# Pump IOB curves/equations **URL:** \ **Category:** Pump\ **Created:** [August 26, 2019, 9:49am UTC](https://forum.fudiabetes.org/t/pump-iob-curves-equations/8193 "2019-08-26T09:49:15Z")\ **Posts on this page:** 6\ **Page:** 1
**Author:** ![Boerenkool](https://yyz2.discourse-cdn.com/flex030/user_avatar/forum.fudiabetes.org/boerenkool/32/6307_2.png) [@Boerenkool](https://forum.fudiabetes.org/u/Boerenkool)\ **Post date:** [August 26, 2019, 9:49am UTC](https://forum.fudiabetes.org/t/pump-iob-curves-equations/8193/1 "2019-08-26T09:49:15Z")
Is it known what kind of formulas pump manufacturers use to determine IOB? I was looking for curves or other info concerning Medtronic pumps, but I couldn’t find much. I hope it’s not just a linear equation. ---
**Author:** ![Jim\_YYC](https://yyz2.discourse-cdn.com/flex030/user_avatar/forum.fudiabetes.org/jim_yyc/32/9553_2.png) [@Jim\_YYC](https://forum.fudiabetes.org/u/Jim_YYC)\ **Post date:** [August 26, 2019, 11:43am UTC](https://forum.fudiabetes.org/t/pump-iob-curves-equations/8193/2 "2019-08-26T11:43:38Z")
I doubt they use anything more than the basic half-life estimate (modified for user input “duration”. Considering that different brands of fast acting insulin’s have significantly different onset and duration profiles, at best it’s just a guess. > **[Drug Half-life Explained: Calculator, Variables & Examples](https://www.drugs.com/article/drug-half-life.html)** > > What is the half-life of a drug, how is this calculated (with calculator), what affects half-life calculations, common drug half-lives and more.... ---
**Author:** ![Thomas](https://yyz2.discourse-cdn.com/flex030/user_avatar/forum.fudiabetes.org/thomas/32/1719_2.png) [@Thomas](https://forum.fudiabetes.org/u/Thomas)\ **Post date:** [August 26, 2019, 1:15pm UTC](https://forum.fudiabetes.org/t/pump-iob-curves-equations/8193/3 "2019-08-26T13:15:32Z")
> [@Boerenkool](#): > > Is it known what kind of formulas pump manufacturers use to determine IOB? We all (FUD) could get the data and plot it out for the pumps we all (FUD) have. Tedious but would be super interesting. > [@Boerenkool](#): > > I hope it’s not just a linear equation. ha ha ha I know. That would be sad. But realistically - yeah - I will not be surprised to see a simple linear with the manufacturer expectation that who will know the difference. ---
**Author:** ![Scotteric](https://avatars.discourse-cdn.com/v4/letter/s/59ef9b/32.png) [@Scotteric](https://forum.fudiabetes.org/u/Scotteric)\ **Post date:** [August 26, 2019, 1:52pm UTC](https://forum.fudiabetes.org/t/pump-iob-curves-equations/8193/5 "2019-08-26T13:52:35Z")
I think Omnipod uses a linear equation but Medtronic doesn’t. Either way I’ve found through guesstimating on MDI my guesses are the same or often better than whatever algorithms the pumps use. There are so many variables including the temperature outside, your stress level, activity level, what you’ve eaten, how your basal is being absorbed, etc, etc that it would be hard for an equation to really predict how much insulin is active in your system. ---
**Author:** ![Eric](https://yyz2.discourse-cdn.com/flex030/user_avatar/forum.fudiabetes.org/eric/32/9209_2.png) [@Eric](https://forum.fudiabetes.org/u/Eric)\ **Post date:** [August 26, 2019, 2:35pm UTC](https://forum.fudiabetes.org/t/pump-iob-curves-equations/8193/6 "2019-08-26T14:35:36Z")
Omnipod does use a linear degradation. It’s kind of weak. IOB **=** bolus amount **-**** [**** ( **bolus amount** X **minutes since bolus** ) **** / **insulin duration in minutes**]** They don’t use onset or peak time. I gave them a better formula a few years ago which would have allowed peak and onset time to be factored in, but I don’t think they made any changes for the new PDM or with their IOB calcs. I had some analysis for Tandem’s formula at one point. I can see if I can find it. They have a much more complicated algorithm. ---
**Author:** ![Boerenkool](https://yyz2.discourse-cdn.com/flex030/user_avatar/forum.fudiabetes.org/boerenkool/32/6307_2.png) [@Boerenkool](https://forum.fudiabetes.org/u/Boerenkool)\ **Post date:** [August 27, 2019, 6:32am UTC](https://forum.fudiabetes.org/t/pump-iob-curves-equations/8193/7 "2019-08-27T06:32:19Z")
> [@Thomas](#): > > We all (FUD) could get the data and plot it out for the pumps we all (FUD) have. > > Tedious but would be super interesting. At some random time points I wrote down the active insulin. Here’s the data plotted it in Excel with a 3rd order polynomial fitted to the data. The equation isn’t correct of course, but it gives us an idea what the curve looks like for a Medtronic 640G. ![IOB](https://canada1.discourse-cdn.com/flex030/uploads/fudiabetes/original/2X/c/c05537da4935ca6f28ecc00e6b323f24eb58e7d0.png) > [@Thomas](#): > > I know. That would be sad. But realistically - yeah - I will not be surprised to see a simple linear with the manufacturer expectation that who will know the difference. Apparently that’s what Omnipoed does 🤨