1 00:00:00,880 --> 00:00:03,220 Today's episode is brought to you by Abridge. 2 00:00:04,240 --> 00:00:06,559 Named best in class for the ambient AI 3 00:00:06,559 --> 00:00:09,359 segment, Abridge is the leading AI platform for 4 00:00:09,359 --> 00:00:10,580 clinical conversations 5 00:00:11,359 --> 00:00:13,315 designed by clinicians and scientists 6 00:00:13,695 --> 00:00:16,495 to meet the diverse needs of the largest 7 00:00:16,495 --> 00:00:18,274 and most complex health systems. 8 00:00:19,135 --> 00:00:21,154 Abridge automates clinical documentation 9 00:00:21,535 --> 00:00:24,335 in real time and integrates directly within the 10 00:00:24,335 --> 00:00:26,114 electronic health record workflow, 11 00:00:26,690 --> 00:00:30,070 reducing the burden and distraction of repetitive administrative 12 00:00:30,370 --> 00:00:32,870 tasks across care settings, specialties, 13 00:00:33,250 --> 00:00:34,070 and languages. 14 00:00:34,929 --> 00:00:36,850 A bridge has been shown to reduce burnout 15 00:00:36,850 --> 00:00:38,149 by 55%, 16 00:00:38,530 --> 00:00:41,804 improve clinician job satisfaction by 85%, 17 00:00:42,204 --> 00:00:42,945 and significantly 18 00:00:43,405 --> 00:00:47,085 increase patient satisfaction scores according to surveys from 19 00:00:47,085 --> 00:00:47,825 Press Ganey. 20 00:00:48,284 --> 00:00:50,445 Bring this impact to your health system today 21 00:00:50,445 --> 00:00:52,924 and join the community of users across leading 22 00:00:52,924 --> 00:00:56,304 health systems, including Duke Health, Emory Health, 23 00:00:56,685 --> 00:01:00,969 John Hopkins Medicine, and Mayo Clinic. Visit abridge.com 24 00:01:00,969 --> 00:01:03,210 to learn more today. Hello, everyone, and welcome 25 00:01:03,210 --> 00:01:05,930 to the Becker's Healthcare Podcast. I'm Scott King 26 00:01:05,930 --> 00:01:08,989 joined by a very, very special guest today, 27 00:01:09,129 --> 00:01:13,155 Ratnakar Lavu, chief digital information officer with Elevent's 28 00:01:13,215 --> 00:01:15,375 Health. Ratnakar, thanks so much for joining us. 29 00:01:15,614 --> 00:01:16,814 Really excited to have you. How are you 30 00:01:16,814 --> 00:01:17,314 doing? 31 00:01:17,854 --> 00:01:20,094 Good, Scott. Thank you for having me here 32 00:01:20,094 --> 00:01:20,594 today. 33 00:01:20,974 --> 00:01:23,795 Really excited to talk through things with you. 34 00:01:24,015 --> 00:01:25,534 Yeah. We have we have some big things 35 00:01:25,534 --> 00:01:27,454 to get into with, you know, on in 36 00:01:27,454 --> 00:01:28,114 the payer 37 00:01:28,549 --> 00:01:31,270 forefront and in health care and technology. But 38 00:01:31,270 --> 00:01:32,790 before we dive in there, can you please 39 00:01:32,790 --> 00:01:34,549 just tell us a little bit about yourself 40 00:01:34,549 --> 00:01:35,450 and your background? 41 00:01:36,230 --> 00:01:37,109 Yeah. So, 42 00:01:37,510 --> 00:01:37,750 I've 43 00:01:38,549 --> 00:01:40,650 my name is Raka Karlavu. I'm the CDIO, 44 00:01:41,405 --> 00:01:43,564 here at Alvan's Health, and I've always been 45 00:01:43,564 --> 00:01:47,164 within technology and really focused on how we 46 00:01:47,164 --> 00:01:50,064 can leverage technology, data, and AI, 47 00:01:51,004 --> 00:01:52,704 to simplify actually, 48 00:01:53,325 --> 00:01:55,584 consumer experiences and member experiences. 49 00:01:56,060 --> 00:01:56,799 And here, 50 00:01:57,180 --> 00:01:58,799 I'm really focused on simplifying 51 00:01:59,100 --> 00:02:03,180 health care to improve outcomes, leveraging technology, data, 52 00:02:03,180 --> 00:02:06,079 and AI. We're embedding AI responsibly 53 00:02:06,540 --> 00:02:08,240 across all our operations, 54 00:02:09,180 --> 00:02:11,099 not just as an experiment, but we're doing 55 00:02:11,099 --> 00:02:12,044 it at scale 56 00:02:12,424 --> 00:02:14,205 to make care more affordable, 57 00:02:14,745 --> 00:02:15,245 accessible, 58 00:02:15,544 --> 00:02:18,584 and personalized, Scott. So that's what's actually really 59 00:02:18,584 --> 00:02:20,905 exciting about the work that, we do here 60 00:02:20,905 --> 00:02:21,965 at Eleventh Health. 61 00:02:22,584 --> 00:02:24,025 Yeah. Yeah. I I feel like in health 62 00:02:24,025 --> 00:02:25,949 care, we're hearing just more and more about 63 00:02:26,030 --> 00:02:27,709 how vital, you know, especially on a payer 64 00:02:27,709 --> 00:02:30,449 side, data is and then across the board 65 00:02:30,590 --> 00:02:31,330 in the industry, 66 00:02:32,269 --> 00:02:34,830 AI and the the uses there, and and 67 00:02:34,830 --> 00:02:36,909 we're just tapping into that. But what do 68 00:02:36,909 --> 00:02:37,569 you have 69 00:02:37,870 --> 00:02:39,629 your eye on right now as far as 70 00:02:39,629 --> 00:02:40,769 opportunities go? 71 00:02:41,185 --> 00:02:44,145 Well, Scott, you know, health care is very 72 00:02:44,145 --> 00:02:45,105 complex. Right? 73 00:02:45,665 --> 00:02:48,625 And, every one of us, who interact with 74 00:02:48,625 --> 00:02:49,365 health care, 75 00:02:49,824 --> 00:02:52,064 see some of this complexity, whether it is 76 00:02:52,064 --> 00:02:53,844 trying to understand the benefit, 77 00:02:54,430 --> 00:02:56,610 trying to connect with the provider 78 00:02:56,990 --> 00:02:58,129 to get care, 79 00:02:58,510 --> 00:03:00,990 and really focused on kind of, you know, 80 00:03:00,990 --> 00:03:03,629 the health care outcomes itself or the clinical 81 00:03:03,629 --> 00:03:06,909 outcome. So in, Eleventh Health, we're really focused 82 00:03:06,909 --> 00:03:07,889 on three things. 83 00:03:08,430 --> 00:03:09,490 One is simplifying 84 00:03:10,324 --> 00:03:13,284 kind of the member experiences itself. We wanna 85 00:03:13,284 --> 00:03:16,025 create personalized and more seamless experiences. 86 00:03:16,884 --> 00:03:18,724 We wanna actually simplify the, 87 00:03:19,125 --> 00:03:21,544 care provider experiences also, the interoperability. 88 00:03:22,245 --> 00:03:25,205 We wanna empower care providers to improve health 89 00:03:25,205 --> 00:03:25,705 outcomes. 90 00:03:26,300 --> 00:03:28,240 And then finally, we wanna simplify 91 00:03:28,699 --> 00:03:31,919 how we work to actually better serve our 92 00:03:31,979 --> 00:03:34,139 customers and members. And I'll give you a 93 00:03:34,139 --> 00:03:35,120 couple of examples. 94 00:03:35,900 --> 00:03:37,199 Let me start with, 95 00:03:37,580 --> 00:03:39,199 the member or the customer, 96 00:03:40,300 --> 00:03:41,280 example itself. 97 00:03:41,764 --> 00:03:43,525 So we have something called, 98 00:03:44,004 --> 00:03:45,784 proactive member engagement. 99 00:03:46,245 --> 00:03:46,564 And, 100 00:03:47,284 --> 00:03:50,164 what this is is it's actually an AI 101 00:03:50,164 --> 00:03:51,145 driven tool 102 00:03:51,844 --> 00:03:55,284 wherein we identify members who could benefit from 103 00:03:55,284 --> 00:03:55,944 an outreach 104 00:03:56,280 --> 00:03:59,959 to proactively address care gaps. So we study 105 00:03:59,959 --> 00:04:02,780 the members, their clinical data. We understand 106 00:04:03,159 --> 00:04:05,900 who could actually benefit from closing care gap. 107 00:04:06,120 --> 00:04:08,459 We do an outreach for to them 108 00:04:08,759 --> 00:04:11,985 to explain to them what would be beneficial, 109 00:04:13,085 --> 00:04:15,485 as they think about going to their care 110 00:04:15,485 --> 00:04:15,985 provider, 111 00:04:16,285 --> 00:04:18,545 what will be beneficial for them to address 112 00:04:19,004 --> 00:04:21,665 so that they can actually have that meaningful 113 00:04:21,725 --> 00:04:23,564 interaction with their care provider, 114 00:04:24,209 --> 00:04:26,050 with a lot more information. And we do 115 00:04:26,050 --> 00:04:27,430 millions of these outreaches 116 00:04:27,730 --> 00:04:28,230 today. 117 00:04:28,610 --> 00:04:31,670 And it's the AI, the in data, 118 00:04:32,050 --> 00:04:34,050 and then the kind of the intelligence that 119 00:04:34,050 --> 00:04:36,894 we build that enables that to happen. The 120 00:04:36,894 --> 00:04:39,634 other example is when members actually call in 121 00:04:39,774 --> 00:04:41,074 to our call center, 122 00:04:41,454 --> 00:04:44,095 we provide our call center agents with the 123 00:04:44,095 --> 00:04:44,595 intelligence 124 00:04:44,975 --> 00:04:46,675 to be able to actually, 125 00:04:47,454 --> 00:04:50,329 address the member issues. So we we correlate 126 00:04:50,550 --> 00:04:51,370 all the information, 127 00:04:51,750 --> 00:04:52,970 the clinical data, 128 00:04:53,750 --> 00:04:55,050 you know, kind of their, 129 00:04:55,669 --> 00:04:56,569 claims data, 130 00:04:56,949 --> 00:05:00,410 their benefits data, and automate the things, 131 00:05:00,790 --> 00:05:02,729 for our, call center agent. 132 00:05:03,074 --> 00:05:05,475 Not only that, but once they finish the 133 00:05:05,475 --> 00:05:08,514 call, we actually do a post call wrap 134 00:05:08,514 --> 00:05:09,014 up, 135 00:05:09,314 --> 00:05:12,035 which is we use AI to understand how 136 00:05:12,035 --> 00:05:13,014 the call went. 137 00:05:13,394 --> 00:05:15,254 How did our members feel? 138 00:05:15,794 --> 00:05:17,014 What were their issues? 139 00:05:17,339 --> 00:05:19,819 And we do millions of these, post call 140 00:05:19,819 --> 00:05:20,540 wrap ups, 141 00:05:20,939 --> 00:05:23,579 every month. And the reason for that is 142 00:05:23,579 --> 00:05:26,160 we actually understand the sentiment of the member 143 00:05:26,220 --> 00:05:27,519 so that we can proactively 144 00:05:27,819 --> 00:05:28,639 address that. 145 00:05:29,100 --> 00:05:31,500 The reason these examples are important is because 146 00:05:31,500 --> 00:05:34,964 we're very focused and obsessed about the member 147 00:05:34,964 --> 00:05:38,564 experience, the member interactions itself. And I'll then 148 00:05:38,564 --> 00:05:40,345 talk a little bit about the care providers. 149 00:05:40,884 --> 00:05:43,524 So in the care providers, we wanna make 150 00:05:43,524 --> 00:05:46,745 the interaction between us and the care providers 151 00:05:46,884 --> 00:05:48,289 easy. So within 152 00:05:48,750 --> 00:05:50,289 UMAI, where they actually, 153 00:05:51,069 --> 00:05:54,289 kind of submit their prior on, we actually 154 00:05:54,430 --> 00:05:56,129 look at what they've submitted, 155 00:05:56,829 --> 00:05:59,009 analyze it with the data that we have, 156 00:05:59,069 --> 00:06:01,009 and our entire goal and objective 157 00:06:01,389 --> 00:06:03,524 is to approve things as fast as we 158 00:06:04,004 --> 00:06:05,544 can, for the care provider. 159 00:06:06,084 --> 00:06:06,564 And, 160 00:06:07,044 --> 00:06:09,044 we use a lot of AI to look 161 00:06:09,044 --> 00:06:11,524 at medical record, to look at their prior 162 00:06:11,524 --> 00:06:12,264 auth submission, 163 00:06:12,644 --> 00:06:15,285 to look at the member's benefits data, and 164 00:06:15,285 --> 00:06:16,985 orchestrate that approval 165 00:06:17,449 --> 00:06:19,610 in a more seamless way for the care 166 00:06:19,610 --> 00:06:20,110 provider. 167 00:06:20,569 --> 00:06:23,209 And another thing is the care providers actually 168 00:06:23,209 --> 00:06:26,250 submit a lot of claims. Our providers submit 169 00:06:26,250 --> 00:06:29,050 a lot of claim. Sometimes those claims are 170 00:06:29,050 --> 00:06:29,790 not complete. 171 00:06:30,214 --> 00:06:32,555 And so we have at applied AI 172 00:06:32,935 --> 00:06:35,014 to look at those claims and see where, 173 00:06:35,495 --> 00:06:38,615 the information is not complete. And the AI 174 00:06:38,615 --> 00:06:39,754 actually help in 175 00:06:40,134 --> 00:06:42,214 completing that information so we can process the 176 00:06:42,214 --> 00:06:45,000 claims on behalf of the providers. Why is 177 00:06:45,000 --> 00:06:47,339 this important? Because a lot of care providers 178 00:06:47,399 --> 00:06:50,139 spend a lot of time in administrative work, 179 00:06:50,360 --> 00:06:52,680 and we wanna take away that or simplify 180 00:06:52,680 --> 00:06:55,079 that administrative work for the for them so 181 00:06:55,079 --> 00:06:56,839 that they can really take care of our 182 00:06:56,839 --> 00:06:58,139 members and, 183 00:06:58,519 --> 00:07:01,535 close those care gaps that are extremely important 184 00:07:01,535 --> 00:07:02,435 for our members. 185 00:07:02,975 --> 00:07:05,294 Now let me come to how we are 186 00:07:05,294 --> 00:07:05,794 simplifying 187 00:07:06,095 --> 00:07:08,115 the work itself for our associates. 188 00:07:08,814 --> 00:07:11,294 So we have we're deploying a lot of 189 00:07:11,294 --> 00:07:14,439 AI productivity tools so that they we can 190 00:07:14,439 --> 00:07:16,220 automate a lot of their workflows 191 00:07:16,680 --> 00:07:19,000 so they really can actually service our members 192 00:07:19,000 --> 00:07:19,659 and providers. 193 00:07:20,120 --> 00:07:22,300 So we have over 52,000 194 00:07:22,360 --> 00:07:23,419 associates today. 195 00:07:23,800 --> 00:07:26,360 We will that are leveraging some sort of 196 00:07:26,360 --> 00:07:27,180 AI capabilities, 197 00:07:28,104 --> 00:07:31,305 within their workflows. But we're also rolling out 198 00:07:31,305 --> 00:07:31,805 now, 199 00:07:32,425 --> 00:07:34,685 chat GPT across the enterprise. 200 00:07:35,224 --> 00:07:37,865 So not only is it for that that 201 00:07:37,865 --> 00:07:40,125 they can do their work more easily, 202 00:07:40,519 --> 00:07:43,259 but they can also kind of understand leveraging 203 00:07:43,480 --> 00:07:45,740 in these chat GPT and other capabilities 204 00:07:46,199 --> 00:07:48,300 to be able to serve our members effectively. 205 00:07:48,920 --> 00:07:51,080 And then the not only are we rolling 206 00:07:51,080 --> 00:07:52,379 out these tools, 207 00:07:52,685 --> 00:07:54,845 we're actually going to take them through AI 208 00:07:54,845 --> 00:07:56,625 certification, working with OpenAI. 209 00:07:57,165 --> 00:07:59,345 Because one of the things we find is 210 00:07:59,884 --> 00:08:02,205 it's not just about the tool, it's our 211 00:08:02,205 --> 00:08:05,004 associates need to understand how to use those 212 00:08:05,004 --> 00:08:08,229 tools, what what it means to leverage AI 213 00:08:08,529 --> 00:08:11,329 to actually do their work better, and that's 214 00:08:11,329 --> 00:08:13,990 where we're we're focused on quite a bit, 215 00:08:14,449 --> 00:08:15,430 with our associate. 216 00:08:15,810 --> 00:08:17,269 So, again, to recap, 217 00:08:18,185 --> 00:08:21,245 we want to actually create a more personalized 218 00:08:21,785 --> 00:08:24,824 and more seamless experience for our members. We 219 00:08:24,824 --> 00:08:27,865 wanna empower our care providers to improve health 220 00:08:27,865 --> 00:08:28,365 outcomes, 221 00:08:28,824 --> 00:08:30,204 and then we wanna simplify 222 00:08:30,504 --> 00:08:33,000 how we work to better serve our 223 00:08:33,379 --> 00:08:35,000 customers and members itself. 224 00:08:35,940 --> 00:08:38,519 Incredibly fascinating. A lot of so much great 225 00:08:38,580 --> 00:08:40,580 information in there right in the car. Let 226 00:08:40,580 --> 00:08:41,700 let me ask you, I guess, to play 227 00:08:41,779 --> 00:08:43,379 and to play devil's advocate a little bit, 228 00:08:43,379 --> 00:08:45,725 you know, even the most practical uses of 229 00:08:45,725 --> 00:08:48,285 AI the average person uses, there's still some 230 00:08:48,285 --> 00:08:50,684 some trial and error. So just curious with 231 00:08:50,684 --> 00:08:53,884 something like the the, AI member outreach that 232 00:08:53,884 --> 00:08:55,804 the elephants has been doing and even those 233 00:08:55,804 --> 00:08:58,524 summaries that it provides, what what's kind of 234 00:08:58,524 --> 00:09:00,629 some of the, I guess, the errors that 235 00:09:00,629 --> 00:09:03,269 there might be, and are they kind of 236 00:09:03,269 --> 00:09:06,149 fixed through the AI self learning? How's how's 237 00:09:06,149 --> 00:09:08,309 that working? How's it getting kinda past any 238 00:09:08,309 --> 00:09:09,450 any hiccups there? 239 00:09:09,909 --> 00:09:11,990 Yeah. Actually, I I wouldn't say it's a 240 00:09:11,990 --> 00:09:14,409 lot, about the hiccups. What we have learned 241 00:09:14,470 --> 00:09:16,764 actually is we need to obsess about the 242 00:09:16,764 --> 00:09:17,904 journeys themselves. 243 00:09:18,445 --> 00:09:20,764 So we are really, kind of focused on 244 00:09:20,764 --> 00:09:21,745 the member journeys, 245 00:09:22,524 --> 00:09:24,764 overall or the provider journey. So I'll give 246 00:09:24,764 --> 00:09:26,384 you an example. In the UMAI 247 00:09:26,764 --> 00:09:27,264 case, 248 00:09:27,819 --> 00:09:30,000 when, a prior auth is submitted, 249 00:09:30,539 --> 00:09:33,339 our entire focus is to how to get 250 00:09:33,339 --> 00:09:36,459 to real time prior authorization. Right? So when 251 00:09:36,459 --> 00:09:37,839 we focus on that journey, 252 00:09:38,459 --> 00:09:40,959 what we have found is when we deploy 253 00:09:41,100 --> 00:09:42,000 AI solutions, 254 00:09:42,975 --> 00:09:45,315 we our our own learning has been 255 00:09:45,774 --> 00:09:48,254 don't just deploy it in in a part 256 00:09:48,254 --> 00:09:49,075 of the workflow, 257 00:09:49,774 --> 00:09:52,095 really think about the entire workflow. So when 258 00:09:52,095 --> 00:09:54,014 we deployed it in the case of prior 259 00:09:54,014 --> 00:09:54,514 auth, 260 00:09:54,980 --> 00:09:57,379 we were only looking at it as the 261 00:09:57,460 --> 00:09:58,279 kind of the 262 00:09:58,660 --> 00:10:01,620 approval for the prior auth. In in fact, 263 00:10:01,620 --> 00:10:03,700 there's a lot more in the process that 264 00:10:03,700 --> 00:10:05,860 we'll have to think through. And so we 265 00:10:05,860 --> 00:10:07,080 went and pivoted. 266 00:10:07,700 --> 00:10:10,284 Instead of just a portion of the workflow, 267 00:10:10,664 --> 00:10:12,985 we pivoted to actually looking at the entire 268 00:10:12,985 --> 00:10:13,485 workflow. 269 00:10:13,865 --> 00:10:16,105 And that has been the big learning for 270 00:10:16,105 --> 00:10:18,584 us, is when you think about AI, just 271 00:10:18,584 --> 00:10:20,825 don't think about it in a silo of 272 00:10:20,825 --> 00:10:21,485 a workflow. 273 00:10:21,929 --> 00:10:23,549 Think about the entire workflow, 274 00:10:23,929 --> 00:10:24,429 and 275 00:10:24,809 --> 00:10:25,309 we 276 00:10:25,929 --> 00:10:27,470 reimagine that workflow 277 00:10:28,009 --> 00:10:28,970 as you think about, 278 00:10:29,690 --> 00:10:32,649 building solutions and deploying AI solution. And that's 279 00:10:32,649 --> 00:10:36,029 how actually you'll make the entire workflow simple. 280 00:10:36,404 --> 00:10:38,485 We we we're re and then we'll also 281 00:10:38,485 --> 00:10:40,404 get them to be more, 282 00:10:40,884 --> 00:10:41,865 kind of connected 283 00:10:42,324 --> 00:10:43,865 as we think about the interoperability 284 00:10:44,324 --> 00:10:45,865 between provider and payer. 285 00:10:46,565 --> 00:10:48,324 Right. It's a journey. So focus on that 286 00:10:48,324 --> 00:10:50,160 journey. I I I think that makes complete 287 00:10:50,160 --> 00:10:50,320 sense. 288 00:10:51,279 --> 00:10:54,019 In in terms of, you know, the potential 289 00:10:54,159 --> 00:10:57,200 maybe that people or payers have even tapped 290 00:10:57,200 --> 00:10:57,700 into 291 00:10:58,079 --> 00:10:59,679 for AI, what are some of the things 292 00:10:59,679 --> 00:11:02,559 that Evolence is kind of looking to in 293 00:11:02,559 --> 00:11:04,339 terms of focus for the future? 294 00:11:05,065 --> 00:11:07,404 Well, I see tremendous amount of potential, 295 00:11:07,865 --> 00:11:09,544 in terms of the focus for the future. 296 00:11:09,544 --> 00:11:13,144 Again, we're deep rooted into creating personalized and 297 00:11:13,144 --> 00:11:14,365 more seamless experiences. 298 00:11:14,985 --> 00:11:17,144 So I'll give you an example. We we 299 00:11:17,144 --> 00:11:18,524 have deployed a virtual 300 00:11:19,199 --> 00:11:21,139 assistant, in our Sydney app 301 00:11:21,519 --> 00:11:23,519 for our members to use that. It's like 302 00:11:23,519 --> 00:11:26,339 a chat GPT experience. So they can actually 303 00:11:26,399 --> 00:11:29,839 ask questions. Is is this particular procedure covered 304 00:11:29,839 --> 00:11:30,659 in my benefits? 305 00:11:31,199 --> 00:11:33,524 Right? And then we we kind of the 306 00:11:33,524 --> 00:11:36,804 AI has intelligence to bring that information back 307 00:11:36,804 --> 00:11:37,625 to the member. 308 00:11:38,245 --> 00:11:40,964 And then not only that, it act it 309 00:11:40,964 --> 00:11:42,824 can then connect that member 310 00:11:43,125 --> 00:11:44,664 to the right care provider 311 00:11:45,139 --> 00:11:47,460 so who can drive quality outcomes at a 312 00:11:47,539 --> 00:11:49,779 at an affordable price. So we have built 313 00:11:49,779 --> 00:11:52,340 that intelligence where the AI understands what the 314 00:11:52,340 --> 00:11:54,980 member is asking for and how to connect 315 00:11:54,980 --> 00:11:57,159 them to get the care and 316 00:11:57,934 --> 00:11:59,154 the clinical outcomes 317 00:11:59,534 --> 00:11:59,774 that, 318 00:12:00,414 --> 00:12:02,654 we we actually want to guide the member 319 00:12:02,654 --> 00:12:05,774 through. And so what I'm excited about is 320 00:12:05,774 --> 00:12:08,674 this creating this personalized and more seamless experiences 321 00:12:09,214 --> 00:12:09,954 at scale 322 00:12:10,254 --> 00:12:12,654 now leveraging AI. The other thing that I'm 323 00:12:12,654 --> 00:12:13,679 really kind of, 324 00:12:14,240 --> 00:12:16,259 excited about, as I mentioned, 325 00:12:16,720 --> 00:12:18,419 is improving this interoperability 326 00:12:19,360 --> 00:12:21,940 between here and provider because we wanna streamline 327 00:12:22,480 --> 00:12:23,700 and make it simple 328 00:12:24,159 --> 00:12:24,879 for the, 329 00:12:25,200 --> 00:12:28,414 providers to actually engage with us. And I 330 00:12:28,414 --> 00:12:31,134 think there's tremendous amount of potential there. And 331 00:12:31,134 --> 00:12:31,634 finally, 332 00:12:32,014 --> 00:12:32,754 I'm really, 333 00:12:33,454 --> 00:12:35,954 kind of excited about how we can transform, 334 00:12:37,054 --> 00:12:38,434 internally our company 335 00:12:38,894 --> 00:12:42,095 so that we can kind of automate a 336 00:12:42,095 --> 00:12:43,794 lot of the mundane tasks, 337 00:12:44,389 --> 00:12:46,629 that are there, within a large company like 338 00:12:46,629 --> 00:12:50,149 ours with, leveraging AI, and really focus on 339 00:12:50,149 --> 00:12:52,549 on the member and the provider and the 340 00:12:52,549 --> 00:12:55,110 high value added task because that's where we 341 00:12:55,110 --> 00:12:56,710 need to be spending a lot more time 342 00:12:56,710 --> 00:12:58,250 to simplify health care. 343 00:12:59,024 --> 00:13:00,865 Yeah. And for anyone listening who wants to 344 00:13:00,865 --> 00:13:02,705 to get more information or or read about 345 00:13:02,705 --> 00:13:05,024 this virtual assistant Ovens is working on, Retina 346 00:13:05,024 --> 00:13:07,184 Carr did a great interview with Becker's, looks 347 00:13:07,184 --> 00:13:09,365 like the end of of October. And, 348 00:13:09,825 --> 00:13:11,345 let let me ask you. How's how's it 349 00:13:11,504 --> 00:13:12,404 how are the 350 00:13:13,009 --> 00:13:16,289 the total number of members that'll have access 351 00:13:16,289 --> 00:13:18,210 to the virtual assistant? How's that tracking as 352 00:13:18,210 --> 00:13:21,350 we're here almost, you know, at the 2025? 353 00:13:21,970 --> 00:13:24,049 Actually, it's tracking really well. So we're getting 354 00:13:24,049 --> 00:13:25,269 a lot more adoption, 355 00:13:26,129 --> 00:13:28,389 you know, as we roll this out to, 356 00:13:29,134 --> 00:13:29,695 kind of, 357 00:13:30,095 --> 00:13:32,575 a large set of our members, we're getting 358 00:13:32,575 --> 00:13:33,794 a lot of adoption, 359 00:13:34,335 --> 00:13:35,554 for the virtual assistant. 360 00:13:36,014 --> 00:13:38,335 What we are seeing is the members are 361 00:13:38,335 --> 00:13:41,500 highly engaged. They wanna understand their benefit. And 362 00:13:41,500 --> 00:13:43,180 the first for the first time, we made 363 00:13:43,180 --> 00:13:45,980 it actually extremely simple for them to understand 364 00:13:45,980 --> 00:13:46,559 the benefit. 365 00:13:46,860 --> 00:13:49,660 Not only that, it is actually making them 366 00:13:49,660 --> 00:13:52,139 easier also to connect with the right provider 367 00:13:52,139 --> 00:13:54,700 because the intelligence that we have built through 368 00:13:54,700 --> 00:13:55,200 this 369 00:13:55,575 --> 00:13:57,735 actually connects them with the right provider who 370 00:13:57,735 --> 00:14:00,134 can drive high quality of care at an 371 00:14:00,134 --> 00:14:02,535 affordable price. And so we're seeing a lot 372 00:14:02,535 --> 00:14:05,595 of engagement. What actually it's helping us do 373 00:14:06,054 --> 00:14:07,654 is the member is able to, 374 00:14:08,909 --> 00:14:09,970 kind of do self-service 375 00:14:10,829 --> 00:14:13,709 instead of calling and trying to understand their 376 00:14:13,709 --> 00:14:16,350 benefits or calling to actually get to the 377 00:14:16,350 --> 00:14:18,990 right provider. Now all of it is moving 378 00:14:18,990 --> 00:14:21,549 towards self-service, and that's what we're really excited 379 00:14:21,549 --> 00:14:24,129 about. Because we really wanna empower our members, 380 00:14:25,034 --> 00:14:27,835 to kind of navigate this complex health care 381 00:14:27,835 --> 00:14:29,835 system, and we wanna make it easy and 382 00:14:29,835 --> 00:14:31,534 simple for them to do that. 383 00:14:31,914 --> 00:14:33,274 Yeah. It sounds like the rollout has been 384 00:14:33,274 --> 00:14:34,014 going great. 385 00:14:35,434 --> 00:14:36,174 As organizations 386 00:14:36,554 --> 00:14:39,375 continue to expand and implement AI driven solutions, 387 00:14:39,929 --> 00:14:41,549 how do they do so in a responsible 388 00:14:41,769 --> 00:14:43,149 and ethical way? 389 00:14:44,009 --> 00:14:45,929 It's a great question, Scott. So, by the 390 00:14:45,929 --> 00:14:47,470 way, we begin with 391 00:14:47,769 --> 00:14:50,750 responsible AI. We begin with transparency. 392 00:14:51,370 --> 00:14:52,669 We begin with explainability. 393 00:14:53,450 --> 00:14:54,990 So we actually have 394 00:14:55,375 --> 00:14:57,235 an entire responsible AI 395 00:14:57,615 --> 00:15:01,615 governance framework within our company. So every AI 396 00:15:01,615 --> 00:15:03,715 solution has to go through that governance 397 00:15:04,095 --> 00:15:06,115 framework where we evaluate, 398 00:15:07,254 --> 00:15:08,355 the AI solution 399 00:15:08,894 --> 00:15:09,375 for, 400 00:15:09,774 --> 00:15:10,274 responsible 401 00:15:10,654 --> 00:15:12,730 AI, transparency, and explainability, 402 00:15:13,590 --> 00:15:16,309 and how the solution is actually designed and 403 00:15:16,309 --> 00:15:17,129 being implemented. 404 00:15:17,750 --> 00:15:20,629 Only then will we actually approve that AI 405 00:15:20,629 --> 00:15:21,129 solution 406 00:15:21,509 --> 00:15:22,649 to go into production. 407 00:15:23,110 --> 00:15:24,495 And then we monitor that, 408 00:15:25,054 --> 00:15:26,514 you know, once it's in production 409 00:15:26,975 --> 00:15:30,355 to ensure that all the transparency and explainability 410 00:15:31,054 --> 00:15:31,855 is are 411 00:15:32,495 --> 00:15:33,475 has been implemented 412 00:15:34,174 --> 00:15:36,894 effectively once it goes into production. So you 413 00:15:36,894 --> 00:15:39,370 need to both have a framework in terms 414 00:15:39,370 --> 00:15:40,590 of the governance framework. 415 00:15:41,129 --> 00:15:43,690 That governance framework has to start from the 416 00:15:43,690 --> 00:15:46,269 beginning of the solution when it's actually designed, 417 00:15:46,730 --> 00:15:49,050 not after the fact. And so it has 418 00:15:49,050 --> 00:15:50,509 to be part of the process. 419 00:15:50,975 --> 00:15:53,875 And then once you actually implement the solution, 420 00:15:54,335 --> 00:15:56,735 you actually have to also monitor it for 421 00:15:56,735 --> 00:15:57,555 its effectiveness. 422 00:15:57,935 --> 00:15:59,855 And that's the way we have actually done 423 00:15:59,855 --> 00:16:01,475 that, and we've been very successful 424 00:16:02,014 --> 00:16:02,995 in doing that, 425 00:16:03,375 --> 00:16:05,235 across all our AI solutions. 426 00:16:06,569 --> 00:16:08,490 In terms of all the uses Elvents has 427 00:16:08,490 --> 00:16:10,649 had for AI, a lot of the things 428 00:16:10,649 --> 00:16:11,949 you've discussed already, 429 00:16:12,649 --> 00:16:14,589 has Elvents been able to pinpoint, 430 00:16:15,049 --> 00:16:17,389 like, you know, just a a number for 431 00:16:17,690 --> 00:16:20,589 efficiency in regards to hours saved between 432 00:16:21,235 --> 00:16:22,855 payers and and and members? 433 00:16:23,875 --> 00:16:25,794 Yes. We do. And so one of the 434 00:16:25,794 --> 00:16:28,615 things we constantly watch is, 435 00:16:28,915 --> 00:16:30,054 is it improving 436 00:16:30,754 --> 00:16:33,634 the outcomes that we we have kind of, 437 00:16:34,115 --> 00:16:34,615 outlined, 438 00:16:35,600 --> 00:16:38,100 you know, in terms of the member efficiency 439 00:16:38,320 --> 00:16:40,579 to actually self serve. Right? 440 00:16:41,039 --> 00:16:43,299 That's a huge efficiency because the members, 441 00:16:44,399 --> 00:16:45,139 are effectively 442 00:16:45,679 --> 00:16:47,519 getting all the information that they had to 443 00:16:47,519 --> 00:16:49,919 go to multiple places in the past. So 444 00:16:49,919 --> 00:16:50,419 self-service. 445 00:16:50,915 --> 00:16:53,634 And then self-service, not only just self-service, how 446 00:16:53,634 --> 00:16:56,274 satisfied are they about with the self-service, so 447 00:16:56,274 --> 00:16:58,675 NPS scores and other thing. So we monitor 448 00:16:58,675 --> 00:17:00,915 all of that. We also look at in 449 00:17:00,915 --> 00:17:01,654 the peers, 450 00:17:03,394 --> 00:17:05,255 in the peer provider interaction, 451 00:17:05,779 --> 00:17:07,880 we also looked at kind of the effectiveness 452 00:17:08,099 --> 00:17:10,419 of our interaction overall. How, 453 00:17:10,899 --> 00:17:11,399 satisfied 454 00:17:11,700 --> 00:17:12,519 are the providers, 455 00:17:13,299 --> 00:17:15,559 both in terms of how we 456 00:17:15,859 --> 00:17:17,960 kind of set them up to begin with, 457 00:17:18,125 --> 00:17:19,404 and then how do they, 458 00:17:19,805 --> 00:17:21,265 actually how do we process 459 00:17:21,805 --> 00:17:24,384 their kind of, prior auths, claims, 460 00:17:25,325 --> 00:17:28,545 and and any other administrative thing. We actually 461 00:17:28,605 --> 00:17:31,330 look at how effective those are, how are 462 00:17:31,330 --> 00:17:33,950 we reducing the time of processing those, 463 00:17:34,529 --> 00:17:35,029 administrative 464 00:17:35,330 --> 00:17:38,309 tasks, and then how satisfied are the providers 465 00:17:38,369 --> 00:17:40,850 there. And then in the associate population, we 466 00:17:40,850 --> 00:17:41,990 do a lot of, 467 00:17:43,009 --> 00:17:43,750 self assessment. 468 00:17:44,049 --> 00:17:45,750 How satisfied are are 469 00:17:46,434 --> 00:17:47,255 our our associates 470 00:17:48,115 --> 00:17:48,934 in actually 471 00:17:49,315 --> 00:17:52,674 looking at, the tools that we've deployed, and 472 00:17:52,674 --> 00:17:54,835 is it actually helping them to do their 473 00:17:54,835 --> 00:17:58,035 job better? So you we monitor a lot 474 00:17:58,035 --> 00:17:58,695 of KPIs 475 00:17:59,075 --> 00:17:59,575 across 476 00:18:00,169 --> 00:18:01,390 the entire ecosystem. 477 00:18:01,929 --> 00:18:03,789 And, again, I come back to 478 00:18:04,169 --> 00:18:04,669 these 479 00:18:05,289 --> 00:18:07,309 these the solutions that we're deploying, 480 00:18:08,009 --> 00:18:09,869 the KPIs that we're monitoring 481 00:18:10,329 --> 00:18:13,230 actually yield to a better and more simplified 482 00:18:13,369 --> 00:18:13,869 experience 483 00:18:14,224 --> 00:18:15,904 for all. And and and I think that's 484 00:18:15,904 --> 00:18:17,845 a way to actually simplify health care, 485 00:18:18,865 --> 00:18:19,605 for everyone. 486 00:18:20,785 --> 00:18:22,945 Yeah. It really seems that for all these 487 00:18:22,945 --> 00:18:25,664 uses, you're you're mentioning that Evans goes to 488 00:18:25,664 --> 00:18:28,279 great lengths to get member feedback. How crucial 489 00:18:28,279 --> 00:18:30,759 is that feedback kinda during the early stages 490 00:18:30,759 --> 00:18:31,819 of some of these rollouts? 491 00:18:32,679 --> 00:18:34,460 It's actually Scott, it's 492 00:18:35,720 --> 00:18:37,799 I'm glad you brought that point up because 493 00:18:37,799 --> 00:18:39,659 it's actually extremely important 494 00:18:40,119 --> 00:18:42,474 to get this feedback, but both from members 495 00:18:42,474 --> 00:18:43,214 and providers. 496 00:18:43,755 --> 00:18:44,575 And we 497 00:18:45,115 --> 00:18:48,075 we actually study so even before rolling out 498 00:18:48,075 --> 00:18:49,775 a solution like the virtual assistant, 499 00:18:50,154 --> 00:18:52,875 we've done a lot of market analysis. So 500 00:18:52,875 --> 00:18:55,269 we brought in members to say, will this 501 00:18:55,269 --> 00:18:56,009 even be 502 00:18:56,470 --> 00:18:57,690 valuable to you? 503 00:18:58,070 --> 00:18:59,369 Is this going to simplify 504 00:19:00,230 --> 00:19:02,309 in how you kind of look at your 505 00:19:02,309 --> 00:19:04,650 benefits, how you can connect your provider? 506 00:19:04,950 --> 00:19:07,109 So we do a lot of market research 507 00:19:07,109 --> 00:19:07,849 and assessment. 508 00:19:08,494 --> 00:19:10,414 And then when we deploy it, we get 509 00:19:10,414 --> 00:19:11,555 a lot of this feedback, 510 00:19:12,015 --> 00:19:15,134 and our teams are really obsessed about member 511 00:19:15,134 --> 00:19:16,674 feedback and provider feedback. 512 00:19:16,975 --> 00:19:18,595 Because we wanna make sure 513 00:19:18,975 --> 00:19:22,195 that we are constantly listening to that feedback 514 00:19:22,579 --> 00:19:24,200 to simplify these experiences. 515 00:19:24,740 --> 00:19:27,140 Because that's the way we actually learn, and 516 00:19:27,140 --> 00:19:29,079 we continue to evolve the solution 517 00:19:29,380 --> 00:19:31,700 because we're in service of them. And and 518 00:19:31,700 --> 00:19:33,960 we wanna make that, very easy 519 00:19:34,259 --> 00:19:35,000 and simple 520 00:19:35,424 --> 00:19:37,585 for our members and providers. And so this 521 00:19:37,585 --> 00:19:38,085 feedbacks, 522 00:19:38,545 --> 00:19:41,184 loop is actually very important for us in 523 00:19:41,184 --> 00:19:42,644 how we evolve our solutions. 524 00:19:43,984 --> 00:19:45,605 Last question I have for you, Radhnikar. 525 00:19:46,224 --> 00:19:48,065 What do you think 2026 526 00:19:48,065 --> 00:19:50,724 will bring for digital technology and AI? 527 00:19:51,380 --> 00:19:51,880 Yeah. 528 00:19:52,420 --> 00:19:53,400 I I think, 529 00:19:54,259 --> 00:19:56,279 you know, as AI is evolving, 530 00:19:57,140 --> 00:19:58,279 as we are evolving 531 00:19:58,580 --> 00:20:01,320 kind of our thinking about data, 532 00:20:01,860 --> 00:20:02,600 our understanding, 533 00:20:04,110 --> 00:20:06,744 better about our members and providers and how 534 00:20:06,744 --> 00:20:09,164 they interact with the tools and capabilities 535 00:20:09,465 --> 00:20:10,445 that we're deploying, 536 00:20:10,985 --> 00:20:12,924 we see tremendous amount of potential 537 00:20:13,625 --> 00:20:14,765 to actually 538 00:20:15,144 --> 00:20:15,644 streamline 539 00:20:15,945 --> 00:20:16,684 and create 540 00:20:17,065 --> 00:20:17,965 more personal, 541 00:20:18,400 --> 00:20:21,440 a more connected and seamless experiences for our 542 00:20:21,440 --> 00:20:21,940 members. 543 00:20:22,400 --> 00:20:23,140 We also, 544 00:20:24,799 --> 00:20:26,420 kind of are bullish on 545 00:20:26,880 --> 00:20:28,259 simplifying the interoperability 546 00:20:28,640 --> 00:20:30,720 with our providers so that we can empower 547 00:20:30,720 --> 00:20:31,460 care providers 548 00:20:31,839 --> 00:20:33,299 to improve health outcomes. 549 00:20:33,835 --> 00:20:36,575 And finally, we really want to accelerate 550 00:20:37,275 --> 00:20:38,654 a lot of the tool adoption 551 00:20:39,035 --> 00:20:39,535 internally 552 00:20:39,914 --> 00:20:41,535 and the capabilities internally 553 00:20:41,994 --> 00:20:44,255 so that we can simplify our workflows 554 00:20:44,715 --> 00:20:46,795 so that we can better serve our members 555 00:20:46,795 --> 00:20:49,329 and providers. And I I'm actually very excited 556 00:20:49,329 --> 00:20:51,809 about that because we've laid the foundation, and 557 00:20:51,809 --> 00:20:53,349 now we believe we can accelerate 558 00:20:53,970 --> 00:20:54,710 this transformation. 559 00:20:56,130 --> 00:20:58,470 A 100%. It's so great to hear how 560 00:20:58,609 --> 00:21:00,565 Elevent is kind of, you know, stepping on 561 00:21:00,565 --> 00:21:02,884 a gas in terms of AI and leveraging 562 00:21:02,884 --> 00:21:04,825 it for the member experience and then also 563 00:21:05,045 --> 00:21:08,085 constantly gauging that member feedback, which is crucial. 564 00:21:08,085 --> 00:21:10,164 The Randekar, this is a great conversation. You 565 00:21:10,164 --> 00:21:11,365 know, we're ending now. I feel like I 566 00:21:11,365 --> 00:21:12,725 could've talked to you for two hours about 567 00:21:12,725 --> 00:21:14,950 this. So thanks so much, and really looking 568 00:21:14,950 --> 00:21:16,950 forward to working with you again soon. Thank 569 00:21:16,950 --> 00:21:18,630 you so much, Scott. It it was really 570 00:21:18,630 --> 00:21:21,429 nice. Really excited about, our journey, and I'm 571 00:21:21,429 --> 00:21:23,829 glad, to share that with everyone. Thanks a 572 00:21:23,829 --> 00:21:24,329 lot.