[81136518]
I use dmes (insulin pump and cgm) to maintain my type 1 diabetes. After (b)(6) years of t1d, i no longer can sense low blood glucose levels and depend on my cgm to alert me so i can treat myself before passing out. On (b)(6) 2017, my historically very reliable cgm receiver unexpectedly quit. The mfr dutifully shipped me a replacement overnight. However, i needed to leave town before the replacement would arrive early the next morning. This should and would not have been a problem - except for the fda. Here's the situation. The mfr has created a smartphone app that provides the basic functionality of the receiver that failed. But when i attempted to download the app from (b)(6), it reported that my device was not compatible and would not let me download it. Thus, i was at night risk for an undetected low blood sugar incident for two days. This is a serious matter; i would be dead now if not for my cgm alarm one morning several months ago. If such an accident had repeated itself that weekend. I would be dead. That's why i labeled this a "life-threatening" incident. Later, i learned that the cgm app was in fact compatible with my phone (and also my wife's phone). But since it had not been run through the testing required by the fda. It has not been added to the approved list. Your mission statement says. "fda is also responsible for advancing the public health by helping to speed innovations that make medicines more effective, safer, and more affordable. " my recent experience indicates the fda's actions are not aligned with their mission statement. Let me explain further. When i contacted the mfr about the fact my phones and could in fact run the app just fine, they responded that they are "strictly regulated by the fda" and "must provide the fda with the results of testing and depending on the fda response" may be able to release an update. (note that my ability to hack the app to circumvent the fda mandated restrictions and software coding that prevented the app from running on "untested" devices was an unbelievably technically difficult task. But it allowed me to test the app and confirm my strong suspicions that the app was in fact compatible with many more devices than claimed. ) let me be blunt. Software will always have bugs. And no amount of testing will prevent that. In fact, the software bug in my insulin pump that nearly killed me went undetected for years and still has not been fixed. Thus, as a practical matter, software testing must be streamlined and expedited (consider monthly (b)(6) updates, for example). In the case of a cgm, the software basically runs or it doesn't. Thus, if the fda should allow it to be installed on nearly any smartphone, accompanied by a warning that it has not been tested on said device. This would benefit a lot of people, and likely harm no one. Much greater harm is done by the delays and increased costs associated with extensive testing that benefits no one. (i challenge you to site any recent software submission to the fda which contained a life-threatening bug which was discovered by the fda. In the case of the cgm app, it is a basic app that either does or does not work. It is not going to mislead users, since it is only an alerting tool. And even in cases where it does not work, "how/why is it better to prevent it from being installed on that device and numerous 'untested' devices on which it does work. " what the fda should do when it comes to software is allow for crowd testing of devices. And allow the mfr to quickly update their software based on suggestion and comments from responsible astute users. In fact, the fda should mandate firmware updates rather than make them non-existent or rare. After hacking the cgm app so it could be run on an 'untested device, my testing discovered the cgm app ran just fine on all three smartphones i own (none of which were deemed 'compatible' on the (b)(6) app store). It even also worked with the smartwatch) i own. One potential issue with a cgm app is the ability to consistently collect data over the bluetooth interface. However, testing this based on different models of phones is almost pointless. The issue, when it does arise, is more a function of radio interference and other applications on the phone completing for bandwidth and cpu cycles. This type of testing is impractical. My point is that the current testing process is quite flawed and provides little if any benefit. The answer is crowd sourced testing, and readily implemented firmware/software updates.
Patient Sequence No: 1, Text Type: D, B5