1 00:00:00,160 --> 00:00:02,319 Hi, everyone. This is Lucas Voss with Becker's 2 00:00:02,319 --> 00:00:04,319 Healthcare. Thanks so much for tuning in to 3 00:00:04,319 --> 00:00:07,279 the Becker's Healthcare podcast series. It's fantastic to 4 00:00:07,279 --> 00:00:10,259 have you. Today, we're talking about AI driven 5 00:00:10,559 --> 00:00:12,740 personalized care for chronic conditions 6 00:00:13,304 --> 00:00:16,984 and how payers are adapting through digital health 7 00:00:16,984 --> 00:00:17,484 innovation. 8 00:00:18,024 --> 00:00:20,265 And joining me for today's discussion, I'm so 9 00:00:20,265 --> 00:00:22,344 excited to have her, is doctor Lisa Shaw. 10 00:00:22,344 --> 00:00:24,504 She's the executive vice president and chief medical 11 00:00:24,504 --> 00:00:26,984 officer of Twin Health. Doctor Shaw, thanks so 12 00:00:26,984 --> 00:00:28,344 much for being here today. It's great to 13 00:00:28,344 --> 00:00:30,800 have you. Thank you, Lucas. It's wonderful to 14 00:00:30,800 --> 00:00:31,460 be here. 15 00:00:31,839 --> 00:00:33,600 For those that might not know you yet 16 00:00:33,600 --> 00:00:35,840 at least, I do wanna start off with 17 00:00:35,840 --> 00:00:36,340 introductions. 18 00:00:37,120 --> 00:00:39,280 If you could just introduce yourself and just 19 00:00:39,280 --> 00:00:40,799 tell us a little bit about your work 20 00:00:40,799 --> 00:00:41,700 in health care. 21 00:00:42,164 --> 00:00:44,804 Yes. That's great. So, my name is doctor 22 00:00:44,804 --> 00:00:45,304 Shah, 23 00:00:45,844 --> 00:00:48,664 Lisa. I've been a I'm an internal medicine 24 00:00:48,725 --> 00:00:50,504 and pediatrics trained physician. 25 00:00:51,045 --> 00:00:53,704 I've been practicing for over twenty years now. 26 00:00:54,350 --> 00:00:56,829 Primarily now internal medicine, and we'll talk a 27 00:00:56,829 --> 00:00:58,350 little bit about what I do now. I 28 00:00:58,350 --> 00:01:00,609 did my training at the University of Chicago, 29 00:01:00,670 --> 00:01:03,149 and for the last, fifteen years, I've really 30 00:01:03,149 --> 00:01:04,130 been focused 31 00:01:04,590 --> 00:01:06,129 on designing and implementing 32 00:01:06,854 --> 00:01:09,254 value based and and next generation models of 33 00:01:09,254 --> 00:01:12,614 care, particularly around chronic conditions and and chronic 34 00:01:12,614 --> 00:01:13,114 diseases. 35 00:01:13,894 --> 00:01:15,415 The work I do at Twin Health is 36 00:01:15,415 --> 00:01:18,454 very personal for me. We've been focused here 37 00:01:18,454 --> 00:01:20,774 on chronic metabolic diseases, and we'll talk a 38 00:01:20,774 --> 00:01:22,155 little bit about that today. 39 00:01:22,510 --> 00:01:24,689 But both my parents have type two diabetes 40 00:01:24,989 --> 00:01:25,729 and obesity, 41 00:01:26,269 --> 00:01:28,430 and Twin has really been the first time, 42 00:01:28,430 --> 00:01:29,950 you know, we we go into becoming a 43 00:01:29,950 --> 00:01:31,709 physician to help the people we love the 44 00:01:31,709 --> 00:01:34,670 most. It's been an incredible opportunity to really 45 00:01:34,670 --> 00:01:35,409 see something 46 00:01:35,814 --> 00:01:38,394 change their lives. They were the first two 47 00:01:38,534 --> 00:01:40,854 patients on this program in the history of 48 00:01:40,854 --> 00:01:42,135 the program. And so, 49 00:01:42,454 --> 00:01:45,655 really doing great, reversed their diabetes, lost combined 50 00:01:45,655 --> 00:01:48,295 45 pounds, and are off medication, and we'll 51 00:01:48,295 --> 00:01:50,250 talk a little bit about that today. When 52 00:01:50,250 --> 00:01:52,090 I started my career, I think one of 53 00:01:52,090 --> 00:01:53,790 the things I really struggled with 54 00:01:54,170 --> 00:01:57,450 was how impersonal medicine felt when I went 55 00:01:57,450 --> 00:01:58,189 into practice. 56 00:01:58,650 --> 00:02:01,130 Yeah. It you know, especially in primary care, 57 00:02:01,130 --> 00:02:03,635 it felt very much as I was giving 58 00:02:03,775 --> 00:02:04,915 generic advice 59 00:02:05,295 --> 00:02:06,034 to people. 60 00:02:06,734 --> 00:02:07,954 You know, eat better, 61 00:02:08,414 --> 00:02:09,555 exercise more, 62 00:02:09,854 --> 00:02:12,814 stress less, sleep more, and then they look 63 00:02:12,814 --> 00:02:14,254 at me and go, thanks. Can I have 64 00:02:14,254 --> 00:02:14,914 a prescription? 65 00:02:15,689 --> 00:02:18,830 And that's really what was hard for me. 66 00:02:19,050 --> 00:02:21,770 And if you fast forward to where we 67 00:02:21,770 --> 00:02:22,430 are today, 68 00:02:23,129 --> 00:02:25,710 the opportunity to use artificial intelligence 69 00:02:26,010 --> 00:02:27,150 and machine learning, 70 00:02:27,849 --> 00:02:29,849 techniques and sensors, and we'll talk about how 71 00:02:29,849 --> 00:02:30,990 Twin does that today, 72 00:02:31,425 --> 00:02:32,164 with compassionate 73 00:02:32,544 --> 00:02:33,444 human care, 74 00:02:34,145 --> 00:02:36,004 has really made this hyper personalization, 75 00:02:37,185 --> 00:02:39,905 continuous real time care possible so that people 76 00:02:39,905 --> 00:02:40,724 can actually, 77 00:02:41,344 --> 00:02:44,384 you know, fix the disruption in their bodies, 78 00:02:44,384 --> 00:02:45,525 fix this metabolic 79 00:02:45,824 --> 00:02:46,324 damage, 80 00:02:46,770 --> 00:02:49,110 not just manage a condition with a medication. 81 00:02:49,729 --> 00:02:52,209 Yeah. And I do wanna talk about those 82 00:02:52,209 --> 00:02:53,729 conditions a little bit more because I think 83 00:02:53,729 --> 00:02:55,650 they're at the core of some of the 84 00:02:55,650 --> 00:02:57,090 things that we're gonna touch on today in 85 00:02:57,090 --> 00:02:59,009 our conversation. Again, we we talk about, as 86 00:02:59,009 --> 00:03:02,215 you've mentioned, obesity and diabetes. Right? Unfortunately, those 87 00:03:02,215 --> 00:03:04,694 numbers are continuing to rise. We have made 88 00:03:04,694 --> 00:03:07,094 some inroads. We don't wanna ignore that, but 89 00:03:07,094 --> 00:03:09,335 they continue to rise. Right? And when we're 90 00:03:09,335 --> 00:03:11,514 looking from the pair lens too, 91 00:03:11,974 --> 00:03:15,275 how are you seeing pairs respond there particularly 92 00:03:15,495 --> 00:03:17,039 when it comes to some of the digital 93 00:03:17,039 --> 00:03:19,360 innovation that you've just touched on? And then, 94 00:03:19,360 --> 00:03:22,340 again, what's gaining traction for you, 95 00:03:22,639 --> 00:03:24,019 and where are you seeing 96 00:03:24,639 --> 00:03:26,019 gaps still exist? 97 00:03:27,039 --> 00:03:28,974 Yeah. It's a great question. And, you know, 98 00:03:28,974 --> 00:03:31,055 I've had been working with payers for most 99 00:03:31,055 --> 00:03:33,375 of my career. And I think, you know, 100 00:03:33,375 --> 00:03:35,854 in this value based care world, we started 101 00:03:35,854 --> 00:03:38,574 with really traditional episodic care models and and 102 00:03:38,574 --> 00:03:40,334 point solutions. Right? And that was a great 103 00:03:40,334 --> 00:03:41,794 start if you look back fifteen, 104 00:03:42,189 --> 00:03:43,250 twenty years ago, 105 00:03:44,189 --> 00:03:46,349 a solution for diabetes, a solution for heart 106 00:03:46,349 --> 00:03:47,810 disease, a solution for weight. 107 00:03:48,110 --> 00:03:50,110 But I think there's a growing recognition that 108 00:03:50,110 --> 00:03:52,449 those episodic and point solution, 109 00:03:53,469 --> 00:03:55,629 models are really not bending the cost curve 110 00:03:55,629 --> 00:03:57,409 enough, and they're not actually creating 111 00:03:57,905 --> 00:03:59,764 a place for long term sustained 112 00:04:00,064 --> 00:04:00,564 outcomes. 113 00:04:01,185 --> 00:04:03,745 So payers are, I think, moving away from 114 00:04:03,745 --> 00:04:04,724 that fragmented 115 00:04:05,105 --> 00:04:05,925 point solution, 116 00:04:07,185 --> 00:04:09,985 environment and landscape and really looking for two 117 00:04:09,985 --> 00:04:10,485 things, 118 00:04:11,259 --> 00:04:14,539 models that, are continuous and and wrap around 119 00:04:14,539 --> 00:04:16,800 multiple conditions so that they're getting, 120 00:04:17,339 --> 00:04:19,040 you know, a lot of value 121 00:04:19,420 --> 00:04:20,240 for one, 122 00:04:20,779 --> 00:04:21,279 solution 123 00:04:21,980 --> 00:04:24,139 across all of these things that a that 124 00:04:24,139 --> 00:04:25,660 a patient is struggling with. And then we 125 00:04:25,660 --> 00:04:28,675 know these diseases do not travel alone. Obesity 126 00:04:28,735 --> 00:04:31,535 doesn't travel separately from diabetes. They're partnered as 127 00:04:31,535 --> 00:04:34,254 as is cardiac disease and some other diseases 128 00:04:34,254 --> 00:04:35,714 we'll talk about today, hypertension. 129 00:04:36,175 --> 00:04:37,774 And the second thing is I think they're 130 00:04:37,774 --> 00:04:40,639 really looking for payment models that are standing 131 00:04:40,639 --> 00:04:43,040 behind the outcomes and are outcomes based as 132 00:04:43,040 --> 00:04:45,520 opposed to these traditional per member per month, 133 00:04:46,000 --> 00:04:48,899 you know, type of models where you're really 134 00:04:48,959 --> 00:04:51,300 paying, but there's no guarantee on the outcome. 135 00:04:52,095 --> 00:04:53,954 So those two things have been really 136 00:04:54,654 --> 00:04:55,235 what has, 137 00:04:55,854 --> 00:04:57,774 shifted, I think, in the in the payer 138 00:04:57,774 --> 00:05:00,175 landscape. I think the second thing is they're 139 00:05:00,175 --> 00:05:02,894 seeing cost go up in the world around 140 00:05:02,894 --> 00:05:03,954 things like medications, 141 00:05:04,814 --> 00:05:05,555 and utilization 142 00:05:06,349 --> 00:05:08,769 and prevalence. Right? So as you were saying, 143 00:05:09,149 --> 00:05:11,870 despite our best efforts, diabetes and obesity are 144 00:05:11,870 --> 00:05:12,849 growing. And 145 00:05:13,149 --> 00:05:15,569 look at all the other conditions, metabolic associated 146 00:05:15,629 --> 00:05:18,430 fatty liver disease growing rapidly, the number one 147 00:05:18,430 --> 00:05:20,349 cause of liver transplants now in The US, 148 00:05:20,349 --> 00:05:21,329 right, and then globally. 149 00:05:21,764 --> 00:05:24,004 And so we're seeing these kinds of shifts, 150 00:05:24,004 --> 00:05:25,845 and and they're scratching our heads. We all 151 00:05:25,845 --> 00:05:27,524 are. So what do you what can you 152 00:05:27,524 --> 00:05:29,524 really do to shift this new standard of 153 00:05:29,524 --> 00:05:31,204 care or or create a new standard of 154 00:05:31,204 --> 00:05:32,485 care? And I think, 155 00:05:33,044 --> 00:05:36,139 metabolic health is now being seen as this 156 00:05:36,139 --> 00:05:36,639 continuous, 157 00:05:37,740 --> 00:05:40,379 suite of conditions that if we could really 158 00:05:40,379 --> 00:05:41,680 figure out how to target 159 00:05:42,220 --> 00:05:44,160 metabolism and metabolic health, 160 00:05:44,620 --> 00:05:47,259 in a way that is very personalized because 161 00:05:47,259 --> 00:05:48,560 it's unique for everyone, 162 00:05:48,875 --> 00:05:51,134 there is an opportunity to really prevent, 163 00:05:51,754 --> 00:05:53,835 both of the rising costs and the long 164 00:05:53,835 --> 00:05:55,375 term health op, complications. 165 00:05:56,235 --> 00:05:57,835 And I think in order to do that, 166 00:05:57,835 --> 00:06:00,654 the traction we're seeing is really with solutions 167 00:06:00,715 --> 00:06:01,855 that are data driven. 168 00:06:02,779 --> 00:06:03,439 They're taking 169 00:06:04,060 --> 00:06:05,759 multiple real time datasets 170 00:06:06,459 --> 00:06:07,439 and personalizing 171 00:06:08,060 --> 00:06:09,680 the engagement for the individual 172 00:06:10,139 --> 00:06:12,060 to something that the person is willing to 173 00:06:12,060 --> 00:06:13,279 do and can do, 174 00:06:13,740 --> 00:06:16,584 that is also overseen by a clinical care 175 00:06:16,584 --> 00:06:19,944 team, particularly for areas like diabetes and and 176 00:06:19,944 --> 00:06:20,444 weight. 177 00:06:21,305 --> 00:06:23,464 You mentioned something very critical there with with 178 00:06:23,464 --> 00:06:25,625 just the member engagement piece, which is so 179 00:06:25,625 --> 00:06:27,865 crucial because, again, it doesn't work if the 180 00:06:27,865 --> 00:06:30,329 member doesn't engage. If there's no engagement, then 181 00:06:30,329 --> 00:06:31,789 there's no solution really. 182 00:06:32,329 --> 00:06:34,589 You talk about the concept of, 183 00:06:35,050 --> 00:06:35,709 a digital 184 00:06:36,089 --> 00:06:36,589 twin. 185 00:06:37,209 --> 00:06:37,949 Can you 186 00:06:38,490 --> 00:06:40,810 walk us through what that experience looked like, 187 00:06:40,810 --> 00:06:43,394 what that is for members, and how it 188 00:06:43,394 --> 00:06:45,574 differs from sort of the traditional 189 00:06:46,834 --> 00:06:49,634 digital health approach that we're seeing a lot 190 00:06:49,634 --> 00:06:50,855 around health care today? 191 00:06:51,634 --> 00:06:52,435 Yes. Yeah. 192 00:06:52,754 --> 00:06:53,894 It's a great question. 193 00:06:54,274 --> 00:06:56,355 So digital twins have been around for a 194 00:06:56,355 --> 00:06:59,490 long time. Twin, Twin Health did not create, 195 00:06:59,970 --> 00:07:02,529 the first digital twin concept, though we are 196 00:07:02,529 --> 00:07:03,269 the first, 197 00:07:03,810 --> 00:07:05,750 application of digital twin technology 198 00:07:06,370 --> 00:07:07,509 to human metabolism. 199 00:07:08,529 --> 00:07:11,430 But digital twins are really they're similar technology 200 00:07:11,490 --> 00:07:12,495 being used in other 201 00:07:17,854 --> 00:07:19,794 self driving, autonomous driving cars, 202 00:07:20,735 --> 00:07:22,675 have digital twins. And, essentially, 203 00:07:22,975 --> 00:07:25,235 when you think of a very complex system 204 00:07:25,694 --> 00:07:26,194 like 205 00:07:26,589 --> 00:07:30,050 self driving cars or aviation or human metabolism, 206 00:07:30,430 --> 00:07:32,689 arguably the most complicated system, 207 00:07:33,389 --> 00:07:34,209 in the world. 208 00:07:34,829 --> 00:07:35,229 The, 209 00:07:35,789 --> 00:07:38,910 ability to use sensor technology, which we've really 210 00:07:38,910 --> 00:07:41,229 evolved with here in health care right now 211 00:07:41,229 --> 00:07:42,220 and in in 212 00:07:42,535 --> 00:07:43,995 human body wearables. 213 00:07:44,935 --> 00:07:47,334 But the the cars, for instance, put sensors 214 00:07:47,334 --> 00:07:48,454 on the outside of the car and the 215 00:07:48,454 --> 00:07:50,935 inside. And each driver is unique, but so 216 00:07:50,935 --> 00:07:53,175 is the external environment where the car is. 217 00:07:53,175 --> 00:07:55,254 Right? Whether you're in a traffic jam or 218 00:07:55,254 --> 00:07:57,495 you're driving through snowy mountains or on a 219 00:07:57,495 --> 00:07:57,899 beach. 220 00:07:58,860 --> 00:08:02,540 So the ability to really predict using sensor 221 00:08:02,540 --> 00:08:04,240 technology that gets smarter 222 00:08:04,540 --> 00:08:06,319 with every drive and every, 223 00:08:06,779 --> 00:08:07,279 exposure 224 00:08:07,740 --> 00:08:10,300 to predict bad things before they happen is 225 00:08:10,300 --> 00:08:12,300 really what a digital twin system is trying 226 00:08:12,300 --> 00:08:13,805 to do, and then making 227 00:08:14,504 --> 00:08:16,764 improvements before those bad things happen, 228 00:08:17,225 --> 00:08:19,225 based on what it's seeing through all the 229 00:08:19,225 --> 00:08:20,125 sensor data. 230 00:08:20,425 --> 00:08:21,404 In metabolism, 231 00:08:21,785 --> 00:08:24,365 we're doing the same thing. So we essentially 232 00:08:24,585 --> 00:08:26,925 create an AI digital twin 233 00:08:27,229 --> 00:08:28,930 of every Twin Health member 234 00:08:29,550 --> 00:08:31,729 that evolves and learns every day 235 00:08:32,110 --> 00:08:35,789 and really pays attention to clinical data that 236 00:08:35,789 --> 00:08:37,089 the sensors are collecting. 237 00:08:37,710 --> 00:08:39,629 So think about those as the inside of 238 00:08:39,629 --> 00:08:42,115 the car or inside the human. Right? So 239 00:08:42,115 --> 00:08:43,975 my blood pressure, my blood sugar, 240 00:08:44,674 --> 00:08:45,735 my heart rate, 241 00:08:46,115 --> 00:08:49,315 my stress levels, my sleep data that we're 242 00:08:49,315 --> 00:08:50,695 getting from all of these wearables. 243 00:08:51,154 --> 00:08:53,334 And then it's also looking at the outside. 244 00:08:53,460 --> 00:08:55,139 So think of the environment where the car 245 00:08:55,139 --> 00:08:58,040 is driving, the choices a person is making. 246 00:08:58,340 --> 00:09:00,519 So am I choosing to eat at McDonald's? 247 00:09:00,980 --> 00:09:01,879 Am I choosing 248 00:09:02,259 --> 00:09:02,759 to, 249 00:09:03,540 --> 00:09:05,460 sleep during the daytime because I have a 250 00:09:05,460 --> 00:09:06,600 night shift job? 251 00:09:07,154 --> 00:09:08,514 And so when it looks at all of 252 00:09:08,514 --> 00:09:11,235 those things, whether controllable in the person's life 253 00:09:11,235 --> 00:09:13,654 or not, but the reality of their life, 254 00:09:14,034 --> 00:09:16,855 it puts the two together and creates models 255 00:09:17,075 --> 00:09:20,054 that can help improve those clinical data points 256 00:09:20,519 --> 00:09:21,340 with environmental, 257 00:09:22,440 --> 00:09:25,240 lifestyle suggestions that the member is willing and 258 00:09:25,240 --> 00:09:27,740 able to do. That's the key, willing 259 00:09:28,200 --> 00:09:30,779 and able to do. Because if they're not, 260 00:09:30,920 --> 00:09:33,399 we've seen this before. We've seen keto diets. 261 00:09:33,399 --> 00:09:34,924 We've seen, you know, 262 00:09:35,384 --> 00:09:38,264 extreme gym programs, and they work for a 263 00:09:38,264 --> 00:09:40,904 few months and then people stop because they're 264 00:09:40,904 --> 00:09:42,125 just not really sustainable. 265 00:09:43,304 --> 00:09:44,845 I like to think that my extreme 266 00:09:45,225 --> 00:09:46,904 gym program would work for a month. It 267 00:09:46,904 --> 00:09:48,179 works for about a week. 268 00:09:48,659 --> 00:09:50,980 So I completely understand. It's very, very important. 269 00:09:50,980 --> 00:09:52,419 I agree. Again, we come back to the 270 00:09:52,419 --> 00:09:54,179 engagement piece. Now the great thing is you've 271 00:09:54,179 --> 00:09:55,860 touched on the fact that data is so, 272 00:09:55,860 --> 00:09:56,679 so critical. 273 00:09:57,379 --> 00:09:57,879 And 274 00:09:58,419 --> 00:10:01,695 you announced results from a randomized controlled trial 275 00:10:01,695 --> 00:10:03,375 with Cleveland Clinic, and it was published in 276 00:10:03,375 --> 00:10:05,154 the New England Journal of Medicine Catalyst. 277 00:10:05,615 --> 00:10:06,975 I'd love to talk a little bit about 278 00:10:06,975 --> 00:10:08,335 that because I think that data is really 279 00:10:08,335 --> 00:10:10,735 crucial to what we're talking about today. At 280 00:10:10,735 --> 00:10:13,054 a high level, can you outline some of 281 00:10:13,054 --> 00:10:14,195 those key findings 282 00:10:14,575 --> 00:10:17,259 and sort of how they shape the broader 283 00:10:17,480 --> 00:10:19,320 conversation about some of the things that you've 284 00:10:19,320 --> 00:10:21,639 touched on specifically when we think about AI 285 00:10:21,639 --> 00:10:23,100 driven personalized care? 286 00:10:23,639 --> 00:10:26,679 Yeah. Yes. Absolutely. So the Cleveland Clinic study 287 00:10:26,679 --> 00:10:28,220 that we did was very intentional. 288 00:10:28,595 --> 00:10:30,754 We really wanted to demonstrate you know, we 289 00:10:30,754 --> 00:10:32,595 see a lot of randomized controlled trials looking 290 00:10:32,595 --> 00:10:33,254 at medications. 291 00:10:34,035 --> 00:10:35,955 We don't see a lot of randomized controlled 292 00:10:35,955 --> 00:10:37,815 trials where we can actually use, 293 00:10:38,835 --> 00:10:40,215 AI digital twin technology, 294 00:10:41,009 --> 00:10:41,509 lifestyle 295 00:10:41,970 --> 00:10:42,470 changes, 296 00:10:43,009 --> 00:10:46,050 and actually eliminate medications, which is really what 297 00:10:46,050 --> 00:10:48,529 twin does with this AI digital twin model. 298 00:10:48,529 --> 00:10:50,070 It gets people off medications. 299 00:10:50,529 --> 00:10:52,470 It reduces their a one c. 300 00:10:52,784 --> 00:10:55,424 It reduces their weight. They lose weight in 301 00:10:55,424 --> 00:10:57,184 a healthy way, which we'll talk a little 302 00:10:57,184 --> 00:10:58,084 bit more about, 303 00:10:58,464 --> 00:10:58,964 later. 304 00:10:59,345 --> 00:11:02,464 But it also is, healing a lot of 305 00:11:02,464 --> 00:11:05,284 different aspects of their metabolism and their pathways 306 00:11:05,345 --> 00:11:06,245 in their body. 307 00:11:06,610 --> 00:11:08,790 So in this study, we wanted to demonstrate 308 00:11:08,850 --> 00:11:11,970 that versus usual care for people living with 309 00:11:11,970 --> 00:11:12,870 type two diabetes, 310 00:11:13,330 --> 00:11:16,070 the AI digital twin intervention was superior. 311 00:11:16,529 --> 00:11:19,809 And, we randomized people to either usual care 312 00:11:19,809 --> 00:11:20,950 at the Cleveland Clinic 313 00:11:21,254 --> 00:11:24,455 or the AI digital twin. Cleveland Clinic ran 314 00:11:24,455 --> 00:11:25,095 the study. 315 00:11:25,415 --> 00:11:27,815 The investigators did all of the randomization and 316 00:11:27,815 --> 00:11:28,315 consenting. 317 00:11:29,095 --> 00:11:32,394 100 people were randomized into the intervention arm. 318 00:11:32,535 --> 00:11:35,470 And what we found is, that seventy one 319 00:11:35,470 --> 00:11:36,830 percent of the people who were in the 320 00:11:36,830 --> 00:11:37,330 intervention 321 00:11:37,870 --> 00:11:39,950 arm normalized their a one c and got 322 00:11:39,950 --> 00:11:41,330 off all of their medications 323 00:11:41,789 --> 00:11:42,690 except metformin, 324 00:11:43,070 --> 00:11:45,009 which is sort of a standard baseline, 325 00:11:45,710 --> 00:11:47,870 compared to only two percent in the control 326 00:11:47,870 --> 00:11:50,424 group. So seventy one percent of the intervention 327 00:11:50,424 --> 00:11:52,264 arm at one year had an a one 328 00:11:52,264 --> 00:11:54,824 c that reduced below 6.5, which is a 329 00:11:54,824 --> 00:11:57,384 non type two diabetes a one c and 330 00:11:57,384 --> 00:11:59,084 were off their high cost medication, 331 00:11:59,944 --> 00:12:02,584 all medications actually other than the metformin versus 332 00:12:02,584 --> 00:12:04,960 two percent in the control group. We also 333 00:12:04,960 --> 00:12:07,220 found sixty six percent of participants 334 00:12:07,679 --> 00:12:08,980 normalized their hypertension. 335 00:12:10,160 --> 00:12:13,200 Eighty five percent of them eliminated the GLP 336 00:12:13,200 --> 00:12:13,940 one medications 337 00:12:14,480 --> 00:12:15,379 and sustained 338 00:12:15,920 --> 00:12:18,254 weight loss. Not only that, but they saw 339 00:12:18,414 --> 00:12:21,215 27 pounds average weight loss that they kept 340 00:12:21,215 --> 00:12:23,294 off versus 12 pounds in the usual care 341 00:12:23,294 --> 00:12:23,794 group. 342 00:12:24,174 --> 00:12:26,575 So we not only took away this, you 343 00:12:26,575 --> 00:12:28,735 know, drug, the GLP ones, which I'm sure 344 00:12:28,735 --> 00:12:30,274 we'll cover later, but 345 00:12:30,659 --> 00:12:31,399 they also 346 00:12:31,779 --> 00:12:33,379 were able to lose more weight, and it 347 00:12:33,379 --> 00:12:35,860 was healthy weight with an increase in muscle 348 00:12:35,860 --> 00:12:36,360 mass, 349 00:12:37,139 --> 00:12:39,159 which, you know, is is a really important 350 00:12:39,379 --> 00:12:40,440 aspect of, 351 00:12:41,299 --> 00:12:43,620 having a healthy lifestyle is making sure that 352 00:12:43,620 --> 00:12:45,459 that weight loss is is a healthy weight 353 00:12:45,459 --> 00:12:48,105 loss. So the study really represented a turning 354 00:12:48,105 --> 00:12:51,785 point in diabetes and metabolic diseases because not 355 00:12:51,785 --> 00:12:53,144 only did we show what we could do 356 00:12:53,144 --> 00:12:54,904 for diabetes, but we showed what we were 357 00:12:54,904 --> 00:12:57,004 doing across a host of metabolic conditions, 358 00:12:57,705 --> 00:13:00,285 including metabolic associated fatty liver disease. 359 00:13:00,779 --> 00:13:02,860 And it offers the strongest evidence to date 360 00:13:02,860 --> 00:13:04,720 that AI digital twin technology 361 00:13:05,339 --> 00:13:09,120 can really actually impact and fix disrupted dysfunctional 362 00:13:09,339 --> 00:13:09,839 metabolism 363 00:13:10,620 --> 00:13:13,279 without adding and actually with eliminating 364 00:13:13,995 --> 00:13:16,975 expensive costly medications that have many side effects. 365 00:13:17,835 --> 00:13:19,355 And just to come back to your earlier 366 00:13:19,355 --> 00:13:20,715 point, I feel like this data is also 367 00:13:20,715 --> 00:13:22,335 so important because, again, then 368 00:13:22,955 --> 00:13:25,355 providers can go beyond what you've mentioned earlier, 369 00:13:25,355 --> 00:13:26,575 which is this generalized 370 00:13:27,550 --> 00:13:30,350 experience of, like, eat healthier, sleep better, which 371 00:13:30,350 --> 00:13:31,410 doesn't really help. 372 00:13:31,790 --> 00:13:34,050 This data backs some of those more precise 373 00:13:34,269 --> 00:13:36,450 approaches up, which is so, so crucial. 374 00:13:37,230 --> 00:13:39,565 You touched on GLP ones, and I do 375 00:13:39,565 --> 00:13:41,084 wanna spend just a little bit of time 376 00:13:41,084 --> 00:13:43,084 on that. We could spend a whole episode 377 00:13:43,084 --> 00:13:45,504 talking about it, really, in this context specifically. 378 00:13:46,365 --> 00:13:49,245 But they, again, they continue to create both 379 00:13:49,245 --> 00:13:51,725 opportunity, but also strain for health plans and 380 00:13:51,725 --> 00:13:54,700 employers. Right? There's there's cost, long term adherence, 381 00:13:54,700 --> 00:13:56,080 some of those things. Right? 382 00:13:56,700 --> 00:13:59,339 When you're thinking about GLP ones in the 383 00:13:59,339 --> 00:14:01,679 context of your work, certainly, but also overall, 384 00:14:01,820 --> 00:14:04,620 what shifts do you think are needed, and 385 00:14:04,620 --> 00:14:07,980 how are you positioning Twin Health, right, within 386 00:14:07,980 --> 00:14:08,480 this 387 00:14:08,975 --> 00:14:11,615 evolving landscape? What what's most exciting for you 388 00:14:11,615 --> 00:14:12,115 there? 389 00:14:12,495 --> 00:14:14,575 Yes. So I really appreciate it. Lucas, look, 390 00:14:14,575 --> 00:14:15,955 I, you know, I wanna be, 391 00:14:16,654 --> 00:14:18,495 clear. I think GLP ones are an incredible 392 00:14:18,495 --> 00:14:21,215 advancement in health care and in science, and 393 00:14:21,215 --> 00:14:21,715 medications 394 00:14:22,040 --> 00:14:24,440 play a really, really critical role in managing 395 00:14:24,440 --> 00:14:25,420 metabolic disease. 396 00:14:25,879 --> 00:14:28,620 And when they're clinically appropriately used, they can 397 00:14:28,679 --> 00:14:30,139 absolutely help stabilize, 398 00:14:30,920 --> 00:14:31,820 disease progression, 399 00:14:32,519 --> 00:14:35,639 improve disease progression, manage the disease, and help 400 00:14:35,639 --> 00:14:36,700 with early progress. 401 00:14:37,394 --> 00:14:38,995 But I have a statement I love to 402 00:14:38,995 --> 00:14:41,154 say, which is GLP ones are not for 403 00:14:41,154 --> 00:14:42,695 everyone, and they're not forever. 404 00:14:43,394 --> 00:14:46,695 And I think what's really important is understanding 405 00:14:46,754 --> 00:14:48,834 that these medications also have a lot of 406 00:14:48,834 --> 00:14:49,574 side effects. 407 00:14:50,355 --> 00:14:52,509 You know, from the hair loss to the 408 00:14:52,509 --> 00:14:53,809 nausea and the GI, 409 00:14:54,669 --> 00:14:56,129 issues that people experience 410 00:14:56,750 --> 00:14:58,350 to even just, you know, a lot of 411 00:14:58,350 --> 00:15:01,230 my patients tell me this apathy toward food 412 00:15:01,230 --> 00:15:04,429 and toward sharing and, you know, breaking bread, 413 00:15:04,429 --> 00:15:06,769 so to speak, with their families. And, 414 00:15:07,174 --> 00:15:10,054 you can also see pancreatitis, hospitalization. So there's 415 00:15:10,054 --> 00:15:12,534 there's a lot to consider, and we need 416 00:15:12,534 --> 00:15:14,615 to have a solution for people who either 417 00:15:14,615 --> 00:15:16,615 don't want to take and now there are 418 00:15:16,615 --> 00:15:18,774 orals, but there are many injectables of these 419 00:15:18,774 --> 00:15:19,274 medications, 420 00:15:20,100 --> 00:15:22,440 who don't wanna take these medications for life 421 00:15:22,740 --> 00:15:24,820 or really don't even wanna start them, want 422 00:15:24,820 --> 00:15:26,100 to try to figure out a way to 423 00:15:26,100 --> 00:15:27,320 be medication free. 424 00:15:27,779 --> 00:15:29,940 And so I think it's really important to 425 00:15:29,940 --> 00:15:32,339 have solutions that do that. The other problem 426 00:15:32,339 --> 00:15:33,080 is that 427 00:15:33,554 --> 00:15:35,154 if you look at the history of diabetes 428 00:15:35,154 --> 00:15:35,654 medications, 429 00:15:36,034 --> 00:15:37,495 they're not actually fixing 430 00:15:38,034 --> 00:15:41,475 the disrupted metabolism. They're not actually fixing insulin 431 00:15:41,475 --> 00:15:41,975 resistance, 432 00:15:42,274 --> 00:15:43,634 and some of these things that are at 433 00:15:43,634 --> 00:15:44,215 the precursor 434 00:15:44,754 --> 00:15:46,774 to the conditions that are surfacing, 435 00:15:47,315 --> 00:15:49,509 with, you know, the clinical metrics that we 436 00:15:49,509 --> 00:15:50,809 see or clinical data. 437 00:15:51,190 --> 00:15:53,990 So Twin uses medications as one tool, but 438 00:15:53,990 --> 00:15:56,649 what we are really doing is helping our 439 00:15:56,709 --> 00:15:57,209 clinicians 440 00:15:57,589 --> 00:15:58,089 and 441 00:15:58,389 --> 00:15:59,610 patients or members 442 00:16:00,149 --> 00:16:02,089 understand their own personal biology. 443 00:16:03,125 --> 00:16:05,384 Watching their app every day and saying, wow. 444 00:16:05,924 --> 00:16:08,245 My digital twin just said that when I 445 00:16:08,245 --> 00:16:08,985 eat rice, 446 00:16:09,605 --> 00:16:11,764 I have this response. But when I eat 447 00:16:11,764 --> 00:16:14,424 bread, I have a very different response. Mhmm. 448 00:16:14,725 --> 00:16:16,325 And, you know, this is something we see 449 00:16:16,325 --> 00:16:18,919 in couples and partners who use our program. 450 00:16:19,139 --> 00:16:20,820 That one person to the next, you could 451 00:16:20,820 --> 00:16:22,259 be living in the same house, but you 452 00:16:22,259 --> 00:16:23,559 have a very different 453 00:16:24,179 --> 00:16:26,600 personalized response to the food you consume, 454 00:16:27,379 --> 00:16:28,820 the order in which you eat the foods 455 00:16:28,820 --> 00:16:29,715 on your plate, 456 00:16:30,035 --> 00:16:32,675 and whether you slept well last night, your 457 00:16:32,675 --> 00:16:33,575 stress level, 458 00:16:33,955 --> 00:16:35,555 or whether you went for a walk this 459 00:16:35,555 --> 00:16:37,235 morning with the dog or your son or 460 00:16:37,235 --> 00:16:39,575 daughter took the dog out. So people have 461 00:16:39,955 --> 00:16:42,835 a completely different response to the same exact 462 00:16:42,835 --> 00:16:45,315 food just based on what's happening in their 463 00:16:45,315 --> 00:16:45,690 lives. 464 00:16:46,250 --> 00:16:48,509 And this is, I think, something that medications 465 00:16:48,649 --> 00:16:50,809 don't really crack the nut on. And if 466 00:16:50,809 --> 00:16:53,929 you want a world where people can have 467 00:16:53,929 --> 00:16:56,429 no side effects, can really fix the underlying 468 00:16:56,570 --> 00:16:57,070 problem, 469 00:16:57,529 --> 00:16:59,985 they have to understand their personal body. They 470 00:16:59,985 --> 00:17:02,485 have to understand what's precise and what's meaningful 471 00:17:02,544 --> 00:17:03,924 to their health, 472 00:17:04,544 --> 00:17:07,184 to their blood sugar, their weight gain, their 473 00:17:07,184 --> 00:17:08,724 stress levels, their sleep. 474 00:17:09,025 --> 00:17:11,664 So we really use that AI digital twin 475 00:17:11,664 --> 00:17:14,960 to to monitor that and give precise recommendations 476 00:17:15,259 --> 00:17:18,220 real time using that sensor data and member 477 00:17:18,220 --> 00:17:21,180 preferences to the member every single day. And 478 00:17:21,180 --> 00:17:23,900 that's what managing your diseases is really about. 479 00:17:23,900 --> 00:17:26,240 It's about daily iteration. Right? 480 00:17:27,065 --> 00:17:29,005 If you go on vacation for a week 481 00:17:29,625 --> 00:17:31,944 and then you come back, you wanna get 482 00:17:31,944 --> 00:17:33,545 back on track and you can do that 483 00:17:33,545 --> 00:17:36,184 if daily you understand or have insights into 484 00:17:36,184 --> 00:17:37,785 your own body, and that's what we're really 485 00:17:37,785 --> 00:17:40,880 shifting the paradigm on in metabolic care. 486 00:17:41,500 --> 00:17:44,320 We're really allowing people this chance to say, 487 00:17:44,460 --> 00:17:46,299 well, I took a GLP one, but I'm 488 00:17:46,299 --> 00:17:48,460 I'm done with it, or I've plateaued, and 489 00:17:48,460 --> 00:17:49,119 I'm not 490 00:17:49,420 --> 00:17:51,980 responding anymore, but I'm so terrified to get 491 00:17:51,980 --> 00:17:55,065 off because I hear about this rebound waking, 492 00:17:55,684 --> 00:17:57,845 and I don't wanna be back where I 493 00:17:57,845 --> 00:17:59,944 am. So is there a solution 494 00:18:00,644 --> 00:18:02,505 that can help me responsibly 495 00:18:03,125 --> 00:18:05,704 have this medication, improve my health holistically, 496 00:18:06,460 --> 00:18:08,220 and give me an opportunity to get off 497 00:18:08,220 --> 00:18:10,140 of it? And is there a solution for 498 00:18:10,140 --> 00:18:12,220 people who don't even want to start? And 499 00:18:12,220 --> 00:18:14,880 that is, for me, really exciting. I mean, 500 00:18:15,100 --> 00:18:15,840 my personal, 501 00:18:16,940 --> 00:18:19,580 hope is that people have choice. And I 502 00:18:19,580 --> 00:18:20,080 think, 503 00:18:20,619 --> 00:18:21,680 health care today 504 00:18:22,075 --> 00:18:24,555 isn't set up to give people choice in 505 00:18:24,555 --> 00:18:27,674 their health, and Twin really gives that to 506 00:18:27,674 --> 00:18:28,174 people. 507 00:18:28,875 --> 00:18:30,555 I want the med, but only for a 508 00:18:30,555 --> 00:18:32,475 short time, and then I want to get 509 00:18:32,475 --> 00:18:34,970 off of it. And that's that's totally reasonable. 510 00:18:34,970 --> 00:18:36,570 And I I think it's important to have 511 00:18:36,570 --> 00:18:38,830 more and more solutions that can support that. 512 00:18:39,289 --> 00:18:40,650 Yeah. I'm so glad that you mentioned that 513 00:18:40,650 --> 00:18:42,490 because, again, we touched on the engagement earlier 514 00:18:42,490 --> 00:18:44,650 in the conversation. But, again, that aspect of 515 00:18:44,650 --> 00:18:45,390 giving somebody 516 00:18:45,690 --> 00:18:46,190 agency 517 00:18:47,049 --> 00:18:48,830 is truly what then enables 518 00:18:49,984 --> 00:18:52,545 them thinking, okay. I can make a change 519 00:18:52,545 --> 00:18:54,545 here. I can do this together, which is 520 00:18:54,545 --> 00:18:56,225 so crucial. I'm so glad that you've mentioned 521 00:18:56,225 --> 00:18:58,424 that. It's it's very, very critical. Doctor Shah, 522 00:18:58,424 --> 00:19:00,065 it's so fantastic to have you on. Thank 523 00:19:00,065 --> 00:19:01,585 you so much for taking some time for 524 00:19:01,585 --> 00:19:03,365 us. Anything else that you think 525 00:19:03,670 --> 00:19:04,950 deserves a little bit more attention that we 526 00:19:04,950 --> 00:19:06,710 haven't touched on in our conversation here? Just 527 00:19:06,710 --> 00:19:09,109 wanna turn the floor over to you. Great. 528 00:19:09,109 --> 00:19:10,630 Well, thank you. Well, I'll just leave you 529 00:19:10,630 --> 00:19:12,150 with one thing. I mean, we talk a 530 00:19:12,150 --> 00:19:14,410 lot about AI in health care, 531 00:19:14,869 --> 00:19:17,085 and I know it's everywhere, 532 00:19:17,545 --> 00:19:19,325 you know, and I I see it everywhere. 533 00:19:19,625 --> 00:19:21,945 But most of the talk about AI is 534 00:19:21,945 --> 00:19:23,945 how do we make health care systems and 535 00:19:23,945 --> 00:19:25,164 people more efficient. 536 00:19:25,785 --> 00:19:26,285 Coding, 537 00:19:26,664 --> 00:19:28,105 you know, how do we get them to 538 00:19:28,105 --> 00:19:30,549 document more efficiently? How do we make providers, 539 00:19:30,549 --> 00:19:32,789 you know, able to do their jobs better, 540 00:19:32,789 --> 00:19:33,609 prior authorizations? 541 00:19:34,950 --> 00:19:37,590 Twin is doing all of that absolutely with 542 00:19:37,590 --> 00:19:40,070 our care teams, but we're doing it to 543 00:19:40,070 --> 00:19:41,289 reengineer metabolism. 544 00:19:41,990 --> 00:19:44,549 We're using AI to really improve how people 545 00:19:44,549 --> 00:19:45,884 understand their own health, 546 00:19:46,605 --> 00:19:47,744 how they can address 547 00:19:48,125 --> 00:19:50,684 the things they struggle with every single day 548 00:19:50,684 --> 00:19:51,904 and redefining 549 00:19:52,285 --> 00:19:53,184 how we could 550 00:19:53,644 --> 00:19:56,125 solve chronic disease and redefining the standard of 551 00:19:56,125 --> 00:19:58,619 care with AI digital twin. So I think 552 00:19:58,619 --> 00:20:00,859 it's really exciting, and I I hope we 553 00:20:00,859 --> 00:20:03,279 can see more progress on using AI 554 00:20:03,660 --> 00:20:06,320 toward the diseases and toward the conditions, 555 00:20:06,940 --> 00:20:08,240 as well in the future. 556 00:20:08,779 --> 00:20:10,380 Doctor Shah, so great to have you. Thanks 557 00:20:10,380 --> 00:20:12,444 so much for being here. You as well. 558 00:20:12,444 --> 00:20:14,524 Thank you for having me. We also want 559 00:20:14,524 --> 00:20:16,365 to thank our podcast sponsor, Twin Health. You 560 00:20:16,365 --> 00:20:18,125 can tune in to more podcasts from Becker's 561 00:20:18,125 --> 00:20:22,704 Healthcare visiting our podcast page at beckershospitalreview.com.