April 30, 2020 Approach to Accessing and Interpreting Pump Data with a Focus on Automated Insulin Delivery

Ilana Halperin: So thanks for joining us again for the Diabetes Canada COVID-19 series. This is part two for Danielle and I talking about leveraging diabetes technology to facilitate virtual care. So just briefly to review my disclosures and I will be clear today when I'm discussing any off-label use of medications or devices and Danielle has no disclosures. So we're going to start by polling the audience to figure out who's with us today. So I think our first question is, what is your profession, so please click on one of those radio buttons that is shown up on your screen and let us know who you are. Okay, so about half of you are registered nurses and a quarter dietitians. Just a few endocrinologists, pharmacists and we have some others. I'm not sure if those others represent endocrinologists in training. That's what I thought about last time when we weren't sure who others were is that maybe we have some residents or fellows joining in. So shout out to those of you who are in that category. Okay, so I think we can close out and find out where you're dialing in from what province. Are you living in currently. Okay, so we've got a spread from all the way across the provinces. Except for Manitoba, we have at least one or two participants across the country with the bulk of you joining us from Ontario. Okay, great. Thank you very much. We'll have a few more polling questions to keep you engaged throughout the presentation. And so I'm going to pass it over to Danielle.

Danielle Goudge: So like I said, today we're talking about leveraging technology, specifically from a pump perspective. So we're going to review some tips and tricks for accessing data from Carelink, Tidepool and Diasend. Those are the softwares we are going to be focusing on today. We'll discuss some key concepts gleaned from pump reports and then some tips for optimizing automated insulin delivery.

So just as an update after we did our talk two weeks ago, I guess now pretty much and it's actually a lot easier to facilitate data sharing with the freestyle a break. So they've made it possible to share data from directly from the phone. So where has historically if you access the shared functionality on the app, it only pulled up Libre link which was sharing with other followers. So patient’s family members, friends. But now we actually have the capacity to be able to share data through the app to the clinic. So whether or not it's an invite from the clinic, patients can accept that, or they can connect to the practice with a clinic code. And I actually talked about how to access that clinic code in the presentation on the… the first presentation we did last week. So if you need to review that will be will be available in the slides. And the next slide just shows accepting that invitation from a practice. So if an invitation has been sent and a patient has been added from practice or clinic account, then you'll be able to… the patient will be able to directly accept that invitation and then share the data which is amazing. The other exciting addition was an AGP with daily views and some of that breakdown of time in range data and then the coefficient of variation. So that's all within the Libre reports view on the side of the clinic there. You can see an AGP report that pulls up this page. So these are two main addition since our talk last week and we just wanted to update everyone that.

So we'll move into our first polling question for our presentation this week. And so how do patients typically share their pump and CGM data with you at your place of work? So either pumping sense or uploading stations are set up for patients to use it a health care provider clinic; patients come in with printed pump and sensor reports; patients upload their pumping sensor data at home and we access this data through a pair of health care provider account and a patient account; or patients email their reports or allow access to the unique accounts by healthcare providers or it’s a combination of all the above. So I'll give you give everyone a minute to just go there and then we'll review the answers together. Yes, let's see what we have in terms of our distribution here. So it looks like for the majority of us 46% we're using a combination of all of the above. And it seems like from the choices, the most common is that we have patients upload pump and sensor data at home and we access this data through a parrot healthcare provider account and patient account. And then some distribution across the other answers as well. But the majority is having patients upload. Okay, so let's walk through that process together.

So the first technology we're going to talk about is actually Carelink and the Carelink platform. So, Carelink is designed for uploading of Medtronic pumps and products. The one we're going to focus on today is the 670g and to access that we would actually go through the carelink.minimed.edu web based platform and this pulls us up to a patient portal. And for the 670g, it's really important to remember that patients are actually in charge of uploading their data because they're using their Carelink personal account. They can also use Carelink personal to upload some of the other pumps, but today we're focusing specifically on the 670g. So important to remember that the Canada… our country, Canada selected as our primary and patients are required to remember their username and password. So it doesn't actually let you know if you log into Carelink if the account is locked, so you have three attempts to log in. After those three attempts, the account will be locked for an hour. So just a reminder to patients if you're logging in, make sure you know your information and a great way to do that, like we talked about last week is just having that information stored as a contact on a phone. If they forget and they don't know their information, best just to call tech support and have them reset the password so we're not having to deal with unsuccessful login.

So a few tips when patients are logged in, they have to upload their device. So to do that, they actually have to install the uploader which is, directions to do that on the main Carelink website. And it's actually right above the button that you would push to upload a device. So the first prompt is to upload or download the installer or the uploader sorry and patients need to remember to use their Carelink or sorry, their Contour 2.4 link meter. So that has to be paired to their pump. If patients need help remembering how to do this tech support can walk them through, but they can also contact their educator and we'd be happy to walk them through doing that. And to remind patients that they can keep their meter and pump close together when they're uploading to facilitate a successful upload. We’ll go to our next slide. I think one thing that we often forget is that a lot of it's very, it's common for us to do this in clinic when we're uploading for patients and so at home it's important to remind patients to actually upload their pump and pair their pump versus pairing their meter because if we just upload the meter will only upload those blood glucose data which will be visible in the log book. But if we upload the pump, then we'll have all that CGM and bolus data, which is really what we're looking for when we're reviewing that information. One other tip that I often remind patients and ourselves when we're uploading is sometimes we upload and then we go into care link and we see nothing and it could be as simple as perhaps the year on the pump is different. And we're uploading today, but the year is actually five years ago, we might not be able to have that successful upload and view that data. So always just reviewing that data in the pump as well. Another thing to remind patients and ourselves is, is setting up appropriate target ranges. So this is for reviewing our time in range information. We would want to be setting those report preferences to our 3.9 to 10 millimole target. And you can do that by accessing the preferences tab in the main menu of the Carelink platform. And just to note here, we've got a few other target ranges selected as well. Those could be your pregnancy ranges for instance, 3.5 to 7.8, so just know that you can modify these and that will modify how the reports are generated in terms of that reporting range and blood glucose.

So how do we actually download or what are we get downloading and this is an area for patients where you can select either bundled reports or individual report pages. This based on health care provider preference. So to add reports to a list. It's as simple as pressing the “Add to List +” button on the right hand side of the of the reports page and to specify the date ranges. So just below the specific reports pages, there is a date tab and you'll have to go through and select each of those individual dates to pull up that data once we hit generate reports and that generate reports date button will be located just on the right hand side of the screen for the patient and for provider, and that's what you'd hit once you're ready to generate those reports. Just below the main kind of bundled reports page, if we select the next slide, you'll actually see that this is where we can… this is where we can select…

Ilana Halperin: Sorry, Danielle and having technical problems. I don’t know why it's not advancing. Oh, here we go. Okay.

Danielle Goudge: So just underneath where those bundled reports were you can actually pull up those single reports and one of the ones that we commonly use when we're looking at our 670g reports is that assessment in progress page. And so this is where we can actually create comparison of data whether we want two weeks current versus two weeks before, it's important to remind patients to actually select that compare… compare progress from tab and then specify that range so that when we review the reports together. And we'll take a look at what those reports look, and we’ll look at that just in a second, you'll actually be able to compare that CGM data and assess and evaluate any changes you might have made for treatment plan.

So this is that pump report that's the assessment and progress page. We're going to take a closer look at this in a minute and Ilana is going to go through some of this information, but you can see you've got to the top upper portion of the screen there to see GGM tracing. If we did have this pulled up with the compare you’d have noticed that there would be two different colors, blue and orange to represent current and historical data and you can use this to compare and evaluate your treatment plan. There's also a lot of information around sensor usage in statistics. Specific auto mode exits are reasons for alerts and alarms. There's hypo and hyper glycemia pattern management assessment bubble. So you can actually take a look at where there might be some noted hypoglycemia or hypoglycemia, and that's drawing your attention there and then if we go into some of the other kind of pump reports that are available, you can select the meal bolus wizard report, it evaluates kind of that pre and post meal blood glucose. You can see that there's the sensor tracing data, there's also that median line there just to evaluate what blood glucose is going into meal and post at the two hour mark. You'll also have your weekly summary. So that would be this is just a snapshot of that. But you can see each day by day, the boluses, the carbs that are entered, and then if we actually go to the weekly report, or sorry, the daily report. This is, this is an expanded view of that kind of weekly snapshot, you can go through day by day and see specific alarms, again boluses, carb entries, and of note. I mean, this isn't. This is a report that's not for someone who's on auto mode. But this is just CGM and pump information together.

Ilana Halperin: Okay, so can everyone hear me okay. Okay, I decided to keep my camera off. Just because I don't know, there's some technical problems in my, on my computer. It seems to be a little overloaded. So hopefully my sound will be better if we don't also try to get to use a computer and use the camera. So just because we're not sure how many of you have actually had experience with patients on a hybrid closed loop system like the 670g, I was just going to spend a few slides reviewing that. So the 670g involves a obviously a pump which looks quite similar to the 630 but if you're not sure if what your patient has or not you'll see this little wave on the front of the pump and then you'll know it's actually able to be a 670g and then use auto mode. And then it requires the Guardian sensor and the Guardian link transmitter, as well as that contour next 2.4 blood glucose meter, all of which are connecting to support this hybrid closed loop algorithm. And so basal insulin when you're in automotive is being delivered every five minutes. And the current sense to determine five minute basal doses are. And so the basal rates that you would spend time her grabbing into the pump usually play ever used or in the predictive low glucose mode. And so what's really important to understand always targets a sensor glucose of 6.7 however you can put in a temporary target of a 8.3… or whatever is appropriate for use for up to 12 hours. And if the blood glucose is greater than 8.3 sensor blood glucose, the pump will ask for confirmation, finger stick and recommend closes above that. Determining the sensitivity factor. So it's taking that old 500 over totally dose and continuously adjusting the sensitivity factor, based on the adjustments of the total daily dose. Directly to give… value…

It considers active insulin on board and it's still required to count their carbohydrates to initiate meal bolus. If anybody… and Danielle. Let me know if there's any problems with my volume in this little message about low system resources again for audio quality. So if it's not working. Maybe we'll switch to you presenting the slides, but…

There we go, carb ratio and active insulin time are the only things that the diabetes professional team is really responsible for. This is just, … trial, though, with the 670g to get their approval, it will sort of before, after, or they had patients on a pump and CGM for a couple weeks. And it's about hybrid closed loop and … all of them for three months.

So very nice graphical representation of what happens to the variability in the time in range once you turn on auto mode. So the orange is the study phase and it's, you know, this is the average blood glucose is in the different groups, it creates, see, because you can appreciate that even in auto mode our patients who start to control, which would be the adolescence, still have more variability than the others. But we know that hybrid closed loop result in increased time in range, decreased time spend low and reduce variability. So here's another polling question.

What are the 670g targets when you're looking at that assessment and progress report that Danielle showed you what are we, what are we hoping to achieve? Sensor wear 90%, auto mode 85%, time in range 70%; sensor wear 85, auto mode 90, time in range 80; sensor wear 100, auto mode 90, time in range 75; or sensor wear 85, auto mode 80; and time in range 70. And if you're like, I don't even know what you're talking about, then feel free not to answer. Hopefully by the end of the talk, you will know what I'm talking about.

Okay, let's see what people think. Okay, good. So the bulk of people do know what I'm talking about. And it would make sense that, you know, to be in auto mode, you need to be wearing a sensor and it needs to be working. And then we're still targeting the same time in range, whether you're in auto mode or whether you're using, you know, injections and not on a multiple not on an insulin pump that time in range target for most people living with diabetes is 70%.

So we'll move on from there and kind of go over that a little bit. So, you know, the goal would be to get your sensor were greater than equal to 85%. I think it's easier to remember round numbers at 90%. And then we want our automode to be greater than 80% which should result in a time in range of 70%. A couple other things that I think are really helpful to remember is that basal total daily dose, so we do not want more than 50% of the insulin to be coming from basal in this hybrid closed loop system. If it is, and they're definitely not getting enough bolus insulin. And we know that the time spent low we're targeting similar to what we would, we talked about in the first talk, which is, you know, less than 4% in that mild hypoglycemia, and less than 1% and the more concerning hypoglycemia. So I'm going to take a moment to orient you to the assessment and progress reports. It is a little bit disarming because you just want to look right here and look at the picture. It's always where our eye goes first. But I actually encourage you to start in the bottom right hand corner and looking at statistics. And so this is an example where we have two different time periods. So we can compare the more recent time period with the more historic time period and look to see how much time our patient is spending an auto mode. Obviously the converses the manual mode. How much of the time are they wearing their sensor. You can get an estimated A1C here. I don't find the average blood glucose to be that helpful, but looking at how often they're calibrating is important because it is possible that you can… calibrate a little bit too much. And so three to four is probably the right number. And then looking at their total daily dose and what percentage of that total daily dose of insulin comes from basal versus bolus and you may recall it in the first talk that I mentioned how important that information is for me, even when I'm assessing patients who are not on a hybrid closed loop. I think it's a really important piece of information to help me decide where I need to make changes if it's in the long acting insulin or the basal rates or in the short acting or bolus calculations. Then you can move over to this tab and look at why are they not in auto mode. So what are the most common reasons that patients are getting kicked out of auto mode. And this can be really helpful to sort of start that conversation with your patients about the behaviors that are required to keep yourself in auto mode which I will speak to a little bit more on the next slide. And then we have our time in range graphs, which we're now used to seeing. You can see the difference between the two time periods. And then finally, I would go up here to look at where they're struggling to keep their time in range. So obviously this particular case the patient's doing excellent with a very nice flat AGP and you can also very easily see what the carb ratios are for the different times of day. So if it was obvious that there was a rise at a certain time of day, you can look there and be like, oh, maybe this carb ratio needs to be strengthened.

So now I'm going to steal liberally from my esteemed colleague and friend Bruce Perkins. This was recently published in CJD, it's an excellent article and takes you through some things to talk about with your patients when they're considering or already using but want to use optimized hybrid closed loop. So, um, it is not a cure. And I think most of us living with diabetes not living with diabetes, but working with people who are living with diabetes can appreciate that. Hybrid closed loop can certainly achieved decreased variability, decreased time below range. But if you're not changing your sites regularly and responding to the alerts and alarms from the sensor to calibrate, it's not going to work properly. So you still for this to work, you still need to maintain your pump. And you need to appreciate that the target is 6.7 and for some people that's a change. They were used to always waking up in the morning in the fives and now they're waking up at 6.7 and that's higher than what they're used to. But we're changing that whole concept of wake up with the blood sugar in the morning between four and six and then after meals eight to 10. Because a full range over 24 hours… 10.

So you can make 6.7 all day long, which even with those can do, then it doesn't matter if you're fastings are not five… will definitely be greater than… You have to trust the system to deliver the insulin. And this is sort of similar to the tip I gave last time, which is, be patient with your insulin. So you have to be patient with the algorithm. And really the time to really be focused on your diabetes is in that 20 to 30 minutes before you're going to eat when you think about how many carbs you're going to take and you think about pre blousing for those carbs. And then between meals, let the system do its job and if it needs you. It will let you know it will let you know that you need to confirm a high blood glucose and provide a correction. But otherwise, don't be too focused on what's happening because the system is doing what it, what it needs to do to bring your blood glucose back into target. And if you see that you're high and you feel like an urge to give a correction and you can't actually give a correction, you end up doing something called pseudo bolusing, which is trying to trick the pump into saying you're eating 15 grams of carbs, just so you'll get a little bit more insulin, then you're going to end up with a problem with stacking hypoglycemia, and more variability. So, as I already mentioned, what that means is that the main focus is accurate delivery of the food bolus and I remember hearing a talk about this before 670 came from one of the leads of Medtronic out of California. And when she talked about was that carb counting down to the gram is way less important than pre bolusing. And that was something that really stuck with me because I don't know how many of you have ever tried to carb count but it is so hard. And so you could be off by five or 10 grams. But if you take your insulin 15 or 20 minutes before you start eating, you're going to have, like, much less of the glycemic excursion then if you're right down to the one gram of carbohydrates but you take it while you're eating and that's just because we're delivering insulin subcutaneously and insulin, it's just not as rapid as we would have hoped it to be.

And then finally, there's a few subtle technical details that are worth mentioning about hybrid closed loop. The algorithm isn't great on the dawn phenomenon. And so, some patients who would have ramped up their, their basal rates sort of around three or four in the morning to prevent that will notice that they wake up a little higher because of that. But again, the, the algorithm is so good. The rest of the day at keeping you in in target that even though you might be a little bit higher when you wake up you're, you're able to achieve that 70 plus percent goal range of four to 10 so it should be okay. It's just important for people to know. The other thing is management of hypoglycemia. So when you do actually have a low in hybrid closed loop the insulin has already been pulled back significantly to try to prevent that low. So if you eat as much as you would have eaten in sort of open loop or manual mode, you will likely overcorrect. And the algorithm seems to work better with a shorter active insulin time. And that sort of, if you think about how corrections are determined between meals, insulin on board always takes into account your active insulin time. So if the patient say that they don't feel like they're getting corrected or they're spending too much time high between meals you may need to shorten that active insulin time to three and a half, three hours occasionally even two and a half hours depending on the patient. Finally, there is a temporary target of 8.3 which is helpful for exercise or other periods of time where you want to really avoid hypoglycemia.

So here's a case of mine from a real patient. One of my first patients on the 670g. And so, unfortunately, this is where I would love to go interactive with the audience, but it's a little bit difficult. So I'll just walk you through here that you know the patients in auto mode about 80% of the time with sensor were of 86% so that's actually really not too bad all things consider we're looking at 69 units of total daily dose, where the basal is taking up a little bit more than the bolus so that's already something that's sort of a flag for me in terms of what I'm thinking about how I might help her increase her time in range. And then when we look and see why is she getting kicked out of auto mode, it's often do to both high sensor glucose is and alarms. So we can appreciate here, she's not having too much hypoglycemia, but there's definitely room for improvement in her time in range. And then when we look at the graph, we can sort of appreciate where that hyperglycemia is happening. And you can see that she's got the same carb ratio all throughout the day. So if we look at that meal bolus comparison and you look at lunch, you can really appreciate that she's got a lot more hypoglycemia happening later in the day. It is important to recognize how confusing this is and I have complained about this to Medtronic before. Medtronic targets of post meal blood sugar of five. So when you see red in these areas, it doesn't mean they're actually less than 3.9 it means they're less than five. And so really for her, she's mostly high post lunch. And so what we decided to do at that visit was changed her insulin to carb ratio across the whole day to fix that basal-to-bolus ratio, but even strengthen that carb ratio at lunch even more and we shortened up her active insulin time. So same patient came back, comparing two different time ranges no difference in her settings. These are the new settings. But what she told me is the main difference between this 56 time in range and the 75% time in range was pre blousing. So once we got used to the carb ratios, the real difference here for her was taking that insulin 15 to 20 minutes before she ate, which really helped to break her time in range. So now I'm going to pass it back to Danielle to talk about our next system.

Danielle Goudge: So we're going to switch gears. Hopefully everyone can hear me okay. We're gonna switch gears from Medtronic and we're going to move into Diasend as our upload program. So diasend.com is where you'd actually go for other patients or clinicians. It's a tool used by both clinics and patients to upload pumping sensor data. We’ll focus more specifically on uploading to the T-Slim with basal IQ and integrated DexCom G6 today. When we start to look at some of the cases. And I think, you know, it's important to remind ourselves and patients when they're setting up their accounts. Obviously, they'll need access to a computer. So if patients don't have a computer to use at home. That's where they may move to more uploading at clinic and I'll talk a little bit more about that, but they'll create an account by visiting the diasend.com website. They'll be prompted to enter in their demographic data and then when they move through that account creation, if they’re a new user, they'll actually be asked for a clinic code. And so if a clinic or you know your area of work or place of work has a diasend account, you can actually share that information with the patient, so that when they're going up and going and setting up their account themselves, they can simply just enter in the clinic ID when they create their account and then they don't have to worry about having a paired account with a clinician moving forward.

So move to our next slide. So for patients to upload at home, they actually have to install a diasend uploader. Most of these platforms are very similar. So they access a website. But then they also are required to install an uploader, so for Diasend and it's the Diasend uploader. They can download this directly through the website, similar to Medtronic and Carelink and then once the driver is installed, they can easily upload their pump or CGM or meter. I think another way for patients to share their data is downloading their pumps and CGMs, meters again through a transmitter box and many clinics. Do you have transmitter boxes available to them. I think in this virtual time given the COVID-19, a lot of clinics won't be utilizing their transmitter boxes and most of this will be done through virtual uploads. So again, this would be just, if we're uploading a patient in clinic. But the information that's uploaded is linked directly to a patient. So if a patient's created an account and they've shared their data or linked their accounts to the clinic account, then every time they upload that information is stored in that cloud platform and it can be reviewed for future use. And patients can access their data at home as well as clinicians accessing the data in clinic. So once we actually have patients uploaded… we’ll maybe go back, back, just one, if we can, I think I animated this slide here. So once we have patients uploaded, we can view their data in one of two ways. So if they're uploading at the clinic, then you'll select the clinic tab from the clinic account and you'll be able to pull up or you'll notice under insulin pumps. There's nothing in this particular slide, but it would pull up a specific pump, either through serial number or a linked to a patient and you'd be able to select that view. And then if patients are uploading at home, we can advance the slide by one you'd actually just select the patient tab and then be able to search them by name. So this will pull them up specifically, and then all of that information that they've uploaded will be available to us to use in clinic.

So some of the examples of the different reports. I think one of the commonly used reports is the log book. So this just gives us an overview of any of those blood glucose data entries, boluses, corrections. You have a view of basal rates in this particular example, this person is actually on basal IQ, so you can see the pumps suspends with a little icon in the square there above the basal rates, but we'll go into this more specifically as we go through the basal IQ reports. Another one that's commonly used is the compilation tab. So this really provides us a quick overview in terms of sensor stats. So time in range, but also total daily dose for insulin average carb intake and underneath this main kind of hub of information you'd find more specifics related to basal-bolus split, which is something we often use and Ilana has mentioned and sensor stats. So this is a really great kind of first glance page to access. And then we'll go into the specific reports and this would be under a comparison. So this is your day by day view. And this is a person who's using basal IQ, so it will actually pull up their sensor tracing it will pull up their basal below and you can see the red bars for those suspends and Ilana is going to talk about this a little bit later, and just to the side of that you can see specific information around the suspends, the time, the duration, how many there are in a day and we'll break that down as we go through basal IQ. Another feature or another report is the AGP. So this is very similar to the DexCom report that we would get in Clarity. You've got your AGP  there and then more specific local metrics below so your time in range breaks down your coefficient of variation as well. So that's a nice kind of capture view of the AGP if you're just looking for that. So that's your AGP and then below your, your statistics. So we're going to shift gears and talk a little bit more about basal IQ and I'm going to leave it over to Ilana again.

Ilana Halperin: Okay, so I was checking out the Q&A and that people were still having problems hearing me, I'm really sorry about that. I don't know. Danielle. If you want to try taking over sharing the slides that will decrease the, I don't know, the demand on my computer, but just type into the Q&A box to let us know because we are going to switch back and forth one more time. So we can try to fix that if you guys think that would be helpful.

So hopefully by now you guys have had a chance to have one or two patients come into your office who's been using basal IQ. So basal IQ is sort of the G6 and Tandem pump integration that suspends before low. And really just to remind people that you know one of the biggest challenges we have is that hypoglycemia results in relative… hyperglycemia. But in the studies of the basal IQ when they turn that on, there was a 31% relative reduction in hypoglycemia without any change in the time in range or mean glucose, suggesting that there's no evidence of rebound hypoglycemia, which is really a move in the right direction for decreasing variability and increasing time in range. So this is a case of a patient I saw recently. She works at Tandem so she was very happy to share all the reports she thinks, she thought I needed with me by email ahead of time. So as you can see, this an AGP that looks quite similar to the AGPs we're used to seeing on our Clarity software with statistics along the bottom. So first of all, shout out to 83% time in range with only 2% low, which is really fantastic. But most of us can appreciate that there's a bit of a challenge with a rise in that sort of early morning hours. Which I've recently started calling, I don't know if I talked about this last time, the feet on the floor effect. Which a lot of my patients talk about now, it's not so much a dawn phenomenon, everything's great until you start moving around in the morning, even without eating less stress and anxiety of getting organized, even though we have nowhere to go these days, still seems to cause a bit of a jump up and blood sugars.

So this is a comparison of a day by day overview. Sort of similar to what Danielle show but obviously very squished together. What I think is nice about this page is that you can just appreciate how much of the time the glucose is in the target range, which is the green bar and how often these small suspensions of the basal delivery are happening. So there's not a single day that goes by where there isn't some pink and then you can appreciate the duration of the suspensions by how thick those pink bars are. I still always go to insulin pumps settings. I think it's really important. I always document the insulin pump settings in my EMR because you know pumps can fail and patients have called me at off hours to say what are my pump settings, I need to program a new pump. And so I still like to get these all this data documented. But in addition, you've also got the CGM settings right here so you can see when are the patients getting alerts and alarms. Where have they set it, and I want to appreciate again that this patient of mine was 83% time in range is getting an alert every time her glucose is above nine so she's very well controlled and being quite meticulous about her glycemic control. This is actually two pages of the compilation report placed it on to one slide. So if you're actually looking at the compilation report on Diasend or on a PDF this page, this part would be below it. I think there's a lot of very useful information on the compilation report as Danielle mentioned. And if I was only going to look at one in the context of Diasend, I think it would be this page. Because what you've got is your time in range, your average daily dose of insulin and if you want to get that breakdown of how much of it is basal or bolus you've got it here on the second page. You know how much carbs, they're eating. Sorry, I clicked that by accident. And then you've got another way of sort of looking at glucose, which is actually the glucose is that they put in which these days is not necessarily meter glucose. It could just be the DexCom G6 glucose that they're putting in, in order to calculate a bolus but that's why it's only 64% and 34 above because when you're high you're putting it in to get a correction whereas when things are going well, you're not necessarily putting a glucose into the pump, manually entering it so just explain why there's a discrepancy here where this is exactly on par with what you see over here. What I think is most interesting, because I’ve only had two or three patients who are actually already using basal IQ that have come into my office or shared their reports with me, is how much time we're spending and when I saw that, I thought what do we need to change. Do we need to reduce the basal rate lot of suspect it turns out that actually it's not on this is tour. So when you see red, you might get nervous. To be worse. Because as you can appreciate the glucose is still on target. Yeah, it's fun, it's skimming low, but it's still within our target range and so red does not necessarily mean hypoglycemia. So we really need to train our patients to focus on the CGM graph on their Tandem screen, rather than the red bars have suggested insulin is being suspended and look for patterns instead of focusing on the suspensions. Consider length and duration of suspensions and frequency and definitely treat lows more conservatively. So, similar to the what we talked about with Medtronic because insulin has already been pulled back, if you take a usual amount of carbohydrates that you would for a low in the old fashioned days you will overcorrect. And one of the main reasons people are still having lows with basal IQ is planning for exercise. So making use of that temp basal and it's really hard because exercise isn't always planned. But ideally setting a temp basal at least an hour before you start exercising or taking some extra carbs, meaning don't bolus for the full amount of carbs that you ate at lunch or have a snack 20 minutes before you exercise and don't bolus for it. Because the pump can pull back on basal but it can't pull back on a bolus that was delivered two hours ago that's already in circulation. And so those are the main times people are still having lows when they're on basil IQ. And this is just some real world data to show that my patients number of suspensions is actually quite in keeping with what we've seen both in the trials and in the observational data that people are having about five suspensions a day for about 15 minutes each time which totals about an hour and 20 minutes a day of suspension. But what I think it's so cool, is how the suspensions are stabilizing the blood glucose. So still very much in a normal glucose range for suspension and resumption.

And I don't think you can talk about basil IQ without talking about the excitement of control IQ. So although it's not yet approved in Canada and has been approved in the US and it's before Health Canada right now. So I just wanted to quickly talk about control IQ. So control IQ. It's a very different algorithm than 670g. And so as we get more comfortable, we're probably going to have to give different sorts of advice to different patients, depending on which hybrid closed loop system they're using. Just makes our job more interesting. So basically when the blood glucose is or between 6.25 and 8.9 it maintains the personal profile settings of the individual. So those basal rates that you programmed before they went into control IQ still really matter. And then when it drops below 6.25 it starts to pull back on the insulin and definitely stops once you're less than four. And above 8.9 it starts to increase the insulin, but above 10 it's going to deliver an automatic correction. This is what's different than 670g which still requests confirmation of the correction. Then it doesn't deliver an entire full correction. It gives about 60% of the correction based on your insulin sensitivity factor because of the fact that they've already increased the basal they I always talk about it like it's a little men in the machine, but the algorithm has already increased the basils. And there is a different target for sleep, which has a tighter range with more frequent basal adjustments and there is an exercise target which targets a higher range. So those are some things to focus on but remembering the differences is really when you're in target you're maintaining your active personal profiles. So this was the study, to get people into the study they did a two to one randomization. Sensor augmented pump therapy just means you're wearing a pump and a sensor. It was not control IQ, sorry, it was not basil IQ. And we did see a nice improvement in the time in range. Noticing it's 71% time in range. So these are not perfect devices. They're not perfect cures because really, you still have to be really good at counting your carbs and pre blousing in any of these hybrid closed loop systems. But we did see a significant decrease in the time span hypoglycemia in the control IQ group as well. So I'm going to pass it back over to Danielle and check in on our technology.

Danielle Goudge: So one of the other platforms are going to talk about today is Tidepool. And so just to get a general sense, we wanted to see who's actually using this software. So some of you might ask, Tidepool, what is this? Or our patients are using the platform but we don't access it in clinic or our clinic has an account but patients upload their data independently or our clinic has an account and we upload their data for them. So we'll just take a minute to see how people are utilizing tide pool. So we'll see here. So for the majority actually 73%, we're not necessarily accessing or utilizing Tidepool. So this will be an introduction, so to speak, in terms of what Tidepool is. I see that about 20% have noted that some of their patients are using the platform but it's not accessed in clinic or the clinic has an account, about 6% of respondees, the clinic has an account but patients upload their data independently.

So let's kind of go through what Tidepool is. It's another downloading software. It actually came out of the States. And if you want to access it here, The website is linked and when we make the slides available, you'll be able to check it out. But it came out of the States out of California. And so we'll just advance to the next slide. It's actually free for clinicians and patients, which is great. Just like any of the software we talked about today. It's a cloud based web application and they're also are mobile apps that patients can use too to pair and link their data. But it was created in the hopes that patients can upload all of their diabetes related devices, whether it's a pump CGM meter or ketone meter into one place. And their emphasis, Tidepool’s emphasis is really on patients owning their data and being the custodians of their data. The nice thing is it's compatible with most devices and they're always updating the list. And one of the advantages is if patients are using mobile devices, so, you know, DexCom Clarity, the information is automatically updated and uploaded to Tidepool as a software, very similar to how we would utilize Clarity for instance where patients who are accessing and paired with a clinic account, the information is automatically available to review in clinic.

So just advance to the next slide. So a lot of supported devices. Some of our devices that we've talked about today, you'll notice here, they mentioned, for instance, the OmniPod Dash. That's not yet available in Canada, but this is the list of supported devices because it is a coming out of California, the devices that are supported are obviously currently available in the US. But most of these are currently also used in Canada. So any device that’s approved by Health Canada is available to download on Tidepool. If you don't see the device there, you can actually simply just reach out to them and they can look at how we can download that information from the pumps or CGMs, whatever it is that the patients using that's not supported. So there are a few ways for patients to share data. And for those of you who are familiar with Tidepool, I think one of the most common ways is to actually have patients set up an account or a personal account themselves. So when you go to that website, patients can create their account, simply by clicking personal account. You have to use Google Chrome. So that's one thing to remember with Tidepool is that it's it really only functions with Google Chrome. Patients are responsible for uploading their data. And to do this, they have to install very similar to Medtronic or Diasend, they have to install the uploader. Patients can also send sharing request to providers and I'll walk you through how you can do how patients can do that in the next few slides. So when patients download the uploader, actually we’ll. Yeah, we can go, we can advance it will go back actually one sorry Ilana. Perfect. So once patients have installed the uploader they will see this type of screen and they can actually select their devices. So for instance, here I've ticked OmniPod and DexCom. And that will be stored and saved as kind of their favorite devices so they don't have to do this each time they go in to upload their device they can change at any time they need to, which is nice. So if any of your devices change along the way they can simply just update the uploader and then upload this devices when they need to.

So from a clinician standpoint, in terms of how to utilize Tidepool you can go about and set up a personal clinic account, so that would be you typing in your email setting it up so that that account is linked to a specific person or you can also create a general kind of clinic account that's shared amongst providers so you might have to talk with privacy in terms of utilizing this technology at your clinic, but you have the option of doing it both ways. And then we'll sneak to our next slide. So once we have a personal or clinic account set up for healthcare providers, like I mentioned previously, patients can actually send specific request to share their data with a provider or clinic and they'd simply do that by entering in the clinicians, or the clinic email to share their data with. The other option is clinics can also upload pump and CGM data for patients and that would simply require the install of the uploader on a clinic computer and then we can guide patients through uploading at the clinic, obviously something we're not necessarily using right now if most clinics have moved to more of a virtual based platform for patient appointments. So clinicians also have the option of inviting patients to Tidepool and actually creating their accounts or creating the general demographic information that's tied to that patient initially and then sending a request. So if you'll notice, just on the right hand side of the screen, there's an option to enter in a patient email when you enter in that patient email and save the page, it will prompt Tidepool to send an email to the patient, which will then prompt the patient to create their own specific account and once they've done this, they can they can take ownership of uploading the data from their pumps and CGMs.

So what are some of the reports and I guess advantages of utilizing Tidepool technology. So you'll find that the reports look very similar to either our Carelink or our Diasend reports, but what you see, first the first report that pops up is trends. So this is the default report with Tidepool. You'll notice there's the AGP report in the center screen. And you'll also see your time in range on the top right hand corner. Sensor usage are estimated, A1C and coefficient of variation is there as well. So this provides you know a quick snapshot in terms of information we might utilize to make changes to therapy when we're working with patients. The other nice thing is at the top, you'll also see the date. So Monday, Tuesday, Wednesday, Thursday, Friday and weekends. So for instance, you know, a person's weekday may look, may look different than their weekend, you can actually select weekends and just focus on a Monday to Friday or you can also just focus on Saturday, Sunday for someone who's working, you know, your standard Monday through Friday work week, it's a nice way to break apart some of that data and then at the top right hand corner just in that that box you'll actually notice that you can copy all of the sensor statistics as text. So if you're trying to enter in this information into your EMR, this is a really nice feature because it takes all of that information, converts it to text and you can simply just plug that in to your note moving forward.

Another report. So this looks… this is our stacked view of our CGM tracing and our basil-bolus information. So you'll notice you've got your CGM tracing at the top and your time frame at the very top of that screen. On the top left hand side you'll also notice this little sticky note feature. And if patients are utilizing the mobile app, so just there with the cursor, they can actually enter in specific notes so they can say, hey, I was feeling on well this was maybe more of a carb guess. And when they're utilizing that app. It'll just pop up on this reports. Here we don't have any specific comments. But if a person was using that feature, you'd be able to see it here. On the bottom of that bar, you'll notice the basal rate. So, these are the basal rates programmed in the pump and then in the center section under bolus and carbohydrates. If you actually click or hover over one of those circles, and if we advance the slide by one, you'll notice it pulls up specifics of that particular entry. So what was the ratio at this time, what's the target. What's the correction factor. What was delivered. Was there a modification made by the patient. Possibly because they were going and exercising. You can see the specific there of the carb entries or the corrections, which is really nice. So we’ll shift it over to Ilana again and she's just going to break down a case with Tidepool.

Ilana Halperin: Okay, thank you very much. What I think is so neat about Tidepool is how it can pull different devices together for us. So this is a patient of mine, recognizing that I am speaking about off label use of sensors in a pregnant patient. But when she first came to me she was at 8 point. Sorry. She was eight weeks pregnant with an A1C of 8.5, feeling very anxious because it was somewhat of an unplanned pregnancy, she planned to get her blood sugar sooner better control but it was a bit of a surprise. And she was using an Animus pump and a Libre. And she she brought it. She had taken the time before she met me to download and she was the first one to add me Tidepool. So she sent me an invitation. I created an account from her and now I have five patients that I follow through Tidepool. So it's very much a patient driven patient motivated type of thing. But the more I use it, the more I appreciate it. I just learned a few things from Danielle, the copy and paste into the EMR is amazing. So you know, I feel that, with time I'm going to be recommending this software more and more to my patients. And so at the time, although the A1C was a 8.5, you can appreciate already that she was doing better with 65% time in range. One thing that I haven't yet figured out on Tidepool, is if it's possible to change these ranges, these seem fixed. And so again, when I am falling pregnant patients that can be a little bit tricky because I do try to focus on that tighter pregnancy rate target range of 3.5 to 7.8. Obviously, we had a concern here with how much time she's spending low. Again, one of the challenges I do experience with my patients with Libre is is to find out from them how much of this they feel is a true low or some of those Libre lows where you know there's 3.6 on the Libre, they double check in, or 4.5 on their finger poke. But you can appreciate hear her total daily dose, very insulin sensitive, total daily dose of 13 units. Split pretty much half and half, but a little bit more basal and maybe not eating quite enough carbs for pregnancy but that's not an uncommon thing that we see in our patients. And then there's a different tab, but I've just put it here for you now to show the profile settings. And so although it says Tandem up there. I think this might have just been as I, for the purposes of the case I captured this later. But you know you can appreciate all of your settings across the top there.

So now we're going to show you. Yeah, I think pretty much the same thing but including in it, you're AGP. So it is a little different than what we're used to seeing. But the green boxes would be your median blood sugar. The dark grey, fiftieth. Light gray, 90th. White area 100% of the blood sugars around that area. And when you use this interactively on the web and you sort of hover over things you can actually trace the blood glucose of that particuar moment. So you can really appreciate where the variability is coming through and how on one day they might be high and then low within two hours of the… you could almost get the dailies on the same page. So this is her picture and you can appreciate that this is low. And so you can see how much of the time she's spending low here. Everything below that white dotted line. I hope that reflects okay. So our last polling question for that for the session is what can we do to increase her time in range. Should we strengthen her carb ratios, should we reduce her basal rates, should we focus on pre bolusling, or should we increase her carb intake? All things that could be the right answer. And as we know it's diabetes. There's always more than one right answer, but what do people think? Okay, let's see what people said.

So focus on pre bolusing. So everyone's been listening to my message, which is to focus on pre bolusing. And generally, I would agree. However, in this and we definitely, I always talk about that, especially, there's some really interesting information out of the UK, Helen Murphy's group showed that as pregnancy progresses pre blousing is needed even more. Insulin absorption slows down. We don't really know why, but some women with type one diabetes in the third trimester after pre bolus an hour in advance to prevent postprandial hypoglycemia. But we'll just close that please. And so what I did do in this case was actually drop all of her basal rates. So I thought that her basal/bolus ratio is off, causing a lot of lows and rebound highs. And that's something you can only appreciate from the dailies and talking to the patients. But in the first trimester, we often have to scale back on insulin, even though the A1C is high. That's already old information and things change very quickly so often we're decreasing insulin in the first trimester because hypoglycemia and hypoglycemia unawareness is a big problem. So this is later in the pregnancy, she did both. Sometimes she downloads to Tidepool. Sometimes we would just look at Clarity, because she got onto the DexCom G6. And you can appreciate that she's put all of these things into play. So she's not… she's pre bolusing. And we've got a good 60% of her insulin from basal and like, look at that. It's so flat. It's so amazing. But yet only 46% time in range. And that's because of this 3.5 to 7.8 target that's incredibly difficult to achieve. So sometimes, just to make them feel better. I go back and I change it to eight. So it's a lot better. I realized, we're sorry about the tech problems today. But one thing that I want to mention that you know, they're using the looping system to create…

And then those patients, I learned so much from. So I just wanted to say, you know, we can't ignore this is this in growing segment of the type one diabetes population. And actually, as we have more information in the sort of regulated hybrid closed loop systems. I feel like I have more useful tips to give to my patients, even in these Do It Yourself algorithms pre blousing,  carb counting that hasn't gone away because the insulins just aren't fast enough. So I'm very open to my patients to taking control of your diabetes and using whatever technology's makes sense for them. And I'm happy that they want to come and get my advice and share their learnings with me. And so that's generally my approach.

So since our last talk till now, we've had the announcement that the camps 2020 has been canceled due to the COVID-19 pandemic. Which is pretty devastating for me and my family and for kids living with diabetes across the country. But let's hope we all get through this together and come out stronger with a great season in 2021. I think we're okay to run over with a few questions. So why don't I start with one for you Danielle. Are you using Tidepool for a patient that's only using a blood glucose meter.

Danielle Goudge: So I think the answer can be yes. I don't have any patients who are currently doing that for our clinic and I think part of that is just because when we think about privacy piece and getting approval in terms of the hospital just being on board with using the technology. I think that's something to consider as well. But the nice thing is you can, so patients have access to a computer and they have the cord that corresponds with their meter, they can easily plug that meter in without having to kind of snapshot screenshots of their meter or review verbally those blood glucose with you. So know that you can utilize the technology simply with meters, as well instead of using pumps and CGM.

Ilana Halperin: So another question was do you need to take Fiasp 10 to 15 minutes before bolus? I think it's really individualized by the patient. I'm certainly I think when I recommend Fiasp, it's usually for patients who really struggle to pre bolus or even taking their insulin after their meals, which causes wild spikes and their blood sugars. But slowly, we move forward in that time frame as they're comfortable and I certainly do have patients who still need to pre bolus with Fiasp by 10 to 15 minutes. So it's very individualized that the amount of time that people need to pre bolus is an individualized thing. And it actually isn't always consistent within one patient from day to day, because it depends on whether they've exercised and of course the types of food they're eating. You know, we talked about pre bolus thing, but obviously if you're having a very fatty meal, the extended bolus can be quite helpful. And so maybe you need to pre bolus but only for 10 or 15% of it and use the extended bolus for the bulk of what you're going to eat. So I hope that's helpful.

Danielle Goudge: Tidepool and trend arrows. And I actually don't know the answer to that. So let me look into it and I can post a response on the TimedRight page.

Ilana Halperin: And then finally, there was a question about DIY and looping. So I have yet to have a patient to DIY looping and pregnant, although I have one who's hoping to get pregnant. So I don't have really any practical advice, except that pregnancy is so tough. Type one diabetes is so tough. And so my goal is always to work with the patient and find a solution that works really well for them as opposed to telling them what they can and can't do. I now have two patients who have a 670g who've become pregnant and they're mostly not using auto mode because they find the 6.7 target too high. I know automode, it's not approved for use in pregnancy but as I mentioned, I'm talking just from clinical experience. I have one patient who actually likes auto mode overnight. And other patients not using auto mode at all. So very individualized, continuing to experiment as is usually the case with diabetes in general. Life is with diabetes is a bit of trial and error to figure out what works for you until it's finally working in and something changes in your life. And pregnancy is just that experience, kind of like in fast forward mode because hormones make things changing all the time. And yes, the audio recordings will be posted as will a cleaned up and finalized version of the slide deck as a PDF for both of our first and second sessions together. So thanks everybody for joining us today.

Danielle Goudge: And you can email us, I guess if you have any questions or emails are there.

Ilana Halperin: Definitely always happy to answer questions.

Danielle Goudge: Have a great weekend. Everyone

Ilana Halperin: Take care, stay safe. Bye.