Hello, and welcome to today's workshop, Using Automated Insulin Delivery Devices in Diabetes Management. Before we begin, the American Diabetes Association would like to acknowledge Medtronic and its support to bring this workshop to you today. The speakers joining me will provide you with key insights to help you effectively incorporate automated insulin delivery devices into your practice. My name is Jane Roosh, and I will be the moderator of today's session. To share a little bit about myself, I am a diabetologist in practice and a professor at the University of Colorado, where I manage many elderly individuals using insulin pump therapy at the Rocky Mountain Regional VA. I am past president of the American Diabetes Association, and I have had the privilege to volunteer for the ADA in many ways, including being part of the standards of care for medical care for diabetes. But most importantly, I have the privilege of introducing you to today's speaker. Dr. Earl Hirsch is a board-certified physician at the Endocrine and Diabetes Care Center at the University of Washington School of Medicine and UW's Diabetes Treatment and Teaching Chair and the UW Professor of Metabolism, Endocrinology, and Nutrition. Dr. Hirsch's career has focused on studying the best strategies for the use of insulin therapy and innovations in technology for diabetes management. Caitlin O'Brien is a clinical pharmacy specialist and a certified diabetes care and education specialist at Boston Medical Center. She works in the outpatient endocrinology, supporting patients through their diabetes management and education. She is also the chair for the Massachusetts Diabetes Care and Education Specialists. And last but not least, Dr. Viral Shah is a professor of medicine in the Division of Endocrinology and Metabolism and the director of diabetes clinical research at the Center for Diabetes and Metabolic Diseases at the Indiana University School of Medicine. His research focuses on improving glycemic control and reducing complications with a special interest in fracture prevention in people with type 1 diabetes. Here are the relevant disclosures for today's presenters. And I also have served in an advisory role for Medtronic. Our learning objectives today are with the intention that when you have participated in this workshop, you will be able to recall the key components and functionalities of automated insulin delivery or AID devices. You will recognize how to address barriers to managing these AID devices, both for physicians, practices, and for the patients themselves. And you will be able to identify ways to implement AID devices in clinical settings. We are going to have three breakout sessions today. And prior to that, we will kick off with a short presentation from each of our speakers, after which we will split into three separate rooms. You will each be assigned to a room and each speaker will join your group to share their expertise and offer insights on techniques to either mitigate wear problems for AID devices, approaches to avoid hypoglycemia from exercise, and strategies to optimize the infrastructure for solutions for sharing data with your AID devices. Participation in the breakout groups is encouraged but not required. Now, before we dive into the introductory talks, I would like you to take a moment to share this infographic, the latest recommendations from the ADA on automated insulin delivery systems. And to use the QR code or the link in the chat box, which has just been delivered now, for you to access this useful infographic for your clinical practice. I will now turn it over to Dr. Earl Hirsch to kick off the presentation portion of the basics of insulin pump therapy. Dr. Hirsch. Thank you very much, Jane. And it is great to be here. Talking about this topic, which is very hot right now. But I think before we get into the specifics of AID, it's very important that we talk about the basics of insulin pump therapy. Let's just talk first about some of the nomenclature. When we talk about open loop, also known as manual mode. In open loop, the pump is infusing insulin. We use glucose monitoring, whether CGM or finger stick, but we don't have communication between the pump and the glucose sensor. And I remember going back to the early 1980s with our first pumps, maybe other people on the panel or in the audience remember that. But what's become so important now is the closed loop or the AID systems, where we use CGM, which now communicates with the pump to modulate the basal insulin needs between the meals. And all current day AID systems require bolus insulin, but can compensate if the bolus is missed or if the bolus is the wrong dose. Now, when we talk about basics and what can be changed from the pump settings with the open loop, first is the basal insulin, where we've been doing this for decades, where different requirements are used at different times of day. And when a pattern of an increased or decreased basal needs are seen, the basal dose needs to be altered two to four hours before that due to the slow kinetics of our quote unquote rapid acting insulin. And we didn't really understand that well until we had CGM to see. So for example, for a dawn phenomenon that may have started four or five in the morning, we would have to increase the basal at two or three in the morning. The insulin to carb ratio is the most common way we calculate mealtime insulin dose. We do that with manual and with the closed loop system. And the insulin sensitivity factor also called the correction factor. And that is how many milligrams per deciliter does one unit drop the blood glucose? And we use formulas to calculate that. What about open loop? What else can we change? Well, the insulin action time, that is insulin on board or IOB. And that's the number of hours the insulin is active after infusion. This number is used for calculations, but is not necessarily based on the actual kinetics of the insulin. And it took me really years to get my head around this. How can the insulin action time change based on what a provider or patient wants to change in the pump? It was originally described to me as an anti-stacking type of number that we use, but now we use it in the AID systems to improve the performance of the algorithms. And that's how we should think about it. The target glucose is where we start correction doses, and that can be changed in some of the pumps, but not all of them. But I want to emphasize it's only by reviewing the download can one determine which of these factors need to be adjusted. And it's surprising to me as an editorial comment, how infrequently these downloads, which are so helpful, are actually downloaded in the real world. Now, as far as the AID settings, there are four AIDs that are on the US market. Each has a separate algorithm and how to adjust and maximize the settings are different for each system. And at least my opinion is you really learn these the best by working with all of the systems because all of them work a little bit differently. I encourage the use of the pantherprogram.org. We use it here at the University of Washington. It's an excellent resource that compares and contrasts the different AID systems on the market because they all work a little bit differently. And you'll be hearing more about that in a moment from Caitlin. We have basal automation. That is the different algorithms for each system only programmed for tandem. So in the other systems that we have, the basal automations occur automatically. With tandem, the algorithm, you actually have to have a basal system already programmed. Again, it's one of the differences between the different systems. Bolus automation, where you have different algorithms giving automated bolus at different times and doses. Now, in some systems, you don't have automated boluses and some you do. Again, it has to do with the differences between these algorithms. The insulin to carb ratio, we use this with all of the systems, but not in the beta bionic islet where you put normal meal for me, small meal for me, or a large meal for me. So again, another difference. And then the correction factor, insulin action time, exercise mode, extended bolus, they all work differently with each algorithm. Now, it seems like this is a lot to understand since we're dealing with four of them. But like I said, when you work with them individually with patients, you learn them all relatively quickly. So again, the detailed comparison of the four AID systems can be seen on pantherprogram.org. And also you'll hear more about this from Dr. Caitlin O'Brien. Thank you, Dr. Hirsch. And it is my pleasure to be here today representing the diabetes care and education specialists in our role as the tech champion on the diabetes care team. So I will be doing an overview on the devices. And as Earl mentioned, what is comprised in the AID is the continuous glucose monitor that communicates with the Bluetooth enabled device holding the algorithm and that adjusts insulin delivery based on CGM values and trends and the subcutaneous insulin pump holding the rapid acting insulin. And I'd like to remind the audience of the current ADA standards of care that do recommend level A evidence. So data from randomized control trials that recommends AID systems as the preferred insulin delivery method to improve glycemic outcomes and reduce hypoglycemia and youths and adults with type one diabetes and insulin pump therapy, preferably with CGM, should be offered as a diabetes management tool to youth and adults on multiple daily injections with type two diabetes. And ultimately, the choice of the system should be based on individual circumstances, preferences and needs. And so things that we should be considering as clinicians and educators when selecting a pump should be what we see here on the slide. And if you use the QR code and in the chat will be a link to this document as well from the ADA on practical considerations for AID. Patients really care about the device size, whether it can be seen when they're wearing clothing, can they clip it on their pants or what if they're not wearing an outfit with pockets or the choice of tubing versus no tubing can be really influential in the decision of which pump a patient would like. The ability to disconnect can be great. Other people don't want to have tubing and that thought about where are they going to put the pump? And so those things can be really important. Also, CGM choice on the following slide will go over which CGMs are compatible with the AID. So if a user has a preference on CGM that might guide the pump choice if they're using CGM before pump initiation and whether or not there's the ability to control our bolus from a smartphone, which can be often more discreet or for me and my population, I'm at a health safety hospital. Not everybody has a smartphone, so it can be really helpful to know which devices can be used without a smartphone. Ultimately, insurance coverage can be a determining factor and other things, including what language the device might be in or if patients with diabetes are currently carb counting or doing meal announcements and if there are pumps that can better suit them if they're having trouble announcing meals. This image has been shown previously and you'll see it throughout the slides, but I did want the devices to be shown to kind of visualize size and you can see the Omnipod 5 kind of next to your typical smartphone, so that gives you kind of a frame of reference. The other pumps are about the same size with the exception of the Mobi being pretty close to the Omnipod 5 in terms of size. And this is adapted from the Panther device comparison chart and so on this slide, I'll point out some key differences and the slides were due to be submitted several weeks ago, so as tech is ever evolving, there are updates to be made to this slide and I will point those out as we come across them. So the Islet beta bionic pancreas is approved in type 1 diabetes. The unique part of this pump is that it doesn't require settings, it only needs a weight to initiate the algorithm and the user doesn't need to carb count, they would just simply announce their meals as their usual meal, less than their usual or more than their usual and the basal will adapt every five minutes based on the CGM readings. It is compatible with the iPhone and Android apps and it has three targets that can be modified 110, 120 and 130. The Medtronic 780G is compatible with the Medtronic CGM. I think, you know, the older CGM, the Guardian 3 required finger sticks, so may not have been, you know, something that was, you know, compared to the other pumps which don't require finger sticks, their factory calibrated CGMs may not have been a pro in some patients' eyes, but things, you know, with the Medtronic 4 sensor that has gotten better, no finger sticks are required, but there are some complaints of sensor adhesion or lasting the full seven days. It's also important to note the, you know, how much insulin volume the pump will hold, especially as we're starting to see pump use and type 2 diabetes, which tend to require higher doses of insulin each day. The Tandem, T-Slim and Mobi run on the ControlIQ algorithm, which just recently announced a newer algorithm, ControlIQ Plus, and the Tandem also recently announced FDA approval in type 2 diabetes, age 18 and older. And the target for Tandem is 110 and that's not to be, that's not something that can be adjusted. So pump settings would be changed based on insulin to carb ratio and the correction factor. And the pump for the Tandem T-Slim is a touchscreen, so the user can bolus right from the pump, whereas the Mobi has a small button to bolus from the pump, but it is run on the iPhone and Android apps only. So this would not be a pump to use if a patient does not have a smartphone. The Omnipod 5 is currently the only tubeless pump. It is approved in type 1 diabetes and type 2 diabetes and is compatible with the G6 and G7 sensors as well as the Libre 2 Plus. And it is programmable in English and as well as Spanish. It has five targets that can be adjusted, 110, 120, 130, 140 and 150. And if the user doesn't have a smartphone, there is a PDM controller that would be an option. And that PDM controller actually will also activate the Libre 2 Plus sensor. And so this is a great option for someone who doesn't have access to a smartphone. Otherwise, it is available on the iOS and Android phone systems. And there is a newer pump that is approved. It should be coming to market sometime between April or June of this year. The Sequel Medtwist, it's for type 1 diabetes, age 6 and older. The compatible CGM was just announced last week. While it was studied with the G6, it's going to launch with the Libre 3 Plus, which is a factory calibrated sensor. It holds up to 300 units. It will only start with the iPhone. So that's something to consider. It will eventually get to the Android system, as well as the Dexcom CGM. And the unique feature of this pump is that it actually has lower targets, which is great in pregnancy or individuals looking for tighter control. And with that, I will turn it over to Dr. Shah to go over AID interpretation. All right. Thank you so much. And it's a perfect segue. Dr. Earl Hirsch spoke about the basics of insulin pump. Kathleen spoke about how you are going to select a pump for your patient. Again, it's a patient's choice and one size does not fit all. But then once they start using the AID system, the key part is that how you are going to download, review and optimize the settings so that everybody gets desirable or recommended glycemic targets, which is time in range of more than 70% for most individuals. So I will take about five minutes and briefly try to highlight how to do that in your clinic in about five minutes or less because we understand that we all are busy, we get about 15 minutes per patient. So you have seen this slide probably four times so far and you may be seeing that because I think this Panther program it's a good resource and if you have not visited this website I would encourage you to do that. But also at the same time remember that each system has its own downloading platform so what I mean by that the data houses into that kind of their proprietary software where you will be logging in and viewing the data and each one looks very different at least in the CGM space we have now ambulatory glucose monitoring profile which is standardized so pretty much all four CGMs that are available in the market they do have an AGP kind of a profiling to their software platform however that's not the case for AID so it's a little bit sometimes feels like an overwhelming information but I will try to make it short and see how can you get that max within that kind of a five minutes. So again the first step is that I would encourage you to to probably you know log in into the each of this software platform and try to familiarize yourself. So ILED Bionic uses the Bionic report platform Medtronic 780G called CareLink. Tandem used to be Tandem Connect and now it's T Source and the Omnipot 5 is currently Gluco but they have their own platform that has been launched now to my knowledge about 20 plus sites in the U.S. called Omnipot Discover so eventually you're going to see Omnipot Discover in a year pretty much everywhere. So these are all different platforms where the data is housed. Now this is just the snapshot that how it looks like so on my left side that's a Tandem Source in the middle it's a Gluco on my right that's Medtronic but what I want you to really do is that try to focus more on which I call it's a summary page or dashboard in a Tandem it's called overview so you will see hopefully you can see my cursor so in a Tandem it's an overview in a Gluco it's called summary and the Medtronic called dashboard if you just focus on that page you're going to get about 90 percent of the information that you need to optimize the AID system for your patient and then the rest 10 to 20 percent of your time you can focus on trying to dig deeper. Now remember that part I'm trying to simplify that here each patient with type 1 or type 2 diabetes using the AID system unique and each one may have some kind of a different challenges so take that with a grain of salt I'm just trying to make it very general very simple but step number one familiarize yourself with this different platform. Step number two is that again try to spend more time on that overview summary or dashboard kind of report. Then this is my kind of a abbreviated four-step method for AID interpretation so once you're in that summary page or overview page the first thing you want to look at is that is someone using an AID system for more than 90 percent of the time if the answer is yes good in that case you might see a good outcomes if not it's giving us an opportunity to ask patients what's going on here there may be an issue with the cgm connections it's something that the pump is kicking out out of the auto mode because of various reasons or maybe something else but it's most of the time it's a good opportunity to talk and try to solve the problem or sometimes people don't like to hear two different devices and if they are not using cgm then the AID system is not going to be you know helpful that much either. After that you're going to look at the time in range and time below range for most my adult patients I would focus on a time in range first the goal is more than 70 percent of the time but remember that that should be achieved with the safety which is time below range and the time below range of less than 70 milligrams per deciliter should be less than four percent of that which we call it's a clinically meaningful time below range of less than 54 or level two hypoglycemia should be less than one percent. Level two is more clinically relevant so I try to focus more on that time below range of a level two or less than 54 unless an older individuals with long-standing diabetes where I would probably focus on both level one and level two. Level two we don't want to see that any percentage in older individuals to be you know making them safe. Then again if the time in range is not up to the goal or too much of time below range then try to think about where is the problem meaning by that try to identify an issue. You can just look at the glycemic pattern again on that summary page or dashboard or overview page and try to just look up where you see ups and downs and then talk to patient. Hey I see there's kind of a bump after lunch or after dinner or the overnight or I see a lot of low blood sugars are happening between 3 p.m. and 4 p.m. and do you know how what's happening here and then patient will tell you that hey I exercise around 3 p.m. and then you know I'm having those low blood sugars after my exercise at about 3 or say 4 p.m. or sometimes people say hey I like ice cream I have a small ball of an ice cream or a scoop of the ice cream before going to bed time and then you see that kind of a rise in glucose overnight right. So most of the time patient will tell you and it's an easy fix. If not you can easily look at those glycemic pattern and try to see how you can optimize that. You want to keep that blood sugar in a narrow and flat range as much as possible right. And also Dr. Hirsch mentioned, Caitlyn mentioned that each system works so differently that it's overwhelming. So what I suggest you to do is that don't worry about anything just try to spend your time focus on the things that matters the most. And what I mean by that this is the most important slide that you want to memorize if possible. What matters the most in each of these four systems? So let's say for example islet. It has three targets. Usual, lower, and higher. The lower target means it's a more aggressive system. Higher target means it's a conservative system. So that's the that's the main thing in an islet. And also the bolus where you're going to have a usual, more than usual, or less than usual okay. But everybody has to bolus in pretty much all the EID system that's why it's called hybrid closed-loop system. They are not fully automated yet right. In a Medtronic 780G system only two things that just try to focus on. Target glucose of 100 and active insulin time of two hours. Once you do that part the only thing you can really modify is insulin to carb ratio if you see those postprandial curves. Otherwise in 90% of the problem fixed when you have a target of 100 and AIT of 2. AIT stands for active insulin time. In tandem control IQ pretty much everything matters that includes the basal and those bolus components. But sleep mode is the lower target. So make sure that most people would have a sleep mode. Generally I prefer to do it about one or two hours after they actually sleep so that it gives an opportunity for one to two hours to have an auto correction before patient goes to bed. If they are not in a sleep mode during the daytime or sometimes some people don't prefer to have a sleep mode because it's a slightly aggressive at nighttime then the best setting to really have a more auto correction is the correction factor. Lower the correction it's going to give you a larger amount of an auto correction every one hour as needed right depending on the target whether the patient's within the target or not. And the Omnipot 5 110 of target and tightening up the insulin to carb ratio that's what it matters. The basal rate doesn't matter. Active insulin time may matter but it's only for the manual not for the automated mode. So in each system I highlighted that by those red two points that if you can just memorize those two points for each system I think you can solve the problems for about you know 70-80% of the time in most AIT system. So now let's take an example. Here is a case study. This is a real case on tandem control IQ. I'm going to just go by step by step here that I described that to you. So step number one is this individual using an AID for more than 90% of the time and I showed you here in this red circle that the time active it says 96% meaning by that this person is using AID system pretty much every day. That's good. Thumbs up. Now step number two. What about the time in range and time below range goal? Sounds like a time in range 56% for the current week. The previous week was pretty good 75% but current week is 56% so suboptimal. We would like that to be more than 70% and good thing is that the time below range is not that high. It generally should be less than 4% for less than 70 and less than 1% for less than 54. So patient is within the goal for time below range but time in range definitely room for improvement. Now in that case I'm going to look at the step number three. Hey where is the problem? So I'm going to just look at that curve and it looks like that there is a blood sugar that goes up after 2 p.m. and stays there until about 6 or 7 p.m. So that's a four to five hours. It's probably a window of an opportunity. I'm going to ask that individual what's happening here. Now you have probably two things. Number one if the patient is missing boluses then probably you can try to encourage the patient to remember the bolus prior to their meals or if they are doing it then you can probably change the insulin to carb ratio or correction factor. Now remember the correction factor will work for auto correction in control IQ right. So that's how I would do that part in simple four step and if you try to do that multiple times in your clinic then you will kind of become a champion in AID interpretation. With that I think I will end my presentation here and we will have a fun in a breakout. Okay we are going to talk about avoiding hypoglycemia. This is a huge issue for so many people and I'm glad I was given the opportunity to talk about this. It's a big deal. Just some general concepts that there are impairments in skeletal muscle and cardiorespiratory fitness in type 1 diabetes in general. We've learned from the inhaled insulin studies that there also is a greater reduction in pulmonary function in type 1 diabetes than there is in age match controls. That is something that we did not know before that will impact exercise but in general endurance exercise activities reduce glycemia. Okay let's think about this endurance reduces glucose whereas explosive activities like a sprint you get a counter regulatory response you actually get an increase in glucose with sprinting. High intensity interval training called HIT and resistance training it's somewhere in between these two extremes. So let's just talk about HIT for a moment. You can see what this um these HIT activities are. This might result in intermediate hyperglycemia especially if it's before breakfast and there isn't much insulin on board and then rebound hypoglycemia later particularly in an AID system where insulin corrections are given. HIT is associated with increased lactate production which has been implicated in the blunting of glucose counter regulatory hormones and I'll just mention that last week at the ATTD conference in Amsterdam there are companies working on continuous ketone and continuous lactate particularly for exercise. Now these are the general trends as I mentioned exercise which everybody thinks about endurance exercise you tend to get hypoglycemia, explosive exercise, sprinting you tend to get hyperglycemia whereas with HIT and with resistance training it is sort of hit or miss depending on the individual and I have to point out everybody is a little bit different. Part of this is also related to what kind of shape they are in and we see this all the time. The people who are weekend athletes tend to have more problem with with hypoglycemia than the people who are regular athletes. Now there are all of these factors and I'm not going to go through them all they are all important but there is one that is more important than the others and that's the fourth bullet on the left the insulin on board. When you have more insulin on board you are going to have a greater problem with hypoglycemia and exercise and no matter which pump you are using we are going to try to mitigate this as much as we can. Now when we think about this insulin on board yes it's true if you measure the insulin in the blood it's both basal and prandial but in somebody on a pump we are generally only talking about prandial insulin and that's what's being measured. The number is a best general estimate for many reasons but mostly due to the fact that we alter this number to impact insulin dosing algorithm. So let's think about this it's the same in everybody and in the same person it's going to be the same every time yet that same person could be on a Medtronic pump and the insulin action time is set at two hours or they could be on a tandem pump where it is set at five hours. Just realize that we're using these numbers as a way to get the best out of that particular algorithm. It has nothing to do with real life. With real life when we talk about this we are talking about not bolusing for one to three hours prior to the exercise because you'll have more insulin on board and greater risk for exercise induced hypoglycemia. So that's the reason why the people who do the best with this they're exercising when active insulin is low the best time of the day is before breakfast but this could also increase the risk of hyperglycemia with sprinting with hit or reduced exercise performance. Now this is a lot of information but this is this is why this is such a complicated situation. Some general principles with AID systems everybody is different it has to be personalized and it's all based on previous experience. As a rule of thumb you want to avoid large carbohydrate meals because that's going to give you greater amounts of insulin on board prior to the exercise for the best practice especially for endurance activities small quantities of carbohydrate during the exercise and you know when you're watching on tv the uh the marathons this is what people without diabetes are doing also it's just the best way to keep to keep the energy going keep the fuel coming in. Pumped discontinuation this is something I was really against for a long time but what I have personally realized with some patients this is going to be the only way to prevent exercise to prevent hypoglycemia with the endurance exercise. You just have to be very careful no more than an hour a lot of trial and error. I think when we get the glucose ketone sensors this is going to be really interesting to see what happens. I would never discontinue a pump on somebody who's using an sglt2 inhibitor off label. I think that is probably too dangerous but again trial and error general suggestions for each device may not work for everybody so we're going to go through these devices but realize this may not work for everybody. Now Medtronic 780g and this is a general recommendation for everybody reducing the meal bolus that occurs one to three hours before activity but ideally you you're not bolusing at all with that for food. With that being said you want a higher target because you are still going to get additional insulin when those blood sugars go up high but you want to make these temporary targets one to two hours earlier than the exercise to bring down that active insulin and that's going to make spontaneous exercise difficult for all of these. Consider leaving this temporary target on for one to two hours after the exercise. For the control iq it's a little bit different. Yes we want to consider reducing the meal bolus that occurs one to three hours before activity but what I would do with this pump is a couple of things. First of all the correction factor is so much more powerful in this pump than the others. You may want to consider making the correction factor more conservative that is increasing that number from 40 to 50 or from 40 to 60 whatever it is. There is an exercise mode on this and again you want to consider turning it on one to two hours prior to exercise and leaving it on for several hours if you're worried about hypoglycemia. But having said that, and although I have not seen any clinical trials on that, what we have done in our clinic very successfully, and it seems paradoxical, putting them on sleep mode, which actually gives a more aggressive target, a lower target, but in sleep mode, they don't get these automated boluses, which increases the risk of hypoglycemia. And we have found this to be quite helpful while in combination with that, making the correction factor more conservative. And for this pump, that has worked for us quite well. Omnipod 5, again, consider reducing the meal bolus. If you have to eat one to three hours before the meals, there is an activity feature on the Omnipod 5 and turning it on one to two hours again before starting and leaving it on after the exercise if delayed hypoglycemia is a concern. And then finally, the Islet from Beta Bionics. What we find out is that most patients, they don't do anything extra. The insulin suspension is usually adequate. If you do get this fall rapid alarm, which is less than a 100 with two arrows going down, you don't wanna treat with big, you know, 40, 50, 60 grams of carbohydrate. You don't need that. Because when you give the big doses of carbohydrates, you're gonna get a big rebound. You're gonna get extra insulin given with this pump, more so, I think, from this algorithm than the others. And you may get more risk with hypoglycemia. When you do this with small amounts of carbohydrate, these patients do extremely well. So you don't wanna over-treat hypoglycemia ever with this pump, especially with exercise. You don't wanna give extra carbs in advance. That, again, will increase the insulin on board. That's the last thing you wanna do. And with this pump, you can always suspend it. And my session is on the infrastructure solutions for sharing data, which can be a challenge, no matter what type of setting you're in, or even beyond AID if you're not quite using AID yet. This can be a barrier for CGM and connected insulin pens, glucometer data. And so to start with some active participation, if you're willing to put in the chat what your biggest barrier is in your clinic or practice for obtaining data for people with diabetes. And I will kind of keep an eye on the chat and summarize some themes. And I'm happy to share themes for my other two breakouts as well if people have limited experience. This one came up quite a bit. No integration with Epic or other EMRs. I know Epic is a pretty common EMR. Patients forgetting, patients not having smartphones, not able to connect remotely their CGM or device. Lack of participation with connected devices. Yes, universal practice login. Another one that came up is several logins, right? So there's not just one agnostic place for all of this to live. There's different websites that you use for different products. And you have to remember your username and password and they don't stay logged in for very long. And so if you have a day where you have 10 or 15 patients scheduled, everyone might be using a different device. And so you may have to log into several things and have several tabs open, which can be really challenging. Internet or download issues in clinic. So these are all great examples. And so hopefully I can kind of walk us through in the next 10 minutes or so, some things to think about if you haven't started using AID or even CGM, like what to think about to get you there. And then I'll share some tips and tricks and workflows. And if people have solutions, I don't have all the solutions and I have some of the same headaches that you have all commented on in the chat. And so I think for me as a CDCES or diabetes care and education specialist, I think often about the ICC framework. And so that outlines a framework for optimizing tech enabled diabetes care and education. And it starts with identifying the right technology for the right person at the right time. And that can be so important because the best technology is the one that the user is willing to use and interact with and knowing what they have access to at home, internet, a smartphone, a computer, a smartwatch, no smartphone at all. And so then working with them to configure these devices with their user preferences and treatment plan and providing ongoing support. And that support could be medication optimization or it could be sometimes being tech support. Even though these companies have tech support, they often trust their care team and are where the first ones that they're calling when they can't log into their app or their alarms are going off or they've lost connection with their CGM. And then it's important to finally collaborate and have those data-driven conversations to have shared decision-making and integrate the care team. But we need to get that data. And so I often think about diabetes as a team sport and how do we get there? And so this is more of just how do we identify patients who would benefit from AAD? Who's doing that? Could that be population health? Could that be us looking at, running a report of patients on MDI, not meeting an A1C goal or patients on a CGM with high amounts of time below range because we know AID can reduce hypoglycemia or looking at the CGM for low time and range and kind of optimizing their insulin delivery. And then working with the team to who's the prescriber, who's helping with prior authorizations and paperwork for medical necessity, where are these orders going to the pump company, DME, pharmacy and how are we getting them covered and who's doing the education? Often people are limited in their time and staff and so is it the clinics team, the CDCS that's going to do patient education or are we outsourcing to the pump companies? And then once we kind of figure all that out, we get to, they're using the device, how do we get the data? And then we can ultimately interpret it like Dr. Shah went through, but getting that data can be the hardest piece. And so some things that have come up and so my role as the diabetes care and education specialist did start as a CGM champion in primary care. And I was able to work with IT about a year and a half ago to, and we use Epic to create a flow sheet. And while this is still a manual data entry of the CGM metrics or the pump metrics that you're using the CGM data, we can enter all of the metrics that are on the pump report or the CGM report and they become discrete data points in Epic. So we can trend time and range over time. We can trend the GMI. We can look at time below range as we're optimizing insulin or pump settings. They become that discrete data field rather than putting a snippet or a screenshot of the report. And that can sometimes get blurry when you're looking back at notes and you're like, is that a six or is that an A, 67%, 87%. And so that's something I've worked on. I've trained the medical assistants on how to use this flow sheet. So when they're downloading patient's data, they can enter it in for the provider and it ends up in their note. It also saves on paper, right? We're always printing these CGM reports and putting them in the exam room, which can be great to look at with a patient, especially if that busy provider might be logged out of source or gluco or doesn't even have it pulled up or forgot their password. So the printed version can be nice, but if we can get this data in the EMR through integration from these platforms, that would be amazing. And I know it's being worked on. So that's one role of me as the CDCS is optimizing this technology, getting the data into Epic, building these workflows, educating medical assistant and support staff and kind of identifying and overcoming these barriers, whether it's IT or time, using billing and reimbursement as something to enhance our practice to allow for more FDEs and get more tech in the hands of the patients. And so this is the main slide here for each device and what either app they need or what clinic account you need. The clinic accounts are free, so there shouldn't be any financial barriers, but there are those barriers of remembering your login, remembering to invite new staff when new staff are hired and making sure they have the access to all the different devices, making sure those appropriate either drivers or installers and software are on the computers and getting IT to approve those where it's going to be uploaded, whether that's with, in our clinic workflow, we have all the CDCSs use the same exam room every day. So we're not worried about where's our cord? Where's this cord? I'm assigned to a different room. That room doesn't have the appropriate software. And so our team is built so that those people get to sit in those same rooms that have the software, have the cords. And in the main areas where that medical assistant, their little kind of huddle area, those computers have all the cords and all the software installed as well. So there's no question of like, where are the cords? What computer can I use? They're the same ones. A provider may have to jump between exam rooms, but they would just be ultimately logging into CareLink or Source or Gluco, Tidepool, Beta Bionics, if they want to visualize the report on the screen with the patient or take screenshots. There is data sharing that can happen with family members, friends, partners, school nurses. And so that is usually the CGM data, whether it's the Dexcom Follow app or the Libre Link up. Bionic Circle is an app for the islet user that can share the data from the pump. It is limited data, but the data that's shared can be the CGM alerts, meal announcements, and things of that nature. For other questions that came up in other groups is, if somebody doesn't have a, they're not connected through the cloud and their pump needs to be manually uploaded and they don't live near the clinic and want to do virtual visits. I learned this recently, actually about a month or so ago from our tandem rep. So the patient gets the cord for the T-SLIM. They can go into a local library and plug it in. It is, they log into their account, they plug it in. There's no software or drivers that need to be installed in the library. Computer, those are all cloud-based. And so for those that don't maybe have access to a computer, a laptop, internet at home, libraries are free to use and the computers are accessible to the public. Other things that I've done in our clinic too, is kind of like remind patients for teaching appointments, please come with your smartphone fully charged. I also have them save the numbers, the phone numbers to their company's 24 hour support or tech support. I have smart phrases in Epic with the phone numbers as well. So that way I can put them in the after visit summary so they can save them in their phone or give them to family members as well. All right. Hello, hello everyone. Sorry, it's now jumping. We are jumping from a different groups to different group. Hope you had a wonderful webinar. Again, I'm Dr. Shah. And I am assigned with this breakout topic of skin integrity and skin issues with this devices. And I think it's very, very important and very clinical practical issue that if you are in a practice being patient, that happens with pretty much every single patient, right? So what I'm gonna do today is, I will spend about a few minutes on sharing my screen and giving you resources that may be helpful for your clinic to keep in and then give those to the patients. The first one, I'm gonna keep here for about a 30 seconds, 40 seconds. So feel free to take your mobile phone, open up the camera and the scan this QR code. This is one of the most or the best written article about the skin integrity with the diabetes devices. And this is about five page, four to five page article, but it lists every single kind of resources or things available in the market that could be helpful to your patients with different skin issues, whether it's not able to retain a particular sensor or device, or some people are having difficulty in removing it, or some people are allergic to one kind of either substances in the adhesive or sensor or something else. So really highly recommended article here. The next slide is that each industry do have their own resources and maybe save that on your desktop as a favorite item and toolbar. That's what I do in my clinic computer is that I have my own personalized like, you know, toolbars and then you just click on that part and give that print out to the patient. So here is the Abirtexcom and the Medtronic. I'm sure if you Google and find it, there may be for others as well, pretty similar. Well, welcome back everyone. And I hope that you enjoyed the breakout rooms as much as I did, and that you had the opportunity to ask some questions to the speakers, allowing for more in-depth understanding of common challenges that are faced regarding the use of automated insulin delivery systems. I'm gonna ask each of the presenters to take a few minutes to recap the highlights and the themes from their sessions. And let's start with Dr. O'Brien, Caitlin. Thank you so much. The breakout rooms were very interactive for the infrastructure portion of getting data downloaded. And I think a lot of us share the same barriers of time, multiple logins, patients without access to technology. And so it was really great to hear from others. And I shared some experiences of training medical assistance staff, building flow sheets for data integration while still manual does help to have discrete data fields in our Epic or EMR to trend time and range or GMI and TBR over time as someone goes from MDI to a pump. And so I was able to share some of my experiences and so there was great participation from the audience around tech disparities and health literacy. Wonderful, thank you. Viral. Yeah, it was a really great and I see in a chat where I think people are mentioning that the breakouts were awesome, useful. I think we shared some of the resources that are available in a public domain that can be used in a clinic with the skin issues. And I think the discussion overall is that, and we know this part very well that the skin issues in people with diabetes with the device is universal. Everyone faces that in their clinic, right? So those resources, I hope that would be useful. And a lot of people are already using those resources. It's really good to know how people do. And yeah, it was a great learning from everyone. Wonderful, and Earl. Mute, thank you. I had so much fun. We talked about different types of exercise and how different types of exercises impact glucose. We talked about all of the things that impact glucose changes during exercise. The most important one we talked about in more detail that is the insulin action time. And what that insulin action time actually is in real life compared to what it is on the pump and how to use that not to exercise when that insulin is peaking is basically what we talked about. And then we talked about strategies with each different pump to minimize hypoglycemia, including if need be turning the pump off for no more than an hour, talking about the fact that it's gonna be very interesting to see what happens with ketones in particular when we have the continuous glucose ketone sensor and how much we're gonna learn. And the other thing I just wanna point out is that as the sessions went on, but I just had a great time and I appreciate the opportunity to talk to everybody. Well, thank you very much to our expert panelists. It was incredible. And we're gonna move next to our slide again in case you missed it at the entry, the ADA infographic, the latest infographic on automated insulin delivery systems. So if anybody missed this, we're gonna give you a few more seconds in order to download the, to take a picture of the URL. And there's also a link in the chat box, in the chat. I wanna thank everyone so much for these excellent and informative sessions. Caitlin, Earl, Viral, your insights are appreciated and very informative for all of us as we incorporate these just really phenomenal automated insulin delivery devices into our practice. I wanna thank each of the attendees for joining us and following the workshop, you will receive an email from us to give us a brief survey to tell us how you liked the breakout rooms, what we could do better, other topics that we might be able to cover. And I wanna thank Medtronic once again for making this possible. And that concludes our program. Thank you so much for joining.

Automated Insulin Delivery

diabetes care

insulin therapy

continuous glucose monitoring

glycemic control

device connectivity

hypoglycemia prevention

personalized care

diabetes management