Omnipod 5: 30 Days Later.
An updated review of my progress on Insulet's hybrid closed-loop system
(Disclaimer: nothing here is medical advice - just my own opinions & experience)
I am now a full month into using the Omnipod 5 (about 5 weeks), so I want to share my impressions now that I’ve had chance to see it fully in action.
The TL;DR version is: I like it, it’s harder than I expected, but I like it.
The first two weeks felt like absolute crap. From what I gleaned on the OP5 Facebook group I’m in, I am not alone in feeling this way. My blood sugar average was marginally better on some days, and on other days, significantly worse.
As someone who had used the Omnipod system since 2013, it was not required or even really recommended to get training. An online training is provided, which simply tells you how to set up an otherwise pretty intuitive navigation system.
The initial pod is clearly the best for everyone, because it’s the pod where the algorithm is still relying on your own manual settings. Once you move over to Pod 2, the algorithm takes over. This makes it unlike the Tandem Control IQ, which still uses your personal basal settings as it’s foundation. On Omnipod (and also Medtronic), the algorithm does responsive, adaptive insulin dosing. You don’t really have a traditional basal rate, because the algorithm isn’t creating a new foundation. It’s taking what it knows about your insulin needs, what your BG currently is and where it thinks it’s going, and your Insulin On Board, and it generates an adaptive, responsive dose - an “auto event” - every 5 minutes. Which is cool except - as I was warned by friends, colleagues and Internet Strangers - the algorithm is very conservative in the beginning. And for some people who already have pretty tight control, it’s going to feel like it’s walked several steps in the wrong direction.
The algorithm is mostly proprietary but what I have gathered the algorithm relies quite a bit on two things:
Total Daily Insulin (the sum total of your bolus and basal rate)
Aiming for a 50/50 split between bolus and basal (that doesn’t mean you actually will have a 50/50 split, just that it’s what the algorithm is programmed to aim for).
For me, knowing these two things really helped me understand how to help the algorithm become better at giving me more insulin. And that took some time, which was definitely frustrating. I don’t think I really felt like the algorithm met my needs and was better than manual mode until the week of July 18, almost a full month in.
You can see the improvement from the 30 day, 14 day, 7 day and 2 day Clarity reports:
When comparing my 30 day before starting OP5 and after starting OP5 shows negligible differences: a 3% increase in time-in-range, a 3 mg/dl drop in average, but a 5% improvement in glycemic variability. But I can see that my trend is changing the longer I am on Omnipod and so I know things will improve long-term.
The thing that helped me the most was correct, correct, correct. And then correct some more.
Because the algorithm relies so much on TDI (total daily insulin), it can feel like a lot of hand-holding at the beginning. It was pretty aggravating. After a few days of seeing huge spikes after meals and long lasting plateaus, I started doing a couple of things:
More corrections. Sometimes a total override of what the system suggested! And I would review my BG once an hour.
Monitoring suspended basal, especially in the hour before a meal. I noticed that in the beginning, a basal suspension was correlated very strongly to a big spike after a meal, so I would replace some or all of basal if it was too close to meal.
Some people navigate this by increasing their insulin-to-carb ratio, lowering their correction factor or decreasing their duration of insulin action (which you need to calculate IOB). For me, I knew those things hadn’t really changed in my body (I hadn’t suddenly become more insulin resistant), and I have a feeling that’s just compensating for a suddenly too low basal rate. I don’t know if I’m right, it’s just a feeling.
I did bring my DIA to 3 hours and my morning ICR down a bit, but otherwise my main avenue for telling the algorithm I need more insulin was through good old fashioned manual overrides!
The main mantra I kept thinking was: “I need to teach it that I need more insulin. I need to show it that I’ll be okay.” Not unlike how I help my counseling likes work on their own hypo anxiety.
So would I recommend system? Sure, but with a few caveats:
If you are already in good/great control, things may get worse before they get better.
Because the algorithm is so conservative at the beginning, it may feel like going backwards at first. I think that’s something people need to know going in. You aren’t going to be sitting in the 80s and 90s anymore and that will be a rude awakening.
This is not a good system for pregnancy.
I also would not recommend this for pregnancy. Postpartum yes. Pregnant? No.
I think this will be great for people with a fear of hypoglycemia.
I have gone low once in the last 14 days, and I only went to 70 mg/dl. Any time I have been trending down with insulin on board, the suspension of basal worked to bring me back up. I think for people who are worried about trending down with IOB, the algorithm is a great safety mechanism to prevent really serious lows.
Want to know something about the system that I didn’t cover? Are you using the Omnipod 5 and trying to figure out why it does this thing or that thing?
Introducing: Ask the Help Desk! This month, Dr. Julia Blanchette, PhD, RN, CDCES and a certified Insulet trainer on the Omnipod 5 will be answering YOUR questions. Please submit your questions by Friday, July 29 at 6pm EST, and I plan to share answers next week!
What is Ask the Help Desk? Once a month (possibly more), I will invite guest experts to come on and answer questions from my subscribers! For this first installment, Ask the Help Desk will be open to everyone. For future installments, this will be for paid subscribers only. (My Monday newsletter will always be free, but other perks include: subscriber-only invites, challenges, essay series, discounts, giveaways, and more!).
Pay-per-month is $5 or subscribe for the year for $50.
This is also a really amazing way to support my work and help make it sustainable for me to continue providing free information and content every week in this newsletter and on Instagram. I appreciate your support!
Did you know we can work together?! I’m now accepting applications for 1-on-1 coaching. Whether you’re looking to understand your blood sugar patterns better, improve your decision-making strategies, or build health-supporting habits, I’ve got you! Fill out an application here.
If you’re thinking about pregnancy in the next 3-12 months, you won’t want to miss my Fall T1D & the 3Ps class! Get acquainted with everything you’ll need to know for preconception, pregnancy, and postpartum. Prices increase on August 1. Register here.