FDA Will Regulate Some Apps As Medical Devices 130
chicksdaddy writes "In an important move, the U.S. Food & Drug Administration (FDA) has released final guidance to mobile application developers that are creating medical applications to run on mobile devices. Some applications, it said, will be treated with the same scrutiny as traditional medical devices. The agency said on Monday that, while it doesn't see the need to vet 'the majority of mobile apps,' because they pose 'minimal risk to consumers,' it will exercise oversight of mobile medical applications that are accessories to regulated medical devices, or that transform a mobile device into a regulated medical device. In those cases, the FDA said that mobile applications will be assessed 'using the same regulatory standards and risk-based approach that the agency applies to other medical device.' The line between a mere 'app' and a 'medical device' is fuzzy. The FDA said it will look to the 'intended use of a mobile app' when determining whether it meets the definition of a medical 'device.' The Agency may study the labeling or advertising claims used to market it, or statements by the device maker and its representatives. In general, 'when the intended use of a mobile app is for the diagnosis of disease or other conditions, or the cure, mitigation, treatment or prevention of disease, or it is intended to affect the structure of any function of the body of man, the mobile app is a device.'"
Re:Woohoo! (Score:5, Insightful)
So you want an app that's controlling an IV pump to be written to the same standards as your average fart app?
That's what this is about. The FDA neither has the time nor energy to look at every stupid 'medical' app in the store. They're only going to deal with ones that have an interaction to hardware that can cause problems or if the app is really, egregariously touting some medical benefit that it can't possibly provide.
Re:Woohoo! (Score:5, Insightful)
So you want an app that's controlling an IV pump to be written to the same standards as your average fart app?
Why the fuck would you use a mobile app to control an IV pump? That's like running a production database for a fortune 500 company off of Arch Linux.
You don't use something that isn't stable to run something as important and life critical as an IV pump. Nor would any hospital do so and any individual that tried would simply be a victim of natural selection.
Re:Woohoo! (Score:4, Insightful)
Why would you put a poorly-tested embedded bluetooth stack in a pacemaker that crashes under fuzzing? Crazy, right?
If it's anything like it was when I was in medical software, the FDA is more concerned with process than specifics of implementation. If you're practicing good software development and QA methodologies and you can impress your ISO9001 auditor, then the FDA won't be too tough. Then again, "those were the days", so maybe somebody can update this - last time I flew to FDA for a development seminar was the first time I got wanded by TSA.
Re:Woohoo! (Score:5, Insightful)
Why the fuck would you use a mobile app to control an IV pump?
The point is really two-fold. First, we already regulate medical devices like infusion pumps and radiology information systems. Under the proposed regime, one does not simply avoid regulatory scrutiny and obligations by offloading them to an app.
Second, if a app makes claims to do things things that would ordinarily be regulated, you don't escape the regulatory regime simply by saying, I'm just an iPhone app.
Both prongs make some sense if you accept the basic assumption that FDA regulation of devices makes sense at all.
Re:Woohoo! (Score:4, Insightful)
You don't use something that isn't stable to run something as important and life critical as an IV pump. Nor would any hospital do so and any individual that tried would simply be a victim of natural selection.
From your mouth to God's Ears.
But just in case someone was thinking of doing this the FDA is going to step in and set guidelines.
Why: Because people are re-purposing cellphones for all sorts of stuff, most of it not well thought out.
You can find examples which contain unvetted information on the Google Play store. Some of this might find its way into actual use in the field, because in all too many cases doctors and computers live in two different worlds, and may come to trust something that looks approximately right, only to find out its horribly wrong in some places:
Examples (with no allegations of incorrectness or danger, simply to show examples that already exist):
https://play.google.com/store/apps/details?id=com.shahlab.anesthesiologist [google.com]
https://play.google.com/store/apps/details?id=net.klier.blutdruck [google.com]
https://play.google.com/store/apps/details?id=com.qxmd.calculate [google.com]
https://play.google.com/store/apps/details?id=com.mobisystems.msdict.embedded.wireless.mcgrawhill.ivdh [google.com]
https://play.google.com/store/apps/details?id=com.sekos.dosagecalc [google.com]
So yeah you might thing such doesn't exist, and these might be harmess infomation apps, but who's to know when nobody is watching.
And with NFC built into every phone these days, who's to say someone won't find another use for it?
Re:Woohoo! (Score:4, Insightful)
These days they have ISO 13485 [wikipedia.org]. It's a lot like ISO 9001, but drops the more marketing continual improvement and customer feedback and adds additional requirement for creation of requirements, something more-or-less non existent in 9001.
So theoretically, 13485 would require you to recognize risks (e.g. fuzzing) and add mitigation (e.g. not crashing) to your requirements, which would then be tested and all the in a similar matter to 9001. I'd say that as a result it's about as good as can really be expected. The only real other steps would be for the FDA to bring in experts to verify the considered risks and possibly verify the testing, which starts to become unrealistic.
For completeness it's worth mentioning that when it comes to diagnostic software (I believe this started in the last couple years) they also require clinical trials to verify effectiveness somewhat like they do for medicine. Therefore if you write an application that, say, helps determine if someone is concussed (the case which started diagnostic software, IIRC) then you actually have to collect data for concussed and non-concussed people (verified by traditional means) and prove that it can provide a meaningful diagnosis.
Re:Woohoo! (Score:4, Insightful)
Nanny State to the rescue!
Bullshit, regulating medical devices isn't being nanny-state, outlawing marijuana is. I, for one am damned glad that the FDA had to approve the device they inserted surgically into my eyeball. I'm damned glad that drugs are regulated, so I won't underdose or overdose on a needed medicine because a supplier is incompetent or dishonest.
Do you scream "nanny state" when the cops go after armed robbers? Did you scream Nanny State when they jailed Madoff?
Sorry, kid, but your comment is just brain-dead stupid. Read a little history, look up "patent medicine" and "snake oil". Read Upton Sinclair's The Jungle (public domain, it's at gutenberg).