OpenAPS and UAM

Now that we’ve had some time with SMBs enabled, I was ready to consider trying the other new feature called UAM (unannounced meals).  I wasn’t previously interested in this feature much, because (I’m lucky) Anna has only forgotten about 2 boluses in her 2+ years of mostly independent bolusing.  So meals generally don’t go unannounced.  I didn’t imagine there would be much value in a feature named “unannounced meals”.

But, then on some random school mornings…I am reminded that I should keep my options open.  School “nutrition” break is at 10:10am.  By 10:15am, I see a bolus come in on Nightscout.  And I think to myself “Huh, I wonder what that will be because I know she didn’t pack a snack today.”  By 10:30am, I see the BGs starting to climb and I wonder how it will go.

Nutrition break is never about quality food.  The things sold by the school for nutrition break usually involve chip bags, mini donuts, twizzlers, and the like.  These snacks are problematic because  (1) we don’t have a lot of practice for these foods because we don’t generally have lots of these particular foods around the house and (2) they are being eaten at a time where she doesn’t have a lot of time or interest in babysitting her BGs or IOB.  Also problematic is that pesky independence that Anna is really quite interested in.  She doesn’t want to tell me what she’s eating, nor does she want to have me telling her how to bolus things much these days.  She will tend to still ask me for input on new food boluses when it’s just the family around…but when she’s around her high school buddies at school?  Forget about it.  So, iteratively improving her bolus techniques for these foods is hard to do given the lack of communication on these snacks.

Anyways, I turned on UAM on Sunday late night with the thought that maybe Anna and I would have some help during those nutrition snacks.  My theory wasn’t that she isn’t bolusing or entering carbs…but that perhaps the carb counting on those snacks was particularly difficult and our desire to keep her independent probably wasn’t going to allow for much improvement on proper carb count/bolus techniques for the snacks. I really do value her growing self-confidence and want to cultivate it by not hovering on every new food and new bolus…but those nutrition snacks test those limits.

I didn’t tell Anna about the new feature, but she must have had that sixth sense because on Monday’s nutrition break…she decided to test it.  My pebble started to alert me around 10:45pm that her BGs had climbed above 130.  When I pulled up NS, there was 50g of bolused carbs entered.  BGs were climbing, but SMBs were started.  I sat patiently as I could.  But, eventually the curiosity got to me when she got to about 180 mg/dl.

Screen Shot 2017-05-22 at 11.16.08 AM
IMG_3399

So nice of her to be testing the new feature with exactly the part that I thought it might help with…some random new food that she was completely guessing the carb count on (without me involved).

Things went awesome…I was so impressed.  Somehow the rapid rise from that mini-McMuffin was halted pretty well and the decline back to target was looking smooth.

Screen Shot 2017-05-22 at 11.51.51 AMSadly though…that experiment would have to end early.  Her rig’s battery came loose and looping stopped.  She couldn’t see that the wire had come loose so we just lived out the rest of the day old-school…no looping. (but look at how well those BGs slid in close to target of 90 even after 43 minutes of no looping.  UAM/SMBs sure did estimate that needed amount of suspension pretty accurately from a fair bit ahead of time)

Screen Shot 2017-05-22 at 12.33.42 PM

When she came home on Monday and I suggested that maybe she could run the experiment again…she was more than happy to oblige.  So, today she got the mini McMuffin again and did the same bolus and same carb entry.  Started the meal looking pretty similar to Monday.

IMG_3414

The UAM/SMBs worked together to control that peak BG as best as possible…looked good.  I like how even though clearly the sandwich may have been a little different carb count (or her prebolus time was a little longer today?), the loop was reacting well to the BG behavior rather than simply the carb count.

IMG_3423

But then I got a little leery when I saw this…was the bottom falling out?

IMG_3430

My worries were unfounded.  The loop’s math was solid.  We gently landed pretty much at our target of 90 mg/dl.

IMG_3443

I’d say the experiment has been a solid success.  One of the biggest things I notice is the way the loop successfully managed this same meal without prejudice.  What do I mean by that?  It reacted to the ACTUAL BGs that it was seeing really well rather than seeming overwhelmed by strictly the carbs entered.  At the peak BG on Monday, Anna was carrying about 5.34 units of IOB.  At the peak BG on Tuesday, Anna was carrying about 3 units of IOB.  To have both of those peaks resolve close to target without overshooting or needing a low treatment or intervention is pretty impressive.

The question is “If it was UAM that did this, HOW did UAM do this?”  I spent a good portion of the day exploring the answer to that question.  And the answer probably deserves its own post.


There’s a line in the movie Labyrinth where the main character Sarah has an epiphany and says to the evil Goblin King “You have no power over me.”  I feel like I’ve just gotten to say to the snack bar “You have no power over me.”

Doritos are still devilish on BGs compared to a low-carb muffin, but at least the BGs are being handled better than if I stared at them and tried to micromanage with 8 text messages and distractions.  Anna comes home and tells me about the things she did with friends and her homework…no need to discuss how to do the snack bar “better” the next day.  What a freedom.

Leave a Reply

Your email address will not be published. Required fields are marked *