Having now run a combination of OpenAPS and Loop for a full month, I thought it was about time to publish some data from looping, and if possible, compare it to a period where I had been using a CGM and Pump, but not running a Hybrid Closed Loop. And hasn’t it done well!
At this point, it’s worth making it clear that when using CGM and Pump, my approach to managing my T1D has been very much of the “Sugar Surfing” ethos, which means that I have spent time “glancing” and then managing my glucose levels. In many ways, this is not so far removed from what a Hybrid Closed Loop is doing, only that the “Loop” is you. During the earlier period, I was using Dexcom G4 with xDrip and the 640G, with around three weeks of HAPP interspersed within it.
So let’s take a look at the numbers. Up first are the two Ambulatory Glucose Profile traces. The first is the 31 day period using a Hybrid Closed Loop. The second the 80 odd days using sugar surfing. The “High” line is set at 8.8mmol/l and the “Low” at 3.8mmol/l.
Throughout both these periods, my food intake is fairly variable, with the October looping period being not particularly well managed, with weekends where beer drinking and carb consumption was definitely in play. My typical daily diet included relatively low carb meals for breakfast and lunch, then dinner depends on whether I am at home or out, whether we are having dessert, etc.
Comparing the two, you can see that the loop did a fantastic job of keeping my 25th-75th percentile zone suitably tight. You can also see that my 10th-90th is a little more spikey, and this is principally because as a person using a CGM and then adjusting, I tend to be more aggressive on my adjustments. I’m just glad that the maximum and minimum aren’t being shown on here. The occasional cocktail plays havoc with that.
Comparing the statistical assessment, we see the below tables:
Remarkably, the statistics are really close. The distribution of highs, lows and normal are pretty much identical. The on;y real differences are in the mean/median values, where the loops managed a lower Normal value, and a higher low value. Unfortunately, the high averages were slightly higher, with slightly greater standard deviation. Having said that, I know the reasons for it and am unsurprised.
Overall, these are really great numbers. The key point the numbers get across is that a hybrid closed loop using only insulin is obtaining numbers similar to an intensively controlled sugar surfer.
But that misses the point rather…
The loop is obtaining great numbers without the user’s involvement. It’s doing the sugar surfing for you. And therein lies the difference. Sugar surfing throughout waking hours entails roughly one glance every fifteen minutes. Assuming an 18 hour waking day, that’s 72 glances a day, or 2,160 a month.
That’s also 2,160 moments where you:
- Read the data
- Assess what the data means
- Decide what the appropriate response to the data is
- Act on the data
It might only take 10 seconds, but that becomes six hours a month. And it’s you doing it.
So the loop has removed 2,160 moments a month, outside of eating, where you, as a sugar surfer, casually interact with your diabetes. It’s doing that for you.
And that’s the key benefit of any form of closed loop system. Freedom. You need to spend much less time focussed on trying to get yourself where you want to be. The system does it for you, unemotionally and efficiently, as I mentioned in this post. And here it is, written down in stark numbers.
This is why so many people are looking to move themselves into this new paradigm. To give themselves a little taste of freedom.
The freedom is what is so liberating. I’m still managing my diabetes, and high and low blood sugars are still going to occur, but with an automated insulin delivery system like Loop or OpenAPS I’m able to reduce the mental burden and for me, that’s incredible and game changing.
Yes, it doesn’t stop you having resistant days or cocking up carb counting, but you know that when you do, it’s got your back and will try and revert you back to normal. Without you having to do anything. Which is huge!
Can you point me to any site where I can read more on how the algorithm for the Rileylink works? There seems to be a lot of info describing Dana’s OpenAPS system but not much explaining how the Rileylink actually works. Thanks.
RileyLink doesn’t use an Algorithm. RileyLink is simply a radio signal transducer from Bluetooth to the 915MHz band to allow the iPhone or other BLE device to speak to a pump.
You might want to join the Loop community on Gitter to learn about the algorithm within Loop. Alternatively, you can go to Loopkit/Loop on GitHub and read the code.
I’m new to your site, but already a fan. I’m would love your opinion. My daughter who is now 8 has been type 1 for 15 months. We are currently on injections. We are making the move to a pump or pod and cgm. We are trying to decide between the Omnipod and the Medtronic 630 and then upgrade to the 670. Most everyone we know uses the pod because it doesn’t have a tube, but our educator informed us that we could get on the 670g hybrid closed loop. The technology is so tempting for me, but is it worth the hassle of living with a tube. My husband and I feel like it would be hard to be a kid with the tubing and all. Please share your thoughts! T1d community is really the sweetest! Can you email so I know to read it?
Hi Sara, in my opinion, any assistance in managing T1 is better than none, even with a Tube over a Tubeless pump. The experience of most on the 670G trials seems to have been excellent, as shown by their wanting to keep the devices post trial end. Personally, as a T1, I would take the tubed 670G with closed loop over the Omnipod, but you should be aware that the target glucose level on the system can’t be adjusted below 120mg/dL. That said, the reduction in overhead of managing T1 that you get from the hybrid closed loop is huge.
Ultimately, I think you should discuss it with your daughter in a way that she can understand. I personally would always choose the Hybrid Closed Loop for myself.
Hi,
My T1 daughter is 10 and was diagnosed T1 at 16 months. She is currently using an omnipod and uses a Libre sensor as a CGM. We have been monitoring her using Spike on an iPhone. We have only just started learning about closed loop systems but I can’t see any that uses the omnipod – is there a system available that uses this pump?
Thanks you for any help,
Stewart (UK)
Hi Stewart,
At the moment there is nothing available for the Omnipod and no indication of when the work to decode it will be available.
Right now, you can close loop with the older Medtronic pumps, Combo and Dana RS and work is ongoing (and in prototype) for the Insight.