April 13, 2020 Interpreting and Accessing Glucose Monitor Data
Join Dr. Ilana Halperin and Danielle Goudge as they discuss the approach to access and interpreting continuous glucose monitoring data. This is part 1 of a 2 part webinar series on leveraging technology to facilitate virtual diabetes care.
By the end of the session, participants will be able to:
- Review tips and tricks for accessing data from Freestyle Libre and Dexcom software
- Interpret an ambulatory glucost profile report and additional reports from CGM data
- Discuss tips for improving time in range
Danielle Goudge: So hi, everyone. Danielle Goudge: My name is Danielle. I'm a nurse educator at St. Mike's hospital and Dr. Lana helper and myself are doing our talk today around leveraging diabetes technology to facilitate virtual care. So it's actually a two part series. So the first part today we will be focusing more on accessing CGM data from Libre and DexCom and then our part to session happening on the 24th of April will be a little bit more around accessing pump data CGM and so on. So thanks for joining us today and look forward to chatting so
Dr Ilana's disclosures here. And for myself, I have no financial disclosure conflict conflicts of interest with the material presented today.
So part one, we're talking about our approach to accessing and interpreting continuous glucose monitoring, which includes Libre index. COM. So before we get into our chat today. We just have to polling questions. So we're going to pull those up here. So the first one is, what is your profession. So registered nurse, nurse practitioner, registered dietitian endocrinologist family physician or general practitioner pharmacist research researcher or scientist or other. Just take a few seconds there to submit your answer and then we'll just kind of see who's on our chat today or who's logged into our webinar. Okay, so it looks like the majority of people registered today are registered nurses, but we've also got 27% other. So I'm not sure not sure who that would be, but we can always see in the Q&A to kind of get a sense of who's on the line. Some NPS registered dietitians endocrinologists pharmacists and researchers. Scientists also on the line. So thanks for doing that. When and where are you dialing in from which province. That's our next polling question here. So I'm dialing in from Ontario, Toronto. Let's maybe see who we have and what our distribution is in terms of which province. So it looks like the majority of people are signing in from Ontario. I don't think we could fit in all of the provinces here on our screen so saying hi to everyone across, across Canada. Thanks for logging in.
Okay, so let's get started. I think we have three main objective objectives for today. So the first is reviewing some tips and tricks for accessing data from freestyle Libre, and the DexCom software. And then the second kind of half of our talk will focus on interpreting and ambulatory glucose profile report and additional reports that we might get from CGM data and then lastly we'll discuss some tips for improving time and range. Just to break the ice, a little bit. I think this is probably how a lot of us are feeling right now in terms of accessing data and technology and our star kind of switch over to what's likely a lot of virtual appointments and follow ups for everyone. So, you know, a, big a, big hurdle that we often face is how do we know we're getting the right information and from where and how do patients also access that information. So we just want to make sure we're on the same page. And I thought this was kind of fitting because a lot of the technologies and tools that we use now have cloud based systems. So how do we actually get that data into the cloud. And then how do we pull that data and extract it for use for review. So the first technology. We're going to talk about is the freestyle Libre. So we've got another polling question for everyone. And so how is your current practice setup for sharing data. Either you know we enroll patients in our clinic Libre view account patients email us there only brave you reports. Patients email screenshots from their Libre link app or patients read off the data from their Lieber a reader and we might have a mixed mix of all on. But what are you using primarily? So we'll pull up that polling question there. We'll just wait for everyone to kind of take a few seconds to respond. Let's maybe see let's pull up the data and see what we have here. So it looks like the majority of clinics are enrolling patients in their only review account, but some of us are still having patients email library reports. It looks like there are a few people that are having patients email screenshots from their library link app and also about a quarter of US accessing data. I having patients read off from their library reader. So during this talk. We'll talk about all of these kind of methods of accessing data and we'll just go through each and if you have any questions you can type, type them into the Q&A box there. So we're probably familiar with both of these tools we have the freestyle Libre a reader on the left hand side of the screen and we have our Libre link app on the right hand side. And so we think about kind of the types of patients or what you know a freestyle Libre a patient might look like. I think we either have someone who's accessing the Libre link up or someone who's using the freestyle Libra reader. And so how we access that data will look a little bit differently depending on whether or not we have a clinic account.
If we don't have a clinic account and we're having patients email emailing us reports or like we mentioned in our polling question if patients are reading off their data from the the library under the review of history screens. So, you know, we'll talk about these types of patients, you know, Libre link app versus Freestyle reader. I think those are kind of the two subsets of patients and how we access the information from there is driven by how they're using the technology. So, a lot of us are probably familiar with this page here. This is the Libre view main website so Libre view calm is what we would access. This is from the provider perspective so you know, for seeing someone in clinic we can upload a one time report or we can actually link that report to a patient. So we have those options there. But what do we do in terms of data sharing. So if we're using that that clinic a clown clinic account. Then we have the option of adding patients to that account so that they can share their data. So to do that we would invite patients to remotely connect and add them to our clinic. So we'd actually enter in all of their demographic information. And at the bottom left hand corner of the screen. Obviously your if your clinic has a practice to link to then you would select the drop down there and add them to that account and then it's really important that we have the patient right email because the email link that is sent to have them pair to the account is sent to that email they provide us. So I know you know from experience I've had patients, maybe provide me an email where one letter was off and I'm sending sending the email to this account here. And then, and then they can't actually access the data because they're not getting that email sent to the right account. So first option, adding patients to our clinic account and linking to practice. Once they've successfully paired when you go to look at them in the patient list you'll actually see that they're connected to the Libre, Libre view patient portal, so it will show green to highlight that they've actually paired up to that account.
So another option is for patients to actually connect their personal Libre view accounts to a clinic account by entering in a clinic code. So naturally, when we set up a Libre review account, there's an A code associated with that providers will access that through their menu tab and their my practices settings. It will generate a code for you. And this can be provided to patients. Who will then go in and enter that code in by accessing your account settings and they're my practices. So when they've entered that code and they've selected add it will actually show that the patient is connected them to that practice, whatever that practice, whatever your practices. The third options third option for patients sharing their data is actually emailing reports through the library link app. So I think we're finding probably more and more that patients are using their phones to scan as their reader versus actually having the physical reader anymore. So they do have the option of emailing us these reports in their library link app through using the share icon. So that's at the top of the screen there and normally this would be found below the reports. So when patients are accessing the reports through their menus.
They'd be able to see that share icon and then actually send that into their provider. So I think it's a useful tool when a clinic is not set up with a clinic account. Because this is very patient driven. So the patients are sending us their information. Important to remember that we want to specify which reports the patients are actually sending us and probably taking into consideration to the time period. So are they sending us a seven day report 14 days 30 days 90 days. We just have to provide them with that information in order to share. I think one main thing to remember is that for patients when they go into their menu. There is a share functionality, kind of at the bottom of that menu. But this is for Libre a link up. So this is for patient followers. It's not actually generating shared reports for us the share functionality for providers is the little icon that you'll see the box with the arrow. So one other option. And this is for patients who don't have computer access would be to either go through the review history icons and select the reports that they want to share with us verbally, but there are some patients who are also taking screenshots of their specific reports and sending those through email as well. I think often we do utilize and I think from the poll about 15% I believe was having a patient verbally report their reports to us so one thing to remember, in all of this is that in order to get a report with the ranges that we would like to see, we want to set up an appropriate target range. And so, remembering that the default range for the freestyle Libre reader is 5.6 to 7.8 millimoles, and for the most part, I think we are as providers, we're using four to 10 million miles as our range.
So when a patient's initially setting up their reader, they will be prompted to enter in a target range. So I think having that conversation up front with patients and letting them know, hey, can we put in that target range of four to 10 versus the default range and just reminding them to change that. Same thing with their app. So there'll be prompted when they, when they create the app Lieber a link account to actually set up their target range as well. But what do we do if we have a patient who's maybe set it up in the community and the target range was still programmed to the default. How do we actually change that so it's pretty straightforward. You, the patient will if they're using the reader access the cog wheel on the top right hand side of the screen on the reader. This will take them to a settings menu, you'll scroll down and when they scroll down and they reach that kind of target range. They'd be able to facilitate entering in whatever target range, they choose and primarily we're using that four to 10 million target range. In the app. It's the same process. So they'll access their menu menu tab on the top left hand of the screen. I know this was for like an IOS, iPhone app. So it might look a little different for Android. But they'll go to the Settings menu at the bottom of that list. And in that settings menu in the second image, you'll see that they can actually select target glucose range and then in the third snapshot. You'll see that they can set that range again by just scrolling up and down on the on their phone. So what types of reports are we expecting to see. So this slide here is just highlighting some of the reader reports that patients are getting from their readers and this is the reader. Specifically, it's not the app. So we're familiar with the log book that's individual scans their daily graphs, what their day to day tracing is 96 average glucose. I think what we use primarily or or something. We look at often is our daily patterns. Sorry. Our daily patterns in our time and target screens and this is what patients might be verbally reporting to us and in clinic or an or over the phone and our phone follow ups.
What are some examples of the library link reports. So again, we have the same reports generated whether the person is using a phone application or the reader. And that includes our time and target low glucose average glucose. Everything is is the same in terms of what it can generate it just looks a little bit different from the phone application to the reader. So when we shift back to accessing this information from the library view website. I think some of the port with the reports that are valuable to take a look at include the glucose pattern insights. So this is the first report generated when we're accessing that patient information. If you look at the the small image at the bottom. It would highlight that the glucose pattern insight is the first report that's pulled up so it provides us an ATP gives us an estimated a one seed or some information around, you know, glucose variability likelihood of low. So it's a good overall picture and report to take a look at this next. The second that we often access or that provides a good snapshot is called snapshot. So this gives us valuable information in terms of time range. It gives us a breakdown of low glucose events. And I think one other important piece to highlight is that sensor data that's captured so you know is that person actually utilizing the sensor and not just necessarily scanning at once a day when they wake up to check their fasting glucose. I think this. This provides a more kind of holistic picture.
Now sometimes what we can do is we can take this snapshot and toggle or similarly to the glucose pattern insights with our daily logs. So this gives us a day to day breakdown. And I think what's nice about the daily logs, is that it also provides insight into when that person might be taking insulin when they're eating if they're entering in comments. So for instance, this daily log is pulled from someone who's actually utilizing the library link app. And so they're able to enter in those specific kind of text comments to just give us some insight into what their day might look like. So I think in general, some tips for successful Libre data sharing are helping set desired target ranges on either their phone app or reader. So really honing in on setting that range of four to 10 million miles or something. Specific for that patient that you're working with having patients set up a Libre view account with you in clinic realizing that right now. This is probably not something we're doing as often. But if you're starting someone initially on a Libre sensor. It's nice to get them to set up that account with you. Saving that information. So having a patient save their username and password. Whether or not that's something they're sharing with you. Or that they're then accessing I think use it utilizing phone contacts is a really nice way to be able to do that. So they have access to that information and they're not always feeling like they can't remember to login when we may be asking them to upload their reader.
And don't forget to add patients to your practice. So creating that new patients so that we can actually generate that share email and invite them to the practice. And if your place of work doesn't have a clinic account, maybe speaking to your local rep, there might also be a conversation that we need to have with privacy or it in terms of facilitating the use of the library view website on on your computer. So we're going to shift a little bit to the DexCom technology and we're going to ask another polling question. So which statement best describes your access to DexCom data. Either I usually have access to the patients home user account someone not me accesses our clarity clinic and sends me the reports I want I access to clarity health care provider clinic myself and look at the reports I print save review live or I have not used clarity. So I'll give you a min, a little bit of time there to respond. Maybe you see the answers. So actually, the majority of SF 59% that are on the live webinar right now have not actually use clarity. So we do have 24% that have access to clarity accounts and they look at the reports themselves. Now about 9% is someone not me accessing our clarity clinic accounts and then the other 9% is I usually access the patient's home user account. So again, similar to how we broke down the Libra will look at the different ways we can access this data. The data from our, our patients. So the main web page or portal that we use for clarity is actually clarity dot x com.edu it's very important that we put the.edu because it won't take you to the right page to access that information. If not, and the to kind of applications that are shown here are DexCom app for g six and our clarity app as well. So what are our options. I think the first option that we often use at least in our practice is that we have a health care provider account so similar to Libre, we can access that healthcare provider account and login at clarity Dex, call me you and then this way, we're able to invite an ad patients to share data with that account and facilitate that transfer of information or access of that information.
So if we were to actually set up a patient to add them to the clinic, it's quite simple. So if we go through the main kind of DexCom clarity page. There is a button at the top right hand corner where it says add patient will have to enter in demographic information and for for DexCom, specifically the date of birth is actually really important because if we don't put in the correct date of birth, the email that sent to patients to link up they will be required to actually confirm their date of birth, and if it doesn't match, they won't be able to set set up sharing once that we've once we've created that patient the tablet. You'll select is shared data. This will allow you to either send an email invite to that patient. And it also gives you the option to print an email invitation with the code and then the patient would then enter that in but once the data once the two accounts are paired so the patients created their DexCom clarity account with the same username and password that they would use to create their DexCom app. The data is automatically transferred. So it's all in that cloud based system. There's no need to upload data center remotely. I think one nice way to access the reports, if we're looking at that report pulling once we have the patients in the clarity account and we generate our list.
We're actually able to save or print reports, you know, some people may be accessing go to interactive reports to do a live review of the reports that are available through Clarity, but we might also saver print to PDF, whether you're attaching that to documentation or having someone bring those reports to you to review. I think it's a matter of how we're accessing the information so another option is that your current practice may not actually have a health care provider account so patients will will and can create a home user clarity account. And again, the login information for this clarity account would be the same, that they use to generate the app on their phones. So they're not having to remember an additional password and additional username. It's just the same information that they've used to create their initial accounts. It does rely on a patient remembering that information. So again, some tips and tricks in terms of maybe having that person set up a contact in their phone list so that they can store that username and password. That's really helpful as well.
Another way that we can have patients, send us data if we don't have a health care provider account is to actually have them utilize the DexCom Clarity app so this is an app platform that's similar to what we would access on clarity through a web browser. And patients can actually send us specific reports through the application, rather than through the the computer or web web browser. Again, it relies on patients remembering their login information and having that app generated, most people I think probably save and store their passwords on their phone. But if they were to access that remotely, they might need to remember. So again, creating a phone contact is sometimes helpful. And with the clarity app, the other nice pieces that patients can consent to time and range and pattern notifications that they may get on a weekly basis. And I think that helps pull people into that self management aspect of, you know, looking at the reports, having access to those reports and the potential benefits. A small subset of patients who are utilizing the desktop Dex column technology might also be utilizing our receiver. So rather than having potentially a compatible phone or a phone that is too far advanced in terms of soft software updates to be compatible with the DexCom app, they may need to use a physical Dex calm receiver which looks fairly similar to a freestyle library reader, for instance. That I found a lot of our patients might be utilizing this if they don't have a compatible device or they they don't have compatibility with the DexCom.
And so they can upload this reader through the DexCom clarity web platform, but they can also utilize other technologies like di ascendant tide pool as an idea. So this just summarizes the different ways of accessing clarity data. So we have the home user account health care provider account and then patients also utilizing the clarity app. And so again, what works best for you in your practice and what what works best for the patient. There are different options in terms of sharing data. And those are all available to us as providers and patients. So what are some of the reports that you can generate with clarity. So I think if you're familiar with clarity. They've recently just had a little bit of a makeover on their site. So the reports and the web browser look a little bit different, but these the same information is there. So I think one of the most valuable reports is actually looking at our trends reports you can pull up patients alarm settings, for instance, what's at the bottom, you can see that that specific patients alarm settings for lows highs, etc. At the top you do get that trends graphic and we also have a bit of that glucose test statistics. In terms of, you know, potential estimated emergency average glucose. And then our time and range bracket. So this is if there's one report to take a look at this might be your first kind of go to when you're accessing clarity information.
Another really valuable report is actually the compare report so we can set specific time ranges, and this might be a time period before you make any changes patient or provider and then looking at a report that that's generated post making changes so it provides a really good access of information to really compare before and after an intervention might have been put in place. So it's a nice capture and graphic to really immediately see the impact of any changes that might have been made. The other one that we obviously go to as the GP report. So at the top, it provides those glucose statistics time and range coefficient of variability
The middle section is really that a GP that ambulatory glucose profile and then the bottom part of the screen is also your daily use. So it really provides a little bit more information that's that's feeding that a GP as well and then this might be a daily view of the DexCom as well. So again, you can see any alarms that are happening as a person's entering in their insulin doses here or carbs in here. You can see that as well. So some summary tips for successful DexCom data sharing. Remind patients that when they download and log into their DexCom app that this information can also be used to access clarity. So same username and password for all platforms. Again, have patients, save that information so they have easy access and easy login. The next time they're using those platforms. Don't forget to add patients to your practice. So they're sent that data sharing email and encourage them to check their junk mail, because that's sometimes where it goes. And if your place of work doesn't have a clinic account you can speak to your local Sales Rep. And your privacy it to help push Clarity onto your platforms. And I think that ends kind of the first part of our talk. And I'm just going to switch it over to Ilana here.
Ilana Halperin: Okay, I'm unmuted and I'm just trying to pull up my slides. Danielle, Can you guys see the objectives there? Yeah, okay. No.
Danielle Goudge: Then I’ll go on mute. We're good to go. Now I muted
Ilana Halperin: Oh, it's a little bit of a glitch as we transition between slides. Okay. So. Thank you Danielle through that I certainly learned some stuff. And I know we had one question come in about the phone and the Libre view. So I'll pose that to you at the end. So in the next two objectives. We're going to work a little bit more on the clinical side of things, less on the techie side and how we interpret an inventory glucose profile report and discuss tips for improving time and range. There we go. Okay, so we're going to do another polling question right now and bit of an advanced question on ambulatory glucose profiles. But when you look at an inventory group glucose profile, what can you not gather from inspecting it patterns of hypoglycemia time and target coefficient of variation or shape of the medium curve. So give people a minute to answer that. And let’s see what the responses look like. Nice. Alright, I have a well educated group from the people who answered. And I was interested to see that over 60 almost 60% of the people on the call. Haven't used clarity and I know we had, we had a nice representation from across the country. And I wonder if that is in some way representative of just to DexCom. So patients not using DexComs or maybe working in a clinic that doesn't have a type one group involved. But hopefully you guys will learn a little bit about how to interpret these reports. Now that you know how to access the data. Okay, okay. So when it comes to constructing the ambulatory glucose profile. I think this is a nice graphical demonstration of you know what the what what's contributing to it. So these are all the possible glucose data that could be a patient could be experiencing at a given time over across the day from 3pm to the next 3am over 14 week period and from their median line is determined and then we see the 25th of 75th percentile around that median line and then finally the tense of the 98th and so you can appreciate that when we take all of those disparate dots and bring them into this ambulatory glucose profile can be a lot easier to start to identify trends. But I think it's very important to pay attention to the width of the those 2050 to 75th intense to 90 of percentiles around the median line. I think when I first got started this, I was always preoccupied with the media in line with the meeting line is kind of like the agency. And if there's too much variability around the median line is really can't tell you much at all. So I like to break down a GP graphic interpretation, similar to how we initially assess logbooks some of the same rules still apply, we always want to identify hypoglycemia. Hypoglycemia is I know can be deadly. And importantly, it leads to variability, because almost everybody who lives with diabetes in his experience hypoglycemia will tell you it's almost impossible to eat 15 grams of carbohydrate and wait 15 minutes and so most people I know one of my patients once said to me, it's like a fugue state and they ate the whole fridge so we know that hypoglycemia leads to hypoglycemia. So we have to find the hypoglycemia, and we have to eliminate it. And we need to figure out, are the readings within the target range. Look at the shape of the median curve because it still can be helpful, and then talk about measures of glycaemic variability and depending on which reports, you're accessing you'll get different measures of glycaemic variability but a true ATP report will report both standard deviation and coefficient of variation but almost all the graphical version views of a GP, even if you're looking at a somebody’s Libre, a reader will show you the width of the entire quarter mile range around that media in line. So what is the coefficient of variation. And I think it's really quite helpful statistics to understand because the problem with standard deviation is it's all based on your average blood glucose. So we know patients whose average blood glucose is 12 or 13 but their standard in their standard deviation would be five or six but their coefficient of variation could still be low, because they're always 12 or 13 they're not going from two to 20 to 20 and so the coefficient of variation basically takes your mean blood glucose and divides or sorry, takes your standard deviation and divides it by your mean blood glucose. So regardless of whether your, your average blood glucose is in the lower range sort of around 10 or sort of around eight or your co your average blood glucose is higher around 12 your coefficient of variation will be the same and the magic number to remember is 36%. I really like looking at the coefficient of variation, because it helps me to know if my next approach is changing the insulin or talking more about habits and behavior. When people have unstable or highly variable blood glucose. It's very dangerous to start adjusting their insulin doses. It's a lot more important to talk to them about what are some of those factors that are contributing to that variability. But if they have pretty standard and less variable blood glucose says, then you can start, you know, adjusting basil rates or bolus doses and that can be the case for you know, long acting insulin for those who are not pumping as well. And you're less likely to cause trouble with hypoglycemia. So we always kind of have to address that variability. First, you can have a high average blood glucose. But if you just increase their insulin and they're still having lows, then you're only going to create more variability… Sorry.
Okay, here we go. So this is the first case, I'm going to share a few cases with you throughout the presentation. And then I've gotten consent from all my patients and change their names. But Molly, I'm sure a lot of you who work in this with this population have experienced a patient who is sort of, you know, really had a fantastic pediatric experience with a great care team that really caught her and then struggled in the adult world. In parks, because she always had good day when sees and so she was very quickly dismissed by the adult endocrinologists that she saw as having good agencies and actually just needed to continue doing what she was doing and this is what Molly's age up looks like when she comes to see me so you can appreciate that for estimated day when see is pretty similar to her actual agency when she came to see me. And her time and target range is reflective of that doesn't look like she's having too many lows. We're going to talk about what the target or acceptable lows are in a little while, but doesn't look too bad. But you do see a little bit of a clue around her coefficient of variation standard deviation doesn't look so bad, but her coefficient of variation is actually higher than we want it to be. And if you look carefully at the graphical a GP. If you're just paying attention to the median line, you might miss the fact that 10% of the time at four o'clock in the morning. She's going low and it was actually her mother who is sitting and saying, I said, look at this, guys. It looks pretty good. I, you know, and her mom says, can you just look at the dailies for me. Just, just look and see what we're experiencing every night because her mom was following her and this is what we're experiencing every night. So she goes high at around two or three in the morning. Her alarm goes off on her DexCom. She corrects she overcorrect goes low and then overcorrect her low and goes Hi again. And so that kind of variability can still give you a nice looking at one. See, but it's terrible for people's quality of life. And the stress and anxiety being woken up by her high alarm every night was a real problem for her. So I actually said maybe we don't need to correct these highs overnight. Maybe we need to wait and see what happens. Why are you going high and what is your body doing here. And so we actually raised her high alarms and when we raised her high alarms. We actually improve things a little bit. Still not perfect, but I think I like demonstrating this because this is the compare tab, and this is why I like not printing the reports and doing things interactively with my patients usually it used to be in person. That I'd be showing them the compare reports and we'd be able to appreciate the changes from when we, you know, she had the high alarm set at 12 to when I convinced her to set her high alarm 18 and just by lowering her high alarms were actually able to get rid of her pattern of nighttime highs and so on. The compare tab. As you scroll down you get sort of comparing the statistics. And then finally, I'll give you some patterns. So, you know, even though her time and range didn't really change very much. And her time hypo overall didn't change. We did get rid of that pattern of nighttime highs, which was disrupting her sleep and both her and her mother's quality of life. So I didn't get a nice demonstration of how helpful this kind of data can be as opposed to just looking at the Wednesday.
My second cases as a woman who came to see me again, a one sees in the 78 range. Most of the time, only checking blood sugar's a few times a day, but really wanted to improve for glycemic control prior to getting pregnant so she got an DexCom and this was her first visit to me after she got her DexCom on and so we're going to go through the HTTP graphic interpretation together at this point in the talk I will usually ask for audience participation. But the little bit hard on zoom, but I'll give people a minute to think about the questions. Do you see any patterns of hypoglycemia. Yeah, so she definitely having hypoglycemia, we can appreciate that almost, you know, 50% of the time she's low if the blue line is touching the low line that means more 50% of the time. She's blood glucose is are low between six and 10am I wrote 10. But I actually think it is 50. And then what about are the targets in the target range. Well, it's not great. She's in the target range for the most part, through sort of six to noon, but then above target range quite a bit as the day goes on and I'm sure a lot of you are used to seeing this as soon as food gets introduced into the day the blood sugar's get higher and higher with each meal and the highest overnight here. And then step three, you know, describing the shape of the meeting and curve, we can appreciate that she has really like a profound drop overnight and then slowly increase with the addition of food. And then what are the measures of glycemic variability. So she's got a ton of glycemic variability. When we look at that coefficient of variation. And we can see that the widest part of her entire quarter mile ranges in that early part of the morning. And so a lot of people can you know make guesses amongst themselves as to what you think might be going on with a patient like this. I'm can't remember if I put down with her original dosing, was it for insulin, but she took lantus every night when I first met her. This was back in 2017. And the other thing that likely could have been happening in this case is a dose of rapid acting insulin to correct a high before going to bed, which was too aggressive and causing her to go low. But what we decided to do was switching from lantus to to jail and put her basal insulin in the morning and encouraged her to stop correcting her highs overnight. And then you can see what happens to her variability. So it's a really kind of nice demonstration of how the second generation basal insulin can help our patients with type one diabetes in their variability. She still has some days where she's high and some days where she's low, but most of the time she's sitting very nicely and target overnight. And you can appreciate the change and what's happened before and after in terms of for time and range. As well as your time hypoglycemic and improvement or agency that's kind of the holy grail of type one diabetes. This isn't it, can we improve the time and range without increasing the time and hypoglycemia to result in a in a better agency. And so certainly with that data. I didn't need to wait for an actual agency to tell his patient that I thought she was good, good to go in terms of trying to conceive. But of course it didn't happen that quickly for her. So here's her case it with the compare tab, about six months later. And I think it's really quite remarkable. And so we're going to pull up a polling question here to get people who have experienced looking after patients with type one diabetes to tell us.
What do you think is different and Tracy's light from the first half of the month of the second half of the month menstruation fertility treatments, low carb diet or an acute respiratory infection. We’ll give people a chance to answer that one. Think we're probably ready for an answer. Up here we go. See, and I agree. I thought it was menstruation or fertility treatments because hormonal changes can do crazy things to your blood sugars and for those of us who work in this space a lot. No, you know, like I encourage my patients to use different doses different rates for basal insulin in the like. During the different phases of their cycle, but in this case it was actually a low carb diet. So she had been noticing her sugars running higher and she just got strict with herself when on the low carb diet and eliminated eating after 8pm and that's what flatten that her curve so remarkably I put acute respiratory infection on there because I think we've all seen that as well. And so something else to consider. But the compared to have can be very helpful when patients that I just had a couple really rough weeks when I was sick, or I was stressed at work or I was on my period. And you can see what a difference. Some these types of factors can be in our patients, day to day life. I think that's the comparison of the two. So she went from 27% time and range to 46% time and range, with an average blood glucose dropping quite a bit, but an increasing risk of hypoglycemia.
So we're going to move on to the third objective now which is talking about tips for improving time and range. So let's talk about what do you are talking about with your patients when you first meet them with diabetes. What's your primary concern glycemic control hypoglycemia diabetes distress or diabetes complications and he recognizes bit tongue in cheek, probably we hope to cover all of these things in their visit. What's the first thing that you talk about when you meet with your patients.You have the answers. Come on.
Hypoglycemia. I think that's excellent. And I think what's really interesting as I've moved through my career, since I started as a staff. I still get stuck in the glycemic control first like what's the agency because so do my patients. So like we've all been trained to talk about. Okay, so what was my A1C like as they're walking in the door before they even sit down. But actually what I try to ask is, how's life going with diabetes, like a really open answered question. I never always get an answer that I'm looking for. I haven't gotten so savvy as to start doing diabetes distress screening tools in the waiting room or anything. I think we should. I mean, it's certainly something we know is super prevalent and probably even more so now in the context of the pandemic. But sometimes I'm worried that I'm going to find stuff. I don't know how I don't even have the resources to help with right. But I think the point of highlighting this is that if we don't undercover the diabetes distress, then we won't be able to improve the time and range. Because if people are having too much stress related to external things or the management of their own diabetes. We're not going to be able to get there. But then of course. Hypoglycemia is our primary concern, like we talked about before. It's the first thing you want to identify once you're delving into the blood glucose records. And one thing that I know is whenever I see an A1C has gotten better from one visit to the next, I always follow up with how are the lows. And so that's regardless of whether or not I have CGM data or knocks and even more concerned about it when I don't have CGM device. I've just seen an A1C improve and we we've all fallen into the trap of saying, Hey, your A1C looks great. And then they're like, oh, but I had a severe hypoglycemic episode last week, and EMS had to come and then you're like, oh, that's why you're A1C looks good. And now I've just told you that it looks great, as if I'm somehow congratulating you on having a severe hypoglycemic episode. So the A1C is just so tricky.
And that was the reason for that polling question. So let's talk about A1Cs. I think I hope I'm talking to a group who appreciates why the A1C so tricky. And that you can have these different patterns that give you the same A1C somebody with medium variability, who's spending a decent amount of time and range but like a little bit too much low there. Somebody who is high variability. But spending, you know, a dangerous amount of time high. And then somebody who has low variability. And I always say that person doesn't have type one diabetes. Number three is like me wearing a CGM for fun, because I think it's really important to recognize, even for our most amazing type 1 well managed patients with diabetes, no one can be 100% time and range that's not realistic. That's not the goal and if you beat yourself up every time your blood sugar goes above 10 then you're going to have a very high level of diabetes distress. So these are the 2019 international CGM targets for non pregnant patients. And that's what I always like to highlight to my patients that the goal is 70% and it's awesome. When we can achieve more than 70% and certainly in our next talk we'll talk about automated insulin delivery, which I think is really helping to move us towards a consistent achievement of greater than 70% time and range without more hypoglycemia, but really we're not looking for A's, we're happy with a solid C’s. So that's something that's important to tell her patients.
And you know the different types of hypoglycemia are also important to recognize those a lot of our patients are not too fussed about 3.8 and 3.9. And generally, you know, getting somewhere around two to 4% in that area is not too bad. But the level two hypoglycemia. We really want to minimize because that's what's going to lead us to hypoglycemia unawareness and severe episodes. So this is just a quick screenshot out of a template within my EMR that I call type one diabetes technology. So I just thought it'd be interesting to just sort of show you guys how I've changed the way I document thanks to these tools. So we used to have, you know, four areas here in the glucose monitoring ACB, ACL, ACD, QHS, blah, who wants to record blood sugars like that anymore. So now I've split up the basil and bolus doses. And I think that's really important to me because this is one of the most important statistics to me. What's the total daily dose and what percentage of it is from bolus. Invariably, it's not 57% from bolus, do you know why this woman's was 57% from bolus because she's using a 670G. So we will talk about that more again at the next visit, but this was just an example of a patient, I just saw last week, and most of our patients are on too much basil and not enough bolus. And so I find with the bolus settings and my patients with type one now, I used to just care what their carb ratios now are. But now, a lot of times I'm saying, okay, but approximately how much insulin. Are you taking it each meal? Because that helps me figure out what's the total daily dose, which helps me figure out if the correction factors right and helps me figure out if they're bolus to basil ratios is right. And then I click off what kind of CGM device they're using. And I document with their time in ranges and their time and hypo, and I actually have it bring out for me what their previous time and range and previous time below range was so I write there. And then, without even going into the other technology. We can talk to them about what's better or different from the last time they were in the clinic.
So when we start patients on CGM I think it's really important to tell people not to do anything different for at least a week. I think that I had a couple early users of these devices true CGM not the flash glucose monitoring Who threw them out within two weeks because they increase their diabetes distress, because there was too many alerts and alarms and they didn't appreciate the accuracy or the accuracy challenges between the CGM and their finger poke so they just thought it was adding to their burden and not lessening their diabetes burden. So it is important to spend a little bit of time talking your patients but accuracy. I could do a whole other talk on accuracy. But I think just letting people know the usual thing I say is the real blood glucose that your eyes and your brain and your heart and your kidneys are seeing is somewhere in between the capillary blood glucose and the and the CGM blood glucose, because both of them have air have an element of error. Talking to patients about how to set their high and low alerts so that they don't get alarm fatigue. And then maybe at the first visit, but maybe not until the second or third visit talking about trend arrows. And finally, understanding what the data is telling you. So on the newest DexCom G6, you can set different different alerts for days and nights, which is a nice feature. And so I really encourage my patients who are just getting started to turn off all high alarms. I just think that that's not extra information you need. And unfortunately, what it does is it leads to overcorrecting and the so called rage blousing. Seeing that you're high not understanding how much time, how you have to be patient with your insulin and then people stack and overcorrect.
How about arrows. You could, I could do a whole other talk on how to interpret CGM arrows. There's a lot of different literature out there. Nothing's particularly evidence based but I really talked to my patients how an arrow could be helpful if they're going out and exercising, before they go to bed, most important, and I just teach them that it predicts where their glucose is going to be in 30 minutes which happens to be when their insulin is going to start to work. So using an arrow at the beginning of a meal is really helpful. But I would be careful about using an arrow. That's sort of where I'm saying so you can consider using a predictive glucose to calculate mealtime corrections, meaning that if your arrow is going up but you know a double up before you start your meal because maybe you had a snack. An hour ago that you forgot to bowl. As for when there was a really good time to add on extra insulin based on that rapid rate of change. But we have to be careful during active insulin time so we should never correct ties within two hours for patients who are using injections, because there's no real way to calculate insulin on board. And really the insulin hasn't peaked yet. And so even though the arrows going up and the numbers, how you have to be patient and after two hours I sort of say trust the OIB, the insulin on board. And if you're in an injector to consider giving half the usual correction. After two hours.
So my final tips for improving time in range for my patients is pre-bolusing. So what CGMS have really taught us is that rapid acting insulin is not that rapid. And to prevent that your high alarm have 11.1 going off, you need to take your insulin 15 to 30 minutes before you eat in order to get it into the circulation. I always sort of say my... It's not like you're taking your insulin IV. So it takes some time to get through the fat cells and into circulation. And we have to be patient with insulin, because it doesn't work as quickly as we wanted to. And so we have to avoid that rage bolusing when we see that our blood sugars are high and not coming down that quickly. CGM sometimes it's dangerous when it gives us too much information. Really, I think that increasing the bolus to basil ratio is something that almost all of our patients need, whether they have type one or type two, but they’re on multiple daily injections. We've over basil’d people for a long time. And that's why if you decide not to eat and go for a walk. You have a hypoglycemic episode. You're on too much basil. That shouldn't happen.
So really we need to give be more aggressive with our bolus insulin and and scale back the basil in order to reduce glycemic variability. I do advocate for a sort of a lower carb intake when we're choosing carbs to choose lower GI foods. I know it's not realistic for everybody, but it certainly does help improve variability and time in range.
And not overcorrecting lows, which I know is a very tough thing for people who live with type one to really manage because the panic and the anxiety that sets in, when you're having a low is real. But I did learn a good little tidbit that at least on a DexCom as soon as that little dot starts to turn the corner before the actual number will change, and certainly before the arrow which is the last thing to change, you're actually on your way out of the low and if you keep eating, you're going to overcorrect. And we really, I encourage my patients to reflect on their data once a week and aim for small wins. And that's where as Daniel had mentioned the, the DexCom Clarity app is kind of neat because you can sign up for these push notifications and it can give you that this like kind of positive reinforcement. Hey, you're up by 13% in your time and range from last week to this week. It's like when my Apple Watch tells me that I'm killing my exercise. It's very similar, but it's really positive reinforcement that I think really helps to empower patients living with diabetes to make changes for the… for the better.
So I always like to finish with a little shout out to one of my favorite places in the world, which is Camp Huronda and all the D-Camp systems across the country. And I know we have people watching today from across the country. So I think that Camp Huronda is the place that really turned me into a good doc for type one diabetes because there's nowhere else that you can really experience what it is to live with diabetes, the highs and lows that happen that really are, you know, not always manageable. I'm not, if I have my camper 70% time in range while they’re at camp, that's really spectacular. I'm sure that's not the case. We're just trying to keep them safe and have fun. And I really hope that everyone will be able to get back to camp this summer. I don't know if that's going to be the case, this year. But if anybody's ever looking for a specific project or organization within Diabetes Canada to donate to I always encourage you to donate to the camps, because it's a really special place.
And with that will open up for Q&A. Stop share. There we go.
Danielle Goudge: I think we have I unmuted myself Ilana. I think we have a few questions. So, one of them was so what's a reasonable CV? So, coefficient variability and the example that we went through. There was a CV of 36.5 and is this a good one?
Ilana Halperin: Yeah, I think everything's relative and I think it's more important to be like relative to your patient as opposed to themselves. So like I have a couple cases that I might share on our on our next webinar in two weeks where the CVs are in the 20s. So it is possible to achieve that kind of glycemic control but it requires a lot of meticulous management. So I think that if the previous CV was 45 and 36.5’s fantastic. 36 is sort of the magic number. But I think that the, you know so 36.5 is really not that far off, but you do have to recognize that there… the, the spread around the CV is going to be much tighter than the spread around the standard deviation, because of the nature of the way it's calculated but yeah, I would say it's it's it's 36 is the number to remember, in terms of stable versus unstable. But really, it's all about comparing to where where the patient was at previously.
Okay, so I have a question for you, Danielle. With a patient is using their phone to track all their Libre results. Do they have to set up a link to Libre view or does it just push to Libre view, similar to the way the DexCom app, which is to clarity.
Danielle Goudge: So essentially all that information is housed in the Libre a view cloud. So if someone has a Libre view phone app like Libre link, technically if they were to access using their same username and password on the Libre View website. All of that information would automatically be there. So everything is cloud based, whether it's phone or they're actually uploading a physical reader through the the web application, all that information is stored there. And so, someone's using a phone, it's just if they're accessing Libre view from a website or in clinic, we're accessing their information through our health care provider accounts and that's all automatically in the cloud and all automatically uploaded I guess or sent.
Ilana Halperin: Okay, so, so, yeah, no. I mean, I think that that's really helpful for me because I've just been using people sending me screenshots. But now I'm thinking they just need. We just need to make that connection sort of that one time connection health care provider account and then I can. They don't have to send me screenshots, because I'll be able to view it just very similar to the way I look at clarity. And be able to see those dailies where they put in those extra notes, which I think is actually very worthwhile and if the patient's take the time to put in all those extra notes about when they took their insulin and ate, it's nice for somebody to look at it.
Danielle Goudge: It's just the one time. It's the one time investment from like a patient perspective or like the provider perspective. Same with Clarity and Libre, it's, it's nice when they do that once and everything's automatically there, which is really helpful. I think there was a question. What I'm seeing here. What is your approach to those patients who do not want to share their CGM data with you as a clinician.
Ilana Halperin: It’s their data. I think I tried to encourage them to understand the utility of looking at the data. So it's really about… the how the… how the… median long, sorry, how the AGP and the time in range data can help us give them useful feedback. It may be that they don't want to do the share feature like they just don't like the idea of their data being on the cloud and who has else has access to it, but maybe they're willing to print and bring in a report. Now in this current climate where we're not really having too much in person visits than maybe they'd be willing to e-mail you their report, as opposed to sharing it. So I'm not sure if their issue is they don't want to share their own personal data with me or they don't want to put their data on the cloud. Certainly have a few patients who've been hesitant to do that, but then they're okay to print out and bring in their own reports. I think that's definitely still worth reviewing together.
Danielle Goudge: That's generally been my shared experience as well with patients.
Ilana Halperin: I guess you might be able to answer this one Danielle. If the accepted time in range is 3.9 to send and why set the Libre at 4.0 to 10.
Danielle Goudge: I know. So I think our default is to often just set the range, to 4-10. I mean, based on the consensus statement in terms of time in range that does encompass like three… a wider range. So the 3.9 to 10. I think you're still getting probably information in terms of time in range percentages that are very similar. Whether that target is 3.9 or 4.0, I think our default is to just kind of us are 4.0 to 10.0 targets based on our kind of historical interpretation of glucose data.
Ilana Halperin: Exactly, I completely agree with you. I mean, I think the point is that the more time we understand about accuracy, we know that whether the blood sugar's 3.9 or 4.0 matters not. So the point is is that we in Canada have artificially said that a low blood glucose is anything less than 4.0, but in the US it's anything less than 70 which is 3.9. So that's where it comes for. And it's really more about like are the patient symptomatic or anything. One thing I would say that as I'm speaking I sort of think about is like maybe because often we have patients tell us it at that 3.9 range they're not symptomatic. They're not low. The Libre says they're low, they double check with a finger poke and they're actually 4.2. We should be doing the opposite and setting it at 3.8 or something like that, because then we see a lot more lows that I often find I'm writing something like time low 20%, patient having asymptomatic low twice a week. So we know that we're like, we're not really sure that they're truly low 20% of the time. So that's one of the challenges I think that's an interesting conversation point.
Danielle Goudge: You know, we have another question here around, do we ever have patients who we need to bolus for things like coffee.
Ilana Halperin: Yeah, so something that I've been reading about recently is this idea of the feet on the floor. Right. So I think we've all learned about the dawn phenomenon for a long time, but I have other patients who are like perfectly in range until they start moving but they haven't eaten anything yet, or sometimes if I questioned them a bit more. They say they haven't eaten anything that they had their coffee, but it was black. There was no sugar. There was no milk, there's no carbs. Why do we need to bolus? so I don't know if it's the coffee and the caffeine and what that does to your stress hormones or just the stress of getting out of bed and running around and trying to get the day going, but both of those things sometimes need a bolus. And because it's very predictable and especially for injectors. Their basil… you can adjust their basal insulin for that phenomenon and so we do certainly have patients who… I certainly have patients who bolus for coffee, it's trial and error. They figure out how their body responds. What do you think Danielle?
Danielle Goudge: Yeah, I agree. And usually we say like, start small and go from there. Just again, trial and error and experimenting.
Ilana Halperin: Yeah, that's, that's, that's life with diabetes. But what I think so fantastic about this, these new approaches to the data management is how the experimenting doesn't feel like we're experimenting in the dark or experiment with a blindfold on anymore. Whereas now, when we do the experiments we can really see the fruits of our and, more importantly, our patients efforts towards improving their time in range. I agree. It's one o'clock. Does anyone, are there any last questions from the audience here?
Well, thanks everybody for joining us from coast to coast. It's pretty cool. And we look forward to chatting some more about more advanced CGM interpretations, in the context of insulin pumps and automated insulin delivery, not this Friday, but next Friday on the 24th. Take care.
Danielle Goudge: Thanks, everyone.