# Loop -- a dynamic answer to a dynamic problem

#243

I’d try a small amount just see how things may work. As an experiment, last night I entered a 10g future meal following that timing approach. The experiment succeeded, but let me explain some more details so you can consider what may or may not actually happen. At the time Loop saw increase in predicted bg I was in range, so Loop started high temping gently because predicted bg was not much above the target. After about half an hour it stopped high-temping because predicted bg dipped below my lower target. When I woke up, I did have some extra insulin going, and I was trending slightly down, so fine. The amount of extra insulin was less than one might expect because high-temping stopped as soon as the predicted bg (at any point in time in the future) dipped below the lower target. Makes sense?

#244

Yes, of course, makes sense. I didn’t know if you had a strategy you applied each time. We’ll experiment and see how it goes. Thanks!

#245

@terry4 - So the problem you were having with trying to preemtively bolus in the AM to counteract your upcoming rise ---- because loops forecasting ability is limited to using (the last 3 (15 mins))[is that right???] - of CGM data to make its prediction, it cant/doesnt predict that high coming in say 1.5hr from now, which by bolusing now, u will be hitting on the nose with insulin delivered immediately. ??? - what was your solution? did you/.can you set a higher temp basal rate in the mornings for that. (which loop will just see as part of your “normal basal settings” ???

// so the largest glaring deficiency with being a diabetic in general, but pumping & looping specifically, is the delay (even the rapid acting insulins (humalog, novolog) are just too slow??? so the best we can do to mitigate that delay is use predictive behaviors (i.e - bolus 20min before meals, preempt highs with corrections) ?? - Maybe in the future we can add a feature set to the pump -(specific insulin profiles) it will know and run on your usual pump insulin - but maybe having an option to tell it your manually delivering 2u of Afrezza (for example) and it would already know afrezzas duration of action, corrF, carbF, all your personal settings for that type of insulin as well. and integrate that info into its calculations… any thoughts? – as newer and faster analogs enter the market, this problem will get diminished and hopefully eventually almost completely eliminated. but ive just been thinking Afrezza might be a halfway decent idea for treating postprandial high corrections, either for just that certain circumstance, or in combination with humalog at meals for a diffferent type/style of “dual wave/ extended wave” bolus… any thoughts??

#246

J’utilise ce système depuis mars 2018 et j’adore !!!