We have accomplished our project and it worked pretty well. In the T1 world that is all about numbers and hard facts, I‘d like first to point out: It was fun! Meeting people, getting to know new places, drinking a lot of good beer and finally just the cycling.
About the numbers. I store my data via Nightscout at 10be.de and use Nightscout Reporter to visualise my data. TIR and more:
My major findings (I think they are not so new) – the keyplayers are: a sensor that works and an Infusionsite & a Pod that are ok. That‘s it.
| Insulin | Lyumjev U200 |
| Pump | Omnipod Dash |
| CGM | Dexcom G6 |
| AID System (Open Source) | Android APS |
I did use AndroidAPS with the Oref1 algorithm with Unannounced Meals (UAM) and Super Micro Bolus (SMB). In addition I have also a dev-feature autoISF included. In order to cope with exercise I started first just with turning off SMB and setting a higher temporary target (tt) and lowered the profile overall to 70%.
Through the days I did change a little the basal settings but not really much. At the beginning I stopped all SMB with the tt. This gave me days like the following charts. With normally cycling between 5-7 hours.
What I didn‘t like about this setting is the need to Bolus manually for carbs intake. In addition to that the system didn‘t cope very well with increase of bg after eating. It was still not bad, but…
What worked best for me in the end was: enable SMB with higher tt, set a higher tt and change the profile depending on the estimated effort between 50% and 70%. That gave me days like in the picture below but with much less effort on my side to manage things.
So what would I change for a next time: Nothing, I think I would go for the same. It was a great time.
