Join Dr. David Ahn, MD, to answer all your questions regarding AID systems and settings for exercise! The conversation may include highlights on new technology and exercise, key features for effective workouts, and pump settings for AID systems.
He will be answering questions live for 1-hour on October 21, 2025 at 9AM PT / 12PM ET / 6PM CEST. The discussion is open now, so you can post your questions ahead of time.
2025-10-21T16:00:00Z→2025-10-21T17:00:00Z
Hi DiabetesWise Team,
Thanks for sharing info about the upcoming Ask the Expert session with Dr. Ahn. I’m really looking forward to it and wanted to submit a few questions in advance:
Exercise and AID Systems
How do different types of exercise (aerobic vs. anaerobic) impact AID performance and insulin needs?
Any tips for avoiding post-exercise highs or lows, especially with intense or interval workouts?
How well do AID systems respond in real time to rapid glucose changes during workouts?
Pump Settings and Personalization
What adjustments are usually helpful for basal rates or temporary targets around exercise?
How can users avoid insulin stacking after exercise while keeping glucose stable?
Technology and Data
Are some AID systems better than others for athletes or very active users?
How can we use system data to fine-tune exercise-related insulin adjustments?
Any new features coming that help with exercise management specifically?
Practical Tips
Any advice for multi-day events or tournaments?
Do things like temperature or altitude impact AID performance.
Really appreciate all the work you do and the chance to ask questions. Looking forward to the session!
Thanks, Harry
I’m looking forward to the session!
I know that managing BG’s during exercise is highly personalized but, after a few years of trying different approaches, I finally found a routine that works for my needs (no having to eat carbs, no after exercise highs or lows, using Tandem Tslim and DexcomG7). I turn off my insulin completely about 60 minutes prior to exercise in the morning. When my BGs increases to approx. 125 mg/dl I turn my Sleep mode Back ON (I’m usually in Sleep mode 24/7) and start my speed walk. This technique allows my BGs to stay between 120-150 and after exercise my BGs have a smooth landing back to normal. For longer or more strenuous exercise or at different times of the day, I need to make adjustments (Exercise Mode, exercise carbs, or longer periods of insulin turned off). Its still pretty much Hypo-suicide if I take a bolus, eat dinner and then go for a casual walk around the block…..any I.O.B. will tank my Bgs. Thoughts?
Hi Dr. Ahn,
I am an avid runner and utilize a T:slim X2 for my diabetes management. I am having an issue where I tend to skyrocket post runs and am having to correct post exercise which can take a few hours to return my BG in range.
I run a temporary profile which is 50% of standard basal, correction, and I/C which I try to start two hours prior to beginning exercise. I then will consume 60-80g per hour to maintain glucose between 100-200mg/dL for the hours I am exercising. I try to switch back to my standard profile as soon as I’m done but it seems like I just sky rocket and I am nervous to bolus until I have stopped exercising as it seems like my insulin works quite a bit faster and more severely when the heart rate is up.
Any suggestions for post-exercise prevention of spikes?
Thanks,
Mike
Hi Dr. Ahn, I’m on Tandem T-slim x2 with Control IQ. I’m a long-distance cyclist and I really wish I could use “exercise mode” but that leaves the auto bolus feature intact. While I’d like to have a higher target, I’d rather not get suprised by a bolus during exercise as I’m very insulin sensitive — so, I use “sleep mode” to prevent that, and I eat carbs freely without taking insulin. I also have an exercise profile that I’ll switch to which is 20% less basal about 1/2 way into my ride. Too little basal only results in highs after the ride. While this works well for me, I’m wondering if you would suggest a different approach? Thanks, Jewels
Hello everyone! I’ve been looking forward to this session. I’m an adult endocrinologist and Chief of Diabetes Services for the Mary & Dick Allen Diabetes Center at Hoag Hospital in Orange County, CA. In addition to seeing patients with Type 1 and Type 2 Diabetes, I also oversee diabetes clinical trials at my Center. I am super passionate about diabetes technology, primarily automated insulin delivery, CGM, and smartphone apps. I also make videos on Instagram (@AhnCall) and TikTok (@DiabetesDoc).
I’m excited to try my best to answer your questions and provide some insight into exercise and AID, which is admittedly probably one of the most nuanced/challenging topics in diabetes technology! 
If you’re not already aware, DiabetesWise has some amazing resources on diabetes and exercise, so it’s a fantastic starting point.
Hi Jewels, I love that you’ve experimented and found some solutions that work well for you! One possible thing to consider to try to achieve your goal of using “Exercise Mode” without auto boluses is a bit of a hack. But if you think about how Control IQ auto-boluses work, it will only deliver an auto-correction bolus if there has not been any bolus (manual or automatic) in the past 60 minutes. Therefore, while this is definitely off-label, you could try giving the smallest possible manual bolus to “delay” the next auto-correction bolus for another hour. So if you give that teeny tiny bolus every 50 minutes or so, you’d never receive an autocorrection bolus. Just something to try and tinker with!
Hi Richard! It sounds like you are approaching things the right way and are already implementing a lot of best practices (exercising in the morning, reducing insulin 1 hour before exercise, making adjustments during your workouts), etc). I think you really hit the nail on the head that insulin on board (IOB) at the start of exercise definitely increases the likelihood of going low during activity. I think awareness of this principle is one of the most important fundamental concepts with exercise. One other possibility to consider for preventing highs during longer sessions where you might want to use exercise mode without auto-correction boluses is a “hack” that I suggested in a different reply to Jules. This might give you some auto-basal adjustment with less likelihood of going low. I’m sure you’re already doing this, but for other people that are reading this, I think creating different insulin delivery profiles for different exercise routines (or using the temp basal feature that was improved with the Spring 2025 Control-IQ Plus software update) might give you a little more fine-tuning control as well.
Hi Dr. Ahn! Thank you so much for being here to connect with our community.
Is there a treatment or technology that is on its way that is really novel or exciting for you? There’s a lot at different stages, but I’m curious what you’re the most excited about.
Thank you!
Your pattern of post-exercise glucose spikes is one of the more challenging scenarios that I see in my patients! The general teaching is that people will be more SENSITIVE to insulin for several hours after exercise, but I also see people experience glucose spikes after exercise as well. I wish I had a more specific piece of advice, and I’m sure you’ve already experimented plenty, but I’d consider trying a few different things: 1) perhaps reducing your 60-80g per hour carb intake during the last hour of your exercise could help, 2) as you mentioned, a mini bolus of insulin right after you finish your exercise might help combat that spike, 3) I’m not sure what “mode” of Control IQ you run, but perhaps switching to Sleep Mode temporarily immediately after your workout might make the auto-basal adjustment a little more aggressive, although you’d be losing the auto-correction boluses, or 4) you could perhaps try using a temp basal adjustment to give 20-40% more basal insulin temporarily after your workout is complete. Those are just a few thoughts that come to mind.
Great questions… I’ll answer them in sections. I’ll start with “Technology and Data”:
That’s a great question. I’ll say I’ve seen elite athletes do well on all the major systems (Omnipod, Tandem, Medtronic), so I don’t think one stands out the most. I think they all have slightly different pros/cons. For example, the Omnipod being tubeless can be an advantage, although it only has one infusion site option which could nudge someone towards using a different system. Alternatively, Tandem’s Control IQ algorithm is the most easy to tinker with because its automation works on top of your manual basal/bolus settings unlike other AID systems. And all three major manufacturers have exercise/activity modes that can help reduce likelihood of going low. As is always the case in diabetes, I recommend trying various systems and finding what’s best for you. This is a great time to remind people that all pumps come with some sort of return policy within the first 30 or 60 days. (Or, you only pay month to month for certain pumps like Omnipod that go through pharmacy benefit).
Hi Dr. Ahn,
Thank you for joining us!! I’ve been watching your videos on the various diabetes technology updates for years - I’ve learned a lot from you! We’re excited that you could join us today.
We’re seeing a big push from companies now to include multi-analyte sensors (like glucose, lactate, ketones, etc.). In the context of exercise, this could be really interesting to study. However, the concerns around increased lag time during exercise are always on my radar. What are your thoughts? Are there any specific use cases where you think this could be interesting to test during exercise? I’m always wondering what would be helpful to integrate into future AID technology. We hope to do some additional testing in our ongoing clinical trials too!
Hope we can meet in person soon (I’ll be at ATTD)!!
Best,
Dessi
So generally speaking, aerobic exercise generally refers to lower intensity exercise such as jogging or cycling, and aerobic exercise generally will lower blood glucose. Therefore, using exercise/activity mode features in AID that raise glucose targets and reduce insulin delivery can help reduce lows in these situation.
Anaerobic exercise usually refers to higher intensity exercise such as weight-lifting or sprinting, and this can often increase blood glucose due to counter-regulatory hormone release (such as adrenaline) and generation of glucose from the liver. This can sometimes be the more challenging scenario in exercise because people are often hesitant to want to give additional insulin during exercise. But perhaps one might give small micro-boluses of insulin or the creation of slightly more aggressive profiles.
Hi Dessi, thanks for the kind words! I just listened to your podcast with Stacey Simms! Hopefully I’ll be able to find my way to ATTD next Spring, and would love to connect.
Oh man, that question about multi-analyze sensors is so intriguing. I think there is so much potential, but I feel like we will really be entering a world of the unknown when it comes to data. I feel like the established medical literature is still struggling to keep up with variables in glucose data (fake pressure lows, real shifts in insulin sensitivity with menstrual cycles, etc), that I fear we won’t have great actionable insights in the early innings after release. Heck, even our current algorithms on the market are all from at least 5 years ago! However, hopefully over time, we can perhaps learn that incorporating lactate and ketone generation might better allow us to better automatically identify what types of exercise are taking place, and in turn come up with better algorithms.
I can’t help but be a little skeptical about how long these things will take, though, to at least come into algorithm development. Hopefully, the user community and resources like DiabetesWise will be able to come up with some experiential best practices to utilize this data and at least inform the users to make more informed decisions.
But that’s why people like you are so important to be championing these technologies, doing the trials and coming up with the innovations to be able to generate evidence to take these theories mainstream.
Not great! While AID has come a long way and has been transformative in helping people improve glucose control with less emotional burden, I think exercise is still the area where AID struggles the most. This is partly because exercise in T1D depends on so many different variables: exercise intensity, duration, insulin on board, adrenaline, time of day, etc. And then on top of that, we know that CGMs have a bit of lag during times of rapid glucose change. So overall, unfortunately, it falls on users to experiment and come up with individualized routines to best work with their specific types of preferred exercise. To many of your questions, though, I’m super impressed with the Exercise section on the DiabetesWise site. Not only does it go into a lot of detail and gather the latest thinking into one site, it also has an expansive list of links to even more resources.
I think one of the most important concepts when it comes to exercise and insulin delivery is to be aware of Insulin on Board (IoB). I explain to my patients that exercise essentially “activates” insulin on board, so it’s generally a good idea to try to minimize IoB. Strategies to lower IoB include: 1) exercising first thing in the morning (as a night owl, I would NEVER be able to do this! 
), 2) turning on exercise/activity mode on your AID 1-2 hours before starting, and 3) being very mindful of any insulin bolus in the 2-3 hours before starting exercise.
Alternatively, and this is much more nuanced than I can go into in this topic, and really only applies to more predictable/configurable systems like Control IQ, but there might be a small role for using inhaled insulin like Afrezza for mealtime boluses 2-3 hours before exercise because it clears faster than traditional boluses of insulin. Like I said, though, there’s a lot of risks/nuance to using Afrezza on top of AID (since the system would not be aware of any Afrezza doses taken), and I would only consider using Afrezza with a more predictable system like Control IQ.
When it comes to basal rates and temporary targets, similar to your other question about anaerobic vs aerobic exercise, it’s so important to understand how different types of exercise impact glucose, and that it seems to be very different based on the individual as well. So definitely check out DiabetesWise’s exercise info pages and experiment with different strategies.
Oh man, this is such a great question. A few things come to mind
And I love that DiabetesWise is such a great resource to learn about and compare all these new and upcoming sensors!
I’m sure I’m missing things, but that’s what comes to mind!
We have now concluded our Ask the Expert session. Thank you @DavidAhn for answering all of our questions, and all of our participants!
Make sure to continue to follow Dr. Ahn on Instagram (@AhnCall) and TikTok (@DiabetesDoc).
