I know that song well from the days I spent designing software as I would think a screen design was something of use and then when doctors put it into use, they told me “it sucked” so again the meeting of the minds here with what a software developer “thinks” is a good model and the reality of the “real”world to use it or not.  Right  now we have software just killing us right and left with so many ideas being tossed against the wall to see what sticks, so that show goes on.

Again too with devices we have to look at “privacy” as well and that’s another issue of it’s own as the “data selling” epidemic continues to grow as people are making money from the virtual world of data and the downside of some of this too is the lack of tangibles we seem to produce in the US.  Will the FDA someday be approving software using facial recognition?  Could be if it interacts with medical data.  FDA Commissioner Margaret Hamburg has her hands full in that area and lucky enough for us she’s a big safety advocate as well and puts that issue in front of the latest and greatest “Thing” out there too. 

FDA Commissioner Margaret Hamburg Addresses “Context” And Safety at mHealth Conference-More Important Than The “Latest New Thing” Out There..
FDA and Smart Phone Devices Struggle With Applications and Regulations–Get More Engineers at the FDA and Don’t Rush the Developers With Medical Software

Again you can’t overlook the impact of additional software included with devices today and this area is only going to grow.  Again if it every becomes a reality, having the ability to scan over the counter products would really be a big plus for consumers as some of those products also require FDA approval.  In addition today we have new for profit and no profit efforts to study and research some of the best performing devices,  especially when it comes to expensive medical devices, like ones that are implanted.  The FDA has their Sentinel Program which is starting to pay off with research information but again there are those ready to make money from the process as well.

FDA Sentinel and “Mini” Sentinel Data Base Is Growing and Becoming A Very Useful Tool

In addition, the FDA does have an app that offers updated information for both clinicians and consumers alike that well worth a look when you are searching for some additional information relative to devices and drugs.  BD 

FDA Creates New Mobile Cellphone App to Help Doctors And Consumers Report Deaths and Injuries From Medical Devices and Drugs to the Agency And They Won’t Sell Your Data Either

Recalls of defective medical devices nearly doubled in the decade from 2003 through 2012, according to a Food and Drug Administration report due Friday.

The total number of recalls rose to 1,190 in 2012, up from 604 in 2003. There was a sharp increase in recalls where the defective product carried a reasonable probability of death. In 2012, there were 57 of these so-called Class I recalls, up from seven in 200


Post a Comment

Google Analytics Alternative