#838 A Look at Tidepool Loop

Howard Look is the Founder, President, CEO of Tidepool. 

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.

Scott Benner 0:00
Hello friends, and welcome to episode 838 of the Juicebox Podcast.

Today on the Juicebox Podcast, we're gonna take a look at tide pool. Did you feel bad pun? That's right, Howard look is on the show. He's the founder, president and CEO of tide pool to talk about tide pool in general, and the tide pool loop app that just received FDA clearance. While you're listening today, please remember that nothing you hear on the Juicebox Podcast should be considered advice, medical or otherwise, always consult a physician before making any changes to your healthcare plan. We're becoming bold with insulin. If you're a person who has type one diabetes, or is the caregiver of someone with type one, please take 10 minutes out of today to go to T one D exchange.org. Forward slash juicebox. All I'm asking you to do join the registry, complete the survey, it should take you about 10 minutes, and you complete that survey. Your answers will help diabetes research to move forward. T one D exchange.org. Forward slash juicebox. I have a moment. So let me tell you that the Juicebox Podcast has an entire series worth of algorithm based episodes, including a ton about loop, check us out on Facebook, or at juicebox podcast.com. This episode of The Juicebox Podcast is sponsored by the Contour Next One blood glucose meter, you can find out all about that little meter at contour next.com forward slash juicebox get the meter that my daughter uses. Speaking of what my daughter uses, you could get your diabetes supplies the way we do from us med get your free benefits check right now at 888-721-1514 or by going to us med.com forward slash juicebox. quite a quite a team actually. And if you don't mind before we talk about everything that's going on recently. Can you talk about that? And you said you found it the whole thing? Was it just you in the beginning? Or is it you and a handful of people? How did that go?

Howard Look 2:17
Yeah, so it actually goes back to 2012 2013. So my daughter Katie was diagnosed in 2011. With type one, I was actually working at Amazon at the time, it was the consumer electronics subsidiary of Amazon called lab 126. And Katie gets diagnosed and my initial reaction was oh my god, I knew nothing about diabetes, I didn't know what insulin was, I didn't know there are different kinds of diabetes. And so it was really a crash course I think, like for a lot of parents. And you know, very quickly into our journey. She got put on a Medtronic insulin pump and originally the the enlight sensor. I'm not exactly answering your question, but I'll get to it in just a second. Because I think it's kind of part of the journey. So she was on the enlight sensor, which for her just didn't work. Like it wasn't comfortable. It wasn't accurate. You know, she would cry every time. It had to go in, we call it the harpoon. It just wasn't right for her. And fortunately, right around that time, is when the Dexcom g4 came out. And so then she's still on the Medtronic pump, but using a Dexcom G four. But we have this crazy situation where at the time, neither the Dexcom software nor Medtronic software would run on my Mac. And I was like, I can't see your data like this is crazy. And so, you know, I literally was saying who writes this stuff? Like, why is it so hard? It shouldn't be so hard to see the data. So I started asking around, I started meeting other people that felt like I did. I always like to give credit to Scott Hanselman. He had this great blog post in 2012 called the sad state of diabetes technology. And if you scroll, you know, 130 people down in the comments, you'll see me going Hi, my name is Howard. This is all kind of crazy. I'd like to do something about it. Anybody else want to do something about it? And so I started getting introduced to people, you know, like lean Desborough, like Brian Maslin people who were also tinkering and trying to make things better for living with type one. I eventually got introduced to a team of folks up at UCSF, Dr. Salia de and Dr. Aaron nine Steen and Dr. Denise Huang. And they had started a little company called Green Dot diabetes with another GeekDad for lack of a better way of saying it like me named Steve McCann. He also had a teenage daughter living with type one, and had been doing a little bit of hacking to visualize her data. And when I met that Team Steve and the doctors at UCSF. I was like, Oh my God, they're doing something about it, I want to do something about it. I initially proposed to Steve that he and I go to a startup together. And he said, You know what, I don't really want to do another startup, I want to go do this other thing. But if you do it, I'll give you your initial funding or his wife at the time, and he gave us our initial funding. And that was how tide pool was born. So it started as Green Dot diabetes for people at UCSF, and me, and Steve McCann, we decided to rename it tide pool in May of 2013. And that's when I became CEO. And we never looked back. From that point on, I started meeting people. The actual first person I hired at the time was Ben West, who I think is, you know, super well known in the DIY community. He's the guy who did the original reverse engineering of the Medtronic insulin pump, which ultimately led to open APs and loop and lots of other great things. I ended up meeting, Brandon Arbeiter, who I convinced to move to California and start typo with me. And the rest is history. As they say, We started small and humble. And we've kind of slowly grown over the last 10 years.

Scott Benner 6:18
It's it's an interesting way to build something just continuing to call out of this group of people who are all very focused for personal reasons. And yet, very technically adept. I'll tell you, there's I recognized almost every name you said. But I want to tell you that I did an interview with Dr. D. Once it's one of the best conversations I ever had about diabetes.

Howard Look 6:39
Oh, he's amazing. Yeah, I love it. We used to call him the basil whisperer. He is so good. And He's so calm, but so insightful. He can look at data and just go, Aha, we should look at your Basal rates between 3:30am and 6am. Or high, your ISF is off by 10%. At this time of day,

Scott Benner 6:58
I remember saying something that he agreed with, and I was so proud of myself. I was like, Wow, all right. But yeah, we had such a what I would consider to be a geeky conversation about using insulin. That was just really fascinating, and very, very informative. I still hear about it from people, they say it's one of the best episodes of the podcast. Oh, that's

Howard Look 7:17
great. Yeah. Solly dr. D, he is amazing. I consider him a close friend. I was actually just in touch with him earlier this week. And he was our original chief medical adviser at typo please tell

Scott Benner 7:32
him I said hello, if you see him again, Burwell. So. Okay, so I understand how you all are then aware of this loop algorithm that exists in the world, it's being made just by people. And, you know, we're gonna use words that if people don't understand the idea of sharing code online, I don't even understand what I do to be perfectly like, honestly, here like, Yeah, I'll give you an example. I have a GitHub account. I don't know what that means. I

Howard Look 8:00
happy to explain it.

Scott Benner 8:01
No, no, if you explain to me to go right through my head, and out the other side, which I think it's good about what you guys did, you know, lovely people will jump on zooms with me and help me rebuild my daughter's loop app sometimes. And as I'm doing it, I think, Oh, I do know how to do this. But I my brain just doesn't work that way. But my daughter has been using loop for a number of years now. And she began, I don't know how it I'm sorry, you my daughter is 18. She's in college. She's a freshman, she was diagnosed when she was two. So you have some context. And a listener of the podcast came to me one day and said, you should put your daughter on loop. And like gave me the big like Sal about it. And later, they came on the show to tell me that they just wanted my daughter to be on loop. So I could figure out how to use loops. So I could explain it to other people. And I was like, okay, but wow, what a big difference. Because I had taught myself I didn't realize that at the time. But I had taught myself to sort of be a living algorithm to to add basil and take it away and Bolus in strange places where nobody would ever think to Bolus and you know, that kind of thing. And I thought I'm so good at it. Like we don't need this. And you know, but who I was so wrong, you know, because I sleep now like a regular person. And you know, and my daughter is in college, taking care of herself. And you know, and maintaining a onesies that are just beyond respectable and eating what she wants and all this stuff. So I'm always very, very grateful that that person came to me and said that but it threw me into a world. That was at first it was off putting like, you know, the first time someone says to you that a bunch of strangers like faceless people wrote code that is going to know it's going to talk to your insulin pump and a CGM and it's going to make decisions about your insulin. It's mind numbing at first like Well, that can't be right. Like that. Like how does that happen? Ben, but then one day you guys came along and said, We're gonna take that code, and we're gonna get it FDA approved. And again, I was like, I don't understand how that works, either. So can you tell me that first? How was it? I know it is I just don't understand the process. How is it okay for you to go online and be like, here, we're taking this code? And we're going to put it through the FDA? Is it because it doesn't belong to anybody to begin with? Or can you explain that whole thing to me?

Howard Look 10:26
Yeah, it's a really great question and really gets to the heart of the ethos of the do it yourself and open source software development community. So you mentioned GitHub, GitHub is just a place online where people like to store their source code. And it happened to be very happens to be very popular with people in the open source community, as a way to share their work with other people. And so to tell the story, I have to back up a little bit, my daughter is in the same zone as your daughter, she's 22. Now was diagnosed when she's 11. She's a senior in college now and doing great. And back when she was a freshman in high school, one of the first open source do it yourself projects that came out was called Open APs. Scott Lee brand and Dana Lewis, along with building on top of the work that Ben West and John Kostik had done, made that happen. And my daughter, Katie, is Dana has the actual list. I don't remember the number. It's somewhere between 15 and 20, was my daughter was somewhere between number 15 and 25. And I don't remember the actual number, but it was very early on. I started meeting other people like Brandon Arbeiter, my colleague and Ben West, who were using open APs and getting amazing results. Like you, my wife and I, we were really struggling, I think is a fair word, you know, we would take turns getting up at night to check our daughter's blood sugar when she was having a rough night, we really knew there had to be a better way. And we went I started seeing what was possible. Based on a software algorithm making the decisions, it was just super clear to me, that software could do a much better job of making the minute to minute decisions like you and I can do a great job or daughter's can do a great job. But we're not going to stop every five minutes and do the math and figure out what the right thing to do is tirelessly, the software can do that. And so I did the work to build open APS for my daughter. In the beginning, it was this clunky rig with a big giant battery and a Raspberry Pi and the CareLink stick. And I every day, my job in the morning was to pack it all up and put it in a little camera case and stick it in her backpack and send her off to school. And then every night I took it out and made sure it all got charged and you know, logged in to make sure everything was working. Okay. And even with all that effort, it was totally worth it. Because like you said, there were really two big impacts to that one was my daughter, for the first time since being diagnosed could go back to living a much more normal life, she could just go about her day as a teenager, and not have to worry that she was going to go low or not have to worry that she was going to go high, because the system would just deal with it. And then the big win for my wife and I is not only could we worry less about her, we slept better. So it was just better for us as parents and better for our marriage and just better for our family. So it was super clear to me early on that that was the way things needed to go. Fast forward a couple of years. Loop becomes a project I was very fortunate to see loop early in its development cycle. I think the first person to show it to me was Ben caimans. At the time, he was VP of Engineering at Cana COMM At Academy. He lives with type one diabetes. And he was friends with Nate rec lift. Nate wrote the original version of a loop. And I went out to breakfast with Ben and saw what he was using saw the Reilly link the little doodad that you need in order to make it talk to a Medtronic insulin pump at the time. And I was like, what is that show me how that works. And he talked me through it. And so I was immediately hooked. I immediately got myself a Reilly link. I went and built it myself at the time early in the evolution of do it yourself loop. It was one big giant Google document that Jeremy Lucas and Katie De Simone had worked on. And I powered my way through that Google document. It took me half a day, but got it up and running. And it was incredible. So not only really didn't have that automation component, the ability to make decisions every five minutes, but because it was iPhone based, it meant I didn't have to bundle up that crazy battery, Raspberry Pi, stuff to send with my daughter, she could actually just control her diabetes from her phone. And it was it was absolutely mind blowing. So the cool thing about the work that all of these people did, you know, going way back to Ben's work, reverse engineering, the Medtronic insulin pump, John caustics, work, reverse engineering the Dexcom je for at the time protocol, the work that peach Suam did in order to design the Reilly link. And then the work that Nate rake lifted to build and publish loop. They all made it openly available, every single one of them said, I've done something good, this is going to be helpful for the rest of the diabetes community. And I'm not in this for the money. I just want to help people. And so they made it available. And to me, that is just one of the incredible stories out of the diabetes communities. How many people have done that how many people have said, This is not about the money. This is about helping our kids live more normal life, this is about helping ourselves to not have to have fear that our kids are going to make it through the night. And so we're just going to put this code out there and help as many people as we can. You asked, So how does it work? So everybody I mentioned, put their code online, most most people use GitHub these days. But it's not always GitHub, it's just a place to store the source code. And they generally make their code available using what's called a permissive open source license. So a lot of people have heard of the GPL, the GNU Public License, that's a less than permissive open source license that requires you to contribute your changes back. Lots of good reasons to do that. There are also permissive open source licenses. The two most notable ones are the MIT license and the BSD license. And those two licenses are really short. And you put them at the top of your source code. And it basically says, use it your own risk, do whatever you want, take it, change it, it's yours. The only requirement is that you put this copyright notice at the top of the file, you can do whatever you want. And that ability is what lets people take all these components and mix them up and do new wonderful things like create open APs and create a loop. And there are lots of other examples of that. So what we at tide pool decided to do was to take that open source code that Nate had published that each one had published. And we brought it in to tide pool. And when I say we brought it in, what does that mean, on on GitHub, you do what's called Creating a fork, it's really just making a copy of the code. But now all that code lives in tide pools account. And so we can make changes to it, we can publish those changes, and other people can take them if they want to. But because all of those people publish their code with a permissive open source license, it also means that we can bundle it up and package it and take it to the FDA and say, Hey, here's this work that we've done. Is it okay for us to call this a product that we're going to ship to the World Within FTAs regulations, and put it in the app store? So that's the work we did between 2018. And now 2023?

Scott Benner 18:48
Wow. Can I ask a question about that? Yeah. What stops somebody now that you have an FDA approved from resharing, the code online as the official tide pool code that you can just do DIY

Howard Look 19:00
does, nothing, nothing stops them. And in fact, we have published all of our code via open source that they can't call it tide pool. Dry pool is a trademark trademark name, and they have to get our permission to do that. They also can't say it's FDA clear. The thing that is FDA cleared is the package of the software compiled built. That includes all of the clinical evidence that they that we submitted all the verification and validation testing that we did all the human factors testing, we did a long laundry list of things that we needed to do in order to submit it to the FDA that we own that FDA clearance is that clearance.

Scott Benner 19:41
It's a that's why I wanted to bring it up that clearance encompasses much more than just the code that is the app. It's That's right. We're working on for all these this time now.

Howard Look 19:53
That's exactly right. In fact, if it were just the code, it would be easy because most of the code existed for us As long as DIY loopers have been using DIY loop, the work in getting to FDA clearance is a whole bunch of things. And I won't go into all the nerdy regulatory detail, but you need to submit clinical evidence. So the way we did that, with the FTAs encouragement was we collaborated with the DIY community and our friends at the Job Center for Health Research. And we did the jig, or sorry, the loop observational study. So the observational study loop was over 1000 people using the DIY version of the loop, many of whom hang out in the loop Facebook group, over 850 of those people ended up submitting at least six months, if not 12 months worth of data that we then gathered, actually, the Job Center for Health Research gathered it, they then did a bunch of statistical analysis on that data, we ended up with a two or 300 Page clinical study report and a bunch of other data analyses that we did. That package of clinical data is part of our submission, and was part of demonstrating to the FDA that the loop algorithm was safe and effective. So that's one big piece of it. Another big piece of it is what's called verification and validation. Most people just call it V and V for short. And that's really the effort to not just test the software, but to document with full traceability that you know that the requirement, the specific user story that is intended to be done, is working the way it's intended, and that you can demonstrate through test documentation that you have completely fulfilled that requirement and have reduced whatever risks may exist in that software. So we did a huge risk management effort, we did a huge verification validation effort that was part of our submission. And we're super proud of that. We also did Human Factors testing, one of the things you're required to do with an FDA submission is demonstrate that your software can be used by the general population. So we actually along with our partners at a firm called core HF and Philadelphia, ran rigorous formal Human Factors usability studies, where they gave people very specific tasks, and had watched them, do those tasks, and then recorded the data. Now, sidebar on this one. This is one of the things that was really challenging about our submission, we were ready to start our human factors test thing right around February, March of 2020. And typically, the way human factors testings work is you bring people into an office, and you sit them at a table and you give them the training, and then you have cameras all over them. And you have a one way mirror and you record the whole thing. That's all fabulous, unless a global pandemic breaks out. So right around the time we were starting, are getting ready to start a human factors testing, we had to completely redo the Human Factors protocol, because of COVID-19. We ended up running are human. So first, we had to redo the protocol to make it a completely remote human factor study. So people did the study from home, we recorded them over zoom, we recorded the screen of what they were seeing on their phone. And we did all that. And it was I don't know if anybody had done it that way before. But it was certainly new and novel for us and new and novel for the team we were working with. So we had to generate all that data and then submit that as part of our submission. And we did it it took extra time. But by August, September, we were doing full remote human factor studies.

Scott Benner 23:53
Can you tell me? Did you learn enough about the process with the FDA that if you had to do it again, do you think there would you be able to streamline what you did the first time? Was there a lot of learning? Oh,

Howard Look 24:03
heck yeah. Heck yeah, there was a ton of learning. The FDA calls us a first time sponsor, tons of things that we would be much, much better at doing it the second time. On top of the fact that now that we have a cleared system, that system becomes what's called a predicate device. So any future submissions we do will refer to this submission and say everything about this is the same except these parts and then they only have to look at the iterative parts.

Scott Benner 24:35
You know why I'm asking you that right? Because absolutely one of the best parts of loop is how, how flexible it is, and how it it morphs and changes so quickly. And you know, people from from my perspective, you feel like that happens more quickly, because you don't have to go to somebody at the end of a change and say, Hey, is this okay? But like, I mean, certainly example like I, what the version you have gotten through like, when did you start? What like what, what? What year did you grab the code and start this whole thing.

Howard Look 25:10
The process started in earnest mid 2018 2018.

Scott Benner 25:14
So now in 2022, my daughter's using loop three, I guess. And in the past, I've used the Peet's branch. I've used Ivan's branch I've used you know what I mean? Like, I've used so many different things I've used the auto Bolus I've used, you know, what this one is going to do? I think was going to make adjustments through basil adjustments. Is that correct? And if so, are you gonna keep going? Like, are you? Like, are you in a back room right now planning on the next thing or like, I know, it's weird to ask before you get the thing out. But you know,

Howard Look 25:46
not at all 100% We are absolutely not just thinking about but working on subsequent versions of tide pool lube. And you're bringing up a really important point. And this is where what we've done, which is to mesh the world of regulated medical device software with the world of do it yourself innovation. We're now seeing what that means. We submitted in December of 2020. It's now January of 2023. That was a long review process. Some of that was due to the pandemic. Some of that was because it took a lot of time to work through with the agency, what it means to build a truly interoperable system. And we can talk more about that. But what we have now becomes the foundation for all future iteration not just for us, but for anybody else that wants to submit a device that refers to type a loop as the predicate device. So yes, we definitely learned a ton and future submissions will go way faster because of that. But also, we can build upon the work we've already done. So by way of example, since your daughter is a looper, you may know about the overrides in DIY loop. So overrides showed up in DIY loop, very much near the end of the loop observational study. And what that means is most of the clinical data that we submitted, does not include data with people using overrides. So we can't we couldn't in our submission in December 2020 include overrides as part of the functionality. Now. It's clearly up and running and working in the DIY community. We have a version up and running in our development branch. And what we can do now is on top of the clearance that we already got go back with a another submission saying, Here's new clinical data, here's New vnv data, here's New Human Factors data, we now want to add this functionality on top of what we already submitted to and it'll be much easier to do that.

Scott Benner 27:57
That's very cool. Okay, that's amazing. I appreciate you giving me that. Like, for everybody listening, honestly, because there's a lot of words like the it's loop. If you don't know anything about it, you're listening right now you're like DIY, that means do it yourself. Okay. And then, you know, like, there's just too many phrases and words that I think people don't even like you say Raspberry Pi. Like, we all know what that means. You know, then I know what it means. But you know, So context is important, I have to say, so

Howard Look 28:26
maybe I can kind of build on that just for one second. Yeah. It's It is one thing I'm gonna make super clear the thing we did, makes typo loop not be a DIY do it yourself project. So DIY loop will continue to exist, the we are not waiting movement will continue doing its thing, innovating and trying new things out. And I love that I encourage innovation. And I'm very, very happy that the DIY community exists and will continue to exist. I pull loop is not a DIY project type of loop will be in the app store, just like any other app that people download to their phone. And I think that's one of the reasons why the work we did is so important. Because there are a whole lot of people that don't want to build the rest their own system. People who are like X code, I don't want to deal with compiling GitHub exco bla bla bla bla bla, I am dude, I am one

Scott Benner 29:20
of those people. I don't I don't want to do it. And you know, the things that machinations we went through to make sure that my daughter was okay when she went away to school and to make sure she was covered if the app should crash or like something like that. I don't want to live through any of that. So some at some point, then you're going to be able to go to your doctor and get a prescription that lets you go to the app store and download your app. Is that right? Is that how that's gonna work? It's

Howard Look 29:47
close. So the way it will work is typo loop will be in the App Store. Anybody will be able to download it just like you download any other app from the App Store. In order to activate it. You're going to need an activation code, which you'll get through your health care provider, we've created a healthcare provider portal where they can go to create a prescription. And this isn't like a prescription that you take to CVS or Walgreens, it's a digital prescription. But it will send you a code through text message or email that once once you type that code into tide pool loop, it unlocks all the functionality so that you can use it. And it will cause your entire set of initial settings, which you'll establish with your health care provider to get downloaded. Now, a lot of people will say, Well, wait a minute, can I change the settings by myself? And the answer is absolutely, positively Yes. It's just that initial set that your healthcare provider will establish. We all know that diabetes is self managed disease, we all know that there are some people that are comfortable changing their settings on their own. And there are lots of people who like to do it in coordination with their health care provider. We've made it so that the health care provider works with you to establish the initial set, but then you can go in and modify it over time.

Scott Benner 31:07
Okay, that's really great. Do you think there's a thirst with the pump companies that you have agreements with? Do you call them agreements? Who's it going to work with right when it comes out? I guess is the way I should ask.

Have you found that getting your diabetes supplies can be a pain in the butt? I have to but not any longer because now we're getting Ardens diabetes supplies from us med to get a free benefits check, just call 888-721-1514 or go to us med.com forward slash juicebox. US med has served over 1 million diabetes customers since 1996. And they want you to know that they're offering you better service and better care than you're getting now. US med always provides 90 days worth of supplies, and fast and free shipping. They carry everything from insulin pumps and diabetes testing supplies to the latest CGM like the FreeStyle Libre two and the Dexcom G six. Arden gets her Dex comps and her on the pods from us med. US med accepts Medicare nationwide and over 800 private insurers. US med is the number one distributor for FreeStyle Libre systems. The number one specialty distributor for Omni pod dash, number one fastest growing tandem distributor, the number one rated distributor index com customer service satisfaction surveys. And they are proud of the white glove treatment that they offer their customers, US med.com forward slash juice box or call 888-721-1514. And that 888 Number is special just for Juicebox Podcast listeners. Now if you decide to go to the website, it's super simple. Here's what it tells you when you get there. Getting started is easy. from the comfort of your own home or office, you can now join over 1 million satisfied customers who rely on our staff of courteous knowledgeable and trained us med customer care representatives to keep you up to date with your medical and diabetic supplies delivered right to your door. Super simple benefits check name, phone number, email, zip code, and then just hit the button that says Request a free benefits check. It's that easy. Let us med take care of your supplies. So you don't have to worry. At some point on your diabetes journey, a person gave you a blood glucose meter. Did they say to you, hey, this is a great blood glucose meter. It's one of the most accurate ones that they ever made. No, no, no one said that. Did they say by the way there are other blood glucose meters you might want to look into it. I'm just gonna give you this one because I have it here in the drawer. No, they didn't say that either. They just gave it to you. And you thought, Well, this must be my blood glucose meter because the doctor gave it to me. But there are many meters and they're not all made equally. You deserve an accurate well made an easy to use blood glucose meter. You deserve the Contour. Next One. The Contour Next One is my favorite blood glucose meter. I know that's a strange thing to say. But we've used a number of them over the years. And this one is my favorite. Why? Bright light for use at night. The screen super easy to read. It's manageable. And by that I mean it's a good size. It's not too big. It's not too small. And I love the way it fits in my hand. It's sort of because of the shape which you'll see it Contour Next One comm forward slash Juicebox Podcast feels like you're holding a like a pen in your hand. I don't know how to put it exactly. You'll see when you get to the website. But the Contour Next One blood glucose meter is incredibly accurate. But you might be worry, Scott all this accuracy? Is it more expensive? Am I going to be paying a bunch more money? I don't think so. Actually, if you go to contour next one.com forward slash juice box, you can actually buy it right now at a number of online venues, Walmart, Amazon, Walgreens, CVS, the list goes on and on target Rite Aid. And so when you get to my link, check it out. Because you might be able to save time and money buying contour next products from the convenience of your home. What am I saying? Well, I'm saying that it's possible that this meter and the test strips could be cheaper in cash than you're paying right now through your insurance company for an inferior product. How crazy is that? You owe it to yourself to be using the best equipment that you can. And there's no reason not to check out the Contour Next One blood glucose meter contour next one.com Ford slash juicebox. Who's it going to work with right when it comes out? I guess is the way I should ask.

Howard Look 36:02
Yeah, so let me tell you the history of that, and then I'll bring you up to speed where we are today. So in 2018, we announced a development partnership with insolate, makers of Omni pod. In 2019, we announced development partnerships with both Medtronic and Dexcom. Now, let me take a step back for a second. In order to use typo loop you need three things. And these three things are all part of what the FDA calls their interoperability pathway. You need a continuous glucose monitor which the the FDA calls an IC CGM for interoperable continuous glucose monitor, you need an insulin pump. And the FDA calls that an ace pump or alternate controller enabled in insulin infusion pump, and then you need what's called an IEC. That's us an interoperable automated glycemic controller. So you need those three pieces. I CGM ace pump IRGC, where the IRGC with tight pull. So Dexcom is still very much a development partner with us. And we're very much looking forward to other continuous glucose monitor companies being becoming compatible with type polu. Medtronic in insolate, at this point in time, have said that they will not be our launch partners for as as pumps, we're very grateful to them for supporting the development of tide pool loop, I will tell you that we are working with another ace pump partner that isn't ready to talk about who they are yet, but we're really excited about them. It's a company that everyone knows and we think everyone loves, and it's going to make a really great combination with tight pollute. The great thing about type ULoop is that it was architected from the ground up to be interoperable, meaning we can add new CGM and add new Ace pumps. And we don't have to submit new evidence to the agency. Part of our submission was a set of plans and processes and nerdy regulatory speak. They're called SOPs, standard operating procedures, we came up with SOPs that we agreed with the FDA. As long as we follow those procedures, we can add new devices over time without a new submission to the FDA. And so for us, we feel like that's pretty groundbreaking. Typo loop is the first truly interoperable system that is designed from the ground up to allow new devices to be added over time. Without a new submission.

Scott Benner 38:37
This might be a question I shouldn't ask you, but do you think you'll get those companies back?

Howard Look 38:43
I'm never gonna give up. Like, I think interoperability is the best thing for the diabetes community. I think if you're a person living with diabetes, you should get to choose what CGM you use, what pump you use, and what the user experience and algorithm is that you use to manage your diabetes. I think interoperability is a great thing. I think the community appreciates this approach. And I think over time, we will see that company companies will start to embrace it as well.

Scott Benner 39:14
It just, I mean, from my perspective, not knowing anything that's going on in any of those companies, but it just more options seems like a good idea. You know, I think so. Yeah. Like I, you know, it took right up until you said that, I sort of had like this little like the 10 year old me inside was all like, oh my god, the world is going to work the way it's supposed to, you know, we're going to, you're going to get this thing and it's going to just work here or here or wherever we want it to and pumps will just become like, like wheels, right? Like I'll have a car and then I'll pick which wheels I want to go on. And I thought that'll be terrific. And then you said that I was like, Oh, okay. So don't give up because and if they're listening, you should do that. It's it just makes sense. And it It's the only thing. I don't know. I mean, I can see why they wouldn't want to, I'm trying to think it through. But the other idea is that it's gonna bring more people to pumping in general. It just that makes sense to me. Anybody who's used lube would tell you I'm sure that's partly what was what happened when you were talking to the FDA. It's, it's astonishing how well it works. It's just it's really astonishing. And so, you know, I hope everybody gets a chance to use it. I have a lot of questions. Keep going. All right, I'll try to rapid fire them as best you can. Although it's the end on a Friday, we've all given up on enjoying this day, right? So we can just keep going. So you haven't joined me? At the moment? Do you have pumps that are on a compatibility list? Or are you not at that point yet?

Howard Look 40:48
We do have compatible pumps, we're not able to talk about who they are yet. Because the pump makers need to make their own decisions about when they're going to let the community know that they are going to be delivering a compatible

Scott Benner 41:02
device, whoever they are, in my opinion, they're going to have a great stranglehold on some great social media and the ability to make some exciting announcement that people are going to get behind. So I can't do you have a timeframe, you think

Howard Look 41:15
it's going to depend a lot on the device partners, and so we're ready to go whenever they are, we're gonna have some work to do with them to coordinate things like customer support handoffs, and pricing is a question that a lot of people want to understand we still going to work through that. How it's going to work, you know, will they subsidize lube? Will that be something that insurance pays for over time? Those are all details that we still need to work out with them.

Scott Benner 41:43
So I have a question. It's gonna sound shady. Sorry, I'm getting off the list here. Oh, they really stop somebody from using it. I mean, they can't stop them from DIY using it, why would they be able to stop them from using your version of it?

Howard Look 41:59
Well, our version, we are the only people that can put it in the app store. We are prohibited by law as a cleared product from distributing this until we've met the requirements that are laid out by the FDA. And one of those requirements is that there is a compatible ace pump that it can work with. So until there is a compatible ace pump, we may not put it in the app store. And as soon as there is a compatible AST pump, we can put it in the app store.

Scott Benner 42:28
My My point was that once that happens, can't I just pair it with a different pump? Like what would start?

Howard Look 42:36
Well, our code will only work with cleared compatible AST

Scott Benner 42:41
pumps. So it's on your side that it'll be it'll be I'm not calling it hobbled. But it'll make sure on your side that it only works with what's

Howard Look 42:49
right or clear. It only works with with products that meet the requirements of a cleared third party, interoperable ace pump. And

Scott Benner 42:56
one of those requirements is that the pump company wants the app to work with their pump. That's right. Yeah. Okay. Got it. Alright, thank you

Howard Look 43:04
that well, actually, let me let me let me correct that that's not completely right. There is a world in which a pump company could deliver a pump, which conforms to diabetes device interface standards, there are Bluetooth standards and I triple E standards. And it is possible that a company can deliver a pump that conforms with those standards, and not specifically say, here are the controllers that it works with. They could just say it will work with controllers that meet these standards. Yeah, that is, hypothetically possible. We haven't seen that happen yet. But I am optimistic about that in the future.

Scott Benner 43:46
I think in the original Jurassic Park movie, Jeff Goldblum says something like life will find a way and I think that's right here. So okay, we'll just keep a good example. Yeah. The algorithm that you have, will it consider digestion times? This is obviously a question from a looper, like, can you tell it like, Hey, this is going to hit more like pizza or more like candy or three hours or four hours? Can you set those with the app?

Howard Look 44:10
Yeah, absolutely. That is one of the unique and wonderful features of tide pool loop. This also exists in DIY loop. When you do a meal Bolus, you not only say here are the number of grams of carbs that I'm ingesting, but in loop you use an interface that uses emojis and the default emojis are lollipop taco and pizza. lollipop has the fastest carb absorption time 30 minutes and then taco and pizza are slower so you can say, here's a slice of Grandma's lasagna and it's clearly more like pizza than it is like a lollipop. So you tap the the the pizza icon

Scott Benner 44:54
can you still post date a Bolus? Can I you can Yeah, excellent. Yeah,

Howard Look 44:59
that's an Another great feature that we love about loop. And again, this exists in both DIY loop and tide pool loop. It's a feature that I think is really great for parents of small kids with type one, for example, you put a sandwich in front of your kid, you guess it's 30 grams of carbs, you Bolus for 30 grams of carbs, and then your kid decides to only eat half a sandwich. Well, it turns out with loop with type a loop and DIY loop, you can go back and edit that meal Bolus, it obviously doesn't pull insulin out of the body. But what it does is it modulates all future delivery of insulin to make up for the fact that it now knows that you ingested fewer carbs than originally planned,

Scott Benner 45:46
I would say that I think two of the best features that I would use all the time is that just be able to like, go back and say, Hey, we said 40, but you know, it was really 30. And you just tell it because maybe it can't get the insulin back. But it can start making decisions in the future with the idea that, hey, I think we have 10 carbs, too much insulin going, maybe I can take the basil away for longer or do something to try to impact it. And then the other thing is to be able to tell it, look, I'm gonna have 20 I'm gonna have a 20 carb impact from fat from French fries, like 90 minutes from now. And then when the loop starts seeing the impact comic goes, Oh, and you don't have to go back again and make a secondary Bolus. It's it's the loop version of, you know, like a square wave Bolus or something like that, which is

Howard Look 46:32
exactly and besides having knowledge of the carb absorption time of what you ingested. The other nice thing about the loop algorithm is that it pays attention to how your body actually reacted. Its dynamic carb absorption is the term it looks at what it predicted what happened and what actually happened. And then it uses that iteratively over the course of the absorption of the carbs in the insulin to make modifications to its prediction.

Scott Benner 47:05
Did you have to get clearance for different insulin surges that lie on the device manufacturer,

Howard Look 47:12
we did have to get cleared for different insulin, so very specifically tied pollut is cleared for use with chemo OG and Novolog. A lot of people ask about fiasco, during the clinical study, it was pretty early in the adoption of Fiesta, there wasn't a lot of data from VSP users. But over time, I think we can collect more clinical evidence and hopefully broaden the labeling to include other insulins. And just while I'm on labeling, it's labeled down to each six. Again, because there was plenty of data in the clinical study to support that. And overtime, I would expect that we can expand that as well.

Scott Benner 47:52
Wow, hey, everyone does but don't forget a Piedra. Don't worry, I know you're not going to but it's my daughter uses fantastic. Target blood sugar's user definable all the way. I mean, with regular loop, I don't know where does it stop you.

Howard Look 48:11
With DIY loop? I don't know the answer to that question with tide pool loop, the target range that is configurable by the user to anywhere between 87 milligrams per deciliter and 180 milligrams per deciliter.

Scott Benner 48:26
He's just trying to make me cry. Howard, is that what you're doing? That's beautiful. Thank you.

Howard Look 48:30
Thank you. I'm pretty happy.

Scott Benner 48:32
You should be. That's me. Hey, let me ask you a question. For the other pump manufacturers that are listening to this right now. And they're listening. So how tough was it to get clearance down to 87? Did it really? Was it really that tough of a slog like because they're all at like 112 or 110? Or something like that?

Howard Look 48:52
Well, you know, I don't know how to answer was it tough. It was a lot of work to gather, analyze and present the clinical data in a way that made it clear to the agency that this would be a safe and effective way to use tight pull loop. So was it tough? I would say it was a lot of work. And we're very grateful to the participants in the loop observational study, because the thing that made that possible was the people that were willing to do all of the work that it takes to participate in a clinical study and to make their data available so that we could include it with our submission.

Scott Benner 49:36
How valuable was that? That you were basically working with pioneers instead of people who were called through and found through like, the regular way they find people for stuff like that?

Howard Look 49:47
Oh, it was to me just an incredible testament to the willingness of the community to pay it forward. Yeah, like it is a ton of work. My daughter has been in several clinical studies. I don't know if your daughter has done that but you It is hard work. You have to answer questionnaires you have to install software to submit data, you have to be willing to let people have access to your data. You have to take home HBA when C kits, like it is a lot of effort to participate in a clinical study. And one of the amazing things about the loop observational study, as I said earlier, there were over 1000 people enrolled in the study, over 850 of which submitted at least six months and in some cases, 12 months of data. That is an enormous quantity of data, which you don't typically see in clinical trials for medical devices. In fact, if you think about it in terms of person days of use, in the study, the loop observational study, had three times the data of the Medtronic 670 G, pivotal study, tandem control, IQ study, and insolate horizon Omni pod five study combined, combined three times the data,

Scott Benner 51:04
do you know that I did a little survey? I, I get a lot of help from wonderful people sort of like us. So I, this MPH student came to me. She just just graduated from Hopkins, and she wanted to help the podcast somehow. And she said, How can I help? And I said, Well, I'd like to do a survey, like a legitimate survey about how the podcast helps people. And she helped me put it together. And we're still collecting at this point. But we've only been at it for about two and a half weeks, we have well over 700 respondents. And she just texted me one day. And she said serious medical organizations struggle to get anywhere near this number of respondents to a survey and I said, Yeah, well, these people are motivated, because the podcast helps them like if they're trying to give, they're looking for a way to give back. And it seems like you had the very same thing. So is it fair to say that the goal to having an 87 target on your system, get through the FDA is your desire? Your desires? What made it happen, right, your desire to do that, to well, it

Howard Look 52:06
was very clear. So first of all, before I answer that, I need to give a huge shout out both to the looped community that people who mostly participate in the loop Facebook group, but very specifically to Katie De Simone. Yeah, Katie was the conduit. She was the person who sat in that community and said, Hey, everybody, you've all been asking me how you can help pay it back or pay it forward. This is how you can do it. You can participate in this clinical study. So Katie's work to coordinate all of that, basically, as the liaison between the loop community. And the Job Center for Health Research that ran the study was just absolutely incomparable. We could not have generated all that data without Katie's efforts. And without the loop community, we

Scott Benner 52:54
don't know each other, personally, but I have such a warm feeling towards her because she's the first person to come on the show and just kindly explained to me while I was going, I don't understand I don't understand, like walking through a loop was to me so many years ago now. Yeah, yeah. He's

Howard Look 53:09
an amazing human. And I love her dearly. Yeah,

Scott Benner 53:11
me too. Okay, so. Alright, so 87. Cool. I don't even care what the top number was.

Howard Look 53:17
Oh, I didn't actually answer your question. Oh, sorry. Yeah. Yeah. So first, how did we come up the data with the data, and that was because of the loop community and the work that Katie did interfacing between the Job Center for Health Research and the loop community participants in that study. At the end of the day, we took all of the data. And we looked for very obvious places where the balance of risk of hypoglycemia and the safety of the lower target range made sense. And so that's just a ton of data analysis work that was done both by our data science team, and also the Job Center for Health Research. And then ultimately, it was a back and forth conversation with the agency where we presented the data we showed them where we thought the lines should be drawn and why we thought those lines made sense. And they agreed.

Scott Benner 54:09
I feel like the answer to this question is going to be no because of the time, the timing when you jumped in. But does the app have the ability to enter non pumped insulin like a Frezza or an injection? And to give it its iteration?

Howard Look 54:22
That's a great question. It currently does not have the ability to enter outdoor, non pump insulin. That's obviously a feature that we know a lot of people would love to have and some will consider for a future version of title.

Scott Benner 54:37
Okay, I'm sorry, I'm scrolling past the the questions that and by the way, it as we're thanking people, the listeners of the podcast sent four pages of terrific questions for me to talk to you. That's great because they probably know how or the what would happen is I would start pontificating, and then you and I like an hour and a half from now we'll be talking about something silly and they'd be like nobody asked any real questions in that would be my answer. They're trying to keep me on brand here. Boston, of for people who plan on just continuing to use the AI. Why does it help you at all? If they get the prescription? Like, here's the this is this question because this is somebody trying to steal they want to give back still.

Howard Look 55:16
Oh, that's super helpful. Does it help tide pool if someone gets a prescription but then continues to use DIY loop? I can't think of an immediate reason why that would be helpful. But let me let me go think about that. I couldn't either.

Scott Benner 55:29
But I just I thought it was another example of just people wanting to do something.

Howard Look 55:33
Here's a thought and this is, you know, one of the reasons again, why I think what we're doing is valuable. We have heard and you know, this is anecdotal, but I think there are some folks I know in Europe, for example, that have studied acceptance of DIY systems in endo clinics. We've heard anecdotally that there are some providers that are really uncomfortable with their patients using DIY systems. So one of the big benefits, I think of tide Palooza, being an FDA cleared product and being available in the App Store, is that for those people that have providers that are a little queasy about their patients beyond the DIY systems, this is a great alternative. So not only does it give the person living with diabetes, who may not want to build their own system, an alternative option that they can look at and go, Oh, great, that's FDA cleared, it also gives their provider a little more confidence because their provider can say, Okay, I'm no longer scared that you as my patient are using a system that I don't know anything about.

Scott Benner 56:40
Do you think that's mostly everybody just, I'm just gonna say what I was gonna say that's just as covering, right? Because like I've, I've seen doctors people mentioned and they go, no, don't you do that I've heard, I've heard people be told, if you do that, I'm dropping you as a patient. And I've seen people, doctors go, Cool, let's do it. And I've seen the one where the doctor pretends they don't know what you're saying they almost lost all their way through it when you tell them and then they just act like you never said it and keep going. I've seen all three of those things happen. So this is going to take any kind of that uncomfortableness for those types for those physicians, and allow them to say, hey, we've got a good, we've got a good thing for you to try here. It's FDA approved, they can get behind it. So then the next question comes, is who's going to help? But this is something I think about all the time? Are you guys gonna get involved with the training? Because we always say like, oh, the doctors will take care of it. But that's not really going to happen. You don't I mean, like, they're not gonna understand loop the way you do, or the way I do, I could, I could grab somebody's loop app right now who I don't even know and probably get it working in a day, you know, like and make adjustments to their stuff. But that's unfair to ask the doctors to do. And, you know, especially when things don't, what do I want to say, I was really good at using insulin. And then I got to watch Nightscout. And then I got way better at it. Like watching the algorithm do what it was, it was like a was like a Master's class in how insulin works. And I already had a good idea about it. But just to watch it, take basil away, and then give a little back and give some extra and then take it away again, I was like, Oh, my God, this is just a dance that never stops, really? Because I learned a lot that the doctors aren't going to get to see that, or are you going to figure out a way to train them? Or how does that gonna go?

Howard Look 58:29
Yeah, so let me unpack that question. There's a lot in there, how's training going to happen? How will doctors get access to the data? So we developed training both for the person living with diabetes, also caregivers, so part of our human factors, testing was testing what are called dyads. So parent and child pairings. So training that is targeted at parents and including training that is targeted at adolescents who are on the transition to self managed care. So we created a whole curriculum that is actually both built into the application, but also exists in traditional user manual form, and then take all of that we recreated healthcare provider training. So the healthcare providing Provider Training exists in traditional forms, we will also have lots of content on our website in terms of knowledge base, and support articles. And we will keep doing something we've been doing for years, which is online, both live and recorded webinars to teach people how to use it. At the end of the day, we feel really strongly that people need to be able to learn how to use it on their own their healthcare provider will be there. But we can't be dependent on the health care provider to provide training to the person living with diabetes that has to be self evident. And the way we do that is by building training right into the app, you have to go through an entire onboarding experience before you can start delivering insulin, and then there's help within the app at all times. If you have questions about something, how something works, you can actually get help from right within the app. So one of the things I'm really excited about is we heard both during the Human Factors testing, and you can also hear over and over from people in the DIY community is how elegant and simple using loop is. And I think that really helps make learning all that material much easier. Because you can look at it. Look at the homescreen of loop and just see, Aha, I see what my glucose is doing. I see what the insulin is doing. I see what the carbohydrates are doing. That all makes good sense to me. Yeah.

Scott Benner 1:00:44
Well, and then I did a thing I talked about earlier, I just did Nightscout as if everybody understands that. So that's an app I use to to, like, I can see my daughter's blood sugar and what her loop is doing. Are you guys going to have like a follow app for for a caregiver? Yeah.

Howard Look 1:01:00
Super, super, super question. And sorry, I forgot to touch on that. So popping back a level, I pull started life as a data management platform, and we are still a data management platform. So tide pool lube will automatically up to upload data to tide pool. So people will be able to see their data in tide pool for web in tide pool mobile. And yes, we will have a follower experience. That'll enable loved ones to be able to see your glucose and insulin data remotely.

Scott Benner 1:01:34
Yeah, I was gonna say you're gonna be like GitHub, but like, right. But nobody will understand that. So I'm not gonna say it. And let me always be clear, I'm always only about 90% sure of any of the things I'm saying around lupus, right, because I just am so not technically adept about about that side of it. But also, I think that gives me a lot of comfort, because, and for other people, because with a little bit of help, I was able to figure out a thing that my brain doesn't naturally do. Like when you were talking about like screens that explain things. I picturing myself sitting there going next, next, next and not reading because by because I'm broken. And and that's not a thing I learned by doing. But it's very comforting to think that it's going to be out there for people like this. And that, that that especially that that that idea that I'm I don't mean it this way. But the idea that I'm doing something wrong, or that it's not like that somebody hasn't said is okay, like just to take that weight off of people saying this is the FDA said this is okay. Like I think that's going to be really helpful for a lot of people and to help them move forward. Because I haven't said it to you yet. But I think algorithms are 1,000,000% the future and the present of how people manage can be managing their their insulin, it's just, it's just such a leap above trying to do it yourself.

Howard Look 1:02:58
I experienced Totally agree. I you know, obviously what we both want for our daughters is a cure for type one. Hopefully, research will pan out and that will happen in their lifetime between now and then I completely agree that the next best thing in terms of both achieving great outcomes. But even more importantly, for me, reducing the burden of living with type one is to have automated insulin delivery. So I'm very excited to that we're participating in that. And

Scott Benner 1:03:27
a tiny little thing that I always throw in, I think Aaron Kowalski sent it to me once it's just even if we could just make the cannula material better. So there's not, you know, so that sites work more the way you expect them to consistently would be such a big deal to I didn't ask a secondary question about the weather. You have a follow up. People also want to know if you can Bolus from that app like will a parent be able to Bolus from their own phone?

Howard Look 1:03:52
Yeah, so the version of tide pool loop that was just cleared does not have that functionality. We know that is something that people really love, and it's something that we will absolutely consider for future versions. I will tell you that. One of the things we spent a lot of time with the FDA on is cybersecurity. And once you open that up, it creates a mechanism by which someone could remotely deliver insulin, which definitely should give everyone pause. That's I think it can be done safely. But it's going to take a lot of work to demonstrate how it can happen safely.

Scott Benner 1:04:30
Do you have um, were you able to get Apple Watch? Can people use your bill? Absolutely.

Howard Look 1:04:36
Yeah, type of loop absolutely has, as part of its feature set the ability to use an Apple Watch and to discreetly manage your diabetes right from your watch. Okay.

Scott Benner 1:04:47
What are some of the well my first question is Where can people go to see a list of the functionality for assuming you have a website where I can see that?

Howard Look 1:04:57
Absolutely, we'd encourage everyone to go to tide pool.org/loop And you can read all about the tide pool loop and also sign up for our mailing list to get updates on things that are coming down the road.

Scott Benner 1:05:11
Do you in your mind, you don't have to share them with me. But is there a list of things like you said, like you, you're going to be able to add things to the to the app as you go. Like, I mean, the one that occurs to me is like auto Bolus, right? Adjustments adjustments through auto Bolus instead of through, like Temp Basal is that like one of the things you're looking at now?

Howard Look 1:05:30
That is absolutely something we're looking at right now, I would say that I talked about overrides. That's clearly the next big suite of functionality that we need to consider for the next version of tide pool loop. And we're working on that right now. And then the biggest thing, quite honestly, that we get asked about, and that we think is important is Android support. So this version of tide pull loop works on an iPhone, we chose that because that's what DIY loop did. So we were starting from an incredible foundation of an iPhone, an Apple Watch app written for iOS. But clearly, if we want to fulfill our mission, as a nonprofit, our goal is to have the broadest possible impact and to enable equitable access to the best possible diabetes technology. And in order to do that, we've got to get bloop onto Android. So that's really important to us as well. I can't say when that will be done. And when that will be ready. I just want everybody to know that it is really important to us. And that's part of our mission. As a nonprofit,

Scott Benner 1:06:35
you just reminded me by saying you're a nonprofit that I have a question here. I think what the person is really asking me is, how do you make money? Like, like, is it like, like, at first I thought, Oh, it was an angel investor that got you started? And then But then how does like you've got a big group of people working for you like, how does that all work?

Howard Look 1:06:55
Yeah, I really appreciate the questions. So over time, our goal is to be a self sustaining nonprofit. And we are generating revenue. Now. We get money from the device makers that we partner with, they pay us to integrate their devices into type of loop. And then over time, they will pay us because the more people that use type of loop with their devices, the more devices they will sell. So they will give us a share of that revenue as well. We also generate revenue through our data management platform, we always have had and always will have a free version of the data management platform. But last summer, we launched what we call tide pool plus, which is an enterprise version of tide pool for web that's very specifically geared towards the needs of enterprise healthcare provider clinics. So it provides functionality, for example, that allows it to integrate with their single sign on mechanism. It has functionality that allows them to manage their entire population of diabetes patients. So there's a dashboard where you can, for example, say, show me all my pregnant moms living with type one, or show me all my new diagnosis patients, or show me all my patients that are struggling with hypoglycemia, so that I can provide better proactive care, okay. And it also lets you do integration with electronic health record systems, for example. So we generate revenue through tide pool plus, which is our enterprise version of tide pool, we generate revenue through tide pool loop, doing both development work, and then ultimately ongoing. And then the final bit is we get incredible support. And I've got to give a shout out to JDRF to the Helmsley Charitable Trust in particular, that have really supported us tremendously over the years, and we would not be here without their support. And there are literally 1000s of individual donors that have reached out. And we get donations everywhere from $1, to $5, to $100, to 10,000, and sometimes hundreds of 1000s of dollars. And every one of those donors makes a difference to us. And it helps us do our work and continue on our mission. So for folks who are listening, who want to help support us, thank you in advance if you have support us, and if you'd like to support us, you can go to type o.org/donate

Scott Benner 1:09:25
That's really wonderful. It's it's actually it's, it's encouraging it really is to just to know that it's that the people don't just say, Oh, that would be nice, but I'm willing to do a thing and I mean, I listen, it's not it's not money, but I have that feeling a little bit like I'm trying to spread the word about algorithms because I just I just know how it will help people. And that's great. It just I think any I think anywhere where people can do something that fits in with their thing is really valuable.

Howard Look 1:09:55
I thought of one other way we make revenue and I'm remiss and talk combat. So I want to make sure your listeners know, one of the things we do is when you make a type pull account, we asked you, would you like to donate your data for research and product development? And 10s of 1000s I think we were well over 30,000 or 40,000. People have checked that box and said, Yes, I'd like to donate my data. That's a pretty amazing thing, because it lets us bundle up those datasets. We call this the tide pool Big Data donation project and create these massive longitudinal datasets, we anonymize them, we strip out all identifying information, but we bundled the datasets together. And we can give them away for free, which we do often to nonprofit academic researchers. But we can also license those datasets to for profit companies, whether they're doing pharmaceutical development, or diabetes, device development. And that's another way that we make revenue. One of the things we do as part of that program, because we're a nonprofit is we want to give back a little to the other diabetes nonprofits that are working so hard to make things better for everyone. So we let you choose which diabetes nonprofit, you want us to share that with that revenue with and we share 10% of the revenue that we get through the type of Big Data donation project back with other diabetes, nonprofits, nonprofits, like children with diabetes, JDRF diabetes sisters, beyond type one, there's a whole list there that people can choose from.

Scott Benner 1:11:34
That's amazing. Thank you. I'm glad I asked that question I almost skipped over. Now I'm like, well, that worked out tide pool just in America or other places.

Howard Look 1:11:45
So the company is based in the US, but we do have international employees. We're a completely remote organization. I'm sitting here in Mountain View, California. Syrah is in Southern California, we've got people all over the US and in Europe. And over the course of time, we've had people in New Zealand and Australia and lots of other places Canada. In terms of use, we officially support our use in the United States. For our type of data management platform. We know we've got tons of users internationally, we are GDPR compliant, which is the data privacy regulation in the European Union. We don't specifically claim that we meet any locales, data privacy requirements, there are some pretty funky ones. Sometimes countries will say, Well, your servers have to be in our country. We don't do that yet. But we also don't shut people off. If we know that they're in another country. We know we've got a ton of people in Israel, for example, there's a ton of people in Northern Europe. And so our goal over time is to much more officially support other languages and other locales. Right now, we're largely US centric, but we definitely have a pretty healthy US population outside the US,

Scott Benner 1:13:05
for people who are familiar with lube and probably familiar with the app that you started with. Has there been anything like we talked about some things you've put into it? Was there anything taken out of it at the behest of the FDA? Or did it pretty much stay the way it was?

Howard Look 1:13:21
It is largely the way it was, I think the constraints on settings were the biggest thing, most notably target range being constrained to 87 to 180. And that wasn't necessarily something the FDA asked us to do. That was something that we knew we needed to do, in order to support support our submission with the clinical data we had. But by and large the features and functionality in tide pole loop map to the features and functionality in DIY loop. It's the DIY loop of mid 2020, however, so we have a little bit of work to do to catch up. But we'll be able to do that much more quickly. Now that we have clearance.

Scott Benner 1:14:00
Did they ask you to add nags? Like after a new site change? Like, don't forget to check your blood sugar like an hour after you made the site change? Did they ask you to put any stuff like that in?

Howard Look 1:14:11
They did not ask us to do anything like that. We did spend a lot of time with them talking through how alerts and alarms will work. And I want to give the agency a lot of credit on this. They really understand that there are just times when you don't want your phone to make noise. If you're getting married, you're in a wedding chapel, you're in a choir singing, you're in a recording studio. You know, it's terrible to think about but if you're in a lockdown situation in a school like there are just times when you don't want your app making noise and the FDA and tide pool iterated on a design for alerts and alarms that makes it possible to provide safe alerts and alarms arms that let you know when there's something you need to pay attention to. But they that also lets you mute them for significant periods of time so that you can accommodate those situations.

Scott Benner 1:15:12
Here's the question. Does this make my phone a medical device? Does that not have to buy me an iPhone knows what I'm asking?

Howard Look 1:15:23
Your iPhone is still an iPhone. Your iPhone is being used as a platform for software as a medical device. I don't think I can answer the question if if it actually turns I mean, the iPhone is not submitted to the FDA as a medical device if that's

Scott Benner 1:15:41
your question, right. But I can't run the app without it. That's right. Oh, I'm totally calling that No, I see. What's that? Yeah. Okay, I'm giving this a shot.

Howard Look 1:15:49
I'm gonna I'm just gonna go out on a limb and say your insurance company is not gonna buy

Scott Benner 1:15:53
Oh, no, they're gonna turn me down. Probably really disappointed. But I'm still gonna ask. Let us know how that goes. Oh, sure. It'll go like this. Hello, hello. I think they hung up. Well, I'm sorry, I lost my space here for a second. Oh, profile switching. People are wondering if you'd be able to like, you know, by using a menstruating woman as an example, like, you know, could I do pre menstrual during the event? I never know what to call it like that kind of stuff. Like, do you have that?

Howard Look 1:16:29
Sorry, I just lost a deal with it later. The work that we are working on now that will add override similar to what you see in DIY loop, I think it's the way that that will be accommodated. So you'll be able to have different forms of overrides for different parts of your menstrual cycle. Okay, but that does not exist in the type of loop that was cleared this week.

Scott Benner 1:16:59
Gotcha. I'm skipping that one. Hold on a second. I did that one. I'm doing good here. Howard. You are actually you're doing really well. I'm basically just reading.

Howard Look 1:17:10
These are great questions. Your your listeners really know what they're listening to and know what they're talking about.

Scott Benner 1:17:16
I agree with that. Can you be a salesman for a second? And answers to answer this question, why do I want this and I don't want control IQ or I don't want the Medtronic one or I don't want on the power? Like why do I want yours?

Howard Look 1:17:30
You know, I'm gonna politely declined to answer that. I think it's a great thing that there are different choices in the world. I think control IQ is going to be the right choice for a lot of people, I think Omni pod five is going to be the right choice for a lot of people. And I think for a lot of people type loop will be the right choice. What I want to see is more interoperability. And more choice, I want people to be able to say that's the right CGM. For me, that's the right pump for me. And this is the right user experience and algorithm for me, I agree with

Scott Benner 1:17:59
you, I was just asking the question, but But I, I 1,000,000% agree with you. I don't care. Honestly, no one should care. One company's not going to get all the business, there's somebody's always going to want to be to bliss or not want to be to bliss or want to use this, you know, want to use a Dexcom or rather use a libre or something. I mean, somebody eventually is going to want to use a Medtronic sensor. Sorry. And, you know, like, like, so when that happens, you don't want to just be like, Oh, now I have to use this thing. And at some point, all those companies have to see that there. I talked to too many people who are like, Well, I have a as an example, I have a Medtronic pump, but I want to use a Dexcom. And so they end up leaving Medtronic because of that. And I'm sure that happens in all different directions with all these companies, you would think that the idea of interoperability would be you think it would be paramount to them. Anyway, that's how I say it. But I'm not going to make the answer that but thank you. Is there any?

Howard Look 1:18:58
I'll answer it. I totally agree. I think that choice is a good thing. I think interoperability is a good thing. And I think the thing that interoperability and choice also lead to is greater access. At the end of the day, what I really want to see happen is everybody, everybody who is living with diabetes to get access to the right technology that works for them. And we just have a lot of work to do. And this is you know, way above my paygrade but as I'm sure your listeners know, the reimbursement system, the health care system, the way access to technology happens today is not fair and equitable. And we are a long way away from the new diagnosis kid with type one who's got a single parent on public assistance assurance, public assistance, insurance, getting access to the best path possible technology. And our small part in that is let's make it easier. Let's make It easier with an interoperable system so that someone can actually get access to the best possible care. Yeah, there's lots more to be done. But that's what I want to see happen.

Scott Benner 1:20:10
Amen. That's well said. Alright, Howard, let me ask you this. Is there anything? I didn't ask you something I should have? Wow.

Howard Look 1:20:17
We covered a lot of ground, Scott. I can't think of anything off the top of my head Syrah. Is there anything that you can think of in your head?

Scott Benner 1:20:28
I know that I was. The one thing I never do understand is that the difference between like, cleared by the FDA and approved by the FDA? Oh, yeah, that's a thing.

Howard Look 1:20:39
Absolutely. They are two different things. Here's the 62nd the digestive what that means. The FDA thinks about medical devices and products in risk levels, risk stratification. There are class one, class two and class three devices. class three is the riskiest and class one is the least risky. So tongue depressors, and band aids and over the counter medicines and things that don't carry a lot of risk are class one. Historically speaking, anything that is new is automatically designated class three because it carries unknown risk. What the FDA did was they created these pathways, and they're called de novo pathways, and de novo means we are going to consider this from new it means from the new and Latin as a class two device. So these interoperability pathways, the AGC, I, CGM and Ace pump designations that I talked about earlier, are all class two devices. When you submit a class three device, you have to go through what's called the PMA process premarket authorization, so are premarket approval. And so that's when you get FDA approved is when you have a class three device with a class two device, you're getting what's known as 510, take 510 K clearance. And what that's saying is, my device is just like this other device that is the same other class two device, and so you get FDA cleared. That's what we just did, we submitted a 510 K application, and we got clearance, or tied pollute our predicate device was control IQ. So part of our submission was going feature by feature capability by capability and saying, here's why our device is what's known as substantially equivalent to the control IQ. What's cool about us getting clearance now as a class two device is we can become the predicate device, or all future devices that want to do similar things to what we've done. Okay. We're pretty proud of that.

Scott Benner 1:22:47
Wow, you should be That's wonderful. This whole thing is amazing. You know, I have to tell you, just to give people the idea of about a span of time, I looked while you were talking, Your Honor, you brought up Dana and Scott, the husband and wife who back when open APs and they were talking about that. They were on my podcast on episode 63 and 64, concurrently. And my best guess is that you're going to be episode like 838 or something like that. Whoa. And when they were amazing, when they were talking back then it was like, I was gobsmacked. I was like, this lady says that, like you don't even know I'm like, she's saying that there's a thing on a thing. And she's saying Raspberry Pi. And I'm like, This sounds like stuff I don't understand. And then she's talking about how great her stuff is, and how our outcomes are. And then I didn't think of it for years. Like it just it felt to me like, like back then it just felt to me like it was three people who were doing this thing, and I happen to talk to a couple of them to see that. I don't know that. I mean, how many years later that is that was the beginning of 2016 that that happened? Well,

Howard Look 1:23:51
if I can pile on that for a second. I think Dana and Scott St. Louis and Scott Lee brand and it may be pronounced Lybrand, I apologies apologies, Scott, if I've got wrong. And before them, John caustic, who did the work to reverse engineer the Dexcom G for Ben West who did the work to reverse engineer the Medtronic insulin pump. And there's a whole bunch of other folks who worked on Nightscout. And I'll miss a bunch of names if I even try, but they all know who they are like those. Everyone who did all of that early work, laid this incredible foundation. And I often say that at tide pool, we're standing on the shoulders of giants. And it's very true. We could not have done this without the work of all the people I just mentioned without the work of Nate rock left, who wrote the original version of loop. And to me it is just incredible testament to how this community wants to make a difference. Like I said earlier, every single person is just saying hey, I think there's a better way to do it, I'm gonna go do it, I'm going to do it, I'm going to use it on myself, I'm going to share my work, I'm going to take contributions that other people make and make it even better. And everybody did that. And I hope in our little way tide pool is doing that too. We're sharing everything we're doing. openly, all of our code is out there, we're actually publishing all of our regulatory interactions openly because we want to help other innovators and entrepreneurs with their interactions with the FDA. And, you know, at the end of the day, it takes a village to help deal with this crappy disease. So we're trying to do the best we can. I

Scott Benner 1:25:38
can't thank you enough for having the idea and bringing it forward like this. And for all the people like us, I can't name them the way you can, but I'm so grateful for all of them. My daughter's life is, is monument monumentally different, like just different than it would have been otherwise? And I'm honestly, I'm not kidding you. I was I was gonna kill myself with the not sleeping thing. You know, the first the first few years I taught my talk myself into believing I was one of those people who didn't need sleep. Like you know, we all tell ourselves that at one point, like, I'll be okay. But man six, seven years into it, and I was I was hanging on by a, by a shoestring you know, and I hear it's a big deal for me. So and yeah, a lot of same people. Same.

Howard Look 1:26:20
My daughter has benefited my wife and I have benefited and 1000s of other people too.

Scott Benner 1:26:24
Okay, tide. pool.org. Right. That's it. All right.

Howard Look 1:26:28
I pulled out org slash loop, sign up, and we'll keep you updated. Howard, thank

Scott Benner 1:26:31
you so much.

Howard Look 1:26:32
I pray thanks for having me on Scott. I really appreciate it, I do too.

Scott Benner 1:26:41
First, I'd like to thank Howard for coming on the show and talking to me today about tide pool loop. And of course, I also want to thank the Contour Next One blood glucose meter, contour next.com forward slash juice box head over there today. See the meter that my daughter uses. It's incredibly accurate. It's one of the reasons I love it the most. Also want to thank us Med and remind you to go to us med.com forward slash juice box to get your free benefits check. Or you can call 888-721-1514. Last reminder, juicebox podcast.com. type one diabetes is our private Facebook group in the feature tab, a list of algorithm based episodes where you can find them at the top of juicebox podcast.com. They're in your episode guides. That's where you'll find the episode guides, or they're in your podcast players. Just use a search like Juicebox Podcast and then you got to find out the names of the episodes. There's like Fox and the loop house and the loop de loop. And you'll see there's a bunch of in there, go find the list. I'll I'll do that. You know what I'm in a good mood. I'll put a list of the episode numbers in the show notes of your podcast player. So in the show notes of the audio app you're listening in right now. Besides links to all the sponsors. I'll put the episode numbers of all of the algorithm based episodes. That sounds good. Alright. Thank you so much for listening. I'll be back very soon with another episode of The Juicebox Podcast.

---

Please support the sponsors

The Juicebox Podcast is a free show, but if you'd like to support the podcast directly, you can make a gift here. Recent donations were used to pay for podcast hosting fees. Thank you to all who have sent 5, 10 and 20 dollars!

Donate