The one that a lot of people seem to neglect is that the factors change. This can be slow or abrupt and often or rare. It happens and learning to recognize(the hard part) and then adapt can lead to better outcomes. Waiting for a MD/Nurse to suggest changes is often too long.
This is an excellent point. What makes this even harder is that some of the changes (like basal rates) the equipment is good about supporting multiple profiles and letting you switch between them. Where as the carb ratios and correction factors are set once and overwritten when you update them, and hard to generate reports on how they are changing over time... so not designed to really be adaptable.
There are micro factors (time of day, activity level, sickness level, gut health) and macro factors (months/years, age-dependent / hormone-dependent, growth spurts, climate / time indoors versus outdoors, etc.) factors that will impact both your basal rate and your carb ratio / correction factors.
So the carb ratio and correction factors are programmed based on hour of the day. 18 carbs per unit from 12am - 8am, 16 carbs per unit from 8am - noon, 14 carbs per unit from 12-8pm, etc... and then all these numbers may need to be shifted from time to time.
The urge is to somehow try to track everything (mood, health, activity, phase of the moon, food intake, and of course your blood glucose level every 5 minutes) and input into an algorithm that will look at how often you ended up too high or too low after eating XYZ, or how much your BG rose or fall when you weren't eating, and then make tiny incremental adjustments to the carb ratio, basal rate, and correction factor curves every week.
Currently the state of the art is the doctors generates a bunch of reports once every 3-6 months, eyeballs them, and decides; "Hey, we're gonna try bumping up your morning basal rate by X, change your evening carb ratio by Y, and ..." and then you make the changes and hope things get easier.
When the ratios are right, you can dose for what you eat, and end up back in range after a few hours. You can give yourself a correction and go to sleep, and wake up in range, etc.
The "artificial pancreas" is going to do a tiny percent of this job. It will basically see you are high or low based on the 5-minute BG reports, and bump your basal rate either up or down slightly to put a finger on the scale and move you in the right direction.
It doesn't know what you eat, so it can't bolus for meals. It might eventually get you back in range many hours later. Even with fast acting insulin, you have to bolus ahead of time anyway to get an ideal glucose response, so it's not something an artificial pancreas will be able to solve without much faster acting insulin.
I use Loop and it takes a LOT of the thought out of it. Being incorrect is ok because the system a) knows about what the work of the carbs should be and b) is often checking the glucose levels vs predicated(from carb/insulin curves) and doing small corrections via either basal rates or bolus's.
The way it plays out those is that being wrong is ok. So the carb count is off, what was important was that it was timely and roughly close. What this makes more difficult though is seeing the patterns of when it is doing corrections more often. There's software for that too. Also, disabling it and seeing what happens can give some insight. Another thing is that systems that use temp basals to reduce insulin introduce a state of low insulin levels and a "bounce back". This is where I would like to see a system that converts that negative insulin back into carbs. I can do it, but would be nice.
The oref1 algorithm in AndroidAPS can detect unannounced meals and together with a very fast insulin such as Lyumjev can automate meal boluses for some people.
I know diabetics in 85% in range using this method. Doesn't work super well for me though.
Indeed. The weather, (lack of) sleep, stress, mood, etc. All can have a big influence on the rates and factors.
For me, for example, summer is entirely different from winter (and spring and autumn). if everything else would remain the exact same, then still I'd have to switch rates at least twice a year because in summer my body works different. apparently.
> Waiting for a MD/Nurse to suggest changes is often too long.
Exactly this! I am single expert on my diabetes. Doctors and endocrinologists are helpful, but they don't know all the factors that effects my BG in both directions on a daily level. I have to adjust how I do things all the time.
