I used the Tandem T:Slim with Control IQ. The pump does not really work for rises in BG because of DP or FOTF. As an example this morning my BG was up to 120mg/dl. I requested a bolus and the suggested based on my CF was 0.5u. I know this was not enough from my past history and so I did a bolus of 1.5u.
I am curious as to how well other pump algorithms deal with DP and FOTF, especially the OM 5 and open loop systems. I am happy with my pump and have no problem overriding the suggested bolus if necessary.
I think you’re doing it right. The algorithms have to use conservative insulin dosing because if they give too much they have no way to recover, so their need to be safe means we either go high or take more insulin than the algorithm would dare to give us (because we have glucose tabs at the ready in case we take too much.)
The deeper issue is that biological BG regulation is a complex system in control theory terms; it’s impossible for an algorithm to accurately model the future effect of an insulin dose. They can model an expected outcome, but it frequently will be wrong, sometimes by a lot. I enjoyed Dan Heller’s discussion of this in Why AID Algorithms Don’t Improve: The Category Error
Bottom line is that BG management has too many interacting factors that influence each other, it’s not a simple machine where if you push exactly this hard, it will move exactly that far. Think of a beginner driver, who thinks that if they hold the steering wheel straight, the car will go straight. They quickly learn to make continuous small adjustments to the steering wheel based on where the car is actually going. Managing BG is like that.
With O5 I started out with the same target all night. I could then see what time DP generally kicked in, and dropped my target beginning an hour before that. It works really well at keeping my trendline steady.
I have the opposite of FOTF. I drop as soon as I get out of bed. I don’t think any regimen is going to manage that.
I use a hybrid closed loop system, DIY Loop. It’s called “hybrid” since it relies on the user to manually input meal nutrition and therefore meal bolus insulin. The version of Loop that I use varies the basal insulin amounts as often as every five minutes to affect the programmed BG targets. I’m aware of an alternate algorithm that uses “micro-boluses” to affect BG levels.
I have observed, for many years, both dawn phenomena and “feet on the floor” morning glucose rises that usually exceed the ability of algorithm to keep my glucose below 120 (6.7). This is my personal allowance for BG rise caused by DP & FOTF. I suspect that in a healthy non-diabetic this threshold is < 100 (5.6).
Here is today’s Tidepool graph which represents a typical morning for me.
Today, like too many other days, I woke up at 1:46 am and saw a 111 BG on my phone. I took 0.3 units of insulin since I target overnight BGs at < 100. I woke up at 6:30 at 86 mg/dL and my DP/FOTF rise started about 30 minutes later. You can see the Loop DIY algorithm already taking significant basal action (the bottom trace above) but from experience I knew that this effort needed support, so I delivered 0.5 units at 7:39.
At 9:00 am my BG is 92, right where I want it. If I hadn’t delivered that last 0.5 unit bolus, it probably would have resulted in a 9 am BG of 120 or more.
This morning is typical but not always. I still live with T1D, wild cards are still present and make themselves known. When that happens, I make adjustments. The sooner the better. My list of adjustments when this tactic doesn’t cooperate? Delaying the meal, changing the infusion site, making a more substantial insulin correction, getting on my exercise rower for 15 minutes, or some combination of these and more.
I’m not sure of the correct terminology, but I think of the DP as part of a circadian rhythm and FOTF as related to when I actually get out of bed. If you get up at the same time every day you can’t untangle them.
I often get up earlier or later than normal. The purely circadian portion (DP) is well controlled by changing basal insulin levels (elevated from 3am to 10am) on my t:slim pump, the actual getting up (FOTF) is not. Which tells me I should work harder to get up at the same time every day.
