Hello. My name is Vivian Fonseca. Professor of medicine and the chronology and the assistant dean for clinical research at Tulane University Health Sciences Center in new Orleans. Welcome to this program on navigating and deploying sense of basic glucose monitoring in the managed care pharmacy and medical city. I want to thank all the previous speakers for introducing you to this topic to china for uh setting it up for me to talk to you about what I'm going to do today and looking at things like real world data, which is actually very important. You often want to think about how does this thing work in the real world? In primary care settings. Immigrant practices across all types of diabetes clinics. And I'm going to show you data from several countries where they've actually looked at good real world data going much beyond clinical trial data which you heard off. And uh that might help us frame where this fits it in clinical practice. So my focus is on the real world and I'll be talking about some cases. How does do these cases help us deploy this kind of technology uh in managed care fantasy. You're on my disclosures and I just want to also acknowledge Miller for providing us with some case studies which he asked me to comment on and I will go through the at least cases. So diabetes remains an important problem in clinical practice is the leading cause of very serious uh huh conditions such as chronic renal failure, heart attacks, strokes etcetera. Major contributor towards death and mobility in this country. Leading cause of same stage renal disease. People with diabetes are two times more likely to develop die from cardiovascular disease including M. I. But increasing a lot of heart failure as well. And uh that there people with diabetes really fear having heart attacks, strokes, amputations, blindness etcetera. Yeah much of this is preventable with good play. Simic glucose control is essential but it's not sufficient. There are multiple other aspects about the overall A. One C control which is what we're focused on the nuances of fluctuations in blood glucose. How people respond to this hyperglycemia etcetera. We'll look at some of these in the context of the cases we talk about much of this is also a driver of cardiovascular disease and diabetes. And uh the risk of CVD is related to overall black Samir with a one C. But there are these other factors also come into play as a one C. Doesn't reflect all these uh at the valleys essentially we need a balance. Uh We have multiple risk factors that we have to address in these patients. But in terms of my semi control we need to be timely effective. Give people stable by semi control achieving targets that will lead to prevention of complications, lower health care utilization with regiments that are flexible that might help them improve endurance which is a major problem in practice at the same time. People fear hypocrisy. You want to choose the therapy that reduces the pure of hypoglycemia. Maybe I eliminate hyperglycemia and there's data to support the use of continuous glucose Margaret too prevent that. This improves the difference in practice which might be driving but the reduction in uh they won't see that we're seeing in practice. So in clinical trials, hypoglycemia double the risk of mortality hit major trials like a court in advance. Since then, we've had some studies with uh huh specific agents for diabetes that have been shown to reduce the risk of mortality and they don't cause that much of a progressive. So it's important yet we often have to use those new agents in conjunction with insulin and self curious. So hyperglycemia remains a problem. Also in observational studies, hypoglycemia, particularly severe hyperglycemia increases the odds ratio for mortality loaded 3.5 times. And so people don't get like experience this, they stopped their medications and it becomes a problem. Technology has really helped it's helped with addressing this problem of hyperbole senior, the vicious cycle. You get there and leading to optimization of therapy as I will show you, I'm going to focus a lot of flash glucose marvin because it's very easy to use and it's been adopted widespread manner across the world. And if used correctly, it can help patients a lot that it's been widely embraced in practice. Uh So here is how see g m has changed the management of diabetes across multiple systems. I don't have time to go into each of these. There are a needle based systems, needle list systems, batches, other other kinds of systems. Increasingly we're going to things that are easy for patients to use easy to switch on, easy to apply easy to read with this work with smartphones or simple readers with good metrics that patients can understand very well. And also perhaps in the most sophisticated patients connect with their pumps and smart pens and help them with uh improving that control and improving their their uh that overall appears. We've also been able to come up with at metrics so called ambulance league ambulatory glucose profile. As you see on the left where the A. One C. Doesn't tell you the whole story. We'll discuss this in more detail some of the ups and downs with the blood glucose that you see on the A. G. P. And giving your patient and understanding of this really helps reduce hype oversee. This was the first a major study that came out of uh clinical trials with these kinds of technologies that showed. The first thing that happened happened very quickly was a reduction in high policy. Later on we started seeing improvements in the Nazi. So let's look at what happens beyond clinical trials out in the real world. Looking at a variety variety of daily basis. We'll start with the UK where they have adopted it very widely in the last couple of years with UK liberal view Taking 9000 patients are using this condition getting sensor readings for multiple days and looking at some time last year in the setting of the covid better and adult patients. The UK report at the start of this time frame had a diamond range, which is a, you know, reasonable target bridge between idiot uh 1 80 it uh Tended to be around 50%,, But that improved very quickly uh to to almost 60%. And uh time below range, which reflects hypoglycemia was also a significant problem and that seemed to get better. Now, a goal for these metrics is a consensus target of diamond range at least 70 and less than 4% below rates that we have ideally, I would like No hyperglycemia sometime below range should be really 0%, but that might not be totally realistic when you're using it. So, here's the UK study looking at diamond range between January 2020 and June 2020 and you see the improvement across all age groups Uh in the 65 plus age group actually jumps to 61%, which is close to where we wanted to be in. There are obviously individual patients that were above 70% and you see that on the right hand side, a large proportion of people Meeting that 70% target and having less than 4% below below range. This is not just in the UK the smallest studies from other countries. He has gone from ITaly. This was a prospective observational study. They designed it to look at a one C is the primary endpoint. And they included patients on basal bolus insulin regiments. That's why we started off with using c g M four and they had poor control despite having to take so many injections per day, one C. 8 to 12%. And they were you to using this kind of technology and many of them could. We're still using self monitoring of blood glucose and after about 3 to 4 months of therapy uh the A one C fell in those who used the freestyle liberate compared to self monitoring of blood glucose. Uh and the reduction uh A one C was 10.3% which was statistically when you take all the patients concerning imputing missing values. The impact will remain statistically significant. Surreal world prospective data suggests that you could reduce uh hemoglobin a one C as has been seen in randomized clinical trials closer to home in Canada. The real world short review study, somewhat smaller scale. And they've now been doing largest prize that this looked at people on basil insulin along these are different population. They were on basal insulin with an a one c of around almost nine Big 65 years of age and have been using insulin for about four years. And when they started using the freestyle, liberate the a one C dropped by 10.8% which is a very significant drop and most of the drop occurred with high baseline able to see. So so real difference there that for the patients and obviously this will lead to better long term outcomes. So that's some real world evidence. I'll come to some newer evidence laying around from some other countries. But let's take some of these cases. But that Doctor Miller has provided. So the first cases a 66 year old male with type two diabetes who has run Olympic and it's still on sulphur neuro therapy. Uh Obviously social areas lead to more hypoglycemia in patients with renal impairment because they are usually cleared by the kidney and A and A one C. Is not very reliable as the reading impairment was. So the a. one c. gives you some discrepancy with the blood glucose reading. Uh The if you look at the glucose management indicator at 7.3 that comes out close to the A one cc. Can anyone see that? You might say this is not bad But there's a lot of variability in the blood glucose as you can see the glucose variability and only 63% of the blood glucose values within the target range. A lot of higher and a few are lower fortunately this particular patient that none that are very low, Which is below 54. So he wouldn't consider this severe. But even so in somebody at that age uh any low blood sugar is that so uh fairly common problem in patients with kidney disease and diabetes. And you're the metrics again in a little bit more detail. But it really makes a lot of sense to look at the G. P. In this particular kind of patient. And what you see here is that a lot of the time the patient is within range. But there are times when the patient's blood glucose is fluctuating much more than you want to see in particular. I'm concerned. Although overall hypoglycemia doesn't look a lot, it's occurring at a vulnerable time for this patient between midnight and two a.m. And maybe a little bit around six years. In addition to getting very high peaks after breakfast after lodge uh around dinner time. So this allows me to modify this patient's therapy may be reduced the nighttime insulin or soften syria in the evening. In the case of this particular case, maybe try to institute some post branded glucose therapy during the day and we now have several agents that can accomplish that. Maybe show this to the patient and get them to change their carbohydrate intake. Mhm. You can go into even more detail and individual patients, you can point out there is where the postprandial is very high at the people of means, usually lunchtime in this particular case, but sometimes at dinner time and the patient might be able to give you some kind of explanation and themselves understanding unhealthy behaviors that are contributing. He said companies, the report that you get gives you a lot of insight into particular times of the day where you're getting a lot of variability. You can see the afternoon time is particularly bad in this respect. Uh huh. And that helps the patients choose what to do uh at particular times of the day And you can see the average glucose at the lower right hand side. You see the low glucose events sometimes going almost 50 in what you would otherwise think is well controlled, relatively mild diabetes. So what in summary years, a case with considerable variability. Very tough day today getting hypoglycemia at night, but he's not recognizing it. 24 postprandial control. And that prompts us to maybe stop the south Muriel changes only to the morning time. Consider GLP one receptor agonist. I consider an S GLT two that we love to hear the day. Others for spreading Lucan speaks. And if you're still getting some hyperglycemia at night, you could uh maybe introduce a small bedrooms. What about another case? This is a 49 year old female with type two diabetes. What is called ketosis from diabetes. These people today actually have Type one need insulin. She's taking basal insulin plus some policies with meals appears to have overall good control. The time in the target range is almost 70 Yet. The a one sees a little on the right side. eight perceptive. Uh there are a number of highs and a few very highs. Uh no hypoglycemia portion. So maybe the patient is just not getting enough, insulin is a little bit more detail of that. And you you have to ask yourself what type of diabetes and suspicion And you can do a get anybody and if the patient has got type one you really need to treat them as type one better basal bolus monitoring. Considering maybe even in its sort of bump book closed loop systems. I think A one C. And G. M. I. R. Missing something very important here. That's the peaks about glucose. Uh Although it's fortunate there's no hypoglycemia. Yeah the peace are a problem. And uh you see a little bit of that on the A. G. P. You may want to go into a little bit more details to find out why some evenings the patient is going very high uh and maybe some days overnight the patient is going very hard. So you may want to see what days those occurred. What uh full did take the patient has been getting and so on. And that can be uh determined from the daily profile that you can see over two weeks and have a discussion with the patient as to what events occurred in particular days. Uh huh. There is no one clear point in time. There's a lot of variation in this particular patient. Uh fortunately no hyperglycemia but a lot of thoughts. So what how does this C. G. M. Help with this particular case. I think what you, what you can see clearly is that the fish is not getting enough insulin secretion or replacement. That means to avoid spreading because the discussion That might indicate that this patient might have type one. Mhm. On some days the carbohydrate intake is high and you identify that as may be occurring at the weekends. You ask the patient to keep food record and you consider things like you know, should we add in a GLP one receptor agonists or an SLT doing a bit of on the other hand, this patient is type one that's off maybe off label or you have a risk of DK. Yeah, you need to have this kind of conversation with the patient. Let's look at another case. This is a 59 year old male with type two diabetes for 10 years. On basic goal of therapy. Uh A one C. Looks great. No, not too bad. 7.3% of the glucose management indicator. GMS actually 6.8. But your patient is getting some hypoglycemia at night just walking him up a couple of times. Also getting some high Diamond range is not bad at 75%. That's what we wanted to be. But we need to eliminate some of those eyes and certainly take care of the looks. So let's look at the profile and here you see the lows very clearly occurring between maybe two o'clock at four o'clock. Very challenging for patients to identify. There's another thing you noticed and that is the steady drop during the night. uh this might be setting your patient up to get this 34 years and then you see the rise after breakfast a further rise after lunch and then a fairly big peak after supper. So this should prompt the conversation about what carbohydrate is being taken. What meantime uh measures are being taken and what's occurring on particular days. You see, for example here, breakfast on monday may not have been very good lunch and suddenly supper on Wednesday was a problem and your patient could identify what what the real issue was. There's a modern degree of variability of blood glucose that can be a problem and you can identify the low glucose events. And so the problem here is considerable variability within days and the cross days unrecognized by globalised senior, particularly at night. And occasionally a little bit of they are four postprandial increase. So again, getting given the patient getting them to read a diary about their food intake might be helpful. Uh Consider adding in other therapies like Shelby one Dario is Shelby to in a bit of maybe too much basil insulin all at one time just before going to bed and then have dropped during the night suit might want to reduce that maybe split the dose of basal insulin or consider a bed lives thing. What these three cases really tell you is that there's so much hetero Geneti in this disease. It allows you to put your patient in different categories. You will find that almost every patient is different. You can personalize their care, choose the great therapy. Play. Civic patterns vary a lot but you can get an insight into the process. Diet certainly plays an impact. Something I have not touched on his stress. You may want plaster patient to stride identified what what's driving that afternoon glucose. Is it stress at work or something like that? That's uh you know, you might need a different strategy. Maybe get the patient to relax a little bit for a short time. Some kind of other behavioral modification and the C. G. M. Overall is allowing you to start taking steps towards an effective personalized treatment strategy. So let's get back to these other real world studies and without, I don't have time to go into details of every study. But there are three studies published together last year from different european countries, went from Austria France Germany and they did a meta analysis and showed a drop in a one C. Uh this is a minimalist for leaders of this amounts to a reduction in a one c of about 1.2% which was just increasing uh They broke it down into what patients benefited and it really didn't matter across the board. Those with very high a one C Over 10 or less than 10 older patients. Younger patients, men and women, those are insulin for many years with or without obesity. They all benefited turkeys. So now we have a range of evidence randomized controlled trials such as the one of the BMJ. You have these real world studies, such as what I have presented to you, showing that it works across different systems. Question that often comes up. What about people who don't take insulin? Does it help? We're seeing a lot of data of this as well. This is a a small randomized control trial that was done and published in the BMJ last year. But they looked at people not taking insulin that they were randomized self blood glucose monitoring all flash glucose monitoring and uh They were not particularly obese a little bit of hypertension. A one C was not well controlled 7.8. And patients were given a lot of advice on how to improve their control. And you see the difference. Those who add the benefit of flash glucose monetary at a greater reduction and a more persistent production in A one C compared to those who are using self monitoring of blood glucose indicating that sometimes patients can self adjust based on what they see from this flash glucose. They know exactly what foods are bad for them when the blood shooters going up and take corrective action overall the when you look at the diamond range, there was significant improvement. When you look at you break down that diamond range and look at the peaks. You see that There was a reduction in time over 1, 80 reduction in time over to 40 and reduction in the really high peaks cuter than 300. Although fortunately that was not as common in this population. Another factor is what do the patients think they were asked to do? Treatment satisfaction question is and this should be a real benefit overall across the board. Uh You know, the patients seem to like that the total school was improved and above all they like the convenience, flexibility and said they would recommend it to others. Uh huh. Another analysis here looking at production in a one c showing that those who have very high, one sees a baseline for 12% and reductions of about focus for uh 4% points and 3.5% points. This is as good as adding in any drug therapy. Probably what you could say, comparison the new drug therapies. This is a very cost effective intervention and it allows patients to get a lot more insight to what's happening with them and this happened when they took uh insulin or didn't take it. The real world demonstrates a number of things to us that C. G. M. Is becoming very much the standard of care. Some of this has been prompted by people wanting to prove care in the face of pandemic. Uh There is an increase in preventive cost but this leads to a lot of benefits and a key benefit is a health care plans. Uh I have seen a reduction in short term costs as well and that needs to be taken into the equation. People are going to the emergency room less be able to make adjustments to their diet and medication that that might drive costs lower. And uh there are obviously still some channel challenges about what kind of benefits people have, how quickly you can observe these savings and coverage remains a challenge in some situations. Finally, there's some new newer technology. Not only is this something that the patients look at, bring it on, the readers were now using more uh phone based card form base applications that allow us to see what's going on with the patient using uh things in the cloud data storage in the cloud that you can access. But some people find that overwhelming doctors offices don't have the time and the resources to keep accessing the club to murder all their patients. So what would happen if we could integrate this with electronic health records and that's actually being implemented rate? So there's a one particular study presented at the meeting that shows that this could be going to the EMR directly allowing you to have easy access and make immediate uh decisions on how things should change. This was done at the international diabetes center in Minneapolis partnering with have it and making all these results available to the three d. clinician at the point of care so that clinicians can place in order for the HR. Within the system when they come back they can see all the results transferred directly from the cloud and liberate view directly to the HR platform. And the clinician can look at the CGM monitoring over the previous few weeks. Make some decisions, discuss it with the patient and make changes. I you've heard about the A. G. P. I can't emphasize more how helpful it is in clinical practice that can now be integrated with the HR and allows 10 issues to easily track their patients of glucose strengths. Not only can this be looked at at one point in time it's possible to track this data over individual patients whole year for example and and groups of patients so you can identify patients with problems remote patient marvin's way much something that's becoming standard of care. So in conclusion C. G. M. Is a foundational approach to diabetes care and this is very easily integrated to manage plans. Diabetes is a major problem for people in these plans costs are very high. There's a lot of morbidity and mortality sensor based. See GM offers an opportunity to change this but both type two and type one diabetes You can improve licensing control low a one C reduce complication rates, improve outcomes and reduced us both short term and law. So it's becoming very much a foundation strategy across the spectrum of diabetes and managed care plans in particular stand to benefit from incorporating this into the standard of care. I'm very excited about the integration into the he char because it has a lot of implications for how we practice medicine, how we approach diabetes care.