Trying to compare BG control using a t:slim closed loop system and shots

I am interested in how much better my blood glucose control could be using a pump system by comparing 90 day Dexcom Clarity blood glucose data (longest it will report) with people using a t:slim pump closed loop system, which is probably the best system currently available, to me being on manual shots (Tresiba/Lyumjev)? I see a lot of people showing outstanding t:slim data over short period but to me the real test is long term.
My Dexcom Clarity reports over the past 90 days (but has very typical even longer) an average glucose level of 100 mg/dl, 34 mg/dl standard deviation, 5.7% GMI and 89% time in range. This could be very interesting and help me and maybe other people to make a decision on getting a pump, thanks in advance!

You might be disappointed depending on what is important to you.
I love my dexcom and tandem loop. At night it targets 112, which is as low as it gets. In reality I stay about 120 through the night with total flat line even in the am where I normally have a dawn phenom.

Running I day made I generally run in the 120-130 range. And only if I miscalculate my bolus for meals, will I go out of range at all. I went 8 days in range one time but it was an effort.

My average time in range is 95%. It’s good enough for me.

The part I love most isn’t super tight control, it’s the stability that I like. I hope they will allow lower targets in the next iteration, but for now it’s fine.

My last a1c was 6.0 and that’s pretty much been how it is in the last year.

I really like that I don’t have to watch it as much because it makes small corrections and basal adjustments on its own.

It isn’t perfect but it suits me well.


Thanks for the information! A 95% time in range is fantastic. What does Dexcom clarity say about your standard deviation?
I personally don’t mind manually controlling my diabetes with shots as my brain has been trained to manage it pretty well and luckily I have not developed insulin resistance/brittleness issues yet. It seems to be working as have been a T1D for 45 years with no complications yet and my control data looks good on shots although lows concern me.
However I am always looking for ways to improve my control and pumps could do that but it also has to be easy to use and managing infusion sets with tubes as well as CGM sensors has me concerned. IMO they really need to figure out how to integrate the CGM and pump into one closed loop compact pod unit and using a phone app for wireless programing and control.

86% in range, 10% high, 1% very high, 2% low less than 1% very low. 90 day average 130mg/dl, 41 SD, GMI 6.4% and last A1C was 6.3%. I’ve gotten as low as 6% but there is only so much attention that I can pay to babysitting my BG so this is with me not really trying.

I am very brittle so I go high very easily followed by low without reason but the control IQ algorithm does a pretty good job of keeping me out of danger and I am rarely awakened by a low alarm at night anymore. I don’t use sleep mode 24/7 like @Timothy does but if I did it might be able to push down my SD. BTW, my A1C is more or less the same as when I was on MDI but with considerably less work.

The tubing rarely bothers me and the only time I really notice it anymore is when I’m changing or using the restroom. If tubing is a deal breaker for you then the Tandem Mobi should be coming out soon (if Tandem gets their act together and actually finishes submitting it to the FDA) and it will be a patch pump with a very short tubing option so you can still use whatever set you like with the bonus of Tandems Control IQ technology.


Thank you! One thing I think the pump could really help me on is controlling lows. Currently I am getting 5% low and 3% very low, but luckily most are during waking hours so I can easily adjust them as I always carry a glucose source (cheap jelly beans I put in pocket size ziplocks).
I am definitely keeping my eye on the new “artificial pancreas” devices like the iLet Bionic.

Before using Tandem X2, I used Minimed pump with Dexcom for many years. My A1Cs averaged 5.8-6.2.

After Tandem X2, reduced to 5.5, and most recently 5.2 with some additional tweaks to settings. Sleep mode helps with tighter targets.

More important is that I spend more than 95% under 160.
On Minimed (without auto/cgm adjusments) I chased more highs and lows. That was with older models.

Thanks and it sounds like your pump does well but I was more looking for the Dexcom Clarity data sync’d to your G6 you use with your pumps over the past 90 days including average glucose, standard deviation, GMI and time in range, which gives a more complete picture to better compare performance of a pump to my shot routine. No worries if you don’t use the Clarity app.

When using Tandem pump, the pump acts as receiver, and T:Connect app is typically used, which combines pump and cgm stats on Tandem based repository. So data not typically loaded to Clarity. But similar stats available.

@Mike_F I’m an agnostic when it comes to which is better for me, MDI or Tandem pumping. So I go by life at the time and the therapy that would give me the best mental and physical health. Here is pretty much what you are asking for. This Clarity report shows MDI vs Pumping as I was on MDI all of March, April and May and reapplied the pump the first week of June. My Dexcom always runs a little high (I’m good with that) and that is why my last A1c, from July 28th, was 5.8. Hope this helps.
Clarity.pdf (187.3 KB)

1 Like

Wow, it looks like virtually identical and excellent control either way! The only reason I haven’t got a pump is because of having to deal with the infusion set versus the simplicity of pen. To me MDI is SO much simpler after doing it for so long. Now if everything including the pump and CGM in a closed loop system was in one small pod unit and simple to use I might change over. That’s why I said early I am keeping an eye on the iLet Bionic they call a artificial pancreas, which is closest yet to what I am thinking. Thanks!

I hear the bionic pancreas still has issues to be resolved, the people who tested it so far have complained of long periods over 180. That’s while it is learning. But hasn’t really been resolved yet. It’s is not a hybrid loop but a closed loop so there is no way give your own corrections.

So in your previous questions you were wanting a smaller system not needing a phone or controller.

I often only use my pump and I don’t take my phone with me to hike or walk around. The pump saves the data and uploads when you come back into range.

Omnipods have a similar situation only after you’ve done enough pods to get decent basal rates.

I really like the hybrid closed loop because I can often correct my glucose before my pump gets around to it.

I can also alter then delivery of blouses if I’m exercising or if I’m sick etc. with me it’s actually very different how I calculate boluses depending what I am doing so more than half of my boluses are overridden to compensate.

I have no idea how the bionic pancreas can manage this, but I would like to see it. My guess is that system is being created for people who really don’t want to manage their diabetes or can’t.

It’s not likely you could get better results than you are already getting or with a hybrid loop pump

Did you watch the video on the iLet Bionic? The whole goal is simplicity, which is could be good or bad depending on how much control the person wants. All you do is answer one question what meal you are planning to eat! It’s an interesting approach that probably uses AI technology that learns what to do?
I am a retired R&D Electrical Engineer and even though I can figure out how to use complex technology and I have lots of experience fixing/improving stuff and I think the whole goal should be to make technology it as simple and error proof as possible for the average person and CGMs are pretty good in that respect but current pumps are not but are getting better.

1 Like

Simplicity is great if the system takes care of business and keeps the BGs in an acceptable range. That makes performance a highly subjective judgment as people with diabetes have wide range of what is acceptable.

I hesitated to comment on this thread as I have no experience with Tandem and limited experience with MDI. I would judge any system on its ability to deliver normal blood sugars for a high percentage of the time.

I think I’d have a hard time with any system that prohibits user intervention. In my mind, that compliance will occur naturally when the system’s performance earns my trust. My DIY Loop does a pretty good job yet I still must intervene on a regular basis. That might mean an external Afrezza dose or an early change to an infusion site due to degrading insulin absorption to name a few.

1 Like

This was my dream pump!!! Been following it since the early 2000’s! A pump that learns and no more carb counting?! I was all in.

I begged and begged the research center I go to, I would do anything if I could get in this study!

I was very fortunate to be in the study and this was after a few years with Tandem basal IQ and now control IQ. I struggled with it. First it was hard not to make mini doses as I have been taught to do to help keep blood sugars in target. This system needs you to just let things ride. I was very good and just let it do it’s thing but it just never worked well for me. It was the first study I decided to end early. My time in range was in the 50-70% range when my Tandem is always above my goal of 80%.

The iLet pump is a great pump for people who are not seeing an endo. It is a great pump for people that want to let go of needing to do things related to one’s diabetes. It is a great pump if “perfect” numbers are not key for ones happiness.

There is no pump that is perfect for everyone. You really do need to list what you must have, what you might like and what you hate in a pump. Try and meet with trainers and hold it, play with it. Maybe get to try it out.

For me the biggest problem with any pump in the infusion set. The one piece and make or break a pump experience!

Good luck with whatever you end up with. Pumps are not for everyone but I do love the flexibility & better numbers I get with my pump and CGM! My new dream pump has become the Tandem control IQ!


For what it’s worth, I think a T:slim would help with the lows.

Yes, having tubes attached to your body is no fun and a pen is easier but I bet you would achieve similar overall excellent control with fewer lows on a pump because it seems like you’re a motivated and disciplined person.

5% low seems like a lot to me (more than an hour every day on average) so cutting that down would be motivating, but if it’s not motivating to you, then I don’t think you would do better than your current overall control which is exceptional and there would probably be a learning period when your numbers would be worse.

1 Like

I don’t have the numbers in front of me as I switched to Looping with AAPS but when I wanted to move to AAPS from T:Slim, my endo said "Why would you want to change when you are getting 98% TIR with T:Slim. I moved anyway and with AAPS I am hitting 97% TIR 1.5 Standard Deviation and 6.0 GMI. There was a drop since you need to complete all objectives before it is a fully closed loop. I loved the T:Slim but Love AAPS with DASH even more. It is a fantastic system. I really needed to work at it with T:Slim but with AAPS, it takes less effort.
It isn’t FDA approved but there are Endo’s that will help you with it, at least in Canada.
To Covert I have a 113 Average BG, 27 Standard Deviation 6.0 GMI. Less than 1% very low, 1% low and less than 2% high. I hope to get better once I am done all objectives and can also set up automation.

1 Like

My 90 day average is 106 with a GMI of 5.9 and a standard deviation of 19. Also my time in range (70-150) is 96%. The best part is I almost never get an alarm after bedtime which occurred regularly before Control IQ. I still have to follow it but I sleep all night.