Wow....great article and here I am, with my new 780 pump being shipped, as of this morning!! I am so interested in the human input for a Type 1 in terms of managing. I have had diabetes for 73 years but have been on insulin pumps since 2000 and CGMs since around 2015. I've had my A1cs in the low 5's for most of my pump years. I use very small amounts of TDD insulin (11-13 units). I am horrified with the idea of "because you are an older person your blood sugars should run higher." As if I'm too old for something like a complication to finally appear...LOL. YES, I am very healthy. So I'm being pushed into using the Medtronic sensor system. Can it really manage ME better than I have been doing? I am keeping your article handy!!!
Barbara -- Statistically, older T1Ds have lower A1c's and narrower CV% than younger ones. I will be posting an article about long-lived T1Ds in the coming months that unravels some of the latest studies on this, but the general theory being that, once you pass 50 years with T1D, you've got some genes that are protecting you from the worst-of-the-worst complications that would have done in others years earlier.
In any event, all T1Ds need to maintain the expectations that they will have to be engaged in their own treatment, and not fall for the fallacy that fully-automated pups will relieve them of any kind of engagement at all. that's just never going to happen. As the article said, even the natural human body can't keep glucose levels within "healthy" range very well UNLESS the person chooses healthy, engaged lifestyle choices (exercise, diet, sleep).
Hi Dan! Excellent article, thanks for your hard work in putting it together. I always learn something new from your pieces. As a fellow MDI-er, I was wondering - what is your personal exercise routine? I've read your previous articles where you emphasize the importance of exercise and physical fitness, but I wasn't able to find an outline of what you do day-to-day. I ask because although I have good control (93% 90d TIR), I would like to be injecting less volume of insulin to achieve these same results - and it sounds like (more) exercise is the best way to get there. Just curious if you have a routine like a weekly step count or miles-run goal, or anything like that, that you adhere to that reduces insulin resistance. Thanks again.
Exercise is such a metabolically healthy thing to do, I don't think one can get too much of it. That said, one can cause self-harm by exercising improperly, which often happens when people begin new routines; it can also happen to those who exercise excessively and get a sense of over-confidence. One has to know their limits to avoid injury. I know that's not the case with you, but it bears mentioning for the general audience.
I run 3mi every morning, and then about 30min walks after every "meal" (or anytime I consume >50g carbs). Beyond that, I also hike or bike several times a week.
The most important thing to watch is your heart rate "zones." Zone 1 is walking, where you can burn carbs sufficiently to offset some insulin needs on a per-walk basis (say, reducing the bolus you need for a meal), whereas Zone 2 and 3 (running) not only does the same thing, but it increases mitochondrial efficiency, allowing more glucose to be absorbed throughout the entire day, not just during exercise. This reduces day-long insulin needs--that is, the basal rates. Once you get into Zone 4 and 5 (much high exertion running or cycling, but also weightlifting), then you're expanding mitochondrial growth itself--that is, you're building new muscle fibers, which imply new mitochondria cells, which further expands efficiency, thereby decreasing overall insulin needs.
But keep in mind that all of this works in reverse as well. I just went on a 10-day trip to new zealand, where I hiked and biked, but I didn't run at all. As the days went by, I got lots of zone 1 and 2, but no zones 3-5, so slowed down because there was less of a demand on it. My insulin intake went from 30-35u/day to 40-45u/day, my TIR dropped from 97% to 91%, and my SD went from 20-25 to 35-40. (SD is just as important as TIR!) In other words, keeping zone 3+ into my regimen was highly important insofar as not just insulin needs, but glycemic variability, which affects TIR and SD.
Now that I'm back home, I'm running again, and experience tells me that it'll take about 1-2 weeks before I'm back where I was.
Note that I'm 61, so this process is much slower than it used to be. When I did all this in my 30s and 40s, it took a much longer time for mitochondrial efficiency to diminish, and much less time for it to recover.
Lastly: genetics plays a roll as well. Some people are more efficient than others and can preserve excellent mitochondrial efficiency, even in the face of direct assaults (like weight gain). As the car companies always say about MPG, "your mileage may vary."
Nice quick summary of all the diabetes knowledge, thanks! People really do expect from automated pumps too much, but as flawed and limited they may be the key thing is it will still ease the burden at some extent.
Hi Dan, My algorithm is Tandem Control IQ, which is one of the three fully automated algorithms commercially available in Australia (the others being the Medtronic 780 and mylife CamAPS FX mobile for the Ypsomed pump ). They all make automatic corrections to basal levels and add extra boluses at all levels of blood glucose, so you are not right in saying that Control IQ is semi-automated. The Tandem pump also allows the user to easily suspend, alter basal or give extra bolus on top of its corrections - so you get the best of both worlds. I have used the suspend type from both Medtronic and Tandem (Basal IQ) as well. In my experience they are not nearly as effective, even at the job they are doing of trying to prevent major hypos. And this is because Control IQ starts tapering basal as soon as you are predicted to be going low and stops basal well before you are low.
My implication was that 'semi-automated' is where the user inputs either data or instruction of any kind. If you completely leave it alone and never interact with it, then that's "fully automated mode."
This article says that fully-automated mode will never perform as well as semi-automated mode, provided the user has any knowledge of how to manage their T1D. That is, they will be aware of conditions that the system cannot know about (usually, future events), where they then give the pump some kind of instruction. Under these cases--which are categorized as semi-automated--the system will yield better outcomes than fully-automated, because the closed-loop system cannot sense the myriad of signaling hormones that can better forecast glucose trajectories. It can only sense current glucose levels, which are inherently insufficient for optimal glycemic control.
My follow-on articles will cite randomized control trials showing that fully-automated pumps are useful for specific individuals who are either unwilling or unable to take care of themselves. And in those cases, these closed-loop systems yield A1c levels around 9%, with occasional cases dropping into the 8's. this is definitely progress, to be sure, but it's still not as good as the user engaging with the system in the semi-automated fashion described above.
I haven't found the distinction you make, Dan, between the Tandem Tslim with Control IQ - a rules-based automated insulin pump - and using a manual pump. I wouldn't go back to any non-automated pump for anything, because it takes away a lot - but by no means all - of the many decisions we need to make each day, even each hour about insulin, carbs, exercise, ... Often from careful observation of what happens in recurring situations, I can do a small boost of insulin or suspend way before the pump does. The automated pump has improved my control but decreased the mental load a lot, so this has improved my mental health a lot. And no more ambulances in the middle of the night from the unexpected low from which I cannot be aroused without drastic intervention. And Control IQ is way better than Basal IQ or manual at dealing with overnight. Mostly I'm between 4 and 6 all night and sleeping all the time.
My concluding comments are referring to FULLY automated pumps, where there is no human interaction at all. Your description is on semi-automated pumps, which (for the most part) are detecting low glucose readings and tapering back (or stopping) insulin delivery. I address those in the next article in the series.
Hey Thomas! That is very interesting....I have not heard of him but thanks for letting me know about that. I did a quick Google search and found him. I am running now about 45 miles a week and that helps my blood sugars tremendously!
Wow....great article and here I am, with my new 780 pump being shipped, as of this morning!! I am so interested in the human input for a Type 1 in terms of managing. I have had diabetes for 73 years but have been on insulin pumps since 2000 and CGMs since around 2015. I've had my A1cs in the low 5's for most of my pump years. I use very small amounts of TDD insulin (11-13 units). I am horrified with the idea of "because you are an older person your blood sugars should run higher." As if I'm too old for something like a complication to finally appear...LOL. YES, I am very healthy. So I'm being pushed into using the Medtronic sensor system. Can it really manage ME better than I have been doing? I am keeping your article handy!!!
Barbara -- Statistically, older T1Ds have lower A1c's and narrower CV% than younger ones. I will be posting an article about long-lived T1Ds in the coming months that unravels some of the latest studies on this, but the general theory being that, once you pass 50 years with T1D, you've got some genes that are protecting you from the worst-of-the-worst complications that would have done in others years earlier.
In any event, all T1Ds need to maintain the expectations that they will have to be engaged in their own treatment, and not fall for the fallacy that fully-automated pups will relieve them of any kind of engagement at all. that's just never going to happen. As the article said, even the natural human body can't keep glucose levels within "healthy" range very well UNLESS the person chooses healthy, engaged lifestyle choices (exercise, diet, sleep).
Hi Dan! Excellent article, thanks for your hard work in putting it together. I always learn something new from your pieces. As a fellow MDI-er, I was wondering - what is your personal exercise routine? I've read your previous articles where you emphasize the importance of exercise and physical fitness, but I wasn't able to find an outline of what you do day-to-day. I ask because although I have good control (93% 90d TIR), I would like to be injecting less volume of insulin to achieve these same results - and it sounds like (more) exercise is the best way to get there. Just curious if you have a routine like a weekly step count or miles-run goal, or anything like that, that you adhere to that reduces insulin resistance. Thanks again.
Exercise is such a metabolically healthy thing to do, I don't think one can get too much of it. That said, one can cause self-harm by exercising improperly, which often happens when people begin new routines; it can also happen to those who exercise excessively and get a sense of over-confidence. One has to know their limits to avoid injury. I know that's not the case with you, but it bears mentioning for the general audience.
I run 3mi every morning, and then about 30min walks after every "meal" (or anytime I consume >50g carbs). Beyond that, I also hike or bike several times a week.
The most important thing to watch is your heart rate "zones." Zone 1 is walking, where you can burn carbs sufficiently to offset some insulin needs on a per-walk basis (say, reducing the bolus you need for a meal), whereas Zone 2 and 3 (running) not only does the same thing, but it increases mitochondrial efficiency, allowing more glucose to be absorbed throughout the entire day, not just during exercise. This reduces day-long insulin needs--that is, the basal rates. Once you get into Zone 4 and 5 (much high exertion running or cycling, but also weightlifting), then you're expanding mitochondrial growth itself--that is, you're building new muscle fibers, which imply new mitochondria cells, which further expands efficiency, thereby decreasing overall insulin needs.
But keep in mind that all of this works in reverse as well. I just went on a 10-day trip to new zealand, where I hiked and biked, but I didn't run at all. As the days went by, I got lots of zone 1 and 2, but no zones 3-5, so slowed down because there was less of a demand on it. My insulin intake went from 30-35u/day to 40-45u/day, my TIR dropped from 97% to 91%, and my SD went from 20-25 to 35-40. (SD is just as important as TIR!) In other words, keeping zone 3+ into my regimen was highly important insofar as not just insulin needs, but glycemic variability, which affects TIR and SD.
Now that I'm back home, I'm running again, and experience tells me that it'll take about 1-2 weeks before I'm back where I was.
Note that I'm 61, so this process is much slower than it used to be. When I did all this in my 30s and 40s, it took a much longer time for mitochondrial efficiency to diminish, and much less time for it to recover.
Lastly: genetics plays a roll as well. Some people are more efficient than others and can preserve excellent mitochondrial efficiency, even in the face of direct assaults (like weight gain). As the car companies always say about MPG, "your mileage may vary."
Nice quick summary of all the diabetes knowledge, thanks! People really do expect from automated pumps too much, but as flawed and limited they may be the key thing is it will still ease the burden at some extent.
Hi Dan, My algorithm is Tandem Control IQ, which is one of the three fully automated algorithms commercially available in Australia (the others being the Medtronic 780 and mylife CamAPS FX mobile for the Ypsomed pump ). They all make automatic corrections to basal levels and add extra boluses at all levels of blood glucose, so you are not right in saying that Control IQ is semi-automated. The Tandem pump also allows the user to easily suspend, alter basal or give extra bolus on top of its corrections - so you get the best of both worlds. I have used the suspend type from both Medtronic and Tandem (Basal IQ) as well. In my experience they are not nearly as effective, even at the job they are doing of trying to prevent major hypos. And this is because Control IQ starts tapering basal as soon as you are predicted to be going low and stops basal well before you are low.
Hi Tim --
My implication was that 'semi-automated' is where the user inputs either data or instruction of any kind. If you completely leave it alone and never interact with it, then that's "fully automated mode."
This article says that fully-automated mode will never perform as well as semi-automated mode, provided the user has any knowledge of how to manage their T1D. That is, they will be aware of conditions that the system cannot know about (usually, future events), where they then give the pump some kind of instruction. Under these cases--which are categorized as semi-automated--the system will yield better outcomes than fully-automated, because the closed-loop system cannot sense the myriad of signaling hormones that can better forecast glucose trajectories. It can only sense current glucose levels, which are inherently insufficient for optimal glycemic control.
My follow-on articles will cite randomized control trials showing that fully-automated pumps are useful for specific individuals who are either unwilling or unable to take care of themselves. And in those cases, these closed-loop systems yield A1c levels around 9%, with occasional cases dropping into the 8's. this is definitely progress, to be sure, but it's still not as good as the user engaging with the system in the semi-automated fashion described above.
I haven't found the distinction you make, Dan, between the Tandem Tslim with Control IQ - a rules-based automated insulin pump - and using a manual pump. I wouldn't go back to any non-automated pump for anything, because it takes away a lot - but by no means all - of the many decisions we need to make each day, even each hour about insulin, carbs, exercise, ... Often from careful observation of what happens in recurring situations, I can do a small boost of insulin or suspend way before the pump does. The automated pump has improved my control but decreased the mental load a lot, so this has improved my mental health a lot. And no more ambulances in the middle of the night from the unexpected low from which I cannot be aroused without drastic intervention. And Control IQ is way better than Basal IQ or manual at dealing with overnight. Mostly I'm between 4 and 6 all night and sleeping all the time.
My concluding comments are referring to FULLY automated pumps, where there is no human interaction at all. Your description is on semi-automated pumps, which (for the most part) are detecting low glucose readings and tapering back (or stopping) insulin delivery. I address those in the next article in the series.
Thanks for your post! I enjoyed it. 🙏❤️✅🙏👍🏻💪 I am a T1D ultra runner. My next big run is 45 miles on January 21st!!
Trying the Daniel Darkes way? 😅 He literally ran out of his type 1.
Hey Thomas! That is very interesting....I have not heard of him but thanks for letting me know about that. I did a quick Google search and found him. I am running now about 45 miles a week and that helps my blood sugars tremendously!
Really terrific article, I learned so much even after 50 years of having type 1, thank you Dan!