So, something that annoys me about the over the counter continuous glucose meter I use: it doesn’t adjust its statistics based on the delta between sensors.
To be fair, a new sensor is dealing with a bit of an immune reaction, so its data can take a day to fall into normal ranges and patterns.
But the app does allow for a “grace period” of up to 24 hours after the end of a sensor’s 2-week lifespan before it starts ignoring readings, and that seems like an ideal time to start a new sensor and look to see how far off they are.
Maybe it’s not medically sound to do so? All I really know is that all meters (continuous and not) can vary from “true” by +/- 15%, and the only real strategy to correct for it is to measure repeatedly. The prescription versions do just this – you can add finger-stick readings to the data, and it adjusts its own based on the comparison.
I kind of wish I had a way to get the raw data from the sensor, as well as the post-session readings. I know that if the sensor loses connection to the app, or gets rebooted (magnet switch) after being told to end its session, it will intermittently look for its phone via bluetooth, (which is surprisingly sad, to have to continually respond to pairing requests from a bit of electronics that’s slowly running out of power.) I just don’t know what to do with that info, to make some use of it.
And of course, I’d love for some functionality to be rolled into the official Dexcom app – they could essentially make these more accurate without having to change anything about the hardware itself. But it just seems like the sort of thing that, if this was open source or even partially open source, you’d long since have had internet people submitting pull requests with code to add this, because they want it too.