Welcome to the ADA quarterly webinar. This is one of the Diabetes Technology Interest Group webinars, Easing the Transition to Automated Insulin Delivery for Non-Traditional Users. I'm Laia Eklaspour. I'm a Pediatric Endocrinologist Assistant Professor at University of California in San Francisco. Thank you, Kerry and Molly, for agreeing to present at this webinar. Kerry Burgett, a Registered Nurse at Barbara Davis Center for Diabetes and Dr. Molly Tannenbaum, Clinical Assistant Professor at Stanford University. Before we get started, I would like to announce an upcoming event in March. So this webinar hosted by American Diabetes Association, it's one of the hands-on webinars, Tips to Improve Diabetes Care webinar series. And you can sign up for this event using the link in the chat. For today's webinar, we would like to hear from you. We encourage you to use the Q&A box in the menu bar to ask questions from the panelists, and we will address your questions at the end of this presentation. And if you would like to comment or discuss the webinar, please use the chat feature located in the menu bar. I'm pleased to welcome our panelists for today's webinar. First presentation will be from Kerry. Kerry Burgett is a Registered Nurse and a Certified Diabetes Care and Education Specialist from the Barbara Davis Center for Diabetes. She manages the diabetes technology clinical trials at this center conducting clinical trials with use using advancing diabetes technologies, including continuous glucose monitoring and automated insulin delivery systems, and also is an AID trainer for the pediatric clinic. She's also the director of the Panther Program, creating free practical resources for clinicians related to the advanced technologies. And our second presenter is Dr. Tanenbaum is a Licensed Clinical Psychologist and Researcher. She's a Clinical Assistant Professor in the Department of Medicine, Division of Endocrinology, Gerontology and Metabolism, and by courtesy in the Department of Pediatrics, Division of Endocrinology and Diabetes at Stanford University School of Medicine. Her recent work has focused on developing onboarding support for adults with type 1 diabetes or starting to use new diabetes technology. Okay, so Kerry, the stage is yours. Okay, okay, great. Thank you, Laia. And thanks to ADA for inviting me to present today. I am very excited about this topic and hopefully we'll have some good discussion and things to think about that relate to all of our practices. Here are all my disclosures. I do various speaking and consulting with different industry partners. And in addition to some grant support from ADCES for the Panther Program, I also have received some grant support from Insulet in Tandem for the resources that we create. Okay, so for my portion of the webinar today, I plan to first discuss, kind of review the evidence that's out there about who can benefit from automated insulin delivery, and then provide some examples of some alternative bolus strategies we can consider when we are working with people using AID. So first, this here is a list of all of the currently available automated insulin delivery systems that we have in the United States. There has been a lot of activity that's gone on just in the last few years. I've highlighted kind of the approval dates, initial approval dates for each of these systems. And you can see that in just the last 18 months, we've had three new AID systems come to the market. We've had Control-IQ since 2019. We also see that we've had all of these devices approved in various age groups over the years as well. So lots of new devices for us to keep up on and lots of great options available for people with diabetes. And a lot of the data that we usually see related to outcomes with these devices are from the Pivotal trials. So the Pivotal trials are the main clinical trial that's conducted with the data that's gonna be sent to the FDA, you know, to review and give approval for use in clinical practice. This table here just summarizes the outcomes mainly for time and range in A1C across the different AID systems that we have. And the point of this table is not really a compare and contrast. You know, you can't compare the outcome from one study to another, different study designs, different populations. However, what's striking is how similar all the results are. So what you tend to see is after anywhere from three months to six months using these devices in the Pivotal trials, for adults, you tend to see an aggregate of about 70 to 75% time and range for children. And it's a little bit lower in the high 60s. And you also see very low levels of time below range as well. A1C change you tend to see is anywhere from, you know, a 0.3% drop to a 0.6% drop. And really the main takeaway here is that this is all in aggregate. So if you look at this data, what you also can see is that the baseline A1C for participants in these trials tends to be pretty low overall. So anywhere from 7.2 to 7.5%. And then the ILADD had a lot more participants with higher baseline A1Cs than we typically see with the average baseline of 7.9%. So I think what this tends to do is it tends to then make us question, well, how generalizable is this data to the broader population? Because we know that most people with diabetes do not have an average A1C in the low sevens. And so how do we take this data and understand what that means for the broader population? And the good news is, is that there actually has been some analyses looking at this. So here we see a control IQ study that looked at glycemic outcomes in the control IQ pivotal trial by baseline A1C group. So first, what you see in this table is, you know, the aggregate where it says all. Overall, the average A1C in the group that was randomized to control IQ, which is 112 participants, was 7.4%. You see that their baseline time and range was 60%, and they had a plus 10.7, so nearly 11% increase in time and range overall. That was for the whole sample. If you break it down by A1C subgroup, what you see here, actually, if you look at the group with a baseline A1C of 8.5% or higher, is you actually see a much bigger improvement in time and range. The table in the paper is, like, really off kilter, so it's a little hard to line it up, but I've circled it here for you to see that the baseline time and range was 33% for this group, and they had a 22% increase in time and range. And then when you look even further at how the insulin delivery is different, so how did that happen? So this graphic here shows that for the different baseline A1C groups, what was the bolus insulin like? So the solid blue bar shows you manual boluses with announced carbs, so that means the user entered carbs into the bolus calculator to give a meal bolus. That's the blue bar. The white bar is manual boluses without carbs, excuse me, so that would be the user giving a correction bolus themselves through the bolus calculator. And then the red bar shows the automated correction boluses. So how many auto boluses was the control IQ algorithm delivering? And the main picture here is that as the baseline A1C goes up, the amount of insulin that person is getting from auto corrections also increases, meaning that algorithm is working extra hard for that individual. The meal boluses tend to go down, and the user giving correction boluses tend to go down. So you have the highest amount of user initiated boluses in the A1C group, less than 6.5%, and the least amount of auto correction boluses. But in the highest A1C group, you have the highest amount of auto corrections. And in this population, you can see auto corrections alone, nearly six auto correction boluses a day on average, and just about two meal boluses per day. So that really shows you how the difference between the insulin delivery in these baseline groups that kind of explains the difference in outcomes that you might see. There was also a meta-analysis done of the three randomized control trials done with control IQ. So there was a trial done, the first trial done in those ages 14 to 72, and then there was a second trial for those age six to 13, and then a third trial for those age two to five years. And this meta-analysis looked at the treatment effect in each of these subgroups, baseline A1C, socioeconomic status, race, ethnicity, and pre-study insulin regimen. Now, the study designed for control IQ was a randomized control trial where participants were randomized to use the control IQ system or standard care. 256 total participants were randomized to control IQ across these three studies, ranging in age from two to 72 years. 22% of them identified as a racial or ethnic minority. 24%, so about a quarter, were using multiple daily injections prior to the trial. And the baseline A1C range was pretty wide, 5.2 to 11.5%. Same with baseline time and range, three to 94%. This graph, while busy, really tells the story because this is looking at the treatment effect based on these different subgroups on time and range. So for those who are using control IQ, what type of benefit did they receive compared to the control group in these various subgroups? So what we see at the top, where I have put the pink rectangle around, is it's showing the treatment effect on time and range, looking about overall, the increase was about 11%, like I had mentioned. And then you see the increase in time and range by age, by race, ethnicity, by household income, whether they were on pump or MDI prior to the study, or whether they were using a CGM prior to the study. And there's no difference. All groups benefited from control IQ the same, regardless of what subgroup they were in in these demographic categories. When you look at baseline A1C and baseline time and range, you see that every group had improvements in A1C and improvements in time and range, regardless of their baseline, where they were starting from. But what's most striking is you see that actually those with the highest A1Cs and the lowest time and ranges have the greatest improvement. So really this question, this idea that, those who are already in pretty decent control are the ones who are gonna benefit the most. The data actually shows us as those with the highest A1Cs are the ones that are gonna see the biggest overall improvements. The very same type of analysis was done with Omnipod 5. This data was presented at EASD conference in October. This was a meta-analysis of the two Omnipod 5 pivotal trials. It had 320 participants between the ages of two and 70 years and you see the exact same thing. You see improvements regardless of age and you see improvements regardless of baseline A1C with those with the highest A1Cs having the largest relative improvements. So what about bolusing? We all know that with most of these systems, you're supposed to bolus and you're supposed to count carbs. That is how we're taught to use these devices. But what happens if you don't bolus? And to be honest, like in the trials, most of the participants in the trials are giving meal boluses and they are carb counting. So this is a study that my colleagues did here at Barbara Davis Center in the adult clinic and they wanted to look at the impact of, or they wanted to look and see what the outcomes are for people who don't give meal boluses on Control IQ. So they found 31 adults in their clinic who were using Control IQ and they looked at autobolus percent. So Control IQ works where it can automate the basal insulin and then it can also give autocorrection boluses once an hour. And then the user is supposed to give meal boluses. So that is the way the system is designed. So if you're not giving meal boluses, then most of your bolus insulin is gonna be coming from autocorrections. So they kind of define those who, like a no bolus group is those who were getting autobolus insulin more than 90%. So the more than 90% of their bolus insulin was from autoboluses. And then they went and matched by age, type one diabetes duration and gender to find individuals who are in an intermediate bolus category which they defined as those getting 50 to 90% of their bolus insulin from autocorrections. And then a high bolusing group who was just getting 10 to 49% of their bolus insulin from autocorrections. And what this came out to be was a sample that was an average age of 44 years with about 23 years of diabetes duration and a baseline A1C of 9.4%. So they looked at the change in time and range in A1C within these three groups across a 12 month period. And what they found is that there was significant improvements in all three groups. So regardless of your bolusing habits, you had improvements in A1C and time and range. And strikingly in the autobolus greater than 90% or the no bolus group, A1C decreased 1.6% and time and range increased 19% over that 12 month period. And this was also done without increasing hypoglycemia. Also interesting, three out of the 10 in that no bolus group actually had a time and range greater than 70%. So we're even able to attain, you know, international consensus guidelines all without giving any meal boluses, which is pretty striking. Same types of analyses have been looked at with the 780G system. So this was a study, really interesting study, small study, but looking at the impact of missed boluses, specifically breakfast. So they had 14 youth who were age 14 and a half on average, and they were instructed to skip their breakfast bolus for 12 days. And they had kind of controlled food content that had varying carb amounts, some with, some without fats, and they all followed this routine. And what they did is they checked their blood glucose two hours before the, I'm sorry, they checked their blood glucose before the meal and then two hours after the meal, and they were looking for like a carbohydrate tolerance level. At what level of carbohydrate do you avoid a 50-point postpartum increase in blood glucose? So I think that's actually a pretty high standard to look for, that you're only, you're wanting to see how many carbs can you eat without a bolus and still not see more than a 50-point rise in blood glucose within those two hours. The other thing they did to help kind of really test this idea is they used the breakfast bolus to help make sure that glucose was more likely to be in range when they did this experiment. And so they only were able to do the MS bolus if their blood glucose was less than 150 at the time of the meal, if they were, and if they did no exercise in the two hours after the meal and also no other meals within that two-hour period, and also no illnesses reported. So what they found was a 20-gram carb cutoff to avoid that 50-point increase in blood glucose. And this had a sensitivity of 0.94, a specificity of 0.46, and an area under the curve of 0.96. So this was a pretty strong finding. And you can see in this graph the difference by the carb snack. So what they did is that crackers were the complex carb, juice was the simple carb, and then they had biscuits with chocolate to kind of make, add in the fat component of it. And so you can see that across any type of snack, the 20-gram carb was the cutoff to avoid that 50-point rise. However, they didn't find any post BGs greater than 250 for any of the unbowled snacks up to 30 grams. And the median time and range was 70% for that two hours for all of the snacks. This was done in another small study that was presented, I think it was last year's ADA, where there were 14 adults who used 780G for 72 days. And they were actually instructed not to bowl this for any meals less than 80 grams of carb. So if they were eating less than 80 grams of carb, they were told not to bowl this for the meal. And after those 72 days doing this, their time and range was 67.6% with just 2% time below range. So we're seeing some data here that maybe you can still have some good outcomes even without bowling. But this here is a real world example I have of a 15-year-old that I've worked with who struggles to bowl this. So most of us are not going in instructing people, oh yeah, no need to bowl this because we know that to get the best outcomes, meal bolusing is important. But this is a kid that we all know that really struggles to bolus. Prior to starting 780G, you know, A1Cs regularly in the nines to tens. And so what we see here though is a 52% time and range, which equates to about a 7.9% GMI with less than two boluses a day. We can see that of his bolus insulin, 71% of it is coming from the autocorrections. With the 780G system, remember, you have autobasal being delivered every five minutes and adjusting regularly. And then you also have these autocorrection boluses that can be given as often as every five minutes if you're already at the maximum autobasal and the sensor glucose is still greater than 120. He's using the lowest target glucose of 100 milligram per deciliter to try to kind of optimize the automation piece, get the most insulin from automation possible. And another even optimization option we have here is we can decrease the active insulin time from three hours to two hours. Using that two-hour active insulin time has also been shown to give you the best glycemic outcomes with the 780G system as well. And we can do that especially because we see that he has very low percent time in hypoglycemia. We can also see in this example when we look at the weekly report that when he's in automated mode, which I have indicated here with this pink rectangle on the upper left corner, you can see in the pink lines that's all the autobasal adjustments. And then those tiny little purple dashes is the autocorrection boluses. So you can see he's getting a lot of autocorrection boluses. The user given boluses are at the very bottom with that kind of purple teardrop and the carb entries are in that orange box. So you can see very, very few meal boluses being given, but yet when in automated mode or smart guard it's called, you see relatively minimal hyperglycemia. Where he runs into problems with hyperglycemia primarily comes from times when he's in manual mode. Now he is 88% in smart guard. So doing pretty well staying in automated mode, but you can see the stark difference of what your blood sugars will look like when in manual mode and not bolusing compared to what they might look like when you are in automated mode and not bolusing. This is another real world, excuse me, example is 780G. This is a 40 year old male and this is really interesting. He only boluses for large carb meals. So he really just doesn't worry too much about bolusing unless it's a quote large meal like pasta or pizza. And you can see this is working decently well for him as well. Just he has 77% time and range, 6% lows. Only 1% of that is below 55. And a lot of these lows are primarily around exercise. And this is also an individual who has a pretty high tolerance for mild hypoglycemia. Most of these lows are in the 60s and that's just not something that really concerns him. This is a real world example of control IQ. This is a 20 year old. And what you can see here is this individual's maintaining about 50 to 60% time and range with actually, I say minimal user given boluses, but really it's like no, none. No user given boluses at all. 95% of their bolus insulin's coming from autocorrections and they're getting 11 autocorrection boluses a day. And in the pump settings profile, you can see that we've set relatively low correction factors one for every 20 kind of throughout the day. And this is to kind of help with that. So with the control IQ, the algorithm will use the correction factor that's programmed in the pump to calculate those autocorrection boluses. So we have a kind of an aggressive correction factor setting to help optimize and encourage and get more, as much insulin as possible while still being safe from the control IQ algorithm. This is the weekly view in source for this individual. And I just do want to point out like, it's certainly not perfect. So you still do see that, in some cases it works relatively well. On the second row there, the January 24th date that I've circled, you can see all the autocorrection boluses being given with the little blue teardrop symbol. And you can see the rise from the meal and then coming back down and then the rise from the meal and then coming back down and it, you know, decently works well. Then you look at the next day in that evening time and you see a lot of autocorrection boluses being given and it just not really coming down. So it definitely is not perfect, but to be able to have 60% time and range with no boluses, I would say that that is definitely successful. And what we also see here is a lot of hyperglycemia overnight. And I can also see that this person is not using the sleep activity, which is an option in control IQ that can also help optimize their overnight control. So, you know, an optimization option for this individual would be to try using the sleep activity for the overnight period. So I've shared with you a few, a little bit of the literature. I've shared a few examples, but I know that for a lot of us, really the biggest concern that we have with insulin pump therapy for those who, you know, maybe aren't as engaged in their care, they're not bolusing, is DKA risk. We all know that infusion site failure is a big problem with pumps and a risk factor for DKA if not addressed. But what does the data say? Like historically, we've always been taught that DKA risk is higher for those who are using insulin pumps than multiple daily injections. And this here is a study that was published a few years ago that looked at 47,000 adults in the DPV registry, which is a registry similar to the type one diabetes exchange but for Germany and Austria. And they looked at data between 2000 and 2016 and actually found that use of an insulin pump was not associated with higher rates of DKA, that there was no difference in the DKA rate that they found when using multiple daily injections compared to insulin pump therapy. However, with that being said, infusion site failure and risk of DKA is an incredibly important safety topic and really, really important that people understand what it is and that they understand how to troubleshoot persistent hyperglycemia. But the idea that by putting somebody with a high A1C in a pump is going to inherently increase the risk of DKA, the data doesn't necessarily support that any longer. I would say if you have somebody who's not giving their injections, their risk of DKA would be a lot higher than potentially getting more insulin if we put them on an automated insulin delivery system. But this here that I'm showing you is a educational tool we have on our Panther Program website to teach people about infusion site failure and the steps to take to prevent DKA, checking ketones, giving injections, monitoring, things like that. Okay, so where does this leave us? So who is a candidate then for AID? And the good news is is that the answer is very easy and it's not complex. AID should be considered for all people with type 1 diabetes. So every single person is a candidate for AID. When you're thinking about which device, that really comes down to what I highlighted here, which I think is personal preference. We really don't have a lot of data to suggest that one algorithm is far superior to another. I think that all of the systems can benefit individuals and it's really more about finding the one that that person wants to wear. It's the one that the person wants to wear, that's the best algorithm for them. And then of course, the one they can access in an affordable manner based on their insurance coverage or cost. But candidacy for AID should not depend on age, race, ethnicity, socioeconomic status. It should not depend on their current A1C or time and range. It should not depend on the number of BG checks they do per day, how long they've had diabetes, whether they've used a pump or CGM before. None of these things should be factors. The problem though is we do have quite a bit of disparity in the United States and actually worldwide in who is actually using and accessing these advanced technologies. This was a study published in 2022 with type 1 diabetes exchange providers trying to identify the potential impact that unconscious bias might have on technology uptake. And what they did is they had some clinical vignettes and a ranking exercise. And they were able to see that 61% of providers did display an implicit bias in recommending diabetes technology based on insurance type. So implicit bias means that it's a bias you hold that is not conscious that you don't even realize you have. And 34% of providers demonstrated implicit bias based on a patient's race or ethnicity. When they looked at the various factors of the providers that were related to having a bias or not, really the only thing that came out was practice years for insurance bias. So the longer you were practicing, the more likely you were to have a bias against insurance type. And insurance type was really public insurance. So if you were using Medicaid or Medicare, that you were less likely to recommend an insulin pump or CGM. Additionally, there's lots of leftover criterias about who can be on a pump that don't really have much evidence base. This was a survey study done with 192 US pediatric endocrinologists that showed that only 30% of them had written guidelines in their practice about how to recommend insulin pumps or who should be offered and encouraged to use them. And what this really means is that providers are using personal subjective feelings and thoughts about who should be eligible for an insulin pump, not objective data and not standardized processes. Three quarters of those surveys had minimum number of daily blood glucose checks required before someone was eligible for a pump. Nearly half required a minimum, a certain number of clinic visits per year. A quarter required certain A1C thresholds and just under half required a minimum duration of time. Additionally, 80% had indicated that carb counting was very, very important to be eligible for a pump and 95% marked that patient and family motivation was important. And I think the challenge with this is when we get into somebody's motivation or desire, it starts to become very subjective. And how do we judge whether somebody's motivated enough to use a pump or not? So we all need to be really careful about these things. And I think the main point to understand is that none of these criterias have any evidence base for them. And so we really need to start questioning who is a candidate and why we think that that might be in order to make sure that we're having equitable access to these technologies. Okay, so to finish up, I'm gonna share a couple examples of some alternative bullish strategies that I've used. So, as I mentioned before, we've talked about how the optimal way to use these AID devices that we currently have is to count carbs and give boluses. That's at least true for Omnipod 5, 780G and Control IQ. With the iLit, which is their newest device, we now have a whole nother way of bolusing by doing the meal announcements. But for those other three devices, that's really been the way that they are to be used. So this is a real world example of a 15 year old in our clinic who we started on Omnipod 5. She's had type 1 diabetes for two years. Her A1C has been 12 to 14 since she was diagnosed. So she was diagnosed and then it has never come down, which is tragically sad. She's currently on multiple daily injections with the Dexcom. Her current A1C is 13.6%. She of course has no lows and she hates giving her shots. And she's been asking about pumps actually for several months now. And we've been really, really hesitant based on her A1C and the fact that she didn't carb count to offer her these technologies. She has a lot of other social challenges like housing and food insecurity. Limited social support. Her mother was actually blind, secondary to diabetes, recently had passed away. She would maybe occasionally get her Levimere at school, but really she was not taking her shots at all. And this is her Dexcom tracing, which is just quite alarming. This is the 14 day tracing that you can see here. Now her strengths are that she really wants a pump. She does not wanna take shots anymore and she has verbalized this very clearly. She also has a sister who wants to help her, yet this sister is also just 17 years old, so not an adult. But the other striking strength that she has here is that she wears her Dexcom. So her CGM glucose is constantly just reading high. However, she's wearing her Dexcom 14 out of 14 days. So she's wearing the device. So we did start her on Omnipod 5. But carb counting wasn't really appropriate. So we had to come up with a different way. So we said, no carb counting, we're gonna do manual fixed meal doses. So we instructed her to tap the use CGM option in the bolus calculator. You can see in the images here what that looks like. And when you tap that purple use sensor hyperlink, it'll pull in the Dexcom value automatically to calculate that carb count. So in this example, you can see that that shows a three-unit bolus. And then we instructed her, if you're gonna eat something with carbs, and we reviewed the basics of what has carbs, what doesn't, increase that dose by six units. And that's what we taught her to do. With the training, we really focused primarily on safety and understanding how to fill the pod with insulin, how to put the pod on, the importance of changing the pod, we focused primarily on those main basic tasks and this very simple bolus strategy. And what we found in two weeks of wearing the pod is the time and range increased from zero to 47%. She was giving some boluses via this strategy, some days zero, but other days two to three. And you can see that she's getting 86% of her insulin from the auto-basal. So this is important for her to understand that she's getting 86% of her insulin from the auto-basal. So this is important to see too, because even though there's no auto-corrections in the Omnipod, the basal can increase significantly. And so she's getting 86% of her insulin all from the algorithm, because it's helping to compensate with those post-meal rises. She's changing her pod regularly. This is where the Omnipod really was a benefit for her specifically, because we needed the simplest device for her to use, given her limited support system. And so the pods are probably the easiest thing to start up and put on for somebody with the auto-insertion being a huge benefit. And she's successfully in automated mode. And she has kept up this positive trajectory with, I mean, she's been on Omnipod 5 now for I think close to nine months. Her A1Cs have been in the high seven. So she went from 13.6 to like 7.8, 7.9 in the last couple of visits. Still using automated mode, still wearing it. And she's now up to bolusing two, three times a day more regularly, which has helped increase that time and range up to near 60. And so for her, this has been a huge success. And I think the simplified bolus strategy has really helped build confidence and engagement in her care that she can do these things for herself, even though it might not be the full optimal traditional way that we are taught to use the device. My next example is a family from Vietnam that we see who has some language barriers and also some limited numeracy. So pretty low health literacy and low limited numeracy skills. And so this is a 12-year-old. Their caregivers only speak Vietnamese. And so when they come see us, they get care through an interpreter using multiple daily injections and a Dexcom. They've been taught to carb count. Caregivers really struggle to carb count. So they've just kind of given up on that. And they're just giving correction shots occasionally after meals. A1C is 9.2%. So it's really not working all that well. They're open to trying a pump. So we decide we are going to do Control IQ with this family because it's accessible to them through their insurance. They're open to it and willing, looking for a different way to manage diabetes other than the shots. But we also didn't want to burden them with carb counting because we'd already tried that and they were struggling with it. So we thought, you know what? We don't need to wait for the carb counting. We can do this in a different way. So to kind of help with the skill building piece is like a place to start from. we implement sometimes what we call like a fixed carb dose. So they're still entering grams of carb into the bolus calculator so that they can get the calculator calculations still and use the carb ratio that we program. However, we assess their diet with them and just determine, get an idea of what they tend to eat and then we determine small, medium and large carb amounts for their meals. So for this family, we typically ate about 30 to 45 grams of carb generally. And so we taught them to use 15, 30 and 45 for small, medium and large meals. And here you can see, this is kind of a follow-up picture of some of the glucose tracings. You can see those little blue bars where you see 30 grams, 15 grams, 30 grams, 45. So they're really using this concept. But you also see sometimes like there's an 18 in there and because they're starting to build skills. So when they feel like they know what the carbs are, they go ahead and put in the real carbs. If they're really not sure, they go with the small, medium, large concept. And they're bolusing two to three times a day this way. And now the A1Cs are down into the sevens. And the family is much, much happier not having to deal with the shots and feeling confident that they know how to manage the meals. So in summary, all people with type 1 diabetes are candidates for AID. Pre-meal bolusing, carb counting, it is ideal. It is the optimal way to use most of these devices. And you're going to get the best outcomes overall if you are carb counting and pre-meal bolusing. However, the important point is that that is not required to benefit from AID. Those with the highest baseline A1Cs are going to benefit the most. Even those who don't bolus are going to see significant improvements relative to where they started with their time and range in A1C. And flexibility is really the key here. We can meet each person where they are and consider simpler bolus approaches to help encourage bolusing. And then this way, these technologies are accessible to all. And we are instead of using criterias that prevent people from accessing them. And the goal is really safety and progress, not perfection. So I would like to leave you with that. And then if you want to check out our resources at the Panther Program, you're welcome to go to our website. And there's a QR code here. And I'm going to pass off to Molly Tannenbaum. And she's going to talk a little bit about ways that we can support individuals as they're trying to make this transition from to automated insulin delivery. Unmute. Thank you so much. Nice to be here. Let me see if I can. Oh, yeah. Let's see. I'm controlling the slides now. So there's a little lag. Well, as Kerry did a great job of showing us, we really want to think broadly and sometimes creatively about how people with diabetes can benefit from AID systems while remembering that what it means to benefit may look different for everyone. So we want to keep an open mind about who can do well on these systems. And I know Kerry kind of already went into this, but I wanted to briefly also highlight this great qualitative study by Lawton and colleagues from a few years ago, where they talked to clinicians about what the kind of the criteria and some of those preconceived ideas of who they thought would do well on AID systems. And then they had the experience of watching people go through and be on the systems. And the takeaway from, oh, yeah, there we go. The takeaway was that really the people who the providers, basically the preconceived ideas often turned out not to be the case. And sometimes it would be even the opposite, that the people who they thought would struggle sometimes actually did the best because they let the systems kind of do their thing and work for them. And so the participants in this study who were the clinicians and providers who kind of witnessed this concluded that individual family and psychosocial attributes cannot be used as pre-selection criteria, and ideally all individuals should be given the chance to try the technology. So then the next question is, how do we as providers support our patients to go from being in a place where they're not using AID to successfully adopting and using AID? Essentially, we ourselves as healthcare providers want to be ready, and we want to be able to help our patients to be ready. And there's a lot that potentially goes into that on the provider side. So depending on what kind of provider you are and what role you play, you may be giving education, you may be helping with more of the nuts and bolts of actually getting the system into the hands of your patients, you may be then reviewing settings and data and troubleshooting, tweaking things, and of course, staying current with the latest updates and technology. And I also wanted to highlight in the ADA standards of care, it says that no device used in diabetes management works optimally without education, training, and follow-up. So we definitely want to remember that starting on a new device or system is not going to be set it and forget it. We may need to tailor this education, training, and support depending on who we're working with. So as Carrie already talked about, we want to be able to take a person-centered approach to introducing advanced diabetes technologies to our patients. And we're really fortunate right now that we have multiple AID systems that are available, since we know that diabetes management is not one size fits all. So we want to be ready to address and work through barriers to initial uptake, to be able to tailor our recommendations for each person to meet their individual lifestyle. And we also want to be there to support the learning curve when someone is thinking about getting started on an AID system. So since we have some folks attending live, I was hoping that I could put this chat question out to the audience, just to think about for yourselves, how do you as a provider know, or how have you known in the past, what are some things you look for to know that someone is ready for an AID system, or what makes you think that someone might benefit from it? And so yeah, you can take a moment, and if anyone has any thoughts, things that they look to that might lead you to suggest an AID system to someone, please feel free to share. And I will just go ahead and move on and share my thoughts. So I would say that the easiest way to know that someone is ready is that they come to us, or they come to you and they ask for it. And maybe they saw something on social media about it, maybe they talked to someone about it, or maybe they saw something on social media about it, maybe they saw something on social media about it, maybe they talked to someone who has had a positive experience. That's all great, and they're initiating the conversation, and then that opens the door for talking about next steps. But there are plenty of people with diabetes who are in a place where they may be able to benefit from this technology, but they may not be coming to you and asking for it. So these are just some things to consider. I'm sure there's other things on the list, but things I might look to for who might be ready, maybe they're already on a pump and a CGM, so they're already wearing two devices and then making the switch to AID wouldn't be as huge of a leap, or they're willing to consider being open to wearing two devices. One other thing that I think about is if someone is using a CGM and getting benefit from it, but then they're also experiencing feeling overwhelmed or overloaded by the data, to me, that could be a sign that while the CGM data is benefiting them, having the data also go somewhere else, so there's actually there's an algorithm that's helping them as well in the background could potentially ease some of the data overload. And then the last thing I wanted to just mention is diabetes distress and burnout. And this one may feel counterintuitive, because if someone is feeling burnt out or overwhelmed, it could seem like maybe the last thing that they need to do is learn a whole new system or technology. But I would say that it could be an indication that maybe they would benefit from some kind of changing things up. And it can be a way to open the conversation about exploring sources of diabetes distress and burnout and potentially starting to think about how an automated system could possibly change up their management in a way that could be helpful. But it's likely that if someone is thinking about making the switch to an AID system that they're going to have some worries and concerns as well. We know from research that there are common barriers that people express about just any diabetes device, not specific to AID, which are not necessarily wanting to have devices on their body all the time, not wanting to spend more time managing diabetes, having worries about what other people will think if they have devices that can be seen or heard by other people, and then also concerns about whether they can trust the device. And then getting more specifically into AID, there are some understandable questions and concerns that might come up. So for example, feeling like I don't want to give control of the management of my diabetes over to some system. I won't be able to trust it. I should be able to do this on my own without a system helping me. I need to be able to carb count to do well on it. What if the pump malfunctions? What if the CGM gives wrong data to my pump? I don't have the time and energy to learn a new system. And more recently, because we have multiple options now, feeling like you need the time and energy to actually choose which one that's right for you. And I'm sure that there are others as well. And I just want to say that these are all very valid concerns. And these are things that will be important to address and work through as part of helping someone who is maybe on the path towards an AID system. So how do we start to do this in our clinical conversations? I really like to take a big step back and start with the big picture of figuring out what each person's why is. Why do they want to try AID? What can they envision getting out of it to help their own life? And it's important for us to know this, not just as the providers who are in the position to provide support, but it's helpful for each person to identify for themselves what would they get out of it? And what do they care about the most? And what do they want it to do? And this conversation can also be a great opportunity to provide some education, because depending on each person's knowledge coming in, they may or may not be able to come up with what an AID system could do for them. So it is a good chance to kind of use a motivational interviewing approach and ask for permission and say, would it be all right if I shared a bit of information with you on the systems that are available now? And you can open that door to provide education that they're interested in and then also ready to learn about and tailor it to what their own challenges are. So, for example, if someone is really struggling with blood sugars overnight, that could be a chance to kind of observe while you're really working hard to deal with your blood sugars overnight and then go into, well, here, let's talk about if you were on one of these systems, how might that change things for you in terms of your overnight management? So, again, everyone's why can be different. It could be, I'm just, you know, throwing some out there, but it could be really about improving diabetes management. It could be just about, like, not having as many finger sticks. It could be peace of mind. So it really depends on each person, and it's helpful to identify that for each individual. And then once we have a sense of that, what each person could really envision getting out of an AID system, that gives us helpful direction for how to work through some of the valid concerns and worries that I mentioned. Again, we'll want to use that motivational interviewing approach of asking for permission to be able to provide education and recommendations if there's areas where maybe we want to dispel some myths or explain a little more. And the idea here is that we're framing the benefits and the potential burdens or worries together in the same kind of against each other because we really want to clarify for each individual, are the benefits going to be worth some of the hassles or worth the energy that it might take to work through some of these potential concerns. And we have, like, five minutes left. I had a case example, but I'm not sure. Well, maybe I'll try to just really quickly mention, or Laia, what do you think? No, I think you should go ahead and present this case, but yeah. Go ahead. Okay. All right. So I wanted to highlight an example of this with a case. This is someone I worked with clinically. I'll call her Erica. She's an adult with type 1 diabetes, diagnosed as an adult, and her A1C had always been above the target of 7%. She was on MDI and had never tried a pump in the past and had recently started CGM. And she was kind of at this crossroads where she and her endocrinologist had kind of started talking about AID. She was kind of interested, but also had a lot of ambivalence. And so that's how she came to me. And one of the things, as I kind of talked about already, oh, here, yeah, I did make those bullets. Yeah. Okay. First, we identified her why. And these were the things that she identified for herself. She wanted to improve her A1C. She was really concerned about her long-term health. She liked the idea that just even having an insulin pump would mean she'd have insulin with her all the time. And then through the providing of a little bit of education about how these systems and the algorithms could work for her, she identified a few more benefits, that the system could give corrections, and then that would mean less mental burden and less pressure on her. And those were kind of really significant for her to identify. And then after that, we identified her worries and concerns. So we kind of opened up the floor for her to get everything out about all the things that she was worried about. She was, in her kind of brief experience with CGM, not 100% confident in the data. So that accuracy, the fact that it would lose signal, those were big concerns for her. She also was concerned about user error, like if she were to press a button and do something that would end up with a negative consequence. And she was concerned about the learning curve. So this was really helpful to learn this from her about what her concerns were, because there was this theme about her both trusting the technology as well as trusting herself to be able to troubleshoot and figure things out. So that opened the door for me to connect her with the CDCES on the team. And that ended up being a very well-timed conversation where she could ask all these questions and address all her concerns. And in the end, she decided to give AID a try. So I wanted to share this to show that we framed her concerns in the context of, well, what would she get out of it? What matters to her? And then being able to provide that timely education then supported her on moving forward. And just to say, this was not one conversation. It happened over several months. So this can be kind of a process. And so just to quickly wrap up, I wanted to stress that when someone is kind of on the verge of thinking about a change in their device or starting on an AID system, it's a big deal. It's a big decision. I don't want to kind of underestimate what the short-term burden can be on someone making this change. And we want to support them along the way and kind of walk alongside them because we know and the hope is that they'll get to a place where they will be able to experience the benefits of an AID system. So I just wanted to kind of boil this down into a few quick tips for having these conversations, focusing on someone's why, which might be different for everyone and different from the provider's why to help highlight that. Then going into exploring barriers and concerns, which can open up the opportunities to provide relevant and timely education, always asking for permission to deliver any education about this. So we don't want it to be like we're lecturing someone on here's why I think you should do this. But so each person is kind of envisioning, well, what's going to make sense in their own life and be best for them. And last, bringing in other resources. So I get to, I'm lucky I get to be part of a multidisciplinary team. So being able to lean on other people's expertise has been really helpful. And also there's a ton of resources online. And I just wanted to highlight diabetes-wise as one that it can be very helpful to kind of go and poke around on the provider side or on the side that's facing for people with diabetes to learn more about the different options that are out there that's in kind of a non-biased low pressure way. So I hope this will be helpful as you navigate having these conversations with your patients. And thank you. Thank you so much, Kari and Molly. I think it's one o'clock. I was told that this webinar that was just recorded will be online very soon for everyone's access. Thank you again. Have a good day, everyone. Bye.

ADA quarterly webinar

automated insulin delivery

non-traditional users

Control IQ

Omnipod 5

type 1 diabetes

education and support

individual needs

healthcare providers

diabetes management outcomes