A better way?

I am a former Animas/Dexcom user. When word came down that Animas was getting out of the pump business, my doctor recommended I switch to the Medtronic 670g. I did so.

So far, I haven’t had a single sensor make it more than six days, and most crap out after three or four.

One consequence of Frequent Sensor Death is that I’ve been using the Bayer Contour meter that came with the pump, and one really nice consequence is that with the pump off automatic, I can control the pump (in terms of bolusing) from the meter.

This has led to an interesting situation in which I can essentially ignore the meter. It’s there, but it’s not like I pay it a lot of attention. I handle things through the remote.

Here are the things that could make that situation better:

  1. Build the remote into a watch-type device.

  2. Improve the fscking sensors so they’re reliable and have sensor results sent to the remote (watch or whatever) instead of to the pump. FSD sucks and is costly in terms of time even when Medtronic is ready to replace sensors ad infinitum.

  3. Allow users to calculate and tailor boluses at the remote/watch, rather than be forced to accept what the calculator computers. Likewise, to the extent they are necessary, alerts should go to the remote.

Yeah, I know this is all 3.14 in the sky, and there are probably patent disputes that would make this difficult, but ALL of the technology necessary to allow this including security to prevent abuse already exists. If Medtronic implemented these suggestions, we could wear the pumps hidden away under our clothing. We wouldn’t have to hunt for the pump to read our the sensor data, or when it vibrates in the middle of the night with an alarm.

The pump itself could be smaller and less complex. No need for an interface on the device itself. Batteries would last longer without the display and backlight.

Somebody please tell me what I’m getting wrong here…


Hi Andrew. The above confused me a little but it could be because I am not that bright.

Anyway I don’t think I can say you are getting anything wrong I think you may have some great ideas here.

There is of course considerations. For example if I have a dumb pump pod that requires a remote, what should that pump do if it can’t communicate with the remote? What happens if the remote gets lost? Am I screwed?

Now you have described what I think is wireless comm between a CGM transmitter and a remote. And another wireless comm between the remote and the pump. What shall we use for these? Bluetooth? How robust do these communications have to be for reliability?

Now consider that wireless comm is not allowed aboard aircraft. What do we do with these pumps if I have to fly?

Now consider that someone is, as of right now, suing Tesla because the car crashed in autopilot mode. What is the liability of a pump that delivers acutely dangerous ( I’ll say it potentially lethal) drug such as insulin? Is there exposure here?

Do you think that if the pump manufacturer allowed you to plug in your own math and deliver a owner derived amount of insulin and if that dose killed harmed or gave the user a headache that the pump manufacturer would be protected by law?

Don’t get me wrong. I like user feedback and totally awesome ideas for improvements but this is a multi variable nightmare with billions of dollars of liability at stake. All American companies have to be cautious because they will be held responsible. In other words: It may not be because they are “stupid” or “tone deaf”it may in fact be because this is an actual mine field.

Anyway my background is making medicines and that makes me painfully aware of the stakes of this game and the limits that have to be established when what you are making could hurt someone. It’s not as simple as it might seem. Cheers!