64% [at the time of this post] - Insulin pump + integrated CGM (commercial or DIY AID system)
In another poll, it might be interesting to further break that 64% into 2 parts. Those who are using a commercial system, and those who are rolling their own.
Now now Eric, we donât do that 
One of the interesting questions is the last one. I have done that in the past when SCUBA diving because I knew the pump was going to fail. So I went back to my âstandardâ Lantus dose, 6+6IU (morning and evening) but only used the pump for the bolus (until it broke). It worked well; I could get 24 hours so do fine-tuned boluses for a while.
Of course that wasnât my answer because that is something I did for very specific circumstances; at the time I didnât do multi-day dives more than twice a year.
However Iâve always considered it a logical step from MDI because the long-acting insulin doesnât require great precision in the dosage, so can be injected with sufficient accuracy, whereas small boluses are impossible by injection.
@Lucia I think Control IQ is great in sleep mode and I know other T1Ds who also keep their TSlim in sleep mode all the time. I always stayed in range at night when I was using the Tslim, during the dayâŚnot so much. I can see why you are microbolusing. â I am now using a different system (Medtronic 780g+Guardian 4 sensor) that works better for my personality.
I would be interested in the poll you are describing here, but thatâs kinda whatâs happening in this thread!! Still reading all the answers. Love it!!!
I love this term! âNuff said. â Now I have a word for it. Thank you @jbowler 
[quote=âjbowler, post:21, topic:15625â]
so the basic AIDS I use still controls correction and bolus, including micro-bolus
[/quote]
What AID system are you using?
Hum. I recently created a Reddit ID called âErold Nuffâ (enough). Iâm pretty sure Iâll delete it fairly soon because Iâm a sucker for rage bait.
I use the Omnipod Dash (the last hackable Omnipod) with the 10-day Dexcom G7 (maybe the last hackable G) connected by AAPS. AAPS (formerly known as AndroidAPS) is, I believe, the currently active development branch of OpenAPS, the original âArtificial Pancreas Somethingâ.
AAPS is something of a sort that Rowland Emett might have designed. Alas the dudes who did design it lack a sense of humour.
I do not feel techie enough to do the aps stuff. I would need a lot of hand-holding to do that. My husband is a computer programmer so I know he would be into it, but I would need to be âall inâ as the user. 
No, you can do this. Itâs pretty easy.
Just get some empty sterile vials (very cheap to purchase), some insulin diluent (available by prescription from either Lilly or Novo Nordisk, and they are pretty much the same, you can use either one with any insulin you want), and whatever insulin you want to use.
Just do the math and load up your sterile vial with the right amount of insulin and diluent, and you can make any number of units on the syringe be equivalent to any bolus amount you want.
Lots of references here on FUD.
https://forum.fudiabetes.org/search?q=diluent
Well⌠thereâs another poll! 6 people (other than me) have upvoted so far, but how many of those people have, in fact, obtained a prescription insulin diluent and used it for micro-dosing, as opposed to those who have used it for reduced dosing for children.
This poll is easier: âDo you dilute your insulin before injection?â
I think the primary take away from Ericâs response is that micro bolusing is; indeed, possible by injection, regardless of why you want to do it. There are 31 different on FUD about the uses for diluent in micro bolusing.
Geez. Please; I understand what he said. Iâm just interested in whether anyone has ever done it and I was VERY careful to exclude you because, yes, I know youâve done it and why. Sorry I excluded you from my question but I was not interested in your answer; I was interested in whether any of the other 5 upvotes have done it for themselves. I know Eric has too; heâs also done micro-glucagon injections, a lot of bike riding and many other things that improve T1 control.
Lol. I actually have NEVER done it. I entertained it way back in 2018 but didnât ever follow through with it. I did learn all about it back then though. Did you read the 31 posts to see any other use cases? If you do that, you may find the answers you seek.
Up votes, btw, donât mean someone has to have done itâŚthey may be up voting because they read Erics info and agreed.
Nope. Iâm lazy and more interested in the actual use in aggregate rather than specific examples. That said, Iâm curious; why didnât you need to use diluents? A two-year-old is the classic example of where they are required.
I might add that the other example I know of is cats and smaller dogs; my impression is that diluents are quite common there because the general treatment, at least in the UK last time I knew, was human insulin and that changed to U100. (This may not be true in the US; in the UK human insulin was much the cheapest approach.)
I wanted to be sure my settings were correct before resorting to it. I spent more time testing and tweaking my settings and found my answers there, so I had no need for it.
Then in the following months, I moved to the DIY Omniloop system which made it even easier to manage, set temp overrides for more or less insulin and even set low/no temp basal rates for set periods of time.
Diluent, while plausible, was more work for me than the other avenues I ended up following to get the same or better results.
I remember .05 being way too much and that being something I battled with early on, and how I solved those issues, Iâve documented here on FUD for future generations for T1D caregivers to read and hopefully be helped.
I use a pump and the FreeStyle Libre sensor. I have, twice now, tried the Medtronic sensors. G3 and G4. They never last the full 7 days and I am better at setting basals than they are at automated corrections.
No. And I have never heard of diluent before today. Microbolusing seems to make WAY more sense with a pump than with a syringe. In fact, all AID systems microbolus automagically, correct? â When I was on MDI, I never gave myself less than 1 unit because I had 1/2 cc needles. Nowadays, when I am adding a correction bolus on my pump when the AID is not doing its job to my satisfaction, I often give a 1.5 unit bump. â I literally never thought about that .5 units and what that would look like with a syringe until just now. Putting this in the You Learn Something New Everyday file.
Micro Bolus are only possible down to the .05 level (with the Omnipod anyway, not sure about the other pumps). If you want to go smaller than that, thatâs where diluent would / could come inâŚMDI or AIDS.
Thatâs true for Tandem T:Slim x2. I just tried to bolus .01 to .05 only .05 would work.
I believe it was originally a feature of AAPS and was surprised to see a reference from a commercial manufacturer. Prior to the AAPS SMB implementation it just adjusted the pump basal rate to implement a correction. It was necessary to explicitly go into the bolus calculator to get a correction bolus and that was a full bolus, not a âmicroâ.
AAPS seems to have invented two important things:
Sometimes when I talk about âmicro bolusesâ I just mean boluses that are below the resolution of U100 insulin even with half-unit pens; less than 0.5IU.
So far as I am aware insulin pumps typically offer a minimum bolus in the range 0.01IU to 0.1IU. The Omnipod has a novel âmotorâ that injects one 0.05IU pulse at a time but it can be programmed so the UI works in 0.1IU multiples.
The general implementation of a dosing pump uses what is called a âstepperâ motor. These things can manage very precise rotations because they use reduction gears to slow down the rotation (like clockwork; the way the minute hand moves at 1/60 the rate of the second hand etc.) Thereâs no limit to the resolution so it just comes down to the performance of the part that drives the plunger in to deliver the insulin; it will stick if the movement is too small.
The twiist pump does things differently. It measures the insulin flowing out of the cartridge so it cares a lot less about the accuracy of the motor etc. It does still seem to restrict delivery to multiples of 0.05IU, or maybe even 0.1IU.
For diabetics with a basal requirement the minimum dose the delivery mechanism can provide is moot. This is because the basal is being delivered continuously in minimum-delivery units separated out by a delay. A technique called digital difference analysis is used to work out the period.
If I have a basal of 0.5IU/hour my pod will deliver one 0.05IU pulse every 6 minutes (for 10 per hour). If I drop one of those 6 minute/360s intervals to only 324s (a reduction of 36s) Iâve done the equivalent of delivering an extra bolus of 0.005IU in addition to my normal basal of 0.5IU/hour.