Good afternoon, everybody. Welcome to today's webinar. I want to thank you all for being here. My name is Stuart Weinzimmer, and I am the current chair of the American Diabetes Association Diabetes and Youth Interest Group Leadership Team. And on behalf of the team, I'd like to welcome you. Here's a quick glimpse at our agenda today. We'll have a few minutes of announcements, and then I will be pleased to introduce and welcome Dr. Diana Isaacs, who's going to speak with us today about automated insulin delivery for kids. After her talk, after her lecture, we'll be followed by a question-answer period. And for all of you, today's webinar will be recorded, and you will all receive an email with the link to the recording for you to view afterwards. Now, this depends on your participation. We encourage you all to share your experiences and bring questions about automated insulin delivery for kids to us. Dr. Isaacs will take questions at the end of the event. However, there's no need to wait until the end of the lecture to send in your questions. You can type your questions into the Q&A box in your control panel, and we will answer them at the end of the lecture. Be sure to use the Q&A box for questions, and then the chat box to share your experiences with other audience members. We will be using the chat box to send you links during this announcement segment as well. I wanted to take a moment to thank all the members of the leadership team for their work throughout the year to provide opportunities to the interest group members. In addition to planning this webinar, this leadership team has planned an interesting symposium at this year's scientific sessions, and we will be hosting a networking event during that weekend in June. We're also offering an abstract award to an early career member. Our team looks forward to offering more to the interest group members throughout the coming year. And as a reminder, registration for the scientific sessions is now open. Early bird registration, which includes registration discounts, ends March 9th, so get your registration in early. Go to the link in the chat or on the screen to hear more, learn more, and register. Also, keep up to date with events or popular topics of this interest group on the Diabetes Pro member forum. This is an ADA member exclusive forum where all members of the interest group can connect. Follow the link in the chat or on your screen to connect with fellow ADA members. And finally, a plug to be a part of the Diabetes and Youth Interest Group leadership team. If you are interested in becoming more involved in our interest group, the ADA is seeking applicants across all 16 interest groups in various positions, but particularly for the youth interest group, we are looking for an early career representative. And as you can see, this will provide our group with the perspective and guidance for what our early career members may need as they advance in their careers as diabetes leaders. You can see the link in the chat or on your screen to apply by February 28th. My last announcement, I want to highlight other ADA webinars that are scheduled for the next month. As you can see here, to register for these, please go to the link in the chat and on the screen. Now, I'd like to introduce today's presenter. Dr. Diana Isaacs is an endocrinology clinical pharmacy specialist and director of diabetes technology initiatives at the Cleveland Clinic, Endocrinology and Metabolism Institute. Dr. Isaacs has served on the American Diabetes Association Professional Practice Committee over the past three years, the committee that updates the ADA standards of care. She's also a national board member of the Association of Diabetes Care and Education Specialists. She advocates access and choice to the latest technologies and therapies for all people with diabetes and speaks on diabetes-related topics nationally and internationally. Dr. Isaacs was the ADCES Diabetes Care and Education Specialist of the Year in 2020. And we are thrilled to be listening and learn from her today. Dr. Isaacs, the floor is yours. Great. Well, thank you so much. And thanks to the ADA and the Diabetes and Youth Interest Group for inviting me to speak today. I am really, really excited about this topic where we will talk about automated insulin delivery for kids. And I will be really be providing my perspective on onboarding and practical tips. So this is not a CE program, but I still wanted to share my disclosures with you. And these are the things that I'm hoping to go through today. So I wanted to review the different types of AID options that we have right now, because it's a little nuanced in terms of onboarding and practical tips based on the specific system. And then I wanted to really get into the heart of, okay, what are those practical things that we can do to really optimize care for people with diabetes? And then also I wanted to share a couple of cases and look through the data in terms of tips and tricks that we could do to optimize the system for those kids. So with that, I wanted to first just highlight this quote from the ADA Standards of Care that says, no device used in diabetes management works optimally without education, training, and follow-up. And this really could not be more true. The tech keeps getting better and better. It gets easier to use. But education and training is still key. Otherwise, there is still a million ways to mess up the technology if you don't have a good, strong foundation. So with that, I wanted to highlight, well, what are we talking about today? And later on, we will go into a little bit more detail about the systems. I have decided to focus on the FDA-approved systems. So we have the Omnipod- and Omnipod-5, and we're going to be talking about Omnipod-5 because that's the AID system. We've got the T-SLIM-X2 with the Dexcom G6 CGM as basal IQ and control IQ. And then we've got the Medtronic 770G system. You may have heard that very recently Tidepool Loop got FDA approval. So that is really, really, really exciting. I'm not going to focus on that today just because it's not commercially available yet, but that is really, really exciting news and offering people more choices. So I wanted to do a little bit of comparison about some of the pediatric-related characteristics of these pumps when kind of deciding, well, which pump are you going to use and what things do you need to know about it? So for starters, there are minimum ages. So either two years for Omnipod-5 and 770G or six years for control IQ. I mean, we know things are used off-label all the time based on the situation, but just understand that's where the data is. That's how these systems were studied. They also all have minimum daily insulin requirements. So that's an important consideration depending on how much insulin the child is using. And so it ranges from five units with OP-5, 10 units and a weight of 55 pounds for control IQ, and then eight units for 770G. So that can come into play when choosing the technology. Also differences in terms of how much insulin that they hold. So on the other spectrum for people that may have more insulin resistance or higher insulin requirements, that is super helpful to know. And then minor differences in terms of basal and bolus increments. I mean, for the most part, they're super small amounts across the board, but slight differences. And then the change frequency comes into play as well. So generally we've always said, okay, change it every two to three days. That has changed now with the 770G where now there is a seven-day extended infusion set option. Omnipod-5 really forces you to change it after 80 hours, but we do see that sometimes people extend it out much further than those three days, which can cause the quality of the insulin to not work as well and cause numbers to go a little bit higher. And then I also wanted to highlight, so saline starts are really nice because it gives an opportunity to experience what the pump is like before it's actually administering insulin. But a caution, we can't do that with Omnipod-5 because the algorithm is really running off of that data. And so if it's saline, that's really going to impact then when you flip it into automated mode, that really will mess up the system. So it's not recommended to do that with that system. All right. So with that kind of overview, and don't worry, we're going to come back to more of the specifics with these pumps, but I wanted to first go through some global onboarding and practical tips. And I wanted to share this paper that I was really privileged to be a part of, and actually your host, Dr. Weinzimmer, also was one of the authors of this paper. It's Consensus Recommendations for the Use of AID Technologies in Clinical Practice. Really, I think a very practical paper that talks about the education, training, and support. It provides a nice overview of the different systems, too. Also talks about some of the psychological issues and person with diabetes perspectives. So a lot of the information I'm going to be sharing actually comes from this paper. If you haven't seen this, this is really great. It's called the CARES Framework, and this was developed by Laurel Messer and colleagues at Harvard Davis. And it's a way to really compare the different AID systems based on these five different categories. So how does it calculate insulin delivery? Because it's different depending on the system. Why that's important is because how you're going to make adjustments depends a lot on how it's calculating that insulin dose and how it's administering that insulin. You want to know when and what type of conditions it's going to revert back to the manual settings. And making sure that the manual settings make sense for times when that may happen, either by choice or just by default. Key educational points that may be specific to the different type of pump. And then sensor and sharing. What are the sharing options in terms of mobile apps and what sensors are compatible with the respective pumps? There are also some great tools. So it's called these Panther tools. And it's also, I've got the link at the end of this presentation, but it's wonderful. They have these wonderful tip sheets on the respective pumps that go through these five categories of the CARES Framework. And also even include going through the data reports and what you're looking at. So I highly recommend using these tools. They're also available in DanaTech, which is a technology resource through the Association of Diabetes Care and Education Specialists. DanaTech is free for everyone. And these tip sheets are also free for anyone and everyone. Another recommendation I have. So one of the things I love doing is helping people to identify the optimal technology for them. And I love having choices and I want people to know about their choices and be able to make the most informed choice. And one of the ways that we can do that are through these different simulation apps. It gives people an opportunity to kind of try it out a little bit. And so for example, there's the T simulator, which works very well. It basically makes your phone into control IQ or basal IQ and you get a really good feel for what the pump is like, how it operates and everything. Omnipod 5 also has a simulator and then Medtronic has a mobile app simulator as well as a website version. That really offers a great feel. There's also a G6 simulator and there's also a great website called DiabetesWise.org, which asks very targeted questions where a person can think through, well, what things are important to me? Am I a very active person? I want maybe something that is tubeless. Different questions that can help guide a person into kind of picking the optimal technology. In terms of readiness, so I'm a big fan of CGM first, like put CGM on the day you're diagnosed, okay? I think that's wonderful and that's really where things are headed, which is great. In terms of the insulin pump though, I think we can do that really early on too, but there are some learning points. We want to make sure people understand before they dive right in, just so they have the most success. I mean, usually the systems work great, but you want to know how to problem solve or troubleshoot in case something does happen. And so having the core knowledge of things like, okay, what is basal versus bolus? How do I count carbohydrates or at least have an idea of the foods that contain carbohydrates for bolusing? What do you do? When is it a time to check ketones? And how do you problem solve if glucose is high and it's not coming down? Also in terms of physical activity and the different factors that can affect glucose levels. Understanding insulin pumps and infusion sets and how they're delivering insulin. And then also just level setting some of the expectations with the system that we are not, and just not even the system, with diabetes in general, the goal is not to have a flat CGM tracing. I mean, flat is basically dead. Even a person without diabetes will have some variation when they eat. So it's really about understanding what the voles are, time and range. We're usually aiming for that 70%. And so just having that background can really help. And unfortunately, there's no one way to do education. Different clinics can set this up differently. But the key is you just want to make sure you've got some program, something in place before you just handed a pump to someone. And I mean, worst case is they self-start and they have none of this information. And anything that can go wrong, I've seen it go wrong. Now, another thing is before someone's starting the pump is you do need to calculate those initial pump settings. Who knows? This could change in the future with systems on the horizon like beta bionics that would just be entering in wait. But with all of our current systems, you need to enter in settings. And this is especially important, even with systems that don't use these settings at all, because there are always times where it can revert back to its manual settings. And so you want to make sure they make sense. In general, a great way of calculating these are you can take the total daily dose before the pump and take 25% of that because generally the insulin is absorbed. It tends to be absorbed and more efficient going through the pump. Or the weight base, taking 0.5 units per kilogram. A lot of times we'll average those two together for your total daily dose. And then you want half of that to be basal. And then we can use these estimates like the rule of 450 or 500 divided by the total daily dose to estimate that carb ratio. For correction factor, that rule of 1700. Sometimes you see at the rule of 1800 divided by that total daily dose. The reason I'm showing this here and I want to review that is because occasionally I will come back to these formulas and say to myself, is this making sense based on someone's total daily dose? Sometimes you will see a carb ratio or a correction factor that just looks really cookie cutter like an ICR of 15 and a correction factor of 50. And it could be that there wasn't a ton of thought going into designing these calculations. And so it is good to periodically come back and say, well, wait, does this make sense based on someone's total daily dose? Oh, caveat though is often with these systems, we find that a more intensive carb ratio works a little bit better. So often we are intensifying it by 10 to 20% of whatever their pre-pump carb ratio was as well. And then also I can't emphasize enough revisiting the basal because a lot of times you'll see with these systems what their person's getting in the automation is very different than what they were getting previously and what is their manual setting. And so it's a good idea to periodically look at that and make sure they align a little bit better for times when they may not be in that automated mode. So some tips and tricks. The pre-bolus is really everything. Pre-bolus is more important than accurate carb counting. I mean, yeah, you can't be completely off with your carb counting, but because these systems are going to ramp up insulin delivery if glucose is predicted to go higher or decrease insulin if it's predicted to go lower, you kind of have some wiggle room. But what we can't make up for is that insulin is just too slow. Even our ultra rapid insulins like our FIASP and our LumeGel, they still are just not fast enough. And so if you can get it in the system faster earlier, then you've got a much better chance of controlling that postprandial glucose. Another word of caution is that if you miss the bolus, and this happens all the time, people are human. So you forget to bolus before you eat. The way you go about it may be a little bit different because that system is going to ramp up insulin delivery, understanding that a bolus was missed. And so one consideration in the paper that I mentioned was that if it's 30 to 60 minutes after eating, consider just putting in half the carbs. And that's because some insulin was already given. And so if you do enter that full amount, you may notice that it drops, it ends up going too low. And that's a common thing we see on these reports that someone bolused, the glucose was already high, and then it ends up dropping after that. And if it's significantly later, if it's like an hour later, two hours later, then you may consider just not putting in those carbs and just giving a correction at that time. And then the other thing is with the treatment of hypoglycemia. So we've long drilled into people the rule of 15, 15 grams of fast acting carbs, but these systems are suspending or reducing insulin already. So often just five to 10 grams of carbs can work really well to kind of ride it out and let it come back up on its own. Otherwise you may have the opposite effect where it goes too high. And then another consideration is not using the reverse correction. So what reverse correction does is it subtracts some insulin if you are below the target. And with some systems, like let's say the Omnipod 5 and your target is set for 110 and the glucose going into the meal is 85. Well, it's gonna actually subtract insulin if that reverse correction's on. And because we know people generally do better if they get a little more insulin upfront, it sometimes can be very beneficial to just turn that reverse correction off. In the case of T-SLIM with Control IQ, what you can do when you put that bolus in, you have the option to basically press the X so that it's not subtracting insulin based on that glucose reading if it's under the target. And then also if glucose is high, we have to go back to kind of our traditional insulin pump thinking that it can just be an infusion set issue. If we're giving several corrections and glucose is not coming down, there's a good chance it's not an issue with the algorithm that actually it just may be that the site, the tubing could be kinked, there can be an issue with the site, the cannula could be bent. And so if it's not coming down, then the best thing to do is to replace the infusion set or replace the pod. And if giving an insulin injection outside of the pump, you just wanna be really careful because that won't be recorded by the pump. And so there'll be active insulin that's not recorded if you're giving future boluses. If you are planning to disconnect for more than 15 minutes, generally you wanna suspend the insulin. I mean, it's a good rule of thumb to just suspend when you're disconnected, but that's just because with many of the algorithms, if it's not connected and the glucose is not going down, it kind of can confuse the algorithm and mess up the insulin delivery moving forward because it doesn't understand why it's giving insulin but the glucose is not coming down because it's actually disconnected. And then with all the systems, we wanna maximize the time in the automated mode. So generally to do that, you've gotta be wearing the CGM. That's like number one. And then if you get kicked out for other reasons, looking at what can be the causes, if it's the maximum insulin delivery or minimum. Fortunately, with all our systems, getting kicked out has really just improved so much throughout the years. And we wanna encourage people not to trick the system, whether that's entering in extra carbohydrates or giving insulin outside of the pump, because that ends up messing up the algorithm and it can just be more frustrating moving forward. So it's better to follow the system. And what I tell my patients is if you're frustrated, maybe the settings aren't working well, at least try to follow them so then I can see what's not working. If someone's always entering in extra carbs, then it may be hard to tell, is it a carb ratio issue, a correction issue? It's hard to tell if they're kind of always overriding it in addition to possibly messing up the algorithm. Now, physical activity is a big thing, especially for kids who can be very, very active. And unfortunately, none of the AID systems have this down 100%, but there's a lot of strategies that we can do. They all have options to have a higher target, whether that's exercise mode, or that's the hypoprotect mode, or setting that higher target at 150. Just like we would do with a temp basal in the past, before we had AID, you wanna start this about an hour or two before the planned activity. Unfortunately, this isn't always enough. Sometimes this higher target is just not enough and people still go low. Sometimes suspending the insulin completely during the activity is needed. With tandem, because it is working off of the preexisting profiles, you do have an option. You could create a separate profile for exercise and have basals that are just much, much lower and other settings like the correction factor that are less aggressive. So that is an option. Also, it's ideal if you can minimize exercise at the peak of bolus insulin action. So generally, after you've given a bolus, like let's say lunchtime bolus, especially if it was a larger bolus, a larger meal, you would expect that insulin's gonna be peaking at one to three hours after, probably really like one to two hours after. So that may not be the optimal time to be exercising, or you may wanna come up with a better plan if you are going to be during that time. You wanna be cautious with extra carbohydrates before exercise, because what can happen is if it's too far before the exercise, like let's say it's 30 minutes, 40 minutes before, the glucose is gonna start trending up and it's gonna give more insulin. So that can totally backfire. If needed, what we can advise to do is to take some simple carbs about 10 minutes before and also throughout the activity as needed. And I took this table from that publication, just offering some additional suggestions about what to do and maybe some of the differences between aerobic and anaerobic activities. So the other thing I wanted to mention are some of the psychological considerations. In general, the idea of AID is it's supposed to be easier. It's supposed to take away a lot of that burden of managing diabetes, especially overnight. It does such an excellent job overnight of keeping the glucose in range. So the whole family is not panicked and everything, but it does involve some letting go. And that can be harder, especially if you were previously on MDI or on a pump that required a lot of kind of manual changes. So that sometimes can be kind of a hard thing to say, well, I'm gonna let this go. I'm gonna let this machine take care of it. Also, if someone was used to keeping glucose a little lower, all of these are designed to keep someone in a very safe range. But for some people that could be higher. They could be averaging 150s, 160s versus previously, some have it where they're averaging 110, 120. And so that's something in general, I would say it's worth it for the peace of mind to be able to run a little higher. And usually that also helps to prevent having as many lows by keeping it a little bit, often if it is averaging a tiny bit higher, although not always. And then we can see if we can increase autonomy for the child. That's really important. And I like to think of it as progress, not perfection. I mean, none of our systems are designed for perfection and we don't need perfection. We want people, we want kids to be kids. We want them to have fun, to be able to go to birthday parties, to eat the cake. I mean, we want kids to be kids and enjoy the activities that their peers are doing. AID, I think the real benefit is it can make it so diabetes is not always center stage. So instead of the parent's child comes home asking from school, asking, well, how was your blood sugar? Like what happened? How much insulin did you take? What did you do? Maybe we can just put that to the side and we can talk about the other things that are happening. Talk about friends, just talk about other activities and not always have to just focus on the diabetes. So with that, I now wanna talk about some of the specific things with the various insulin pumps. So I'm gonna start with the Omnipod 5. And of course, one of the big, wonderful things about this is the fact that there is not tubing. And so for active kids, this is a huge, huge plus. The algorithm works off of the total daily dose and it does use the last four to five pods for adjustment. So it is good. The first pod, generally, people do run a little bit higher because it's very, very cautious. And one of the benefits about it being so cautious is that if someone's going from MDI, especially with a longer acting insulin, like Glargine U300 or Degludec, that it's often it's safe without having to hold tons of doses because it is much more cautious in that first pod. But for some, it's really annoying because they don't wanna be running so much higher with this system. Now, the system can be controlled through a compatible smartphone or PDM. And right now, smartphones, they're all Android, but eventually iPhones should be compatible too. Now, one of the things that you have to be aware of is you do need to have the Dexcom G6 on some type of compatible smart device. So what that means is the child will need to carry either a smartphone or a device that has the G6 mobile app on it. You cannot use the receiver for the G6 and you need to have the app at the very least to be able to start and stop the sensor. But also the app is where the alerts are gonna come through. And that's also facilitates data sharing. So that's why that is really needed. I included charging cables required with the PDM because I think it's important to know when do you need a charging cable versus when do you need batteries and making sure you have backups because if it stops charging, then you can't give a bolus. Fortunately, the Basel will still continue to automate in the background even if the PDM is misplaced. I do recommend putting like one of those Apple chips on it so that you don't lose it or some type of chip. It doesn't have to be Apple, but you do need that to be able to bolus. And then I have had people lose their PDM. So that's unfortunate, so you can't bolus. And then the glucose targets range from 110 to 150. You can change those in 10 point increments and you can set different ones throughout the day. So if there's gonna be a lot of activity, worried about going low overnight or certain times of the day, it's that option to have that higher target as well as that HypoProtect, which also has some, it's just less aggressive with that insulin. And then this is an interesting thing. So you can adjust the active insulin time with this pump, although it only affects the bolus dose. What's really interesting is the insulin on board, the way it's calculated with this system. So normally insulin on board, we think of as, okay, you recently gave a bolus an hour ago. And so that insulin, it's still working in your system. So if you go to give another bolus, it's gonna subtract some, understanding there's still some in your system. But the way this calculates it, is it also includes the basal. So if the basal, that background insulin is higher than kind of what the usual background insulin is for the system, it will put that towards the insulin on board. And so it's very possible that someone, a child could wake up in the morning with some insulin on board. And you're like, wait, what happened? We didn't give a bolus the last eight hours or 12 hours. Well, it's because for whatever reason, that background insulin was ramping up overnight to keep the glucose at target. And so that's gonna be factored in then when giving that bolus the next time that there is more of that insulin on board. And here's just an example. So glucose is what's used to be able to view this data. And one of the things with these reports are you can't, you might be wondering, well, how do you know what that max basal is? How do you know when it's gonna show up as insulin on board? We don't know. I mean, it's proprietary. It's part of the algorithm. What you do see on the reports are that the red represents times where insulin is suspending. And then orange is representing times where it reaches its maximum delivery. That's when you see that OPS basal. Beyond that, we don't really know exactly how much it's giving. It does say though, in each day, you do get a breakdown of how much total insulin and how much total basal. So at least you have an idea of what the person is getting each day. Okay, next, I wanna talk about the Tandem. And I've got some fun charts that compare these as we get a little later on. So Tandem, it's a touchscreen. It is a rechargeable pump. Only this pump, if it's not charged, you will not get your background insulin. So super important that it's charged a little bit every day. It's also integrated with the Dexcom G6 and it loads through T-Connect, although you also can use Gluco or Tidepool, but T-Connect is its own platform. And then there's the two algorithms. There's basal IQ, where it's suspending if it's predicted to go low, and then control IQ, which is the fully automated system. Just wanted to show you the basal IQ example here. So here, all it's doing is suspending glucose if it's predicted to go under 80 in the next 30 minutes. So that can reduce hypoglycemia, although you can see it's still, in this example, it's still gone below 70. So I like to let people know it's not gonna fully eliminate hypoglycemia. No system fully eliminates hypoglycemia. Now, control IQ. So what's really unique about this algorithm is how it is working off of the programmed manual settings. And so this is good. It's really good if you like adjusting settings and understand how it's working, because you have more ability to kind of make a difference with the algorithm. Like if you want it to be a little more aggressive, you can increase those basal rates. You can affect those correction factors. And so there's a little bit more that we can do to kind of to play around with it. Also, you can set multiple profiles too. So thinking about like if a person's sick during menstruation, sports, if insulin requirements are going to change a lot, you actually can set a different profile. And so like with that exercise, you can set a much like lower basal rates, which can help to prevent lows. Also, what's unique about this system are automated correction doses. So in addition to just adjusting that background insulin, up to once every hour is a correction dose is predicted to go over 180. And it corrects it down to 110. It is cautious in that it just gives 60% of that dose. And just so you know, it only does this within one hour of the last bolus. So let's say you give your own correction dose, then it's going to wait one hour from that time period. And another thing about this pump is that the insulin action time is set at five hours. That's part of the algorithm. So that is not something that you can change. Now, sometimes people are up in arms about this because with traditional insulin pumps for kids, we often set that insulin action time to three hours. However, what I want to make people aware of is it's not like a linear five hours. Most of the insulin action is used up by three hours. However, there's still a tail. There still is a little bit of insulin that is working through that five hours. So I don't look at this as a drawback, but just understand that that's how the algorithm, it's using that to calculate it. And then there are sleep and activity options. So where you can adjust those targets, you can have a more narrow target for sleep and a wider one which is less aggressive for exercise. And then with our other AID systems, you can't do extended boluses because the idea is that the system will kind of ramp up around it, and it just isn't an option. With this, you actually can. So you can do an extended bolus if maybe a meal is going to be eaten much slower, or you're not sure how much is going to be eaten. So you do have that option to extend it out over two hours. And then also there is now, there's the phone app, there's the T-Connect where you can remote bolus through there. So you don't have to take the pump out to bolus from it. As long as you're within like the Bluetooth range, you can actually use that to be able to bolus. In terms of this algorithm, so it's a little bit different instead of having just that one straight target of 110 or 150, it is a range. And so I love this chart because it describes it very well how this range works in terms of 112.5 to 160 for the usual settings. And then for that sleep activity, that tighter range of 112.5 to 120. And something to note is when it isn't sleep activity, you will not get automated correction doses, but you do have that more intensive range. So some people like to keep it in sleep all the time to have that tighter range, but just know then you're not going to get the automatic correction doses. And then that exercise has that higher range of 140 to 160. You do still get correction doses though with the exercise. And so for many, many kids, that 140 to 160 isn't quite enough. And so that's where either suspending completely or using alternate profiles can be really helpful. Okay, and then the Medtronic 770G. So this is the smart guard with auto mode using adaptive basal rates, also based on the total daily dose like we see with Omnipod 5, except it's mostly using the last two to six days for insight into that insulin delivery. The usual target is 120, but there is that temporary target for 150 for activity exercise. And then one unique thing about this pump is the fact that if you do want to go out of automated mode, you can still keep on the suspend before low or suspend on low options, which is pretty nice. With Tandem and Omnipod, you can't, those are just not options. Tandem has basal IQ, but once you have control IQ, you don't have the basal IQ features. This works with the Guardian 3 CGM. So one of the downsides is that unlike with the G6, which does not require finger sticks and calibrations, this does require really two to four per day and three to four tends to be the most optimal. And this one does run off of batteries. So you want to make sure you have extra batteries and the CareLink app is what's used for data sharing and for viewing. One really nice thing is all of these now have the Bluetooth options where data sharing, which is really nice for a clinic, isn't so much easier. And we're all kind of anxiously waiting for the 780G, which is available in many, many other countries, which has a more aggressive algorithm. The target glucose could be taken down to 100 and also Guardian 4 is their updated CGM, which no longer requires the calibrations. So, and this, just so you know, with this algorithm, really the only things that really affect the algorithm are the carb ratio and the insulin action time. And with this algorithm, it tends to be the shorter you make that active insulin time, like all the way down to two hours, it tends to be a bit more aggressive getting to those lower glucose values. Here's an example of what the 770G looks like. And here you'll notice this person spending a lot of time in hypoglycemia. And this is a good example of someone putting in basically fictitious carbs to get the system to run a little lower out of frustration, it's running a little bit higher. And so that's what's happening here. And a good kind of giveaway for that is, well, there's 403 grams of carbs entered, but no judgment. I mean, someone could be eating that much. But also the fact that even overnight, there's more glucose checks, more carbs kind of being put in. And usually when the system is running optimal, you don't expect that there's going to be so many lows. And also the bolus percentage is 75%. Okay, this is a great comparison. I'm not going to read this off, but it's another one of the Panther tools from the Barbara Davis website. So beautiful chart comparing all of the systems. So, because I know this is a lot of information to try to keep straight all of the time. So now I created these charts because I thought there's so many questions of like, okay, well, wait, what setting do you adjust? How do you optimize this pump? And so what I did was I compared the three AID options and then what settings will make a difference to affect that basal insulin delivery when it's an automated mode. For all of them, the glucose target, the way you set the targets a little different, different choices with OP5, you've got the all the way 110 to 150 in 10 point increments. With control IQ, it's really a matter of, are you choosing sleep or you're choosing the exercise or the usual 770G, it's that 120 or that 150, but all of them are going to impact them. Insulin action time. So only the 770G, this is the most important thing that will really impact the algorithm with the 770G. With control IQ, it's defaulted to five hours. So that's not going to matter. With OP5, it only matters with bolus doses. So if you're just talking about that basal, changing this is not going to impact it. So if you see, oh, it's going higher overnight, this is not going to change that, but it will for boluses. Correction factor. So only with control IQ is the correction factor going to matter with basal. And that's because it only matters a little, but it's kind of one of the levers in the algorithm where if it is more intensive, that can impact how rapidly the glucose changes. So it's kind of one of those factors that has a tiny impact in that basal. For all of them, the correction factor is going to matter much more for bolus with the exception of 770G, where it does not use the correction factor, that process is completely automated. Basal rates only control IQ, is it going to work off of those existing basal rates? And max basal. So sometimes people say, well, I set this max basal lower, or I'm going to set it higher, so it'll be more aggressive. I'm sorry, it matters none in the automated mode. Now let's talk about bolus doses. When you're bolusing, what settings will matter and what will make a difference? All of them carb ratio. So that's something we should always be looking at and we can optimize, in addition to paying attention if someone's pre-bolusing or bolusing late. Insulin action time. So with the 770G and Omnipod 5, it will impact how those boluses are given, but not with control IQ because of that default to five hours. Correction factor. OP5 and control IQ, yep, it will impact that. With the 770G, it will not. Not when it's in the automated mode, it will decide through its own algorithm how much to give. Max bolus. So unlike max basal, the max bolus actually will matter, and this is important because a lot of pumps, they'll default to like 30 units and that, wow, I mean, that could be a lot for a kid, right? So this is something where we can, it's basically a safety feature. Let's say someone accidentally puts in, they're having 150 grams of carbs and meant to just put in one five, 15. It's very, I mean, I wouldn't say it's so common, but most people like once or twice in their life, like they accidentally press through all the prompts and end up giving themselves that max 25 or 30 unit bolus, which can be so dangerous. So this is something we can adjust it. I usually look at, well, what's kind of the maximum amount the person usually may give and try to set it around there. I could set it at 10 units, eight units, it just kind of depends. The ability to override doses. So with OP5 and control IQ, it's an option, and this will show up on reports. You can see the percentage of times someone's overriding, and we should look at that. With the 770G, it's not possible to do that. And what I mean is you can't just like increase the dose and say, okay, it's recommending 2.1 units, I want to give three units. But what you can do with 770 is you can put in extra carbs. And so that tends to be what happens, although even with the other systems, sometimes that's how people go about it. They might just put in more carbs if they want a bigger dose. Extended bolus, only control IQ and reverse correction. So on OP5, you can turn this off. In control IQ, you can press X to not incorporate it into the bolus. 770G, it's just you don't kind of see what it's doing, so there's not any way to target it off. And then I've also included there what glucose target it uses for the correction dose. 770G is correcting down to 150, and control IQ is correcting down to 110. And then reasons you can get kicked out is important, because you want to be in it as much as possible to maximize the use. So with OP5, if there's, I mean, with all of them, if there's not CGM readings, but fortunately, like with 770, there's a safe basal. With OP5, there's a limited. So you're still getting some automation. It's just done in a safer way. OP5 and 770G do have maximum insulin delivery. And like with OP5, if you hit that, you actually, you'll get this advisory alert. You need to go out of automated mode for five minutes. Usually you verify your glucose, give a correction, and you can go back, right back in it. With 770G, this is improved from the original 670G. But if you are over 300 for one hour or over 250 for three hours, then you will get kicked out. And the best thing to do is give a correction and then try to come back in. And then 770G does require those calibrations, which could be another reason to get kicked out. These are the respective data platforms. So you guys can have a copy of this and viewing the slides. And then I am running out of time, but let me just quickly at least go through one case and then we can turn it over for Q&A. So this is a Omnipod 5, and this is kind of, I would say a typical, can be like a typical first week where the person is 54% in range, which is, or sorry, 58%, which is like fine. A lot of times these systems can really help us to get 70% or more, but that first week, it's really common to be running a little bit higher. And as the system's kind of adjusting and we need to make some adjustments to help kind of increase that time and range. So as I'm looking at this, I can see the person's actually is bolusing really, really frequently, 12 boluses per day, 40% are overrides. So that indicates, assuming the person, we can look deeper, but is taking more, that indicates there may be an issue with the current settings that it's not aggressive enough at giving enough insulin. I can see the number of carbs entered per day, the fact that it is an automated mode 100% of the time, and 62% is basal. So a little bit more lopsided toward basal compared to bolus. So how do we approach this? Well, I love looking at the day by days to see what's happening. And here's where you get that better idea of, okay, when are these boluses occurring and what's happening after the boluses? And so what's happening here is that it's going high. The boluses are just not quite effective enough. Also we can pay attention to what the glucose target is. Sometimes we'll start off with a higher target because just kind of being cautious about hypoglycemia. For many people, we can take that down to that 110, the most aggressive target. And so very common, we're going to intensify those carb ratios and try to counsel on using it to kind of, so we can see what's happening and how to make adjustments. And so in this case, the person actually improved to 77% time and range. That average came down to 145. And that was after just a couple of weeks. So there's still a bit of lows. There's still a little bit of work to do, but that improved a lot. And then I'm going to real quickly show you this one. So this is someone using control IQ, 70% in range, 5% low. So like doing okay, a little bit too much low, averaging 147, but the family says, well, we don't trust the settings. And often we're actually putting in much less carbs because we're so afraid about the lows. And I can see here 21% basal, the rest is correction, or sorry, the rest is bolus and control IQ corrections are actually 23%. So those are those automated corrections. And so let's kind of try to see what's going on here. So this is the CGM summary from the pump. And I can see that it seems to be rising a little bit more around breakfast. There's lows that are occurring at different times of the day, but especially overnight that 11% kind of going into the early morning. And so I take a look at the settings and these are the current settings. And what I like to do, especially when I have a new patient referred to me is like, go back to the basics, go back to those rule of 450, the rule of 1700 and say, okay, is this making sense? And when I did that, when I do the rule of 450 and I do that rule of 1700, I calculate a carb ratio of 30. This carb ratio is 18 or 23 and sensitivity factor. I calculate one 10 and this is 40 and 45. So what do you think is happening here? Well, this, these are the day by day reports. And so what is obviously what's happening is those auto corrections are contributing to lows because even though those auto corrections are cautious and they're only giving 60% of that calculated amount, they are still enough when it's, when the ISF is so far off, they're, they're dropping it. And so what you see is kind of this yo-yo, this up and down of going low, treating it then going higher and just happening. And so also she was going low into the night because she started off the night low and I was contributing to lows overnight. And so just kind of this pattern. And so what we did was we just took a step back. I didn't change all the settings to what I calculated, but I did. I was more aggressive in terms of, okay, let's change these carb ratios. Let's cut back on them so that you can feel more comfortable to use your carb ratios and let's dial back this correction factor. And so this is what I initially set it at and it took a few more visits, but I'm happy to see that we really, we improved this a lot because this was a very motivated, motivated family. And so 85% in range, average 138, and just, just much less of that rollercoaster all around. So, and then I do just want to say the future is looking so bright. So Islet Beta Bionics being able to just put in usual large or small meal is like very much on the horizon with weight-based startup dosing. Moby, which would be basically like a tandem that doesn't have all the tubing, plus they just bought this patch pump company. So who knows? And then Tidepool Loop, which will give people the freedom to essentially choose their algorithm, choose their CGM, choose their pump, and just have the utmost choice. Also interoperability is likely to expand, Dexcom G7 has the FDA approval, which would actually minimize, right now there's a two hour warmup. So there's always that two hours every 10 days that you don't have, you don't have your CGM readings. Well, they've cut that down to 30 minutes and also they've made it so that even the 30 minutes you can put on a new sensor, you don't even have to deal with those 30 minutes. So that's going to be great. And then also likely we'll see Libre 3 and maybe other CGMs having, being able to work with these systems in the future. So just some summary points, we've got lots of options. All of the options do require education and training and review of the data to optimize use. And while AID systems are becoming better and easier to use, safety and backup plans are essential. Here are some resources. And with that, I am happy to open it up to any questions at this time. Wow. I'd like to thank Dr. Isaacs for an amazing discussion and amazing presentation. And I'll just vamp for a couple of minutes so I can give you the opportunity to take a couple of breaths. I'll put my video back on so that everybody can see me, which may or may not be a good thing. But in any case, we have just a couple of minutes and I wanted to hit a couple of questions that came through the Q&A. One was for people using either the OP5 or the 770, how does that take over the basal rates? How do people adjust the basals using the closed-loop data to adjust the basals in open loop? Yeah, that's a great question. So what I do is I look at the total daily dose that they're getting. I like to look at like a two-week period and make sure you have the full days, that you're not like having half days. But I look at that and generally I'll divide it by 24 to come up with a basal rate. Because you are hoping they're going to be in automation most of the time, it's really not necessary to have like eight different basal rates. So usually we'll do one. If there is a big discrepancy where someone is just more likely to go low overnight or something, then maybe we'll have two. But usually we just divide by 24 and use that. Okay, thank you. And I know everybody is listening. We had a lot of requests for your slides and I just want to let you know that Dr. Isaacs was happy to share the slides. So I really want to thank you for that. I have a question, particularly since the, do you tend to use the extended bolus for CTRL-EQ for high-fat meals or not? And what do you do for high-fat meals in the other two pumps? So yeah, it is something that it's on the differential and we use. I don't use it for everyone. But sometimes we will use it, especially if there's certain foods that just really are causing that delayed rise. So that's a great option. For the other pumps, a strategy is to bolus a little bit upfront and then bolus a little bit at the end of the meal. So kind of splitting it up. I also use that strategy, like a lot of times people say, I just, I don't even know what I'm going to eat. I'm not sure. I don't know how many carbs I'm going to have. And I say, well, just, if you know, you're going to have at least 15 grams of carbs, 20 grams of carbs, just put those in 10 minutes ahead. And then whatever you end up eating, we can add that in, in the middle or at the end. Great. Thank you. One other question, are there any other calculations that you use besides the rules that you mentioned already, like 415 and 1700? Those are the ones that I use. I mean, you can always also come back to like the weight-based dosing, 0.5 to one unit per kg. But if there's insulin resistance or other things happening, those sometimes those go out the window. But those are the main ones I use. And I do want to give a word of caution, like those are just estimates, they're starting points. So you can use that calculation and someone may end up being far off, but I do like to have that in my head, especially if someone's new to me and I'm like, wait a second, they're on like five, like in that last case I showed you, so I do, but yeah, those are the ones that I use. Okay. Another question from the floor is, in exercise mode in the control IQ pump, does it correct down to, what does it correct down to? So it's still, I believe it still corrects down to 110. Let me just verify on the slide, but I think it still does correct down to 110. I have it right here. Yeah, the thing is it does suspend, it's more cautious with suspending that if it's predicted to hit, go below 80, it will suspend. But that's what I think, but we'll, we can double check. Or if anyone is on here from tandem, you can, you can let us know for sure. Okay. And another question from the floor, the reverse, reverse bolusing. Some people have talked about disabling that reverse bolusing. What are your thoughts on that? I'm a fan of doing that. So yeah, I like to turn that off because it tends to be with these systems that you, people do better when you get a little more of a stronger bolus and often we have to intensify those carb ratios. So that is something I've been doing for most patients now is turning off that reverse correction. Okay. That is good to know. All right. So last question, I think practical suggestions to getting downloads, if you can't do them in clinic. So fortunately now everything, all of the systems have the ability to be Bluetooth connected and with OP5, you can't even plug it in anymore. Like you have to get them connected. It is true. There sometimes are issues getting them connected at first. Once you get them connected, usually it works very well. So, but that becomes very important. So I mean, you've got T-Connect, you've got Gluco, you've got Carolink. And so I find if you do spend that time upfront and usually the company reps are so helpful with this. But if you get them connected, then you will have the data and it's just a matter of logging into that system. It really facilitates virtual visits because you can just log into that system. They don't have to send you anything. You don't have to plug anything in. Yeah, I definitely agree. Getting people to be getting that data in advance of the visit, so you're not wasting time during the visit. And also it really helps with the telehealth visits. All right. Well, I can see we've reached the end of the hour. I'm sure there's a lot more questions, which means that any good presentation generates a lot of questions. I want to thank you, Dr. Isaacs. I always learn something. I do this all the time, but I continue to learn things. And for those of you who are still on, you will get a copy of the presentation in your email and you can reach back to the ADA about getting copies of the slides, which we will work out after the talk. So any parting words of advice? No, I just want to thank you so much for having me. And I encourage you guys to use those resources because it's a lot to keep straight. So being able to refer to those things can help a lot. Okay, well, thank you all so much and wishing everybody a wonderful rest of your day.

automated insulin delivery

AID

children

Omnipod-5

T-Slim-X2

Medtronic 770G

insulin settings

education and training

blood sugar control

pre-bolusing