Scott Benner 6/8/24 Scott Benner 6/8/24

#1219 137 and Counting

Erin is 44 years old and was diagnosed last year after being misdiagnosed as a type 2 since 2014.

You can always listen to the Juicebox Podcast here but the cool kids use: Apple Podcasts/iOS - Spotify - Amazon Music - Google Play/Android - iHeart Radio - Radio Public, Amazon Alexa or wherever they get audio.

+ Click for EPISODE TRANSCRIPT

DISCLAIMER: This text is the output of AI based transcribing from an audio recording. Although the transcription is largely accurate, in some cases it is incomplete or inaccurate due to inaudible passages or transcription errors and should not be treated as an authoritative record. Nothing that you read here constitutes advice medical or otherwise. Always consult with a healthcare professional before making changes to a healthcare plan.

REPLACE

#1218 I Have Vision

Donald is a software engineer who has type 1 diabetes. We talk about technology and how it may be valuable for people living with type 1. - I HAVE VISION

+ Click for EPISODE TRANSCRIPT

Scott Benner 0:00
Hello friends and welcome to episode 1218 of the Juicebox Podcast

Donald has type one diabetes here software engineer and he is a person who embraces new technology. So today we're going to talk to him about looping, different things that you can do for your diabetes and something called Vision AI. Nothing you hear on the Juicebox Podcast should be considered advice, medical or otherwise, always consult a physician before making any changes to your health care plan. When you place your first order for ag one with my link, you'll get five free travel packs and a free year supply of vitamin D. Drink ag one.com/juicebox US residents who have type one diabetes or are the caregivers of someone with type one T one D exchange is looking for you go to T one D exchange.org/juicebox. complete the survey and you will be helping with type one diabetes research. All you have to do is complete that survey your answers to easy questions stuff that you already know. We'll help to move type one diabetes research forward T one D exchange.org/juicebox. If you're looking for community around type one diabetes, check out the Juicebox Podcast private Facebook group Juicebox Podcast type one diabetes this episode of The Juicebox Podcast is sponsored by touched by type one find the Met touched by type one.org on Instagram and Facebook and in person sometimes, as a matter of fact, I just committed to their next in person event. Find out more at touched by type one.org. This episode of The Juicebox Podcast is sponsored by the ever sent CGM and implantable six month sensor is what you get with ever since. But you get so much more exceptional and consistent accuracy over six months, and distinct on body vibe alerts when you're higher low on body vibe alerts. You don't even know what that means to you ever sent cgm.com/juicebox Go find out. Today's episode is sponsored by Medtronic diabetes, a company that's bringing together people who are redefining what it means to live with diabetes. Later in this episode, I'll be speaking with Mark, he was diagnosed with type one diabetes at 28. He's 47 Now, and you're going to learn a little bit about him in just a moment. To hear more stories from the Medtronic champion community or to share your own story, visit Medtronic diabetes.com/juice box and check out the Medtronic champion hashtag on social media.

Donald 2:50
Hey, I'm Donald. I've been a longtime software engineer, programmer, technology enthusiast was diagnosed with diabetes type one diabetes a little less than 10 years ago, started fully looping and, you know, really getting into all the tech and open source and all that good stuff. A few years after being diagnosed, so I've definitely gotten familiar with all that. Also listener to the pot. So I'm excited to be here.

Scott Benner 3:20
I'm excited to have you I appreciate you listening. 10 years ago. How old were you?

Donald 3:24
Oh, man, the math. I was about 18 You're 28 now? Yeah,

Scott Benner 3:30
it's not a lot of math. That's fair. So at 18 years old, you were diagnosed? You're 28. Now, is there any type one in your family tree?

Donald 3:40
No, that's the weird thing. We closest we had was type two, grandma, but no type one.

Scott Benner 3:46
Okay. How about other autoimmune for you? Other autoimmune?

Donald 3:49
Yeah. No, nothing really out of the ordinary. Okay. Celiac?

Scott Benner 3:54
Thyroid, anything like that in the family? None of it. Interesting. Yes. Odd. Were you sick before your diagnosis? Do you have any kind of virus? coxsackievirus? Anything like that?

Donald 4:08
Yeah, I don't think it was anything specific. It was the first year that I was in college. And just like, you know, the first few years, or I guess first few months that you're in a college dorm and just, you know, run a bunch of guys. And you know, the hygiene is definitely not kept up from the others. I I feel like I was pretty solid, at least. Yeah, I think just getting sick a lot from that. And the flu. Just stress being away from home caught up. And yeah, I think that's what did it.

Scott Benner 4:34
You were ill a number of times in college. Yeah, just

Donald 4:38
you know, like, flu colds and things like that. Was that uncommon for you? It was it was more frequent. You know, I don't think it was anything that hadn't happened before. But it was definitely more than an apt app and you know, before that, okay,

Scott Benner 4:53
that's interesting. Let's talk about how so you said you're doing what are you looping now? I am looping into z classically, enthusiasts, Stickley looping could actually be your episode title. Good one. But but that's not how it started, obviously. So when you were diagnosed, what did they give you? How long did you do it for?

Donald 5:11
Yes. So when I was diagnosed, it was actually, when I was I remember like, so clearly, I was on winter break. And I was, you know, hanging out with my parents were eating some good Chinese food. And my mom noticed that stinky just kind of water. So eventually that got me over the hospital. It was my family doctor, I see my whole life he was, he was actually pretty pretty with it and gave me like a really good overview of things. Like, I feel like if you have to tell someone like, Hey, here's his whole life, disease, you're gonna have to deal with your whole life, I'm only gonna be able to talk to you about it for a few days. So here's the Crash Course. So he did a good job with that. I was started off on, you know, the manual injection pens, pricking my finger, I think it was Novolog I did that for you know, even a few years after I got to college and got another endo who's really awesome. You know, it's okay, you know, your, your tech guy, you can upgrade to a continuous glucose monitor and insulin pump. But I was still a little in denial. You know, I was like, Oh my gosh, this thing's always gonna be on me. Yeah, that makes it permanent. What if somebody sees it? You know, I'm in college. So I need to, you know, try to feel cool and everything, but ended up getting all the tech and, you know, everybody else thought it was cool, too. So it was a huge quality life improvement.

Scott Benner 6:34
Download what for your friends into that they thought your insulin pump was cool.

Donald 6:40
That's a good question. I don't know. But I felt good about it.

Scott Benner 6:44
I feel like you're telling me that everybody at the d&d game liked your insulin pump?

Donald 6:49
Where you just call the

Scott Benner 6:51
did I get close? Oh, no,

Donald 6:53
actually, I was on like, both sides of the spectrum. I was president for our packing club at my university. And then also in a fraternity. No kidding. So I got to Yeah, I got united both. And it was pretty, it's pretty accepted in both spots

Scott Benner 7:10
about that good for you. The hacking club. What did that consist of?

Donald 7:15
It consisted of, you know, a couple of late nights, and pizza, making some really cool projects. Yeah, we actually had a hackers club. And usually, you know, when most people hear hacking, you think of some guy wearing a hoodie. And you know, he's trying to break into the government's computers or something. How we use the word hacking is for building. So you know, one example is, you know, we might hack together a app that lets you connect to your window blinds, so that, you know, at certain times of the day, you can have that automatically open your buy blinds and close them and, you know, just whatever cool projects, we felt like making it actually one of my early ones was a really basic way to track the last finger I had picked. So you know, it's kind of like, okay, you know, need to cycle through them evenly. And as a fun way to make that kind of stuff easier.

Scott Benner 8:14
So it just kind of kept track of your, your finger picks. Yeah, like, I

Donald 8:19
think I also used it for my dosage calculator, because I didn't like any of the ones that were out there that I found at that point. Interesting. So, so I think it knew, you know, oh, you had just entered your, you know, you just put your finger to get your glucose and use that to Bolus or something. So I'll rotate you know, the next finger you need to go to

Scott Benner 8:40
interesting. That's really something. Have you made other things for yourself that you want to use personally since then? Or is your work been mostly, you know, related to your job?

Donald 8:51
That was towards the end of I did you learn that there was a competition? Competition is probably not the right word. There was a open kind of like bounty where someone said, hey, you know, we know sometimes some of the diabetes sharing servers from popular apps can go down. So you know, we'll pay someone $500 If they can send a notification whenever it was, like the Dexcom, clarity sharing servers went down. So I put it into submission for that someone. I think someone else beat outline, though. They got it in earlier. But yeah, yeah, I remember putting that together.

Scott Benner 9:29
Donald's like, it's not that mine didn't work. It just has got there first, that's all. Yeah,

Donald 9:34
yeah. Mine was like, you know, kind of cool, I guess, I guess isn't really cool.

Scott Benner 9:38
too interesting. Really is. So growing up with type one really happen writing college for you? It did, yeah. And so we're your parents probably weren't involved. Or

Donald 9:50
were they? No, they were not. They were like, very hands off. You know, I remember the firt the start that I had going back To My doctor was he'd written all my formulas on a, on a little index card. Explain what sliding scale was. And like, my parents were like, well, you're better. You're better at math than we are at this point. So we're going to let you roll with it. And it worked out pretty good. Nice.

Scott Benner 10:17
Did you have any? I mean, any odd feelings about that, like just going off to college, but this whole new thing, or did it feel very doable to you? Right now we're gonna hear from a member of the Medtronic champion community. This episode of The Juicebox Podcast is sponsored by Medtronic diabetes. And this is Mark.

David 10:36
I use injections for about six months. And then my endocrinologist at a navy recommended a pump.

Scott Benner 10:42
How long had you been in the Navy? Eight years up to that point? I've interviewed a number of people who have been diagnosed during service and most of the time they're discharged. What happened to you?

David 10:52
I was medically discharged. Yeah, six months after my diagnosis.

Scott Benner 10:56
Was it your goal to stay in the Navy for your whole life? Your career was? Yeah,

David 10:59
yeah. In fact, I think a few months before my diagnosis, my wife and I had that discussion about, you know, staying in for the long term. And, you know, we made the decision despite all the hardships and time away from home, that was what we

Scott Benner 11:13
loved the most. Was the Navy, like a lifetime goal of yours?

David 11:17
lifetime goal. I mean, as my earliest childhood memories were flying, being a fighter pilot,

Scott Benner 11:22
how did your diagnosis impact your lifelong dream?

David 11:25
It was devastating. Everything I had done in life, everything I'd worked up to up to that point was just taken away in an instant. I was not prepared

Scott Benner 11:34
for that at all. What does your support system look like? friends,

David 11:37
your family caregivers, you know, for me to Medtronic, champions, community, you know, all those resources that are out there to help guide the way but then help keep abreast on you know, the new things that are coming down the pike and to give you hope for eventually that we can find a cure.

Scott Benner 11:50
And you can hear more stories from Medtronic champions, and share your own story at Medtronic diabetes.com/juice box. This episode of The Juicebox Podcast is sponsored by the only six month where implantable CGM on the market. And it's very unique. So you go into an office, it's I've actually seen an insertion done online like a live one like, well, they recorded the entire videos less than eight minutes long, and they're talking most of the time the insertion took no time at all right? So you go into the office, they insert the sensor, now it's in there and working for six months, you go back six months later, they pop out that one put in another one, so two office visits a year to get really accurate and consistent CGM data that's neither here nor there for what I'm trying to say. So this thing's under your skin, right? And you then wear a transmitter over top of it. Transmitters got this nice, gentle silicone adhesive that you change daily, so very little chance of having skin irritations. That's a plus. So you put the transmitter on it talks to your phone app tells you your blood sugar, your your alerts, your alarms, etc. But if you want to be discreet, for some reason, you take the transmitter off just comes right off no, like, you know, not like peeling at or having to rub off at he's just kind of pops right off the silicone stuff really cool. You'll say it. And now you're ready for your big day. Whatever that day is, it could be a prom, or a wedding or just a moment when you don't want something hanging on your arm. The ever sent CGM allows you to do that without wasting a sensor because you just take the transmitter off and then when you're ready to use it again, you pop it back on. Maybe you just want to take a shower without rocking a sensor with a bar of soap just remove the transmitter and put it back on when you're ready. Ever since cgm.com/juicebox You really should check it out

Donald 13:46
you know I think I did really walk out from you know some people have probably already heard of the honeymoon phase you know where you're still getting your your pancreas is still putting in some work but you need to kind of help it out with some insulin in the back. So I remember going into the dorm rooms and being able to just crush a ton of cookies and I was like yeah, this seems like the right amount of insulin and they matched up pretty well like as I got diabetes things pretty easy.

Scott Benner 14:12
You think you're a honeymoon maybe?

Donald 14:15
Yeah, I was definitely honeymoon and yeah. And so then after I came down from that it was you know, like, oh, you know, I need to be careful whenever, you know we go out for some drinks because then I learned you know, your liver can't put out glucagon to bring your blood sugar back up. So you know how to learn the hard way from some scary episodes never end up in the hospital fortunately, but I can't say that was up to you know, that was from my own skill or anything.

Scott Benner 14:42
Yeah. Can you give me an example of a scary episode?

Donald 14:46
Yeah, no, this is like a typical night out and we were on our way back and I was not really all the way with it. Probably more from the drinks than anything I had already gave you know all Have my fraternity buddies a good good lowdown of hey, you know, if I ever need anything, here's how you can prick my finger. Here's glucagon, you know, here's some Gatorade to keep under my bed. Here's the signs of hypoglycemia and everything. I let them know. And I got back and I think it had dipped down to like 38. And they were helping me take down some Gatorade and everything, but just

Scott Benner 15:22
you and a bunch of drunk 18 year olds taking care of your 38 blood sugar.

Donald 15:26
You know, I'm sure from like, an outside perspective, it was a beautiful thing at the moment. Yeah, yeah, I it was, it was definitely scary. Well,

Scott Benner 15:35
you know, seriously, letting other people know about your situation was huge for your property right there. Yeah, absolutely.

Donald 15:42
And, you know, it's like, I if I wasn't doing all the other right things, I guess that was one of them. For sure.

Scott Benner 15:49
Well, let me ask you this. What were your agencies like in the first couple years? Oh, jeez.

Donald 15:54
I think the first one when I came in, after being diagnosed was 10 and 11, or some, something pretty high up like that. And then I think my average glucose, you know, I might have lucked out with keeping it under 200 in the honeymoon phase, but it was probably averaging out to like, you know, 190 through 250. Back, like, you know, you know, if, if it really wasn't going well, yeah, I think I think once I had figured things out a bit more, back then. No, probably one ad was like, you know, doing good.

Scott Benner 16:33
Okay. How long do you think it took you to get to that level?

Donald 16:37
A couple years? Yeah.

Scott Benner 16:39
And what about those years? Were they maturing years? Or were they experienced with diabetes? Until you understood it years?

Donald 16:47
Maybe a blend? Oh, man. Yeah, you know, it probably was a blend, you know, kind of going back to feeling a little bit of denial and overconfidence from the honeymoon phase, you know, didn't think it needed a ton of mental effort or just, you know, much thinking from my end. So, I think after I kind of got through that, and just matured up and it's like, okay, you know, this is something that takes effort. This is something that's really important. That's when things got a little better.

Scott Benner 17:18
Okay. Can you tell me what made you feel like it was important to understand? Do you remember, like an aha moment? Or what made you think this isn't good enough? I need to do better.

Donald 17:32
You know, I think part of it, there's actually a programmer is a type one diabetic from Microsoft. And I'm just really, really, really ashamed releasing my inner nerd here. I think it's Scott Hanselman, on Twitter. And he's really awesome influenced through that stuff. And I remember seeing him, you know, some really good consistent glucose numbers. I think he keeps it, you know, like, pretty consistently around 90 or something. Or he showed in some like other graphs, and I was like, Oh, my gosh, like, I know, my endo had said one thing, you know, but I can see someone else actually hitting that, I probably need to do that. So that influence made a big difference.

Scott Benner 18:12
Someone you had respect for and knew about was having outcomes that you saw is hopeful. And you thought if they can do that, I could probably do that as well. Yeah, definitely. Interesting. That's great. I know Scott's name by the way. I don't know if I've ever met him. But it's interesting that you said somebody else like, Oh, I know that name. It's

Donald 18:32
not that I know him personally, either. But by now it kind of feels like it.

Scott Benner 18:35
That's something he's been added a long time as an ad on Twitter. Yeah, yeah. I mean,

Donald 18:41
even with the diabetes scene, I remember seeing him posts about it was like a newspaper clipping from the 90s, or the 80s, when he was in college, and he was doing a little bit of hacking on his PDM to, you know, get it to do some more cool stuff. Amazing.

Scott Benner 18:55
So you're using what at that point, you're pumping and do your work? What pump was it?

Donald 19:03
Oh, I was bravery for me at that point was upgrading to the continuous

Scott Benner 19:10
glucose monitor. So you're MDI with the CGM? Yep, manual injections with the G six, okay, five, six, whatever it

Donald 19:18
was back then. But it wasn't until I got out of college that I actually began looping to Omnipod didn't have to in which was a big win for me. And I think by this point, I heard about the looping scene. And I just saw that I could, maybe it was just the app interface or something. But I was like, Oh, I'm not even gonna bother with the, you know, like the factory app. I'm just gonna use this like open source ones. I think I

Scott Benner 19:47
do it right. From MDI, to looping. Yeah, yeah. Good. Yeah, that's amazing. Well, also it shows that you weren't, I imagine put off by the technical aspects of setting up loop that probably seemed obvious and Easy to I would imagine, oh

Donald 20:01
man, you know, that's a funny thing. I mean, you know, my full time job is setting up technical projects like that, you know, you have to download all this stuff you have to get it set up and then you know, put it on your phone. But it was still a challenge is such a good community that it gets past all those technical issues, which is super incredible. You know, just all the looped Facebook groups and everything else that's out there all those people are making a big difference right? know for sure. That was actually part of why I ended up going from those manual setup steps to the T one Powell service where they just do it for you. Okay.

Scott Benner 20:42
Oh, so Oh, hold on a second. Done. This is strange. I don't usually record this late at night. But my my, like, chameleons getting automatically missed it. So let's hold on a second. Well.

Donald 20:55
That's awesome. You're chameleon.

Scott Benner 20:56
Yeah, I have a chameleon behind me in my office and his tank gets doused right before his lights go out. And it's on a timer. So I didn't Can you hear it? Yes. Okay. Yeah, let's change

Donald 21:09
my insulin pump. It was I usually always do it at night. But there is I just got back from that trip. And I had to do it at a weird time. So my I changed my pump like right before, otherwise, I would have went off.

Scott Benner 21:21
Yeah, so he likes to sleep with more humidity. So for about a minute before the lights go out, the tank gets missed it very heavily, gives him one last chance to, you know, get a drop of water if you want, helps the plants grow, etc. It also helps when later tonight, I can't believe I'm saying this, you thought you sounded like a dork Hold on a second. Wait till later tonight, when fogger that I made myself will come on and bring up his humidity to like 100% for about six hours overnight. Because he actually hydrates through breathing the air. So he doesn't really drink like you would imagine animal drinking out of like a basin of something like that. So he gets all of his hydration through humidity overnight. That's awesome. But no humidifier that you can buy can handle that or holds enough water. So what I have here is a 10 gallon dog food container. So it's, it's opaque so the sun can't get in. So we don't have any problem with getting you know the waterjet and funky, then there are two holes drilled in the top of it. One is for a low speed fan and one is for a three inch tube that goes right up into the bottom of the enclosure. So then there's Jesus guys do sound like an idiot. This is cool. So two inside the water is a UV bulb that keeps algae from growing. And there is also a bubbler. So like if you ever saw a pond that has like fog going across the top of it. So that comes on a timer. It starts to make fog inside of the container and then the air pushing through. It's actually electric the way it does that. It's crazy. It vibrates and heats. And it makes it into fog. It's really interesting. Well, that's so cool. I want to let people know I got that disk from the house of hydro.com. The guys fantastic if you have any needs and then the fan comes on pressurizes the tank and then the the fog escapes through the tube and fills the the enclosure. That's awesome. Yeah. Anyway, I

Donald 23:28
want to make one now just to have a fogger I'm the man oddly proud of it. eautiful long dwell and in our hacker clubs,

Scott Benner 23:36
I would have thought okay, so because they you know, when you set these up, I must have bought four different humidifiers and return them. Well, every time I got one, I was like this won't do it. It's not set up correctly. It doesn't hold enough like blah, blah, blah. It needed to be cleaned too often. You know, there are all kinds of different problems. And one day I just sat down I was like, I'm just gonna make one. Then after I did it, of course, I found somebody online who's like, you should just make one yourself. And I was like, Where was this information? While I was beating my head. I felt like I reinvented the wheel. And apparently I was like three years late. But anyway, I love it. That's

Donald 24:09
that's cool project. Yeah.

Scott Benner 24:11
So okay, Donald. So, when we're done, I'll send you a picture of of Durbin, which is my chameleons. They love it. Yeah. Anyway. So what I like here is that you go to move to loop. And even with your extensive background, which I'm going to ask you about in a second, it still was daunting enough that you reached out and said I need to get somebody else to do this for me. You're not fantastic nowadays, can you do it yourself? Are you still paying someone to do it for you? Well,

Donald 24:41
you know, and I'd even say I did it for myself for a while. I think it was when ironically it was when Heroku had their plane their plan change where you'd have to upgrade from the free

Scott Benner 24:54
seminars a month now or something like that, right? Yeah. And

Donald 24:58
that was ironically I was working at Salesforce at the time who owned Heroku. So I was hearing about some of those comps all well received stuff too, which was, which was good to see. And fun to see like who is in the diluvian community inside of Salesforce? Yeah, I was able to set it up. It was just real pain. When the upgrade happened, I was like, okay, I can do all this stuff, but I just don't want to spend my time doing it. And so when there was messaging out from I can't remember where, but they mentioned T one palace service, they can just set it up for you. So like, yeah, that'd be great. And I pay someone to do that. I

Scott Benner 25:34
actually have a guest on the show who works there. If you ever go find type two stories, John. I think John. Yeah. John, I

Donald 25:43
guess to add on to the history. When I on boarded, I had met Ben West. I was like, Oh, my gosh, you know, you helped me connect Scout that is ultra cool. So from there, I actually did a little bit of work with them to just kind of help out on the few things. And, you know, I tied it up that that just keep working on some other projects, it was a really awesome way to hang out with all of those guys. They're just super brilliant. And, you know, it was fun to work with them for a little bit. Yeah,

Scott Benner 26:14
I hear that. Okay, so you're using loop and you're going along? Does your does your diabetes get easier to manage? Does your agency come down? Does your time and range get better? What do you learn from being automated like that?

Donald 26:28
You know, it was really cool. It was my fiance actually noticed the major change was that I wasn't going low at night anymore. You know, I'm trying to think back to like, you know, what was really causing it, I'm sure I was still just trying to figure things out. You know, just having that that Basal be able to change in I take care of you when you're sleeping? Oh,

Scott Benner 26:52
yeah. Oh, Donald, listen, if if all an algorithm did was let you sleep better, I think it was amazing. You know, the rest of it's a bonus in my eyes. So that's amazing. So okay, so if you were Lowe's overnight, it's a good experience for you. Now, this is a little like, a weird minefield for me to walk through here. So I'm going to try to give a little bit of background, you're going to fill in some gaps that I don't understand. And then we're going to talk about something completely different than what we've been talking about before. Okay. Now, what I have to say is, is that oddly, I find myself in a situation where I have a financial stake in vision AI. And I've never had anything on the show before where I have a connection like that. So I don't really know the right way to handle it, other than to just say out in front, that this is true. Now, having said that, I have made no money from it whatsoever. And it would probably only benefit me if the company were to get very popular, or get sold or something like that. So let me give you some backstory, and then you can fill me in on the things I don't understand. Okay, a man named Justin came to me one day, and started telling me about his his company, Vision AI, and how he wants to use AI to basically give people like a doctor in their pocket. And he said that he wanted to have very good information about diabetes there. And he asked me about integrating the actual podcast into what his AI understands. And I thought, well, this is very weird and interesting. But also Donald, what I thought was, I've already been working on this for like, a year and a half. So behind the scenes, I had been picking through everything I could, keeping in mind, I'm not a programmer, and I don't understand AI at all. All they had going for me was a podcast episode I listened to on a long ride, where this guy got on and described how AI worked, and how he saw it working in the world. And all I could think while he was talking was Oh my god. It trains itself on the knowledge. It's available online, and therefore it has answers based on all this data. I kept thinking, what if it only trained itself on something specific? Like, wouldn't that be amazing? And isn't my podcast very likely the largest repository of diabetes conversations that's ever existed? Yeah. Is there a way for me to get my audio transcribed, and I was actually getting it accomplished. It was expensive, and I couldn't really do it the way I wanted to, but I was doing it. I started with the Pro Tip series and some stuff like that. And I'm telling you, Donald, this is probably a no surprise to you whatsoever. But I could ask an episode of The Pro Tip series and incredibly specific question about diabetes. And the answer that came back was not only insanely accurate, but it felt like I was talking to myself. Isn't that incredible?

Donald 29:59
It's It's crazy.

Scott Benner 30:00
I would think I wrote this. As a matter of fact, the last speaking gig I did, I asked it to go through the Pro Tip series, and break it down into categories and give me talking points. Love it. And when it gave them to me, I looked at them and thought, this is exactly what I would say about this. If I were to stand up and promise, like, this is amazing. So anyway, I was working towards it. But when Justin's like, hey, we kind of already did that. I was like, well, that's better than me doing it, because I don't think I could have afforded it because there would have been a lot of back end cost for me. And it would have been a nice value add for, for the listeners, but I couldn't have asked him to pay me for it. Like the only mean, so I was like, how am I gonna do this, it's gonna, it's gonna break me they're gonna get value, and I'm gonna go, like, you know, broke trying to keep it going for them. So anyway, Justin comes along and says that he has this opportunity, would I be interested? And I said, Absolutely. You know, he asked me to be an advisor for the company, which is the thing I did. And again, full disclosure, I guess, if somebody comes along and buys Justin's company for money, I'll get a very, very tiny piece of it. So I want people to know that, but here's the thing. I'm starting to understand that you're the reason Justin knows who I am. Is that right?

Donald 31:12
It is through a funny turn of fate. Well,

Scott Benner 31:16
thank you. I owe you nothing. Donald, I want to say that upfront. Tell me tell me, you cannot have a peace of mind. Nothing. Okay. But, but tell me how this all happens. Please. Yeah,

Donald 31:27
I mean, well, being here is is you know, I think that's pretty good. Pride does pay off at all. So Oh, man. Yeah, I'm trying to piece that together myself. You know, back to, to me using T one pow. One day, I had gotten an email saying that, hey, there's a new service that is integrated one T one Powell comes out of the box, you click a few buttons and sign up and you know, pay their fee and it's on like, oh, okay, well, let's take a look into this. And I open up the website, oh, this is this actually looks pretty impressive. I sign up and get a call to kind of get integrated and answer any questions I had. So funny enough, just how I had like how I had met Ben in the folk, Ben and John and the folks at u one, Powell is how I met Justin, when I was just getting set up for the product. And after that I had spoke with Justin for a bit, you know, I can talk about this maybe a little bit more after, but I'm just, I was super excited about it. I mean, I had thought about this kind of product for years, before we had the breakthrough of AI models that we have now. And so when I saw someone doing it, I was like, All right, you know, let's see if this is the person or this is the company that's going to do it. But I'm at least gonna be like the waters and see what's going on out a lot of great conversations and eventually mentioned, hey, you know, have you listened to the Juicebox Podcast, it's the best place for advanced advice is really no one else is doing it. You know, a lot of people cover the fundamentals. But you know, when you're 10 years into, and you're really trying to get your eight, a one C down and deal with all the variables of life, like, no one else covers it better than Scott and his crew. And that's

Scott Benner 33:18
so nice that I appreciate that very much. Justin said something to me and tell me if he was just being kind he said that you said or somebody said, it would be cool if I asked it a question. And Scott's like knowledge came out of it.

Donald 33:31
Did you say that? Verbatim? Yeah.

Scott Benner 33:34
That's so nice. Thank you. You saved me a lot of hassle. I appreciate that. Because I was like slaving away behind the scenes trying to make this thing work, because I saw the value in it. Like, you know, I mean, if people don't understand how AI works, I think I did a fine enough job talking about it a second ago, but but all I could keep imagining was like, what if you just trained it on a finite amount of information, but I never imagined that it would actually call from conversation. And where it really threw me for a loop is there's an episode Virginie. I do the math of setting up like stuff, the math of Basal insulin, the math of insulin to carb ratio, that kind of stuff. And we never actually like straight through step by step describe the process for coming up with say your you know, your insulin sensitivity factor. But you can ask the episode how to do it. And it tells you and it's like, if you say to it, like I weigh this, I'm an adult I you know, I move at a average activity level, it will tell you like your insulin sensitivity is about here to here. And I'm like, these are the kinds of things that people struggle with forever and never get the straight answer to imagine you get a range. You start at the low part, you know, the safest part and you work your way up and you at least you have like a dial in idea like, you know, I can turn this knob about this far and try it on my own. Man, I'll tell you, that's when I was like, oh my god, it picked that out of a conversation. Like it really just it flipped my mind a little bit that it could do that. And it made me very excited just like you were describing before. Yeah, absolutely. So then you meet your, your user, you become a user. And how long ago was that? Oh, man,

Donald 35:23
not long, I'd say a few months. How are you enjoying it?

Scott Benner 35:26
What do you find yourself using it for?

Donald 35:28
It's incredible. I mean, you know, I'd say, for what I enjoy about it, it's probably good to mention what I expected, you know, what I would kind of expect a copycat to look like. So what I expected to see was someone take, you know, the existing chat GPT technology, and just kind of, you know, hey, when someone asks you a question, give diabetes related advice, you know, that would kind of be the most generic thing someone could do, and it wouldn't be great. But when I use vision, it's connect. So you know, probably the one of the main parts is that it's connected to your Nightscout and Nightscout, it's already been out there for so long to help collect all this data about your diabetes, you know, it knows your overrides, and knows your insulin that you're putting in your Basal. So your therapy settings, everything, when you ask it questions, you know, first off, it already knows a ton about you. And it's it's snuck that information in and conversations in a bunch of really interesting ways. We can talk about a little bit. But then the other part is that, you know, like, you're mentioning how they want to put your knowledge from all of these podcasts into it. It's really well fine tuned. I mean, you know, I was kind of expecting to give it a little bit of grace. I'm like, Okay, I'll ask it, you know, this really complex scenario that I'm in, you know, like, the hairy scenarios of diabetes. Yeah. And for it to just give a subpar answer, but it was good. Yeah. You know, it was something I would, it's the advanced answer. That handles the scariness of diabetes.

Scott Benner 37:09
I did the same thing you did when Justin, let me try it. The first time, I asked it a very convoluted question about my blood sugar. Now, you know, we were actually in somebody's, you know, account. I was like, okay, you know, what's my blood sugar right now? And it came back and told me and I said, Are you afraid that I'm going to get low? And it said, it didn't expect me to get low. But that if I thought I was going to get low, here's how many carbs I should take to stop that. And then I said, Okay, what if I want to take this many carbs and eat? And then it said, Oh, well, then you'd want to Bolus this much. And I was like, Man, that's not what I expected. either. I was right there with you. I thought for sure this guy was going to show me a sock puppet. You know what I mean? Like, like something that's just like, I know what diabetes is. And it's this and I was like, and when it wasn't that like, while you and I were talking earlier. So I'm in Arden's vision AI account right now, I asked her what my BG was said, the most recent update your glucose levels 159 to stable if you have any concerns, or notice any changes in how you feel it's a good idea to keep monitoring your glucose, like, you know, it does that. I said, I have a low grade fever and avoid a very sore throat. I'm sorry to hear that you're not feeling well. A low grade fever and sore throat can be uncomfortable. And it's important to take care of yourself, especially with type one diabetes, this illness can affect your blood glucose levels. Okay, so you think that's pretty basic, but then, one, stay hydrated, and describes that it says to monitor your blood glucose closely, and it describes that adjust your insulin ISness, if necessary, rest, soothe your throat, consider an over the counter remedy, seek medical advice. Under each one of those bullets gives you like, you know a description of what that is, in case you don't know. And then I said well, how much insulin would bring my blood sugar to 99 then it tells me to calculate how much insulin is needed to bring my blood glucose down to 99. We can use your insulin sensitivity factor, which by the way, it knows what Ardens insulin sensitivity factor is when you're it's in your Nightscout and I'm like, Holy Hannah. And then I said based on your data provider of your insulin sensitivity is this right now, per unit of insulin blah blah blah determine the drop is this I would use a unit of insulin to bring your blood sugar to 99 and then it gives you the please remember, this is a calculation based on typical responses, you know, for people who don't know what they're doing, oh, my god, like this levels you up three levels, you know? Really? Yeah, just fantastic. I

Donald 39:39
know and even you know like that it mentions the hydration that that cracks me up because that was one advanced suggestion I got from my Harry scenario that I was also impressed by it and like, that is a basic thing that I feel like honestly, even if you were talking that a really great endo in person about this very specific situation. They probably forget it You know, anybody what might forget? It's just something you overlook. But to have all of that knowledge in this model, it's not going to forget. Yeah. Yeah. It's it's incredible.

Scott Benner 40:11
Do you see it being something you would need forever? We're like, forever? And do you use it for other things other than diabetes? Hmm.

Donald 40:17
Other than diabetes, so, like, these models are going to be involved in our lives, like how a lot of, you know, website, websites, the internet, you know, all those things are involved in our lives. Now we're talking over over zoom right now. So specifically, I I wouldn't see myself using envision forever. I think, you know, your, your glucose can always, not always get better. But I feel like there's always variables that pop up, you know, right when I've got it, man, best week of the year. And then, you know, the very next week, oh, the community is picking up, you know, why is there anything out of whack? Oh, I'm, you know, instead of putting on muscle right now, I'm trying to lose weight? Oh, you know, it's, it's, it all starts over again. I think it does so much already. You know, some fans, I would be really excited to see, you know, the things I'd always wanted for years was, you know, I had asked, Why can't I track a bunch of events somewhere, and have some system kind of keep track like, okay, you know, I just wanted my walk at 9am. I just, it was, you know, 70 degrees outside, I felt a little sick today. But those over and over and over again. And then have some system some app, give me suggestions based off of that. Yeah. And this is really the first thing that in take the incredible amount of variability from your life, and put it all together and come out with a really simple suggestion. Everything

Scott Benner 41:50
I asked Justin like, is it possible to get to do this is that he goes all that is possible. Like I said something like, Could we get it to a point? I don't know if it does this now. But could you get it to a point where you could say, hey, last Thursday, I ate pizza, you know, just nappies. I'm going to do it again. How was my Bolus ng strategy last time? Anything I asked him, he's like, I don't see why I couldn't. And I'm like, Jesus, it's incorrect. Listen, while you were talking. I've been tired and my hair is falling out what could be wrong with me? Number one, thyroid issues both hypo and boom, like right like that. That's a common complaint for people with type one diabetes. A lot of people have unmanaged thyroids that need help, and they don't know. And they spend their life struggling with these problems, right? And because you're not going to think of that, like, come on, Donald who's gonna go you know, this is probably my thyroid. Yeah, boom, but also gonna be Yeah, to nutritional deficiencies, like vitamin D, vitamin B, 12, and iron, stress and mental health, auto, other autoimmune diseases could cause these symptoms, especially if you have one autoimmune condition, you might be more susceptible to others. There's a thing that you think people know, but they don't. And it's just hormonal imbalance, medication side effects, given these possibilities, bah, bah, bah, like, Dude, I falling short of wondering what I could ask that I couldn't get a reasonable answer back for like, I would like people to know that I don't sign my name on two things like, easily. And a lot of people have come and asked me to partner with this podcast over the years, and I just like, I shoot them all away. And I swear to you, I thought I was getting on a call to shoot Justin away. Yeah, I really did. But here we are. So

Donald 43:31
and one thing I have to say about that, too, is that I had totally thought the same thing, but just just him mentioning the quality of people in the community, you know, like, like, you included, obviously, it was just like, oh, my gosh, you know, I had thought for years that this product will come. You know, it's like how I think back to this really funny term that people use, you know, when insulin was first invented, people were like, Aha, you know, diabetes is now cured. We have insulin that, you know, from a, from a cow that we can put into people, that's it, then they came out with the home glucose meter, aha, it's cured, you know. So you know, really all along, it's breakthroughs that just make it a lot more manageable. When I saw that someone is harnessing the awesome power of these, LLM these AI models. And they're doing it with the high quality folks in the community, as Oh, my gosh, you know, we're looking at a breakthrough right here. This is the very beginning of the next big diabetes tool. Yeah. It's hard not to want to be a part of it.

Scott Benner 44:32
I also think that if I've been paying attention and reading between the lines for the last five years, every time I interviewed Dexcom, they mentioned how valuable data is. Yeah, it's so valuable that we have everyone's data like one day we'll be able to help you make decisions and I feel like that's, that's this right here. I think a lot of people might, might push into this space. I think that's pretty obvious that we're going to see a lot of AI companies come in, some of them will stick and some of them won't as we go along. But This very well might be a situation where being good and first might really be a hell of an advantage. Yeah,

Donald 45:09
and, you know, I think an important thing is, you know, there's an old saying of, you know, garbage in, garbage out, you know, if you have bad data, you know, you're not going to be able to get good outputs from it, you know. So having all of your knowledge in there, and just all the other good stuff is going to be important. Hey, it

Scott Benner 45:29
filled my heart, I'm going to sound like a dope for a second here. But the reason I've been looking into the AI so much, is because I have been having and continue to have this one singular feeling that I put so much effort and time into getting these conversations in one place. But I know the way social media works, and the minute I stopped making this podcast, it's just going to disappear. And so, you know, sometimes people ask me, like, Why do you put out so much content, I'm like, This is how you keep the podcast alive. Like everybody, you know, people are putting out like, one episode a week or every other week, I'm like, Yeah, you have a podcast, you have a hobby, which is nice. And I'm not saying it's not helping somebody, but the minute they take that candle off, that stone is gonna go stone cold, I so I know, keep coming. But I also know the information has to be good has to be interesting. You can't get filler, when you start putting filler content out, you lose people, you know. So all I could think is that one day, all this great information is just going to disappear. And I can't let that happen. You know, like, that's just it's a disservice at this point.

Donald 46:34
It is, and you know, it's like, you know, for 1000s of years, people have been able to kind of preserve their thoughts through writing, which is great. But you know, the paper is kind of hard to manage. And, you know, you have to get the piece of paper from one part of the world to the other, and come up with the breakthrough, you know, the internet, you can get that piece of paper from that piece of information, you're writing what's in your head, someone else in the world right away, you know, it's like someone else can talk to you. And you know, you might be from 100 years ago, so and then 100 years might be, you know, talking to Scott about their diabetes. Yeah,

Scott Benner 47:08
we can keep building on ourselves instead of starting over all the time. And listen, my example that I'll sound like a crackpot for half a second. But at some point, someone new had to build those pyramids, and I don't see it written down anywhere. And I don't think it was easy as like sliding rocks together. What happens if, you know, like, for whatever reason, like just on a small scale, the podcast, just go I get old, and I just stopped making it, and it goes away. And then you'd be surprised how quickly it'll go back to a bunch of people getting diagnosed. And now there's no like doctor telling them, Hey, you should go check out this podcast, Donald's not telling his friends, you should check out the podcast, like that doesn't happen. And then those people struggle, and they start over and 10 years from now, you hear more and more of the same stories we heard 10 years ago. And I'm not saying that I'm the pinnacle of diabetes knowledge or anything like that. I do think that I have found a way to make it very accessible to a lot of people in a way that makes them want to listen to it. Totally. Yeah. And I think that's really, maybe one of the things I can be more proud of, because there are times like, I think to myself, like this is not like brain surgery, what I'm talking about here, but how come no one else has captured an audience this large before? I

Donald 48:23
mean, I know that seeing that's the incredible part with that information. I mean, you know, I'm sure someone else out there might have that a lot of good info written down. But where is it? You know, I'm looking around for it and not just gonna be able to magically find it, where

Scott Benner 48:38
you put it in a book and you know, God bless readers, but most people don't read. Yeah, yeah. Oh, and I say all the time, like you can have like the secret of life in a podcast. And if it doesn't sound good, people won't listen to it. Right, you know, forget that. They'd have to pick up a book and actually stay focused and read it and forget that these people are looking for this information at some of the most dire points in their life. And you're asking them to now read a book about it. You know, like, it's just, this isn't the time. You know, if I walked up to you, Donald, I was like, hey, Donald, guess what, and I smacked in the head with a shovel a couple of times and handed you a book about algebra. And I was like, I need to know what x is. You'd be like, Oh, what, and I think that's how people feel after they're diagnosed. And even if you've had diabetes, for some time, if you've been struggling with it, you still have that, like, shook feeling all the time. And it's just, it's not reasonable to ask somebody to go read a tome, and then come back and remember everything and put it into practice. And I mean, if we learn nothing about people, you know, we like storytellers, and we like listening to stories. And that's how a lot of us learn. So anyway, I guess what, it's a long way of me saying, I threw in with Justin, and this idea of AI because this already seemed like a natural path to me. And I was trying in the background to accomplish it all by myself from this room and I was When getting anywhere, I don't know how to say this, which doesn't sound pompous. So I'll just say it. I've been write a lot over the last decade about how to help people with diabetes. I feel very strongly about this, that this is a good idea. Completely.

Donald 50:13
Yeah. And I think that's why you're going to see so many, so many respected people and the community kind of blocking around this and trying to, you know, bring it up, you know, as the, as one of the next great tools in our tool belt. Yeah. Well, I

Scott Benner 50:30
agree. So what else? Would you want to tell me about it that I don't know to ask you about it?

Donald 50:33
Oh, man, I had some good use cases come out of it. It's funny, every time I it's like, I love the word, you know, delightful when you're talking about, you know, using something because like, I'll definitely get some like, just just delightful. Like, it feels good. Like, oh, wow, I did not expect this machine to do this. You know, there's kind of the more saints you'd mentioned. You know, I had, hey, vision, I just did a spin class. And now I'm going to go do a long walk to my favorite bagel shop, you know, how can I not go low from doing my workout, but still managed to not go high from my high carb, delicious bagel, those are really cool responses, they go through a lot of things you mentioned on the pod love, they can call out, you know, a call out my specific overrides, because it saw, hey, you have a cardio override, you know, if you're doing spin, that's kind of a cardio exercise, I would use that then. Which just you know, the language tie together is on that blew my mind. And while also being extremely helpful, I think the one that you might have not thought of let me know if you did, but I had, it helped me review my therapy settings. So like, all my different you know, my ice my basil is all of that. The most basic example I had for it was usually I'll chart out my icy and my basil, and trying to make sure I'm not doing anything too crazy. Like, you know, I want to make sure that if at noon, you know, I'm using more basil, I want to make sure I'm using more ice to and they don't have to be perfectly insane. But, you know, that's kind of genuinely how I run it. This is such a pain of kind of going to Google Sheet, it'll pull up all these charts for it. So when I was using vision, during my kind of weekly self appointment with myself, I always have vision, pull up another tab when I'm doing that, like, hey, you know, look at all my therapy settings, give me a text based table, and just kind of helped me understand what my ups and downs from my settings look like. And it did it. You know, I didn't have to go through all of this effort to put that together. So I think having a kind of holistically review your therapy settings is going to be a really big win. And yeah, you know, there's there's so much more they can add on to it. But that was a that was a fun use case. No, it's fantastic.

Scott Benner 52:57
Like, even when I asked it earlier about Arden like moving her 159 It said that her insulin sensitivity, it knew her insulin sensitivity was 61. But that's not really Arden's insulin sensitivity. And let me explain Arden's using dynamic ISF on Iaps. So it changes her insulin sensitivity based on what it sees his needs. So my my point is, is that her set insulin sensitivity right now is like 80. But it knew that right now, because it's looking at the settings that the algorithm has been working off and ISF of 61. That's awesome. And the truth is, I wouldn't have thought that if like you made me think about it right now, and said, hey, you know, move Ardens blood sugar Baba blah, I don't think I would have, I wouldn't have thought to look to see what Iaps was calling her insulin sensitivity right now. Like that's a that's a pretty deep layer of complexity. Most people might not even understand what I just said. But it's just very cool. It has access to that information.

Donald 54:05
Yeah. And it's like, it's so hard to like, I don't know, 10 years ago, I would have never asked, you know, for AI for a website, or an app to be to have the rigor of a computer, you know, where it's always going to consider everything, but then to be flexible enough, like, you know, a chatbot or just, you know, a person to be able to, you know, communicate it simply and just kind of understand what I'm asking it to have been able to have both of those just, it feels too good to be true. Yeah. You know, you use it. And you see it's not

Scott Benner 54:40
it makes me wonder as we're talking which pump company is going to have the foresight that track Justin down and try to get an amen. Yeah, yeah.

Donald 54:49
So maybe maybe it'll be all of them at once.

Scott Benner 54:52
Well, he should lease it to everybody, not just one of them. I would as just spectacular. I'm not surprised by it. Right, like I've been on the side of this, it was hard not to pay attention when they made chat GPT. Public, you know, and you could see its limitations and everything. But you know, you keep hearing over and over again from the people really understand, wait till you see how much more quickly this is going to double over itself. You know, what else did you see that happened recently? That was kind of brilliant. What's the oh, wait, Tesla gave their self driving away for a month to everybody for free? And I thought, Oh, they did that to get more data back? Ah, right. Like they didn't? Yeah, yeah, maybe it's a sales pitch. But they also wanted more people to drive with it so that their computers could crunch more scenarios and come up with ways to make the self driving faster. Data scoop? Yeah. And I thought, Oh, that's really smart. Right? And then all I could wonder is like, How much is it going to improve just from that, then? How much is it going to improve if they, you know, a year from now, like, and that's what I keep thinking about all this? Like, I don't know, if there's a cap to this, I would assume that getting the right answer is the right answer. I don't know how much more better, you know, it's like you can only get so cold, I think you can only do so well, like, you're getting information from somewhere. But all of a sudden, what if you could make it more human or more relatable? Or I even asked Justin I thought I asked him an off the wall question. I said, is there ever a time where someone could ask the podcast the question, and it would respond back in my voice? And he said, That's not out of the question. You could have it to your face, too. It's insane. Like, that's feels insane to people. But the point is, is he's got so much of my voice recorded, that it could approximate my voice. I was like, that's not that I started having like big questions like, do you think my kids will talk to it? When I'm dead?

Donald 56:45
All the existential questions start to drift, then would it be me anyway, I

Scott Benner 56:49
got passed pretty quickly. But like, my point is, is that at some point, these things are going to return back, you know, answers that are, are so good, they're just kind of perfect. And then it's going to become more about speed, and, you know, experience and that kind of stuff. I just I wonder how it's gonna double over itself as

Donald 57:09
well. And it's like, you know, diabetes is such a big problem, it's, you know, I'm sure there's some things that just the problems are crazy hard to solve, you know, like, if you take your, your sixth grade writing assignment, you know, write a two page paper on William Shakespeare, you know, it's already gotten to that, right? Not not like the biggest problem that have to go through, you know, factor in all of these environmental variables about your life and the diabetes that like, you might not even be able to, you know, write down, think of in your head, that's a big problem. So I think having something that's going to double over and kind of almost be hard to imagine is like, just what we need,

Scott Benner 57:52
the next level becomes people knowing how to prompt it like that, that's really going to because Could you ask it for example, like and I don't know, right? I'd have to ask Justin. But like, Could I say to it, you know, my a one C has been 7.5. For the last two years, you see my data? What am I not doing? You only mean like, like, where am I seeing excursions? And how could they be minimized? That'd be amazing. If it just had a look, you know, what if he just started Bolus in your meals like six minutes earlier? Or you know what I mean, like when you see a rise to this, you know, historically, once we get to here, it feels like we didn't cover the meal, well, you probably try putting in more insulin or like those kinds of ideas that peep those are not leaps people would make. So it's going to be about like asking the right questions. I even wondered, like, Could you just say to it, at some point, I'd like to brush up on my skills about diabetes. Can you write a 50 question quiz about type one diabetes, for me, that would help me? And I could do it now. Yeah, it's, I mean, that kind of stuff is all like, it's going to be about the people who teach you how to use the AI. And

Donald 59:00
that's why I was excited, you know, when chat TPT came out, I mean, you would think me being really into tech, I'd be the first person to try it. You know, I still kind of gave it some time for like, the hype cycle to go through because I'm like, Oh, is this gonna be a big thing? You know, right. In the beginning, it wasn't very good. So like, I was admittedly even kind of slowed picking up but once I saw, hey, this is gonna be a thing. You know, open up chat, GPT play around with it, and like, Oh, hey, this is pretty good. And I had the same feeling about vision, where, you know, we're fortunate, it's already helpful and useful. But I want to practice, you know, asking the right questions, kind of seeing what the limitations are and the things that it can bump into. Yeah, yeah, to get those skills up. Now.

Scott Benner 59:44
I'm just I'm with you. I imagine a time I don't know what were in the future. But I imagine a time that you have an app that is, you know, even from a company that you know, a retail app that you can ask questions to about your care about like, I mean, it's a big leap for a company to make, like a pump company would have to get out of their own way a little bit to be willing to do that, where maybe it's going to be a third party app. And you know, it'll just connect to your data. And that keeps the companies out of it. Because there is, you know, I don't know how much people understand, but the FDA doesn't let a pump company give you treatment advice. That's not actually a thing they're allowed to do. It might always have to be a third party situation, but that you just, you know, as long as you can get your data into it, it can make decisions based on it.

Donald 1:00:32
And, you know, it's really cool how, you know, fate has kind of brought us to where we're at where, you know, Nightscout is a really popular open source tool that lets you put all your data in one place. And then you know, you have something like vision that comes along, you don't have to have, you know, months of, you know, the people from Vision trying to muck around and talk to the right company, so that you can pull your data and it's already right there. Yeah. So I just, I just feel like it's so teed up to do what it's doing really well. It's going to work out,

Scott Benner 1:01:06
I can't wait to find out where it goes, I really can't. And I again, like I said earlier on, I'm happy to say it again. You know, if Justin becomes a bazillionaire, I will make some money. But you know, if just two becomes 1000 a year, I'm you know, I mean, am I telling you, I think you should try this? I think it's interesting. And I think you'd be amazed to see it. I can't lie, I think it would be really, it would be a really an eye opening thing for a lot of people. I felt weird about it. Because you know, I do, I haven't connection to them. And I've never really brought somebody on the podcast before to talk about something that like directly benefited me before, like, I mean, but in the end, like you guys know, like, click on the links for the advertisers, the advertisers are happy with me, they buy more ads, like it's, you know, probably not dissimilar, but it just still made me feel uncomfortable. So I appreciate you like letting me blurt that all out at the beginning. Is there anything? I'm not asking you that I should have?

Donald 1:02:00
Yeah, you know, I think I think the last thing that I've kind of picked up from some conversations with Justin, when I on boarded, it's helpful for further folks here. You know, it's really funny, for a long time, I would, I don't know if other people have done this, I'm sure they do. But like, you kind of get this, like, these custom little code that you have when you enter in carbs. So like, for, you know, my food, if, you know, I know, if it's a Greek yogurt, you know, I'll just put like, yolk, you know, or if I had protein shake, I just put, you know, put shake, you know, some like silly things like that. But so whenever I was onboarding with Justin provision, we picked up my nightstand, he's like, Oh, you actually like, enter in, you know, what you ate, you know, you didn't just like, you know, have it where there's like, no icon that you picked or whatever. Oh, yeah. Like, you know, kind of helps me like, look back and know what it is. And he's like, Oh, that, you know, that could be helpful for vision to know what you're eating. And now I'm thinking like, Oh, holy cow, like, I had always kind of thought, you know, I'm giving, you know, I'm kind of being the nerd and making sure that all the data is there. And you know, I'm labeling know what I'm entering and things like that. But it's cool to see that there's finally a use case where all of this information that you throw at it can be useful. Yeah, I think once people try it out, they can see what, what's helpful to have what's not, it was just kind of a reminder for me that, yeah, the more information I'm going to put into this, and, you know, kind of keep all of my stuff curated, the more it's gonna be able to help me out.

Scott Benner 1:03:36
I agree. I've had the same thought, like, you're making all these keystrokes and we're not getting anything out of it. I mean, the algorithm just knows, like, slower digestion. You know what I mean? It's just not exciting enough. I had somebody come on once and talk about Iaps. And we went over that idea of, you know, he used a pizza place as an example, but you know, pizza place a and pizza place B hit your blood sugar differently. Why couldn't your algorithm know when you're a pizza place? A and when you're at pizza place? B?

Donald 1:04:07
Yeah. And, you know, it means little, I think everybody probably has this wish list of things. Like, you know, someone could do this, why aren't they doing it? That's the really cool thing about these large language AI models is that you can just ask it, and see how close you can get to it. You know, if I'm typing in, I'm going to my favorite pizza shop, you know, or you know, look back at my times I've written in pizza shop a pizza shop B and you know, my going high is one of the others or something different I can do that my usually entering like too few carbs for that meal. It's crazy to think about

Scott Benner 1:04:43
Yeah, no, I can't wait. I honestly can't wait to see what happens. So alright. Well, Donald, I can't I can't thank you enough for doing this. Can you I didn't really ask you to tell people but can you give people a tiny bit of background about what you do? You don't have to say where you do it. But what do you do for a living?

Donald 1:04:57
So I guess as we kind of hinted at You know, I'm a software engineer, left technology, I'm currently working at Amazon. And before that I was at Salesforce, also work on a couple things on the side, you know, hopefully get it to

Scott Benner 1:05:12
get away from your overlords at Amazon. Hey,

Donald 1:05:16
you said it, not me.

Scott Benner 1:05:18
It's amazing. What did you do in college? Like, how did you get to this? What would you major in computer

Donald 1:05:22
science? You know, pretty, pretty typical path from from that part

Scott Benner 1:05:27
you just took to it? Or was it always something? Like, I mean, how do you know? I mean, 10 years ago, did you even know what it was you were going to be doing one day when you were studying it? Or is, is today so much different than what you imagined? I guess, is my question.

Donald 1:05:40
You know, I don't think I expected getting into no entrepreneurship and starting my own thing, but I was definitely always, when I was just starting off in high school that I got into it. And you know, just did it for a little bit. I was like, Oh, my gosh, this is really fun. I can just, I can make things, you know, it was a little different than something like, you know, I done some woodshop. And that was fun, because, you know, I like building things. But I'm like, oh, gotta have other materials. I gotta have a shop. But then whenever I was doing software, like, well, pretty much always have a computer available to me. And I can just, you know, type on the keyboard and make something in front of me and share it with everybody. That's definitely what pulled me into it.

Scott Benner 1:06:22
Amazing. All right. Well, I appreciate you doing this very much. I know it's the evening and you probably had a long day already. Can I ask you after we stopped? Can I ask you to hold for one second? I want to ask you another question. Yeah, for sure. Off the recording, though. Okay. Great. Thank you so much. I really appreciate it. And

Donald 1:06:38
I had to I have to call out here too, that you know, if people want to check out vision that I have vision.com That's solid prices. 1999 a month, which I think you're just getting a huge bang for your buck for all the stuff you're getting able to do with it. So that was

Scott Benner 1:06:52
like you're not much of a Salesman, Scott. You didn't mention the website. Gotta give a shout out. I have vision.com

Donald 1:06:59
I have vision.com Yep,

Scott Benner 1:07:02
I appreciate it. All right, hold on one second for me.

Want to thank the ever since CGM for sponsoring this episode of The Juicebox Podcast, learn more about its implantable sensor, smart transmitter and terrific mobile application at ever sent cgm.com/juicebox Get the only implantable sensor for longterm where get ever since mark is an incredible example of what so many experience living with diabetes, you show up for yourself and others every day, never letting diabetes to find you. And that is what the Medtronic champion community is all about. Each of us is strong. And together, we're even stronger. To hear more stories from the Medtronic champion community or to share your own story, visit Medtronic diabetes.com/juice box. I'd love to see you in Orlando for the touch by type one event go to touched by type one.org and get yourself some tickets are absolutely free. The events coming up soon. Don't wait. It's filling up. And I'm not just saying that. Hey, everyone, it's me, I just want to be clear again that if you should go to I have vision.com And subscribe, I will get a piece of that money. And if the company should go public one day or be sold and make any kind of income from that I get the tiniest, tiniest little quarter of a percent of the value at the end. So I just want to be completely clear about that. There's no pressure for me for you to go to I have vision.com Please don't feel like that. But if you do, I appreciate it. Sure that Justin appreciates you helping him getting his his business off the ground. No pressure from me. But I do benefit if you spend money at that website. And I want to be a 1,000,000% clear about that. Thank you. If you or a loved one was just diagnosed with type one diabetes, and you're looking for some fresh perspective, the bold beginning series from the Juicebox Podcast is a terrific place to start. That series is with myself and Jenny Smith. Jenny is a CDC es a registered dietitian and a type one for over 35 years. And in the bowl beginning series Jenny and I are going to answer the questions that most people have after a type one diabetes diagnosis. The series begins at episode 698 In your podcast player, or you can go to juicebox podcast.com and click on bold beginnings in the menu. Thank you so much for listening. I'll be back soon with another episode of The Juicebox Podcast. The episode you just heard was professionally edited by wrong way recording. Wrong way recording.com

#1217 iLet from Beta Bionics

Steven Russell, MD is here to talk about the iLet pump from Beta Bionics.

+ Click for EPISODE TRANSCRIPT

Scott Benner 0:00
Hello friends and welcome to episode 1217 of the Juicebox Podcast.

Today I'm going to be talking about the islet insulin pump with the Chief Medical Officer of beta bionics, Dr. Steven Russell. Dr. Russell and I will go over questions from the audience as well as his feelings about the eyelet pump. Nothing you hear on the Juicebox Podcast should be considered advice, medical or otherwise, always consult a physician before making any changes to your health care plan. When you place your first order for ag one with my link, you'll get five free travel packs and a free year supply of vitamin D. Drink ag one.com/juice box. If you are a loved one has type one diabetes and you are a US resident. I'm asking you to take 10 minutes to fill out the survey AT T one D exchange.org/juice. Box completing that survey helps significantly with type one diabetes research. T one D exchange.org/juice box. If you're looking for community around type one diabetes, check out the Juicebox Podcast private Facebook group Juicebox Podcast type one diabetes. And as a matter of fact, we have quite a few users using the islet pump so if you have questions, join the group is absolutely free. This show is sponsored today by the glucagon that my daughter carries G voc hypo Penn Find out more at G voc glucagon.com forward slash juicebox. This episode of The Juicebox Podcast is sponsored by us med U S med.com/juice box or call 888721151 for us med is where my daughter gets her diabetes supplies from and you could to use the link or number to get your free benefit check and get started today with us met. Today's episode is sponsored by Medtronic diabetes, a company that's bringing people together to redefine what it means to live with diabetes. Later in this episode, I'll be speaking with Jalen, he was diagnosed with type one diabetes at 14. He's 29. Now and a little later, he'll tell you about his life with type one. To hear more stories with Medtronic champions. Go to Medtronic diabetes.com/juice box or search the hashtag Medtronic champion on your favorite social media platform.

Steven Russell, M.D. 2:39
Hi, this is Steven Russell. I am the Chief Medical Officer of beta bionics. I am also a associate professor of medicine at Harvard Medical School and I see patients at the Massachusetts General diabetes Center part time but most of my time now is spent working at beta bionics. I have been working on the bionic pancreas project for about 16 years now been working with that Damiano. And for us Alkhateeb from the early days of the bionic pancreas project and, and directed most of the clinical studies that were done three pivotal and pivotal and joined the company in the end of 2022 to help to get the product approved by the FDA and prepare for launch.

Scott Benner 3:28
Okay, so you have a good, healthy, long relationship with it. So you're gonna have all the answers, right? Some of them well, I'll do my best. What's your educational background?

Steven Russell, M.D. 3:38
So I have an MD and a PhD in biological chemistry. And I trained at UT Southwestern for my MD and graduate degree. And then I came to Massachusetts General Hospital for my internal medicine residency, and stayed at Mass General Hospital for an adult endocrinology fellowship. Okay, so

Scott Benner 4:00
did you do most of your work in endocrinology, or is it internal? So

Steven Russell, M.D. 4:04
endocrinology is a subspecialty of Internal Medicine. So you have to do internal medicine residency first, okay. And then you can sub specialize and I chose to sub specialize in endocrinology.

Scott Benner 4:16
I always like to know what made you choose it? Well, I

Steven Russell, M.D. 4:18
think it's partly the kind of work it's you know, diagnostic work, solving puzzles, but also long term management, you know, trying to figure out how to help people live their best life with diabetes is appeal to me. And from the research standpoint, I've always liked intellectually curious, I like the fact that in endocrinology, there's control loops, that are sort of intellectually satisfying. And so that was appealing. For that reason, I think, and I knew I wanted to do some research, and there's just this great history of research that comes out of endocrine I think something about hormones, they were accessible to work on, early on. If you identified a hormone, not only did you have the hormone but you that you had a drug, you know all almost all of those hormones were also immediately usable as drugs didn't have to go find a small molecule. So when insulin was discovered as the key element to control glucose, all you had to do was purify the insulin, not a trivial matter, but you had to purify it, figure out how to use it, you didn't have to then go and find some other small molecule drug that could act on the insulin receptor. It's a, perhaps a shorter path from figuring out how the mechanism works, to being able to do something with that information to help a patient. Okay,

Scott Benner 5:45
so then is it fair to say that those things that light you up are maybe one of the reasons why somebody came and found you for for this 16 years

Steven Russell, M.D. 5:54
ago? Well, we sort of found each other. So Edie had started to work on the bionic pancreas algorithms, because, as his well known now, he had a son who developed diabetes at 11 months of age. He was an applied mathematician doing fluid mechanics, but thought maybe he could use his applied mathematical skills to to improve management of diabetes. And he was to the point of doing studies in pigs that he made diabetic with bras and they were doing these studies with an earlier version of the algorithm. And he came to the Joslin diabetes Center to report on that work. And I was there doing a postdoc trying to understand how insulin signaling effects longevity and aging. But I also had a side project looking at accuracy of CGM and in people in the hospital and critically ill. So we reported these data using both insulin and glucagon. And I really liked that approach. It was kind of contrarian at that point, because people said you couldn't use glucagon as part of a control strategy. And sometimes when people say something can't be done, it's true. But in many times, they they just haven't opened their mind and really investigated it. So I was intrigued by that possibility. So I approached him and said, hey, you know, you're doing these animal studies? Could we work together to do studies in humans, and he was enthusiastic about that prospect. And from that point onwards, we started writing grants and planning studies, and, and we've been working together ever since it's

Scott Benner 7:37
interesting. It's interesting how you can ask a question, and somebody will say, No, absolutely not. Or maybe, and, you know, like that the No, absolutely not people like, are they so tied to what they know that they can't see the rest of it? Is it just comfortable for them? It must make all of these things difficult in an academic setting, right? Because so many people tie their ideas to who they are. You know what I mean?

Steven Russell, M.D. 8:04
Absolutely. I mean, if somebody has been working on insulin, only automated insulin delivery or insulin only automated glucose control for a long time, and then you come in and per proposing using a second hormone. I think that that can be kind of threatening. Yeah. And can feel like, oh, wait a second. You know, you're you're encroaching in my territory. This, we're going to do this this way. Right. At one point, I had somebody not to be named come up to me and say that nobody should be working on by hormonal glucose control until there was an insulin only system on the market. And I thought that doesn't make a lot of sense to me. I

Scott Benner 8:49
know someone told me one time that I couldn't make a podcast and talk about how we manage my daughter's blood sugars. If you take insulin or sulfonylureas you are at risk for your blood sugar going too low. You need a safety net when it matters most. Be ready with G voc hypo pen. My daughter carries G voc hypo pen everywhere she goes because it's a ready to use rescue pen for treating very low blood sugar and people with diabetes ages two and above that I trust. Low blood sugar emergencies can happen unexpectedly and they demand quick action. Luckily, G voc hypo pen can be administered in two simple steps even by yourself in certain situations. Show those around you where you storage evoke hypo pen and how to use it. They need to know how to use Tchibo Capo pen before an emergency situation happens. Learn more about why G vo Capo pen is in Ardens diabetes toolkit at G voc glucagon.com/juicebox. G voc shouldn't be used if you have a tumor in the gland on the top of your kidneys called a pheochromocytoma. Or if you have a tumor in your pancreas called Have an insulinoma visit je voc glucagon.com/risk For safety information. And now 17 million downloads later, it helps people in 48 different countries. So if I were to listen to that person, none of this would have happened. So here's the people who don't listen. That's great. So now we've got a fast forward all this way you go through this entire process, you get a device, it gets its way through the FDA, it's it's available on the market now. I always, when I'm thinking about it, let I want to know if I am. If I'm clearly understanding what it is. And I'll just tell you that from my perspective, it seems like a fantastic device. If someone has an elevated a one C, and struggles and can't figure things out about how to dose or count carbs or anything like to me this seems genius. I mean, if you're a person walking around with a 1011 1213, a one see, are you telling me I can put the eyelet pump on and announced meals? Like I mean, what are the announcements? It's snack? How was the

Steven Russell, M.D. 11:03
Wreckfest? Lunch or dinner and usual for me more or less? Okay,

Scott Benner 11:07
so I make those two announcements. And I get an A one see where? Well,

Steven Russell, M.D. 11:13
on average seven, about half of the people in our pivotal trial wound up with agencies below seven and about half above seven, and most of those were below seven and a half. Okay,

Scott Benner 11:25
that's astonishing. So that did. So that's my point. For every person who listens to this podcast and firmly understands Pre-Bolus eating their meals and adding insulin for fat and protein and all the other things that you need to understand to keep stability. I mean, I don't know the number, but there's got to be 100 people who don't understand it are never going to even intersect the information that if they did, they might not get it. So am I thinking about that right? Or am I undercutting islet? By thinking about it that way?

Steven Russell, M.D. 11:49
I think you are under cutting a little bit, I think you're right. It's definitely great for people like that. And we wanted it to work for people like that there's a lot of people out there who aren't being served by the current or previously available diabetes technologies before islet. And we know that that's the case, because only 20% of people in the US with type one diabetes at expert centers have an A one c less than 7%. So that's a pretty sobering statistic. The average a one sea of people with type one in the US is about 8.2%. So that tells you that the current methods are not sufficient. And that hasn't changed. By the way very much as we've gotten CGM and AI D. I think a lot of new technologies that we've gotten have allowed people who are already had pretty good glucose control to get even better glucose control. But what they haven't really done is allowed people who weren't getting good glucose control to get good glucose control. And I think that's where the islet is, is really different. One of the ways that the sense that I think you may be underselling it is even people who have good glucose control, are often spending more time managing their diabetes than they would ideally like to do. There's an enormous mental overhead associated with managing diabetes just takes a tremendous amount of thought and effort and time and consideration to achieve those kinds of results for even for the people who are achieving good results. And I think that there's a real opportunity for those people to continue to get good results. But with less effort on their part, what we found in the pivotal trial is that for people who already had an agency less than 7%, at baseline and the trial, their average a once he did not change on the eyelid. So if they had an agency below 7%, on average, they still had an agency below 7% was about six and a half percent for people on the island for people who already had an agency below seven. Okay. And they did that with a lot less effort. And you don't really capture that in the in the numbers. But we did do patient reported outcomes where we asked people using these standard sorts of questionnaires, essentially, how are you feeling? And what we found is that they had less diabetes related distress, less fear of hyperglycemia, better quality of life. And those are things that don't show up in those agencies or those average glucose numbers but are real and matter. And

Scott Benner 14:40
so are you saying that people who had a better understanding to begin with, generally speaking didn't see a rise in a one see when they moved to Island?

Steven Russell, M.D. 14:48
That's true. Yeah. What we did see is that as I said, on average there anyone see didn't change on the eyelid if they started off already meeting goals for therapy if they already had anyone see less than 7% And if you break it down further and look at that group, some of them had no change in their agency from baseline, some actually saw further decreases in their agency on the eyelet. Some actually saw increases in their agency, although by and large, they still stayed below seven. But what was interesting about those people who saw rises in their agency, they generally had significant reductions in their time less than 54. So they were seeing an increase in their agency, because they were having a lot fewer lows,

Scott Benner 15:31
people who were coming to their agency, sort of, not honestly, right, like they it was being offset, their standard deviation was being offset by low blood sugars.

Steven Russell, M.D. 15:39
That's right, okay. And on average, the eyelet didn't change the amount of time less than 70 or less than 54, there was no difference in the control group and the islet group, and time less than 54. Median was point 3% of the time less than 54. And that was true in both groups. But that does conceal a little bit that conceals interesting things. So for people who had very high levels of hyperglycemia, at baseline, the eyelet, decreased them quite a bit. For people who had none at baseline, it tended to increase it a little bit from say zero to point 2% of the time. And that might be the cost of you know, somebody having an agency of 10 or 11, or 12, with zero hyperglycemia, you bring them down to an agency of seven or seven and a half, they're going to have a little bit but still acceptable amounts. And then people who had hypoglycemia below 1%, at baseline, typically no change.

Scott Benner 16:38
Okay, so I have a ton of questions here. And if you don't mind, I don't I don't mean rapid fire, like we're going to be exhausted and sweating when it's over. But I'd like to maybe lose the conversational nature of this a little bit and go to more of a question answer if you don't mind. Of course. Okay, that's excellent. Thank you. Let's start with kind of the basic stuff. Let's just start with dosing. Like you just said, breakfast, lunch, or dinner. So you you pull up your pump and say, hey, it's breakfast, and I am having an average meal for myself, do I Pre-Bolus that,

Steven Russell, M.D. 17:09
we suggest that you announce the meal at the time you're starting to eat

Scott Benner 17:14
it. So that's a no then to the Pre-Bolus. Right? If

Steven Russell, M.D. 17:17
you are some people really like to do that and find that, you know, they might get some people say that they get better glycemic control less of a postprandial excursion. If you're going to do that, though, you need to do it consistently, because it will affect how the islet learns how much insulin you need for that meal.

Scott Benner 17:34
Okay, let me jump to that word, then learn what is learned mean, in regards to eyelet? How does it learn? What is it do with that information? How does it change the future?

Steven Russell, M.D. 17:44
Sure, it depends on which part of the insulin dosing you mean. So there's basil, there's a basil algorithm that learns, there's a correction algorithm that learns. And then there is the meal dosing algorithm that learns, and they, it is unique in that all of those do truly adapt and learn the individual's insulin needs. Most of the other AI D systems don't actually adapt or learn anything. The one exception I would say would be Omni pod five, because it it does actually determine the basil dose, it updates it every time you change a pod based on the average basil for the previous pod. But all of the other systems and all the other aspects of Omni pod five, are just reacting to the circumstance at hand, right? It's not changing how it doses in the future. But with the island, it truly is learning and adapting. So you ask about the meal doses, what we do is for each of those meal types breakfast, lunch and dinner, which are which adapt independently of each other. The island initially the first time you announced a meal, it gives just a very small weight base dose, which is point o five units per kilo. So if you're an 80 kilo person, it would be roughly four units, which for most people isn't enough, then what it does is the corrections algorithm automatically provides additional insulin to correct the blood glucose back down to the target. And it keeps track of how much insulin it needs to do that over the next four hours. Then the next time they announced another say usual for me breakfast, it looks back at that previous meal and say Oh, I gave for at the beginning but then I had to give another for correction. So it looks like a usual for me breakfast needs eight units, and then the eyelet gives 75% of that. So six units. Again, that's because we want usual for me to be most of the meals we want it to be a range so that people aren't counting carbs or just qualitatively estimating. And then the next meal keeps track of additional correction insulin maybe this time it uses a total of nine units. So the next time You announced the usual for me breakfast, it averages those two previous meals. And it keeps doing this up to seven meals. And so it's constantly looking at the average of the insulin dose it thinks you needed for the last seven meals once a day, once a day to as eight days old, it's forgotten, and you've brought a new day in. So we're always looking at the average amount of insulin needed for a meal announced that way over the last seven days. Okay, we give 75% of that.

Scott Benner 20:28
What happens if in the middle of that week, I throw a breakfast sausages or bacon in with my meal that I haven't had before. And suddenly my digestion slows down. It'll address that as a higher blood sugar when you get the fat rise. This episode is sponsored by Medtronic diabetes, Medtronic diabetes.com/juice box. And now we're going to hear from Medtronic champion Jalen.

Speaker 1 20:51
I was going straight into high school. So it was a summer heading into high school was that particularly difficult, unimaginable, you know, I missed my entire summer. So I went, I was going to a brand new school, I was around a bunch of new people that I had not been going to school with. So it was hard trying to balance that while also explaining to people what type one diabetes was. My hometown did not have an endocrinologist. So I was traveling over an hour to the nearest endocrinologist for children. So you know, I outside of that I didn't have any type of support in my hometown.

Scott Benner 21:25
Did you try to explain to people or did you find it easier just to stay private?

Speaker 1 21:30
I honestly I just held back I didn't really like talking about it. It was just it felt like it was just an repeating record where I was saying things and people weren't understanding it. And I also was still in the process of learning it. So I just kept it to myself didn't really talk about it.

Scott Benner 21:45
Did you eventually find people in real life that you could confide in. I

Speaker 1 21:49
never really got the experience until after getting to college. And then once I graduated college, it's all I see. You know, you can easily search Medtronic champions, you see people that pop up and you're like, wow, look at all this content. And I think that's something that motivates me started embracing more. You know how I'm able to type one diabetes, Medtronic

Scott Benner 22:10
diabetes.com/juice box to hear more stories from the Medtronic champion community. I used to hate ordering my daughter's diabetes supplies, and never had a good experience. And it was frustrating. But it hasn't been that way for a while, actually for about three years now. Because that's how long we've been using us med us med.com/juice box or call 888721151 for us med is the number one distributor for FreeStyle Libre systems nationwide. They are the number one specialty distributor for Omni pod dash, the number one fastest growing tandem distributor nationwide, the number one rated distributor in Dexcom customer satisfaction surveys. They have served over 1 million people with diabetes since 1996. And they always provide 90 days worth of supplies and fast and free shipping. US med carries everything from insulin pumps, and diabetes testing supplies to the latest CGM like the libre three and Dexcom g7. They accept Medicare nationwide, and over 800 private insurers find out why us med has an A plus rating with the Better Business Bureau at us med.com/juice box or just call them at 888-721-1514 get started right now. And you'll be getting your supplies the same way we do?

Steven Russell, M.D. 23:42
Well, there's going to be really two effects of that right. One is that it may slow down the absorption of the carbs. But it may also make you have to get more insulin ultimately, because the fat is interfering with the action of insulin. That's part of the reason that we only give that 75% Because we don't want to overdose right up front. If somebody eats on the lower side or for instance, eats something that's going to slow down their absorption. So we tell them not to consider the amount of protein not to consider the amount of fat. That's part of why we're conservative with that initial dosing knowing that our corrections algorithm can come in and fill in any gaps. So there's an element of adaptation. It's learning how much it thinks you need, but then we're only giving three quarters of that. And then there's an element of being reactive to how this meal might be different from previous meals. Because the correction algorithm it doesn't, you know, give 75% and then give the other 25% it gives 75% And then just sees whether more is needed. Maybe none will be needed, or maybe it'll be needed later in that four hour period than earlier because they had more protein. If

Scott Benner 24:54
I flip that on its head then and I eat that fattier meal to three days in a row and then the fourth day I just I have the same amount of carbs, but just nothing that's going to slow down digestion am I going to see a low blood sugar there?

Steven Russell, M.D. 25:04
Well, if you have nothing that's going to slow down the absorption, the glucose may be absorbed faster, you may get a sharper spike, actually. And so the correction algorithm may jump in sooner to give more correction insulin, but then, because there's no fat or protein slowing it down, it may not need to give any more after that initial dosing. But

Scott Benner 25:26
in that situation, if the carbs are the same five days in a row, the first four are fat laden. And so it's it's correcting, correcting correcting on the fifth day, does it not think breakfast is going to need more? And then suddenly, it doesn't is that's where I'm asking. Like, almost like with Right, right? Yeah,

Steven Russell, M.D. 25:42
but that, but that's part of why we only give 75% That's what helps that? Yeah, exactly. Okay. Yep.

Scott Benner 25:48
It's interesting. Now what happens? I mean, do you see I guess the question should be, do you see in testing or with real life use high blood sugars that it just can't affect? Like, what happens when a person who's used to taking care of their blood sugar's looking and saying, God, I have been to 50 for three hours, I want to Bolus here like what are they supposed to do then?

Steven Russell, M.D. 26:09
Right, I think that does sometimes happen early on before the meal doses have had a chance to adapt. So it's almost by design, that you probably if you're somebody who needs more insulin than that sort of low initial estimate, almost by design, you're going to see hyperglycemia after that first meal. And the island needs to see that hyperglycemia to give correction insulin to know that you're going to need more insulin next time. Okay. And so if you are patient and wait, and assuming your infusion set is working, and it hasn't failed or anything, the the eyelet will bring your blood glucose back down into range. And then the next time you announced the usual for me breakfast, it will increase the amount of insulin that it gives you for that meal. So you do have to be a little patient over that first day, or two or three or four days in some cases, or the islet to learn how much insulin you need and to adapt up and if you short circuit that process by say, I can't stand this my blood glucose 250, I'm gonna have to do something. Let's say you give insulin that the island doesn't know about, you give an injection and the island doesn't know that you gave that injection. Now the island doesn't think you need more insulin. So the next time you announce a meal in the same way, it's not going to give more insulin, because as far as it's concerned, you didn't need more insulin, you came down on your own. Or if you announce a fake meal, you know, one, do another meal announcement, even though you're not eating more. Well, we have certain rules about our adaptation, if you announce another meal in that period, that will cancel the adaptation because we can't tell which insulin we're going to give as ascribe to that first announcement or the second announcement. And so we just decide, okay, we're not going to learn from that announcement. You know, we're conservative about which meals we use for learning. And that's fine. As you're using the device over a long period of time, you don't have to adapt on every single meal. But if you consistently are using extra insulin, either by telling, you know giving insulin, the eyelet doesn't know about or doing additional announcements, you will completely block its ability to learn that it needs more insulin the next time,

Scott Benner 28:26
how long do I need to be on the device before I don't expect it to struggle?

Steven Russell, M.D. 28:31
Well, it varies a little bit from person to person, it depends on how high your total daily insulin dose is. But I would say that if you follow the recommendations for allowing the island to learn you, most people have have gotten to a pretty good place in terms of their meal announcements within five to seven days. Okay. Now, what we found in the pivotal trial was that on average, people got to their new time and range within 48 hours. But that doesn't mean that the meal doses it adapted that fast. What was happening is that the islet was relying more on correction, insulin, the correction algorithm. And they they were getting under pretty good glucose control, on average within that 48 hour period. But over time, there was a shift as more insulin was coming in from the meal algorithm and less insulin was being delivered by the corrections algorithm. Okay.

Scott Benner 29:28
So here's my next question. Then. Exercise. I have a message here a question. So this person says or a camp director to T one camp. He says we have many families that are on I let the they're debating on sending their kids to camp due to concerns that there's no way to reduce basil for demands of camp. These people are citing to him that they are having issues around activity with no way of announcing activity or reducing Basal. So what do you think about that?

Steven Russell, M.D. 29:56
My recommendation to kids go into camp is to think about how much their activity is actually going to change a camp. Because some kids are much more active at Camp than they are at home, and others are not. So if they're not going to be more active, it's, it's probably not going to be a different issue than whatever they've been addressing at home. If they are dramatically increasing their activity, then one of the things they could potentially do is raise the glucose target. That's a lever that to that we've haven't really talked about to date, or at this point in the in the discussion, but there is the possibility of changing the glucose target. I would say that, you know, exercise is a whole conversation. Yeah, and right now, we don't have an exercise mode for the eyelet. What we recommend is that people, for the most part, keep the eyelet on. And most kinds of exercise actually don't cause the blood glucose to drop, you know, weightlifting, high intensity, interval training, and so forth typically don't cause a drop in blood glucose. And so it makes sense to leave the eyelet on and running. For exercises that do cause the blood glucose consistently to drop, the recommendation is to do one of two things, either leave it on, you'll, it'll see your blood glucose drop, and it'll suspend insulin. And you may still need to take some carbs to cover that period, until the insulin levels of the insulin on board decreases. And that takes some time because it takes, you know, on average, five or six hours for insulin to really clear. And so you should definitely have some carbs available. And then take whatever amount of carbs is needed to keep you from going low, although try not to over treat below, so that you're not provoking the islet to give more insulin, or you can just disconnect from the eyelid. And that way, the insulin levels will obviously be much less, much lower. And if you disconnect from the eyelet, then you have the option to take some carbs to raise your blood glucose. Okay, going into the exercise, just like any other AI D system, it's really important not to re carb load for exercise without disconnecting from the island. Because then what you'll do is you'll cause your blood glucose to go up the Island will see that and respond to it with additional insulin dosing. And so then you're going into exercise with even more insulin on board. We are aware that there is I think, a need for an exercise mode for the island. And we have designed one, and we're working on the strategy for for getting it out there, we have to talk to the FDA about you know what, what the appropriate way is to get that update out. And once we have clearance for that it will come out just like any of our other over the air updates. Okay, is that,

Scott Benner 32:51
uh, you hope for this year or you can't even say inside of a calendar year?

Steven Russell, M.D. 32:58
I think that I better not have any chances of getting it out in a particular timeframe. I certainly hope for it this year.

Scott Benner 33:06
Okay. You're working on it. Now. You hoped for it this year. Okay. Well, that maybe doesn't answer that question. Hey, you talked about targets. What do you mean that you haven't been telling people that there's targets? What does that? Well, we hadn't talked about it in conversation, one of the targets that I can set with eyelid.

Steven Russell, M.D. 33:20
So there are three levels of target, there's the usual target, the lower and the higher target. We intentionally don't talk about the what glucose level those are right up front, because we're trying to simplify it and to allow people to manage their diabetes without thinking about numbers. But everyone wants to know what they are. And they are on 120 is the target for usual 110 is lower, and 130 is higher, I just want to make the point that they aren't directly comparable to targets with other systems, because the algorithms work differently. So Ed likes to joke that if somebody asks what the target says, he's like, Well, what is the gear ratio in your car, what matters is not actually what the target is, but what kind of average glucose you can achieve with it. And what we found in the pivotal trial was that about 60% of the time was spent with the usual target 20% with the lower 20% with the higher. And with that mix of targets, we had an average glucose of about 155 and adults. So you know, right at an average a one C of 7%. But of course, the target for a particular person is what's appropriate for that person, right. And I should also point out that there you can have two different targets a day so you can have one target during the daytime and one target at night. And you can choose the timing of the start of that daytime target and shifting back to the nighttime target they can be at any time. So that's something that we think people shouldn't just do on their own, they should talk to their healthcare provider about but it's not uncommon to have two different targets one during the day and one at night, can they be automated? Yeah, they are. So once you set the pattern, it's just a recurring pattern happens. Okay. So, you know, probably the most common target is just to be at the usual target the whole time. But probably the most common after that is to have a target, that's one step lower during the daytime when people are eating, and a step higher than that at nighttime when they're not. And that means the eyelid can be a little bit more aggressive at dosing, any postprandial any after eating rise in the blood glucose, okay. And it just tends to make it more efficient at managing any hyperglycemia during the day for some people.

Scott Benner 35:51
My next question is from a woman who says that her husband has a co worker, a friend who's using ILF, that the guy's a snacker. And the pump doesn't seem to be handling that well. So what should I do, if I'm the kind of person who walks through the room and grabs seven carbs of something and throws it in my mouth, what we

Steven Russell, M.D. 36:07
recommend is that people announce carbs that they're eating, snacks that they're eating, if they have as many carbs as a meal. So the usual for me meal is one, a half a less meal is half of that. And a more is one and a half times that. So those are sort of the levels of what we call a usual for me a more and less meal. So if a snack has as many carbs as a less meal, so it has as many carbs as half of your normal breakfast or half of your normal lunch, or anywhere down to a quarter. So that that range is like a quarter to three quarters, that's a last meal. If it's anywhere in that range, they should go ahead and announce it. We say if it's less than a quarter of a usual meal, they shouldn't announce it. And they should just let the correction algorithm manage it. But if it's in this case, if if they're finding that they're having hyperglycemia, after not announcing those snacks, my guest is that they have significant carbs in them. And they probably should be announcing them as meals. So

Scott Benner 37:15
you're thinking already about the activity addition? Do you think about like a grazing mode? Or like you don't? I mean, there's, I mean, it'd be Listen, you've been an endo for a long time, right? So Thanksgiving comes, I tell people all the time, like put your basil up 20% Like you don't you mean, like you're gonna be eating throughout the day? Like, let's lay a heavier, you know, blanket of insulin over the situation? Like, is there thoughts about like, I guess my, my real question should be, as you're seeing it in people's hands. Are you thinking like, oh, this would be good to add? Like, are you having those thoughts, we

Steven Russell, M.D. 37:50
actually have designed a snack announcement that will be even less insulin than a less meal. There'll be a kind of a different category. I would say that for most snacks. If they're gonna if they're causing a significant rise in blood glucose, it's a meal, they probably should have been announced as immediate. Right, right.

Scott Benner 38:10
But I'm talking about like, the real like, you know, like, I don't know, a candy bowl and you grab seven gummy bears, and it's, you know, it's eight carbs or nine carbs, something like that. And like, you have no ability to maybe the biggest, like sticking point right now. And I am I'm certainly not leaning into it. I am not a person who says I don't want to change, right. So I don't get upset when Facebook changes the way it looks. You know, if a car company puts out a card, it's not quite done yet. I'm like, Oh, this is what they need to do to get to it. Like I'm okay with it, like, so I'm flexible about that stuff, and about diabetes as well. So my question, I guess is, how do people who are just so accustomed to doing something one way? Do you think if they just tried it? Eventually they they'd be like, Alright, I'm good with this? Or do you think that there's going to be a happy medium in there? So you're looking for a snack button, you're thinking about exercise? I mean, I love the idea of eyelet. Like, it's it's a romantic idea about taking good care of yourself with less input. But what does that mean? I guess, like, I'm going on a little bit, but if you told me my daughter's blood sugar was going to be 155. I'd say oh, God, what did we mess up? Right? So like, I'd be looking at like, is the basil wrong? Do we have the insulin to carb ratio wrong? Like there's part of me that feels like we're having this initial conversation with beta bionics, who is now looking at how to go to a by hormonal pump. And I'm thinking like, I'm going to look back on this one day and just see this as the infancy of the pump is is that maybe how I should be thinking about that?

Steven Russell, M.D. 39:46
Well, let me say that I think we we really see tremendous value in by her model. And the reason is, I like to think about it in terms of a population health reason with the island we can get About 50% of people to have an agency less than 7%, most of the remainder to have anyone see less than seven and a half with the buyer hormonal, we expect to be able to get 90% of people to have an agency less than seven, and 100% to have an agency less than seven and a half. That's what our pre pivotal data says. So it's it's lowering the average glucose, if you will, by about another 15 MCs per deciliter or about another half percent of a one C, you know, why do I keep talking about seven and seven and a half, obviously, seven is the recommended a one C target that ADA has come out with, and that's based on the diabetes control and complication trial, it's worth pointing out that the diabetes control and complication trial started with people who had an average valency of nine. And what they found is that if they got their control group that had and I want to point this out, on average, a one C of 7%, not everyone got lower than that, the intensive control group had an average a one C of 7%, they found that that was sufficient to cut the the progression of complications or the development of new complications to the background level. So if you look at that hockey shaped curve of complications versus a one C, below seven, there is no signal for further progression of complications. Since that time, there has been some data like registry studies from the Scandinavian countries that, you know, look at people who have been in there, who everyone who gets diabetes is immediately tracked. It's like a registry studies the whole country. When they go back and look at that, they find that people who had an A one C below seven and a half from shortly after the time they were diagnosed throughout their entire lives never get complications of diabetes. And one more piece of information is people who have what's called Glucokinase Modi, it's a particular single gene mutation. That means that people have a average glucose from birth in the 160 to 170 range. They're born, their their pancreas works, it just works for a higher setpoint. So there's several 1000 of these people in the world. They've been carefully studied, followed throughout their lifespans, they don't get any complications of diabetes. So I think that probably if we were able to get everyone below and a once you have seven and a half, diabetes, complications wouldn't be a thing.

Scott Benner 42:29
You bring up something that I never know how I've been doing this a long time. And one of the questions that people asked me that I just don't know how to answer is what is the benefit of being lower? And I don't know, I have, I always say I have no idea. Like, I don't know, when it always comes down to if you can, without it being too much trouble. I'd err on the side of doing it. I can't tell you. And I mean, those are interesting studies for certain. I think the problem ends up being and this is obviously more philosophical than anything else is that if it happens for a person, they live their whole life with a seven, for example, and then they have a problem. You can't go backwards and try again. Like, right, so there's the that's the rub. And I also feel like if we were alone in a bar, and I asked you, if you were working on the algorithm having a lower target, you would be like, oh, yeah, of course, Scott, we're on that, don't worry. I'm just super interested in where this is all going. I love the form factor. I love how it wants to work. And I appreciate you sharing that information, because that's the closest I've ever come to an answer to that question like, what's the what's the benefit of lower?

Steven Russell, M.D. 43:36
I would argue that there isn't, there isn't evidence that there's a benefit for lower. So what do I tell people? I tell people that if you can achieve a lower ANC without hyperglycemia, then great, and you want to then great, then fine. But if there's a significant trade off in terms of increasing hypoglycemia, it's not worth it. No, I don't I hate to either. And I would never target below, say 6.5. I mean, there's, you know, you just can't make an argument and evidence based argument for for shooting for an A one C lower than that in my mind. Well,

Scott Benner 44:09
you're also not trying to sell a pump to somebody who's walking around with a five, five a one C to begin with, by the way, none of these algorithms without some real like, intervention, tinkering, like deeper understanding. Like you're not you're not getting there unless you unless you really know that hey, yeah, that makes me wonder, how does the algorithm work for people who are very low carb works

Steven Russell, M.D. 44:30
great. I mean, that's, that's an easy case. Okay. Those people tend to do well, no matter what they do. So the algorithm works really well in that context. Okay. I kind of want to go back to that issue of, you know, what's the advantage of going lower? And, you know, as I said, I think that there isn't much evidence that there is an advantage in going lower and if we go to the sitting in a bar scenario, the reason that I give for wanting to use The by hormonal system is that we can just get more people, in this case, everyone to have that a one C that I think will pretty much eliminate complications, it's not about taking somebody who already has an A once fee of 6.5 and bringing them down to six, I do you think that for individual people, because we can be a little bit more aggressive with the algorithm, it probably will give them the capability of lowering their agency by another half percent, whether they should is a separate issue. I would at that point, if if somebody you know, has an agency of six and a half, the way I would use glucagon is not to further lower their agency, but to further lower their hyperglycemia. Okay, that's where the, that's where the benefit comes from my perspective perspective. And also, it has the additional advantage of meaning that you don't have to take oral carbohydrates, with all these insulin only systems, they're all pretty good at suspending insulin, if it looks like the person is going to go low. But they all occasionally fail in that, that you still need to take carbs to treat a low. And, you know, it's frustrating to people to be, you know, doing exercise because they're trying to, you know, trim down or something, and then they wind up having to take a bunch of carbs, to treat a low that happens when they exercise. So the glucagon has an advantage from that standpoint, and it also just the spontaneity standpoint, you know, not having to take carbs, just seeing your blood glucose going low, and then just going right back up without you doing anything, that certainly increases peace of mind. And, you know, whereas with the insulin only system, we found that we were able to lower that agency by half a percent, on average, more for people who are higher, obviously, we didn't lower it for people who already had agencies below seven, and we didn't increase hyperglycemia. With the bio hormonal system, we're able to lower it even further. And actually also reduce hyperglycemia.

Scott Benner 46:58
You're, you're testing it now.

Steven Russell, M.D. 47:01
We are not testing it again. Yet. We just recently announced an agreement with Cirrus pharmaceuticals, to test a new formulation of stable glucagon. And we're moving forward with that with all haste, because we really do think that that by hormonal system is going to provide terrific advantages.

Scott Benner 47:24
People are listening right now if you go back a week in the podcast, you'll see an episode with zeros and eyelet talking about that agreement that they that that Dr. Russell, just, yeah, I've done an interview about that yet that's gonna go up right before your stuff. I'm excited by it. Like, I have to tell you, I believe my daughter's 20. Right. For the most part, my daughter is away at college right now using IEP s. And a she's got an A one C like six one, I think and that's at college like and not just the college like you're thinking of it, my daughter is not like a crazy party person, she's working incredibly hard, like seven days a week doesn't take off long quarters that go into six months, times when she's at school, and that she's managing that is insane. It's absolutely fantastic. But we also do a really good job of her not having lows. Having said that, I've seen really emergent lows. And I've I've watched her have a seizure when she was two years old, she had one when she was a high school senior, it was both times variables that were not common everyday stuff. I just think in my heart if her pump would have noticed that and given her some glucagon, it might have been maybe the greatest thing that I've ever seen in my life. Like, seriously, I don't know how people who are using insulin are not Ultra aware of this constantly. To take that brain worm out of the back of my head would be amazing. You know what I mean? So I'm with you, I that. This is probably an unfair question. But if you ever talked about would you license this to other pump companies if they were interested? Or are you guys gonna keep glucagon, your your, your algorithm and everything? Like it's some point? I keep thinking at some point one of these systems is going to show and everyone's gonna say, Well, yeah, that's how we should be doing it. And then what are we doing? Then we're in what a retail game. Like, that's ridiculous, right? Like, at what point? Do we just say, because listen, I have my questions here for you 50 people, this is nice. We'll never be tubeless. So like, you know what I mean, there are people who they don't care if it comes with a free pack of bubble gum. You don't I mean, in a bus pass, if it's not tubeless they don't want to hear about it. Like that was my wonderment like, you know what I mean? Like, I wonder if it'll ever it will really get into that interoperability that we all kind of got teased about with tide pool. It doesn't seem like it's ever gonna happen, I guess is what? Yeah,

Steven Russell, M.D. 49:52
we're certainly aware that people are interested in the tubeless form factor. All I'm gonna say about that right now, but okay. where you're we're certainly aware of that. And we realize that that's really important for some folks. So all right,

Scott Benner 50:07
I take your point, don't worry, I can read between the lines. I'm good.

Steven Russell, M.D. 50:12
As far as as far as licensing the algorithm, I mean, you know, Ed, would, it would be a better person to talk to you about this. But we definitely, were interested in making the algorithm available to other pump companies. And there were no takers. That's why beta bionics got started, you know, the, the idea of having a system that's this automated was something that I think was scary to other pump manufacturers, and they preferred a system that still left more of the responsibility in the hands of the user, I say, so that if something went wrong, they could say, well, you know, we were not responsible on

Scott Benner 50:52
the pods gotten away from that they're basically telling you now, we don't even want you to understand how it's working. You know what I mean? Like, it's, you know, it's going to do what it's going to do and let it do it. And I mean, you're saying that to you are saying, like, let the thing work, right, like don't? Well,

Steven Russell, M.D. 51:05
I think that's that it's different, I would argue, because we're very clear, we have a lot of information about how our algorithm works. And I think that's a little different than than Omni pod. But Omni pod also still requires the user to do a lot. What I've learned clinically with the Omni pod is that if you want to get good glucose control with that system, you need to make sure that you fully dose your your meal, your your meal, carbs, sure, because it's a fairly conservative algorithm in terms of giving corrections and adjusting basil, and so forth. So you can get good glucose control with that, but you need to be pretty aggressive about giving the full amount of insulin for carbs, and also giving manual correction bonuses to get the best glycemic control. So it's still quite dependent on the expertise of the user in being able to count carbs accurately, and to choose the appropriate times to give corrections and you know, to to give the corrections that are needed and not give the rage boluses. So it still is quite dependent on the user, whereas the eyelet is not with the island, it's really more about not doing certain things that can confuse the island. And and, you know, sort of teach it the wrong things, right. It's often about doing less with the island, not more than and in that sense. It's really unique. Every other diabetes control system, the more you interact with it, the more you pitch in, the better glycemic control you're gonna get. And the island is unique. And that's not the case. What's

Scott Benner 52:37
our average excursion, I expect with an eyelet? Like, what number am I going to see a spike go to, for example?

Steven Russell, M.D. 52:47
Well, if you have a meal and you and you do announce that meal and your meal doses have adapted, it might be quite small, might be 150. But it might also get to 200 are transiently above 200. It just depends on where you are in that in that range of what your usual for me meal is for instance, how much of the insulin actually came up front and how much is coming as correction Bolus. I see. I

Scott Benner 53:14
mean, it's all like relative like I when I tell people my daughters they want to see I think they imagine like an ad three blood sugar that goes on for weeks and months, but she has spikes at meals just like everybody else.

Steven Russell, M.D. 53:25
And soda. Why, by the way? Yeah, I don't have diabetes. Yeah, but I wear CGM. And my blood glucose goes above 200. Sometimes no kidding. Absolutely. And yeah, I mean, that's not uncommon for people without diabetes. And I think we're only now appreciating it because you can wear CGM and you can see that which you normally wouldn't have seen right. Now mine automatically comes back down and I don't have to do anything about it. But it's not uncommon for it to go above 200

Scott Benner 53:53
Can I ask how old you're 5066 Okay, I'm 52 I have to wearing a CGM. I have to eat intentionally to get my blood sugar to go over like 160 But I'm talking about like pizza on top of fast acting sugar that's like it's I had to like gorge myself to make it happen. So it's interesting. Look at give me I'm showing off. Steve varies

Steven Russell, M.D. 54:15
varies from person to person to person to person. And I think you know, breakfast is the meal if I eat a high carb breakfast, I'm definitely going high. Okay, because it's you know, and it changed the way I ate you know, I would in the morning would have you know, some waffles and some peanut butter and maple syrup. And when I saw what that did to my blood glucose, all of a sudden I started eating more eggs.

Scott Benner 54:39
Everyone who's ever come on here, as Kevin Sayer was on here one time he's like, I wear my own product and just like four things I just stopped eating. Yep. So listen, we talked about like, okay, but even a person without diabetes is going to see an excursion up but there's also truth is there not that a person without diabetes also might see a blood sugar into the 60s throughout the day. So I mean it So it begs the question like, why were you guys not a little more aggressive with it?

Steven Russell, M.D. 55:04
Well, I think it's the idea is to try and tune the system so that we're improving or lowering the average glucose, we're lowering the a one C without increasing hyperglycemia. That was, effectively the way it was tuned over those pre pivotal studies that we did. And that's what we found in our pivotal study, we lowered aid, what CD didn't change the amount of time less than 70 and less than 50 for sort of a pragmatic choice to tune it that way. And of course, that means that, you know, in some people are going to have more hypoglycemia, and some people are going to have less, right. And we're just trying to kind of pick a middle path where we get the best glucose control we can without putting people at risk.

Scott Benner 55:52
Yeah, in fairness to every company. And every algorithm, there's the nebulous part of the conversation that we don't talk about, which is some people just eat differently. And so you're trying to make a thing that you can just handle the world without knowing each person individually. So there has to be buffer, I guess, on either side of it. I guess my more maybe, maybe my, my bigger question is, if the mind if the if the brain trust that is beta bionics now existed on day one. Do you think the algorithm would be more aggressive today? Like, have you learned more now? Or you're still very comfortable with how it's done?

Steven Russell, M.D. 56:25
I don't think it would be different. Okay. I think we might, you know, we're seeing that, we would like to add a, a, an exercise button on the island. And that wasn't built into it from the very beginning. And so now we want to add it. And I think one of the reasons that wasn't built into it from the beginning, is because the island kind of started as a by hormonal system. And we didn't really need it for the bio hormonal system. And then we got to insulin only. And we found that we were still getting very good, sort of surprisingly good results with the insulin only system. And it was quite differentiated, it was getting results similar to what other ad systems were getting, but with a lot less work. And we thought, you know, that is, that is a differentiated thing that we should get out there, even though, it's going to take us a little bit longer to get the buy hormonal product out because of the complexity of doing a drug trial as well. So that's why we chose to move forward with the insulin only system, because it it is a big difference. It I think it does bring a tremendous amount of value. But that's also why we're not stopping on the by hormonal system, because we think there's additional value that we can bring by bringing in that that by hormonal piece. So back

Scott Benner 57:45
in the day, the glucagon wouldn't hold up in the pump, right? That was the biggest.

Steven Russell, M.D. 57:49
That's absolutely right. So we did all of our pre pivotal studies, or most of our pre pivotal studies using the red kit from Lilly, and we would just reconstitute that glucagon. And we had to change it out every day. Because it wouldn't, it wouldn't last longer than that. And shout out to Lily for providing us enormous amounts of that red kit glucagon for free, they were fantastic about that, we would literally get boxes that were three feet on a side full of nothing but those red kits to be able to do our pre pivotal studies. But that wasn't going to be practical in the long run. Yeah. And so that's why we needed a stable glucagon. And fortunately, two came along. And we're, you know, moving forward with with one of those a new formulation of the zeros that the actual zero Asal glucagon that's available right now for rescue treatment. That one won't work in a pump, because it actually damages the plastic. But a different version of it a slightly modified version, it looks like it's going to work just fine and a pump. And so that's what we're going to be moving forward with.

Scott Benner 59:00
Paul was explaining that to me in that recording, because I said, Oh, you're just gonna put g voc into the pump. He's like, no, no, we have to do a formulation that changed. And I was like, Oh, that's really interesting. Like, and you know, pretty cool. Honestly, Steven, that an idea? That is how old now? I mean, when when did that study happen with those red kits?

Steven Russell, M.D. 59:20
For you guys? Well, the very first one happened in 2008.

Scott Benner 59:24
Okay, so a lot of years later, like 17, that sounds like to me, you know, you're still on it, which is pretty great. I also want to say this too. I recently just floated out into the world, like if anybody wants to come, I think I said online, if you have a device or a product that helps people with diabetes, and you want to come on the podcast and talk about it, come on, and you and I are having a pretty honest conversation I'd say about, you know, about eyelet. And I appreciate that because not everybody stepped forward and was interested. So thank you. I really do appreciate talking about it like this, because it's fair to say all these algorithms and They can all be better, obviously, right? It doesn't it's not a denigration of what they are now. And if we don't talk about it like this, then where's the onus to do the work? You know what I mean? Like I all these companies should be in a room right now looking at their data thinking, how can I make this better? And if they're not I, I hope, and I hope they'll consider doing it. And it sounds like that's what you guys are doing. So I appreciate it. I

Steven Russell, M.D. 1:00:23
totally agree. I mean, look, AI D is so impactful, it makes a huge difference. CGM was huge. And AI D is huge. And I think that probably everybody with type one should be on an AI D system. I think that beta bionics with the eyelet is trying to be available and reach a group of people that may have a harder time working with some of these other systems that require more sophistication on the part of the user ability to carb count, choosing the right times to do corrections, and so forth. Yeah. And so I think that I talked about democratizing good glucose control, it shouldn't be just for the elites, it shouldn't just be for the small percentage of people who are very numerous and, you know, have the best executive function so they can be on it all the time, and paying attention and, you know, not working two jobs, because, you know, you really with diabetes, it's like a second job. So if you already have two jobs, then you know that your ability to spend the amount of time it takes to get that excellent glucose control is limited. So I think one of the things that we bring is bringing the possibility of good glucose control to people who might not be able to do it with the other devices. And for people who already are able to achieve good glucose control with other devices, it's another option that can reduce the amount of time they have to spend with it. Yeah, not everybody wants that. You know, I think it's important to note, it's not for everybody, that kind of people who are want to be on loop or one of these other systems and are the people who are have control IQ, but they've turned it off. And they've just got it on, you know, sleep mode at all times. Because they, they they feel like they can do a better job. Those are probably not the people for the eyelet. Right. Frankly,

Scott Benner 1:02:14
the button the button pushers in the dial. Turner's might not be your your bread and butter. But there's a listen, I say it all the time. Well, let me go back earlier, when you were describing how the algorithm works, I thought, Oh, my God, that's how I managed manually, right? Like, I think back to like the Pro Tip series in this podcast, which is just me telling you how I got my daughter's a one C down and stable, and how I think about insulin and diabetes. It's all timing and amount. And, you know, understanding the variables like that, that is literally what the algorithm is doing. I didn't realize that till the first time, I saw my daughter on loop, and I could watch through Nightscout. I watched it work. And I thought, Oh, I would have Temp Basal here. Or I would have done this, like you don't even I was like, This is so interesting. It's doing everything for me. And I'm getting asleep. This is insane. You know how happy is about it. But it's never lost on me that even though I watched the podcast help, like, seriously, I'm not trying to, like blow my own horn here. But But this podcast charts in 48 countries around the world, right? Like I wake up to a dozen notes every day from people are like, Oh my God, look at my one say, it's amazing how many people it touches. And yet, if I stopped myself in a quiet moment, I remind myself, I'm probably not touching 5% of people with type one diabetes. Like it just I'm not reaching nearly everyone. This is for the people who have the time to listen to it, that have the acuity to understand it, and have the patience to put it into practice that have the patience to watch it go over and over again until it finally makes sense to them. And then they can kind of do it blindly without even thinking about it. That is not something that's going to work out for everybody. I just know it isn't. And everyone should be thinking about that like, because if not, if not, then what we're doing is we have a bunch of different pump companies who are saying to themselves, there's a block of people with type one diabetes, this percentage of them can figure out how to use this pump will market to them. And then the rest of them just they get left behind. That to me is like I have thought from day one about this algorithm like this is for the people who got left behind. And if you can turn it into something as for everybody, not just for them, then in my opinion, you know, you just jumped ahead. So, you know, everyone else should try to keep up. But you know, I would love to see this on more people and see how it works out. Speaking of it being on more people and seeing things, have you seen it on somebody who's using a GLP yet? I would imagine that's like a low carb vibe. Right?

Steven Russell, M.D. 1:04:46
Right. We have I don't have statistics for you. Sure. And I would say also in our we do post market surveillance all the time. You know, we we can look at the whole population of people on the island and Look at what average glucose they're getting and how much hyperglycemia all that stuff we look at. Unfortunately, we often don't get any information about other medications that they're on. That's not part of the information that we get. But I am aware of people on GLP one agonists with the islet. And as far as the eyelids concerned, that just means that the person is using less insulin. And their glucose excursions are less acute, less sharp, or eased, and they're easier to manage. So and you know, on the island, of course, is continually adapting to the need for insulin. So if somebody starts on a GLP, one, and their insulin needs go down, the islet just adapts to that online. And if they are absorbing food a little bit slower, the meal dose adapts to that. So I don't think there's any reason why you think that it wouldn't play nicely with a GLP one agonist and in the cases I'm aware of, it seems to play just fine with GLP, one agonist,

Scott Benner 1:06:00
I'll say here that my daughter is using Manjaro. At the moment, I I've said it over and over again, but so that, you know, I estimate show you 16,000 fewer units of insulin this year because of it. And it's, it's, and we were doing fine before like, you know, we in fairness, and we're very good at using insulin like so it doesn't matter how high or low carb she has we know how to manage. But this is just like her stability is even more stable. The it gives the algorithm that makes the algorithm look like it's better what it's doing. You don't I mean, like, Yeah, I used to say to people, you know, if you if you came to my house that I have to give one of these things back, I probably fight for the CGM over the pump. But then once they got integrated now I kind of see them as one in the same thing. But I'd have a long, hard think, before you ask me, What can I have back the algorithm or the GLP. I'm doing a small series right now with a guy you might know Tom Blevins. He's an endo in Austin. And Tom and I are recording a small series about GLP I am literally hoping to be one of the voices that pushes JDRF and other big institutions into pushing for coverage and testing, you know, for people with type one diabetes, just based on what I've seen, and what I'm hearing from other people, which has nothing to do with our conversation. But anyway, sure,

Steven Russell, M.D. 1:07:21
I will say that, you know, this is entirely separate from my role of antibiotics, but I'm still a practicing endocrinologist. And I personally use GLP one agonists in people with type one diabetes, who also have the physiology of type two diabetes, and there's plenty of those people, right? There's nothing about having an autoimmune attack on your pancreas, or B or beta cells that modifies those genetic risk factors for Type two diabetes. So as far as I'm concerned, you can have both types of physiology. And for those people, absolutely GLP. One agonists makes sense, I think, perhaps more controversial is the idea of using even people with type one diabetes who are lean, not insulin resistant. And I think that's where, you know, the companies just chose not to pursue an indication for it. But clearly, it does reduce the amount of insulin use and it and it does tend to appetite. Yeah, and appetite and it smooths out the glucose excursions. So it does make a lot of sense that, that it just makes diabetes easier to manage,

Scott Benner 1:08:33
I have to thank you, because you just put that I've been trying to make that point. I keep fumbling with it. Every time I tried to make it. The idea like they used to hear people say like, my doctor says, I have type one and type two diabetes, then people would say that and you'd like you can't have both and it would kind of like, you know, people get upset. But my daughter, for example, like very clearly has PCOS symptoms, which is just the thing you really can't even get diagnosed, you know, weight gain, acne, high need for insulin no matter what she's eating or not. And she didn't, you know, she wasn't overly you shouldn't gained a ton of weight. But she was also 20. Still, whereas I know a person who doesn't have type one diabetes is a young girl in her mid 20s, who has PCOS. And is a clean eater who exercises daily who gains weight every week, just can't stop, right? And goes on a on weego V, and boom, it all just ends and the weight starts coming off and everything goes back to normal. It's absolutely amazing. But the what you're starting to see right now and you probably know this because it sounds like you're doing it is you're you're having to diagnose your patients as having insulin resistance and type one diabetes to get around the quagmire that is the insurance at the moment on this. But I just I really, Dr. Blevins believes that if you don't go out there and do the studies that it's going to be very difficult to push through insurance and it it's going to take a big voice to push the studies along. So Oh, I'm trying to magnify that if I can, just like I'm trying to do with you and anybody else like I listen, there's part of me, Steven. Yeah. I don't want everybody fighting. And I take ads from a pump company. Like, I have no trouble saying that. But I don't want anybody get uncomfortable. Like, I make a podcast and you know, people pay me for ads, but I got a kid with diabetes. I need you guys all out there hustling? You know what I mean? You know, for anybody, you might be listening, thinking, God, I can't believe Scott had this these people on when you know, Omni pod buys ads from him. I want Omnipod to work hard. And I want you all to work hard. You know what I mean? So I mean, absolutely.

Steven Russell, M.D. 1:10:35
I think it's incumbent on all of us to continue innovating. And I'm, I'm thankful that we have multiple different AI D Systems, because I think they're not all perfect for everybody. And I think everyone should be on an AI D system. And it should be one that fits their lifestyle and their predilections and their interests and, and it works for them. And so I think there's no way that any one system is going to be all things to all people. So I think it's great that there are choices. I agree.

Scott Benner 1:11:06
I also think that when you're being diagnosed, the next thing you thought you should have is, what's that you're putting on my arm? Is that a CGM? What is that, but you shouldn't leave the hospital without a CGM. If you had a heart issue, they would put a monitor on you. They would not let you out of the hospital without it. But

Steven Russell, M.D. 1:11:22
yeah, 100% agree. Yeah. And I, you know, one of the things I do is I still work in the hospital. On the weekends, we have this inpatient diabetes service. And it's getting pretty big. Now it takes three doctors to cover it every weekend, we see all these patients in the hospital. And, you know, we're seeing the patients who are already on our service with consults to us. But we also get new onset people in, you know, people get new onset type one diabetes all the time as adults. So I see these folks. And I really try and take it upon myself to try and get them on CGM as soon as possible, you know, send the script down to our pharmacy, and get them on a Dexcom or an Abbott sensor before they leave the hospital. And put in a statement of medical necessity for an AI D system if they're willing to consider it. Yep, as soon as possible. I'm really grateful that there, there are more than one systems out there that are automating glucose control. I really like to get people on CGM as soon as possible after they're diagnosed. And, you know, there's adults getting diagnosed all the time. And, and I really want everybody to think about an AI D system. On the day they're diagnosed when I see them in the hospital. Yeah,

Scott Benner 1:12:42
I completely agree. And I like again, not only is it good for people to have choice, but from a different perspective, it is very good if the space is not controlled by one company, we've seen in the past that company gets comfortable, and then nobody gets served. So that's

Steven Russell, M.D. 1:13:01
absolutely true. And I and I think it's really, it's really important for us to continue innovating. You know, once you have something that's approved in a very highly regulated industry like this, there is the tendency to feel like well, we, you know, it would be too costly, too time consuming. To continue to innovate, let's just keep pushing the product that we already have. But I don't think any of us should be satisfied with that. We should always be looking to how we can make it better.

Scott Benner 1:13:28
Yeah, listen, I have a two year outboard for podcast, they should be doing the same thing. As soon as everybody I always say you get your thing through the FDA, everybody gets a vacation, then we come back. And we started thinking about how do we push this farther. And if you're not doing that, then I think shame on you a little bit. So seriously, like, it's just, I mean, you saw it in the past, one company had control of everything. And they were just like, whoo, and now look at them rushing around trying to get back in the game, you know what I mean? So it

Steven Russell, M.D. 1:13:55
can, it can be that kind of behavior winds up coming back to burn you in the end, but it tends to be years later in their case. I think now that there's more people in the game, it's doesn't take as long right? For that kind of behavior to loop back and burn. It's

Scott Benner 1:14:14
not just cyclical now because it's cyclically happening, like four or five different in four or five different buildings. And that really is valuable for us because listen, if you're gonna talk about the business of it, we got the big thing through now we're going to make the money I'll cut costs right now I'll be a big hero bringing a nice bonus for myself, I'll retire and then the problem will be on somebody else when we're not innovating anymore. That's the backside business problem. So I guess what I'm saying is if you're in a business where that's happening, you better speak up or 10 years from now your job is going to be worth much anyway there. I just want to make sure everybody's working Steven, that's all I got. I got a I got a daughter I got to worry about you know, well you know,

Steven Russell, M.D. 1:14:49
we are we are working very hard on it and and as you as you know, you know, we launched the product with G six and shortly after We put out an update that made us compatible with g7. And, you know, we've been very clear that we want to be compatible with any CGM that is able to meet the ice CGM stand there. It's because we think people should have choices. And we're going to continue to innovate in other ways as well. Cool.

Scott Benner 1:15:19
It sounds like maybe you've reached out to the people ever since then, perhaps they just got their IEP s. Distinction, didn't they? They did. Or I used the wrong terminology. I CGM. Excuse me, but yeah, CGM.

Steven Russell, M.D. 1:15:34
Yeah, yeah. And I actually a fan of that technology. I, one of the things I do in clinic is to replace and replace those Eversense sensors. And I think it's a it's a good system.

Scott Benner 1:15:46
That's excellent. Good. Well, hopefully, I mean, I, they buy ads from me too, like, just so I'm being clear. But I always say to them, like, you know, like how the transmitter vibrates as like, Wouldn't it be cool if the pill vibrated? And they're like, Wouldn't it be? And I was like, it would be wouldn't it be? Like somebody work on that? That's a great, I mean, can you imagine that an implantable CGM that gave you a little shake if you were low or too little shakes? If you're getting high? That's insane. I mean, I think it'd be fantastic. Anyway. Is there anything I didn't ask you that I should have something we left out?

Steven Russell, M.D. 1:16:18
Hmm, let me think I know there was some because I had some things on our because I have here.

Scott Benner 1:16:25
Can you tell me about the you

Steven Russell, M.D. 1:16:27
know what, there was one that was on there? What are the best practices in the first two weeks? I'd love to address that, please. All right. Well, one of the questions that comes up is what to do when you first start on the island, what are the best practices to get the most out of it. And those best practices really are derived from the way the algorithm works, what you want to do is give the algorithm the best chance to learn you as quickly as possible. And because the way the meal algorithm works is it looks at the the four hours after the meal, we do ask people to space out their meals by four hours, initially to allow the islet the chance to learn how much insulin they need for the meal, because we have these rules that if you get another meal announced within that four hours, it will cancel the adaptation, okay, now there is an exception to that, which is if you announced multiple meals in the first hour, it will actually combine those meals and adapt on the sum of those meals. And that's quite useful for announcing meals by courses. So you go out to eat for dinner, and you have an appetizer. And that appetizer has sent you out to eat maybe as many carbs as your usual meal. So you announced the appetizer, and now your main course comes out. And that may have more carbs than your usual meal. So you announced that and maybe you then have dessert, you want to announce that it will combine all those announcements and adapt on the some of them. And that may be better than trying to anticipate at the very beginning of the meal, how much total carbs, you're going to have for the entire period, for two reasons. One, because it may be hard to do you not sure whether you're going to have that dessert or not at that point. But to because if you took all of the insulin, if you told the islet, you're going to have all those carbs right at the beginning, it might give too much insulin too early. In other words, you know, an hour later, when you actually have the dessert, you want to have that insulin, then not at the very beginning, it might be too much too soon. But after that first hour period, if you announce another meal in that two to four hour period, it'll cancel that adaptation. So if you never space out the meals by four hours, it's never going to have a chance to learn after that initial learning, then you can be much less regimented about it. You know, if you decide to have a snack a couple hours after the meal, have at it, go ahead and announce it. That'll cancel the adaptation for that meal. But so let you know there you have lots of opportunities to sort of update the meal announcements, and most people tend to eat about the same amount, you know, same stuff anyway, right. And so you probably don't need to update it. But in that first week or so, spacing them out is good. Also, eating meals that really are typical for you or usual for you is good so that it has a chance to learn that there is always this temptation that you've got a new system, like a new fancy car and like let's try this thing out, see how it goes. And we definitely see some of that. Yeah, I remember in one of our first outpatient study, a long time ago now, we this was early on when Twitter was new and shiny. And one of our users had a Twitter account and he was tweeting all of his meals that he was having. And you know one of them was a plate with a burger fries, a couple beers and then stack of of little dessert dishes hmm And, and he's like, you know, I just had this meal. And, you know, seven bowls of ice cream hashtag bionic pancreas?

Scott Benner 1:20:09
Can I tell you how old I am? I know that photo. Yeah, I can't believe. I just thought have I been doing this that long? I know exactly what you're talking about. That's crazy. Yeah, yeah. So what's the downside isn't doing that?

Steven Russell, M.D. 1:20:25
Well, the downside, of course, is that the chances of being able to adapt on that meal are not very good. And certainly it's not a usual meal. least not for most mortals. Yeah. So really, if you want the eyelet, to learn you, you've got to, you've got to give it the right information. And this is when we see problems. These are the kinds of problems we see somebody, like, let's say they, they, they have habits that they've developed over the course of a long time with diabetes. That totally makes sense with some other systems, but don't make sense with the eyelet. But they find it hard to readjust. So I'll give you an example. Somebody has that first meal with the eyelet. And their blood glucose goes high. And that's a common occurrence, because it's starting off with this very low dose, they see that go high. And they think, Oh, this thing isn't going to give me enough insulin. And so the next time they announced a meal, they announced it as a more meal instead of a usual meal, even though they're eating a usual meal, right. If they just announced it as a usual meal, it would have given them more insulin, because it learned from the last time, okay, but now it doesn't have a time, it doesn't have a chance to show them that. And they will get a little bit more insulin that one time because, of course, they announced it as a more than usual. But that will actually make it think that they need less insulin for a usual meal. So it has the exact opposite effect over time that they intend. So it works wants to get more insulin, but in the long run, they'll actually get less. Yeah. So the islet, because it's adaptive, it has this unique characteristic that you can, you can't trick it, but you can confuse it. And if you tell it things that aren't true, it will learn the wrong thing, right. And another example that is totally understandable, given you know, how people have been managing their diabetes is they're about to eat a usual meal, but they're like, but my blood glucose is dropping. So I'm going to announce it as a lesson meal instead of a usual meal, so I'll get less insulin. The problem with that is that the islet is already taking into account that dropping glucose in terms of its Basal dosing, it's correction. It's like, if you're in a self driving car, and it's turning around the corner, but you grab the wheel and turn it to, and now you at the curb, you're both trying to take care of the same thing. And you're overdoing it, I've

Scott Benner 1:22:46
come to believe that we've reached the level in some technology, you can outsmart it, you can't even understand what it's thinking about. And you know, your little thoughts are gonna get in the way I learned that with loop when my daughter started using a loop. Like if this isn't working, our settings are wrong. Our timing is wrong. Me trying to fix a problem is just going to make it more confused. And it wasn't like a learning stitch. Yeah, I'm going to try to hack through. Plus, why am I putting myself in that situation where I'm constantly having to do more, when it should be automated as much as it possibly can be? Yeah, no, I mean, it's, it's it's AI, right? Like, it's it's a it's a learning model. It's considering just more things than, you know, exists, let alone that you could consider and and it's on different timelines, which is That's right. Yeah, yeah, all

Steven Russell, M.D. 1:23:37
that adaptation is going on. And if you try to outsmart it to hack it, it'll just screw up the adaptation, things will get worse, not better. But that is a really hard thing, especially for the knob, Turner's button pushers, folks, it's just so hard to sit on their hands and and wait for the system to learn. And so we, you know, some of those people do want to try the eyelet. And we're encouraging of that, if they want to, as long as they're, you know, willing to at least try re considering how they manage things. And some of them find that they can push through, they can sit on their hands for long enough for it to learn and do well. And then they think, Oh, great, I, I'm fine with not doing all that extra work as long as the results are good. Yeah. Interestingly, we've also seen some folks who got undeniably great results on the eyelet in some cases better than the results that they were getting with all of their work. And they just still find it an uncomfortable thing. They want to be able to go in and give that insulin, even if they're not going to achieve a better result. That's just who they see themselves as being Yeah, and that's fine. That's fine.

Scott Benner 1:24:52
It's interesting. It is it's the psychological. We I've had these existential conversations with type ones before about Like, what would you do you know if diabetes disappeared, and the number of people who say that they wouldn't even want that to happen, because they don't know who they are without diabetes is fascinating. Like, you know what I mean? Like, it really is interesting how much it becomes a part of, of who you are and what you do every day. Anyway, and there's

Steven Russell, M.D. 1:25:17
just sort of a pride of being able to manage it, like you're really good at this, this is something that I have got down. And so taking it away feels like a loss almost. Yeah,

Scott Benner 1:25:27
I did all this work to understand this thing. I let it go. I have to tell you, it's even between you and I, I don't want to sound pompous. But I can manage insulin within an inch of its life. I'm really good at it. I kept my daughter's a one C like 5355. No problem through like middle school, high school, no big deal. And then I was like, oh, god, she's going to go to college. And you know, like, she's not going to do the stuff I do. But luckily, prior to that, we started to use a loop before Omnipod five was even available. And I got to sleep. And once I got to sleep, I thought, Oh, I was gonna die. I didn't even know it. You know, like, like, I was my my No kidding, that by the time I got my daughter off to college, like I had to go to a doctor and say, Okay, now we gotta save me. Like, you know what I mean? Like my health had gotten poor. Just it's not sustainable. It just really Yeah. So

Steven Russell, M.D. 1:26:20
that's right. I mean, that's very impressive. And I'm always impressed by people who can do that. And there are plenty of people out there who can do it, but there is a cost to it, there is a real cost to it. It just takes years off your life. You know, I

Scott Benner 1:26:33
was not going to make it the whole way. Like I just if I had to do that for another 10 years, I have had times in my life where I've sat in bed in the middle of the night, and my brain is vibrating because I hadn't slept. And I've actually had the conscious thought, I'm going to have a heart attack. Like I'm gonna die like I have to go to sleep. And then you look at that blood sugar and you go I can't right now because we made a correction and like Boba, like that whole thing. All that is gone. Like I just, and I'm telling you, if you add glucagon to that, I might, I might, man, I might sleep with a noise machine. Like I might be like, I really don't want to hear a damn thing. Well, this is let me sleep. Anyway. Good luck to you. Godspeed. Seriously, I hope it goes exactly the way you guys are envisioning.

Steven Russell, M.D. 1:27:18
All right. Thanks a real pleasure talking to you. Oh, sincerely thank you for thank you for taking the time to have me on now. It's a pleasure.

Scott Benner 1:27:24
Hold on one second for me.

A huge thank you to one of today's sponsors, G voc glucagon. Find out more about Chivo Capo pen at G Vogue glucagon.com Ford slash juicebox. you spell that GVOKEGL You see ag o n.com. Forward slash juicebox. A huge thanks to us Matt for sponsoring this episode of The Juicebox Podcast. Don't forget us med.com/juice box. This is where we get our diabetes supplies from you can as well use the link or call 888-721-1514 Use the link or call the number get your free benefits check so that you can start getting your diabetes supplies the way we do from us med Jalen is an incredible example of what's so many experience living with diabetes. You show up for yourself and others every day, never letting diabetes to find you. And that is what the Medtronic champion community is all about. Each of us is strong and together we're even stronger. To hear more stories from the Medtronic champion community where to share your own story visit Medtronic diabetes.com/juicebox And look out online for the hashtag Medtronic champion. If you're not already subscribed or following in your favorite audio app, please take the time now to do that. It really helps the show and get those automatic downloads set up so you never miss an episode. Thank you so much for listening. I'll be back very soon with another episode of The Juicebox Podcast. The episode you just heard was professionally edited by wrong way recording. Wrong way recording.com