Transcrição completa de The Shocking Truth About Navigating The Healthcare System: Insurance Appeals, AI, and Patient Advocacy

where you want to be taken if you know you have to go to a hospital. 00:05:01.279 --> 00:05:06.399 So having all of this in one place, I think is very helpful. 00:05:06.560 --> 00:05:24.959 Also having your insurance information, understanding what your insurance covers, and especially for older people, when you have multiple medical conditions, you're on multiple medications, knowing what your insurance covers is critically, critically important. 00:05:25.360 --> 00:05:30.160 The second thing is being prepared specifically for your appointments. 00:05:30.399 --> 00:05:42.160 I think oftentimes we all go into our appointments kind of expecting the physician to tell you what to do, but it's really a partnership. 00:05:42.319 --> 00:05:48.720 And so I think knowing what you want to know before you go into that appointment is really important. 00:05:48.879 --> 00:05:59.920 So writing down your questions and not being afraid to ask them because physicians really, really want their uh patients to be full participants in their care. 00:06:00.399 --> 00:06:10.800 I think another thing that is really important and is on people's minds, and I've seen this covered a lot in the media recently, is about appealing decisions of your insurance company. 00:06:11.120 --> 00:06:18.399 So I even have a letter in there, a sample letter for appealing a prior authorization, for example. 00:06:18.560 --> 00:06:36.720 And where this impacts older Americans is if you're on multiple medications and if your insurance company decides one year that your medication is no longer on their formulary and there's multiple tiers in that regard, you may be responsible for having to pay for that medication. 00:06:37.279 --> 00:06:41.920 And what the insurance company may tell you is, well, try another medication. 00:06:42.160 --> 00:06:49.199 Well, if you've been on that medication for a good number of years and it's working for you, you don't want to change that medication. 00:06:49.360 --> 00:06:51.199 So you should be appealing that. 00:06:51.439 --> 00:07:00.480 And I think what people need to know is that most people do not appeal these decisions, but when they do, most people prevail. 00:07:00.720 --> 00:07:02.560 So I think that's really important. 00:07:02.959 --> 00:07:10.480 I think the other thing that I talk about in my book, as well as talking about Medicare, because that's something that can be confusing for people. 00:07:10.639 --> 00:07:14.000 I learned a lot in writing that particular chapter. 00:07:14.160 --> 00:07:19.759 But I think it's also helping your family and your caregivers be prepared for your end of life. 00:07:19.839 --> 00:07:35.519 And this is where you probably know a lot more about this than I do, but having your proxy forms and your wishes known, having that documented and amongst your papers is critically important. 00:07:35.839 --> 00:07:40.639 I know we as physicians ask for these these uh documents. 00:07:40.879 --> 00:07:46.319 And so it's important that these are done well before you actually need them. 00:07:46.399 --> 00:07:50.560 So those are some of the things that I talk about, amongst other things. 00:07:51.040 --> 00:07:53.839 You know, I think that is all really excellent advice. 00:07:54.000 --> 00:08:04.000 And I I think, you know, I it there is sort of a disconnect sometimes, I think, than when we go to doctors, especially for people my age and older, I'd say. 00:08:04.240 --> 00:08:15.199 I don't, you know, I think sometimes younger people are more they they're less impressed by a white coat, they're less impressed by a person in a suit and tie or things like that. 00:08:15.360 --> 00:08:22.160 But especially, you know, some of the older generations, the expectation was you go into the doctor and you just listen. 00:08:22.319 --> 00:08:29.680 And the truth is the doctor can't really fully do their job without you fully informing them. 00:08:29.839 --> 00:08:36.320 And so there's going to be things in your medical history, there's things that are symptoms of yours. 00:08:36.559 --> 00:08:43.840 They're not going to be able to see that big picture unless you speak up, unless you kind of become that advocate. 00:08:44.080 --> 00:08:49.039 And I have a couple of questions about what you said, but I want to focus on this one first. 00:08:49.279 --> 00:08:59.519 One of the things that I've noticed is that if you have that, and I'm curious if you have a better thought on the way to relay that information. 00:09:00.000 --> 00:09:09.679 So a lot of people might just go to a doctor's appointment or even the ER, anything like that, with the expectation that they're just going to answer whatever questions are asked of them. 00:09:09.840 --> 00:09:17.039 But I really like your idea of having, you know, something where you kind of that already has your history written out there. 00:09:17.279 --> 00:09:20.799 Is there a better way that that can be accessible? 00:09:20.960 --> 00:09:26.720 I know sometimes one of the things that we experience is we've got different hospital systems in our area. 00:09:26.960 --> 00:09:35.039 And so if you went to one hospital system for the ER, but your doctor's not part of that hospital system, they're not going to have those records. 00:09:35.200 --> 00:09:48.879 And so is there a way that you can make sure that your doctor not only has access to, but really has the medical background and your medical history in a usable form? 00:09:49.039 --> 00:10:10.000 Because if you just come in with a three-ring binder with 100 pages in it, and you know, you're having a 10-minute meeting with your doctor, or even a you know, 20-minute meeting with your doctor, they're not going, that's not, in my view, I think that's going to be very difficult for the doctor to look at that three-ring binder with 100 pages in it and and really have a uh accurate history. 00:10:10.159 --> 00:10:19.039 So is there a better way to make sure that we're sharing that information and

advocating, you know, with our medical providers? 00:10:19.440 --> 00:10:22.639 Yeah, you asked two really good questions. 00:10:23.120 --> 00:10:29.759 The first one is about older generations maybe feeling like they need to be passive. 00:10:29.919 --> 00:10:34.320 And to be a good patient, they should just listen to what their doctor has to say. 00:10:34.399 --> 00:10:39.440 And I'm reminded about my father when he was very sick. 00:10:39.679 --> 00:10:41.600 And I found him a physician. 00:10:41.679 --> 00:10:44.399 And after the first appointment, my father said, He's great. 00:10:44.480 --> 00:10:45.200 I like him. 00:10:45.360 --> 00:10:47.600 And I said, Well, you didn't ask him anything. 00:10:47.759 --> 00:10:49.360 How do you know you like him? 00:10:49.519 --> 00:10:51.679 Whatever he says, he seems so smart. 00:10:51.840 --> 00:10:52.879 I'm like, oh God. 00:10:53.039 --> 00:10:54.159 So you're exactly right. 00:10:54.320 --> 00:10:55.360 That is pretty common. 00:10:55.519 --> 00:10:57.440 And I think people are afraid too. 00:10:57.519 --> 00:10:59.440 There's a power dynamic, right? 00:10:59.759 --> 00:11:04.000 The physician who supposedly knows everything, and then the patient who doesn't. 00:11:04.240 --> 00:11:09.600 People might be afraid of upsetting their physician knowing it's a short amount of time. 00:11:09.840 --> 00:11:12.399 They may be embarrassed or whatever those feelings are. 00:11:12.559 --> 00:11:18.720 And I think this is one place where AI might help level that playing field for people. 00:11:18.799 --> 00:11:20.240 So that's one thought. 00:11:20.480 --> 00:11:28.240 The other thought about 300-page binder, I thought you were describing somebody's medical record coming out of the electronic health record. 00:11:28.639 --> 00:11:36.720 I had to go through my mother's phone, her record, and seriously, it was like 3,000 pages of gobbledygook and garbage. 00:11:37.039 --> 00:11:45.039 So what I encourage patients to do is follow uh the patients that I was inspired by in writing this book. 00:11:45.200 --> 00:11:56.240 When I was earlier in my career, the parents of children with uh medical complexity and special health care needs, they knew everything about their kid, right? 00:11:56.320 --> 00:11:59.919 They knew when their last feed was, they knew their medication dosages. 00:12:00.080 --> 00:12:08.480 It's different in adult world, but in child world where the stakes are really high, they do come with their white binders. 00:12:08.960 --> 00:12:12.399 And as an emergency physician, I really appreciated that. 00:12:12.639 --> 00:12:25.440 But I think one of the things that all of us can do is to have in one place not extraneous information, but that real core information that's important for your doctor to know. 00:12:25.600 --> 00:12:46.320 All of your medical diagnoses, all your past surgeries on any allergies or problems that you've had with anesthesia, your uh medical insurance information and a copy of your card, any medications that you're on with the doses, where you get your medications from. 00:12:46.639 --> 00:12:51.759 Just the core information like that is so very useful to your physician. 00:12:52.000 --> 00:12:59.039 And in fact, I created uh for my website on under resource, and actually I have it sitting right here. 00:12:59.279 --> 00:13:17.679 I created a workbook for patients that has and I'm I'm gonna make it fillable as soon as I uh am able to do that, but it has what are the most important pieces of information that you need to bring to your physician, right? 00:13:17.919 --> 00:13:25.360 You bring up a third issue, actually, that is really critical for people to understand that I think most patients don't. 00:13:25.440 --> 00:13:32.080 I I remember once I was working in the ER and a father was transported from another hospital. 00:13:33.039 --> 00:13:36.000 His him, his son was transported from another hospital. 00:13:36.320 --> 00:13:41.200 And I began to ask questions about what was going on, what happened. 00:13:41.279 --> 00:13:43.919 And the father said, Well, isn't it in that computer? 00:13:44.000 --> 00:13:45.360 Isn't all your information there? 00:13:45.519 --> 00:13:50.240 And I had to tell him, I said, I'm really sorry, but you were seen in another healthcare system. 00:13:50.480 --> 00:13:52.240 I don't have that information. 00:13:52.480 --> 00:13:56.159 And so that's why you get asked these questions over and over again. 00:13:56.240 --> 00:13:59.759 And it is frustrating and tiresome for patients. 00:14:00.080 --> 00:14:19.759 But right now, even though our federal government has tried and tried, there really isn't the full connectivity and coordination of medical records that would make care seamless for patients when they go from one practitioner to another practitioner. 00:14:20.000 --> 00:14:26.639 So you've really picked up on some of the really big frustrations for physicians as well. 00:14:27.519 --> 00:14:27.840 Right. 00:14:28.080 --> 00:14:28.480 Yeah. 00:14:28.639 --> 00:14:35.120 And I I wanted to mention you had brought up Medicare and insurance, and you talked about appeals. 00:14:35.279 --> 00:14:46.320 And I I want to make sure that everybody understands when we use these words like appeal, it seems some people might be just intimidated by that very thought of an appeal. 00:14:46.559 --> 00:14:53.840 And what everybody needs to understand is appealing in the with regard to Medicare in particular is not that difficult. 00:14:54.000 --> 00:15:01.519 Sometimes that's a phone call, you know, sometimes there might be a form, but and in Medicare in particular, it's usually a phone call. 00:15:01.600 --> 00:15:02.480 And that's the appeal. 00:15:02.639 --> 00:15:07.279 You don't need a lawyer, you don't need to go through and write a brief, you don't have to go to a judge. 00:15:07.440 --> 00:15:11.519 It is a phone call, and then they'll investigate it further. 00:15:11.759 --> 00:15:15.440 And so

Medicare has a lot of patient rights in it. 00:15:15.679 --> 00:15:29.759 And many of us, or many of the folks that are on Medicare, are moving more towards this Medicare advantage, where it's really the insurance company has more of the authority over what traditional Medicare. 00:15:29.919 --> 00:15:36.639 But certainly, if you have traditional Medicare and a supplement, you have a lot of rights to appeal that aren't, and it's not that difficult. 00:15:36.879 --> 00:15:43.759 And you still have those rights if you have a Medicare advantage, it's just might be a little bit different how depending on how the insurance company handles that. 00:15:43.840 --> 00:15:47.919 So I wanted to clarify that for folks because what you say is is great. 00:15:48.000 --> 00:15:50.159 And don't be intimidated by the process. 00:15:50.240 --> 00:15:51.679 Sometimes it's easier than you think. 00:15:52.000 --> 00:15:54.399 Yeah, and I think that's a really, really important point. 00:15:54.639 --> 00:15:59.679 And I do have a chapter on knowing your rights, uh, which talks about exactly this. 00:15:59.919 --> 00:16:04.080 I was helping somebody appeal an insurance decision on their medication. 00:16:04.159 --> 00:16:05.759 And I'm very tech savvy. 00:16:05.840 --> 00:16:07.440 So I wanted to send an email. 00:16:07.600 --> 00:16:08.960 They don't make it that easy. 00:16:09.039 --> 00:16:11.679 You have to either make a phone call or write a letter. 00:16:11.840 --> 00:16:18.159 So I mean, I know that sounds easy, but it it could be intimidating, I think, for a lot of people. 00:16:18.320 --> 00:16:21.840 I was just ticked off that I had to write a letter and I couldn't write an email. 00:16:22.000 --> 00:16:27.759 But I so, you know, hopefully one day they'll make it easier for patients to do that appeal. 00:16:27.919 --> 00:16:38.399 But I encourage everyone that if if care is denied, a medication is denied, and you believe that you're entitled to that care and you need it, there's several things you can do. 00:16:38.559 --> 00:16:44.559 One of them is to engage a physician because your physician wants you to be on that medication or have that treatment. 00:16:44.639 --> 00:16:49.120 I mean, there's a reason why your physician prescribed that treatment to you. 00:16:49.279 --> 00:16:52.480 Um and they're they're set up to help you with that. 00:16:52.639 --> 00:16:57.440 It could be a wrong code, it could be um that didn't have the right amount of justification. 00:16:57.600 --> 00:17:06.240 In some instances, they may need what's called a peer-to-peer discussion about the care that your doctor's prescribing. 00:17:06.480 --> 00:17:07.359 So that's number one. 00:17:07.519 --> 00:17:17.440 Number two, if it's something simpler, you've been on a medication for a really long time and they want you to go to some other generic drug and you're doing well on that medication. 00:17:17.599 --> 00:17:23.200 I would definitely appeal that and write that letter and say exactly that. 00:17:23.359 --> 00:17:28.400 And you do have the right to write that letter and appeal that decision. 00:17:28.880 --> 00:17:36.559 So you mentioned something else that I think I we should dive into, and you have a particular expertise about, and you mentioned AI. 00:17:36.799 --> 00:17:46.480 So I wanted to point out that you are also editor and lead author of digital health AI and generative AI in healthcare, which I find fascinating. 00:17:46.640 --> 00:17:57.920 So tell us a lot of, you know, at this stage, there's going to be people that have a really pretty good understanding of AI and how it's everywhere now. 00:17:58.079 --> 00:18:02.079 And there's going to be folks that have avoided it as much as they possibly can. 00:18:02.319 --> 00:18:12.640 So if you can help us start out on just a real simple level of how AI is affecting healthcare, and and then we can maybe dive into it a little bit deeper. 00:18:12.720 --> 00:18:19.359 But just for those folks that don't really like thinking about AI, talk to us a little bit just on a very basic level. 00:18:19.839 --> 00:18:20.160 Sure. 00:18:20.319 --> 00:18:20.799 Thank you. 00:18:20.960 --> 00:18:27.200 And actually, that book is a textbook, and it was supposed to be a it is a primer for clinicians. 00:18:27.359 --> 00:18:39.119 And I wrote that book because when open AI came out at the end of 2022, people started making predictions that AI was going to replace your physicians. 00:18:39.200 --> 00:18:40.799 And I'm like, oh no, it's not. 00:18:40.960 --> 00:18:42.960 But so I wrote that textbook. 00:18:43.039 --> 00:18:43.920 It was for clinicians. 00:18:44.079 --> 00:18:49.279 I do have another book that I'm in the middle of completing right now, that is for patients. 00:18:49.599 --> 00:19:02.079 And I think it's really important for all of us to understand how AI is being used within their care, as well as how patients could use AI to help with their care. 00:19:02.559 --> 00:19:19.200 So right now, some of the big areas that patients can expect their physicians to be using AI include, for example, as support, for example, to radiologists and helping to read X-rays, CT scans. 00:19:19.279 --> 00:19:21.839 Um AI is helping with that. 00:19:22.079 --> 00:19:24.799 There's uh ambient AI scribes. 00:19:25.279 --> 00:19:41.680 A good number of office physicians are using these where a physician just talks, goes through the appointment with the patient, and then AI automatically transcribes it, puts the note in the right order. 00:19:41.839 --> 00:19:46.880 And that's very helpful because for many years we used electronic health records. 00:19:47.039 --> 00:19:48.720 Our face was in the keyboard. 00:19:48.960 --> 00:19:52.559 We were typing away and not paying attention with the patient. 00:19:52.640 --> 00:19:59.440 And I know that could be very frustrating for a patient to that their doctor is not looking them at them while they're talking to them. 00:19:59.599 --> 00:20:03.839

So that's something where AI is being used and a patient would notice. 00:20:04.000 --> 00:20:07.119 It's also being used for clinical decision support. 00:20:07.359 --> 00:20:22.240 So it helps a physician, for example, if they come across a patient who has a certain condition, they could find out very quickly what the guidelines say about how they should work up that patient, how they should be treated. 00:20:22.559 --> 00:20:27.440 I think too, and for me, this is kind of a really important area. 00:20:27.519 --> 00:20:35.920 I think there's a lot of nonsense, I'll call it nonsense, a lot of hoops that you have to jump through administratively in healthcare, right? 00:20:36.240 --> 00:20:47.359 In making an appointment for physicians and having to do the coding, having full staffs of people to send to insurance companies, you know, they call that RCM. 00:20:47.519 --> 00:20:56.160 So having to do all of that work uh just to uh take care of that patient and get compensated for it. 00:20:56.480 --> 00:21:04.160 There's a lot of really good innovation, AI-powered tools that are being used in that regard. 00:21:04.400 --> 00:21:06.400 It's also being used for good. 00:21:06.480 --> 00:21:12.960 Um, I advise a company that uh uses AI-powered uh medical translation. 00:21:13.119 --> 00:21:22.720 And it's really, I've used it and it's really super helpful for people that where English isn't their first language. 00:21:23.200 --> 00:21:24.799 So these are some of the things. 00:21:24.960 --> 00:21:27.200 Now, how are patients using it? 00:21:27.359 --> 00:21:37.440 You know, a lot of people are using open AI, Claude, Perplexity, all these other um what we call LLMs, large language models, using them to ask their questions. 00:21:37.599 --> 00:21:46.079 They get a test result back, and you know, they immediately go to one of these LLMs to find out what that test result means. 00:21:46.480 --> 00:21:51.440 And one of the things that I encourage people when they do that, I'm I'm not against them doing that. 00:21:51.519 --> 00:21:55.680 It sort of reminds me early in my career when people started Googling all their stuff. 00:21:56.160 --> 00:21:58.559 Um it's kind of similar, but on steroids. 00:21:58.799 --> 00:21:59.759 I I think it's really. 00:22:00.400 --> 00:22:05.599 important to then turn around and talk to your physician about what you found out. 00:22:06.000 --> 00:22:06.400 Right. 00:22:07.039 --> 00:22:09.680 Because AI, they don't have any context. 00:22:09.920 --> 00:22:11.440 AI doesn't know you. 00:22:11.839 --> 00:22:18.400 And I and I tell patients, please, please, please do not put your medical records into any LLM. 00:22:18.799 --> 00:22:21.920 So I I have a book where I talk about a lot of these things. 00:22:22.079 --> 00:22:23.759 And I think it could be exciting. 00:22:23.839 --> 00:22:35.119 We're kind of in the Wild West, but when we get to a point where we have good guardrails, good guidance for how to use it, I think it'll help level the playing field for patients. 00:22:35.279 --> 00:22:38.720 And then I think it'll help them become more engaged in their care. 00:22:38.880 --> 00:22:41.920 So that's what I hope AI will do for patients. 00:22:42.400 --> 00:22:44.640 Well I think there's a couple things that you mentioned there. 00:22:44.799 --> 00:22:53.440 And one of yes I imagine people are going to go to open AI and and others with regard to their symptoms, with regard to their results. 00:22:53.599 --> 00:22:56.400 Like you say, you want to make sure you maintain your privacy. 00:22:56.559 --> 00:22:58.960 There's a real importance of medical privacy. 00:22:59.119 --> 00:23:13.920 And there are ways that you can do that within especially if you have a paid subscription to one of these services, you can you can put up sort of a firewall there that it's not populating the rest of the known universal knowledge with what you put in there. 00:23:14.079 --> 00:23:21.839 And so but the thing that I think is most useful about that is having AI help you phrase your questions for the doctor. 00:23:22.000 --> 00:23:42.400 I know personally when I go to the doctor I'm stumbling over my words trying to explain my symptoms and if it can kind of clarify my thought process or clarify the words so that I can say it in the right way, it's going to be a much more effective consultation with the doctor if I go in and are able to better articulate my questions. 00:23:42.559 --> 00:23:55.759 So I think that's one of the most useful things with AI and and you know you don't want to take advice from AI, medical or legal advice from AI, but it can help you form the questions to when you are talking to your professionals. 00:23:56.000 --> 00:23:58.400 Yeah no that's exactly right and that's a really good point. 00:23:58.720 --> 00:24:00.720 We use a lot of jargon in healthcare. 00:24:00.880 --> 00:24:08.480 And so I think it really could it could be a partner in helping patients get prepared for their appointment. 00:24:08.640 --> 00:24:10.079 So that's a really good point. 00:24:10.559 --> 00:24:19.039 The other thing that you know we talked about earlier is being able to make sure the doctor has your medical history. 00:24:19.279 --> 00:24:41.599 And I'm not sure if we're there yet but I could see AI being very helpful in that process to be able to you know streamline and share your information more effectively by some of the things that AI can do to appropriately summarize and kind of weed through to get to sort of the critical issues faster. 00:24:41.839 --> 00:25:00.799 So I don't know that we're there yet but that issue about you know being able to convey your previous history and and your previous medical appointments to a new doctor or even your existing doctor that I believe in in the near future AI might be very helpful in that regard. 00:25:01.119 --> 00:25:09.680 Yeah and I think what I would add to that is right now I would not put any of your records or medical information

into an LLM. 00:25:09.920 --> 00:25:12.880 I to me I it doesn't matter what they say. 00:25:13.440 --> 00:25:15.920 I think once it's there they have it. 00:25:16.160 --> 00:25:17.680 So I would not do that. 00:25:17.839 --> 00:25:27.680 What you can do and what AI is doing for physicians especially AI ambient scribes it a lot of them are providing summaries. 00:25:27.920 --> 00:25:31.599 And those summaries a patient can ask for them. 00:25:31.839 --> 00:25:46.559 And I think it would be great you know and maybe physicians are doing this now sending those summaries to patients so that they have them that to me at least for the time being is probably the better way to get those summaries. 00:25:46.960 --> 00:26:21.119 Excellent so you know I think we all see this in so many different areas can you give us an idea on how AI is likely transforming healthcare and what that actually means on a patient level so whether folks are comfortable with it or or not how that the you know this is how how the patients are going to see this and interact with it and how things might change in in the future because of this new technology advancing technology let's say yeah no that's a great question. 00:26:21.279 --> 00:26:27.519 And I think we throw around the term transformation which is kind of a vague term like what does it really mean? 00:26:27.759 --> 00:26:33.119 And to me transformation is better outcomes for patients. 00:26:33.440 --> 00:26:37.759 That to me you know everything else is not necessarily transformative. 00:26:38.000 --> 00:26:48.799 And I think one of the really exciting areas for patients especially older adults is uh the movement of hospital at home. 00:26:49.200 --> 00:26:53.599 And I see that getting better and better over the next several years. 00:26:53.920 --> 00:26:56.319 Currently we can do remote monitoring. 00:26:56.640 --> 00:26:57.599 So we do that. 00:26:57.759 --> 00:27:10.000 Right now hospital at home includes some remote monitoring but also having nurses for example visit the home there's a lot of things that could be done at home that don't need to be done in the hospital. 00:27:10.160 --> 00:27:30.079 So use as an example an older person who has CHF heart failure they may have a mild episode do they need to be admitted to the hospital for that particular episode probably not because what they would be doing would be monitoring weight monitoring blood pressure doing lab studies. 00:27:30.240 --> 00:27:40.480 All those things can be done at home imagine having a bed that could that could detect changes in weight right blood pressure monitoring. 00:27:40.559 --> 00:27:52.240 I mean we have rings now and smart devices that that do track certain heart rhythms and some of those are FDA approved that could track blood pressure. 00:27:52.480 --> 00:28:00.640 So I think you know in the future a lot of the stuff that's being done in the hospital is going to be done at home. 00:28:00.720 --> 00:28:08.000 And I think for older Americans that will be a transformational use of innovation. 00:28:08.559 --> 00:28:22.880 I think too you know for physicians I think what would be transformational for us would be reducing the complexity of all the hoops we have to jump through in order to provide that care to patients. 00:28:23.119 --> 00:28:51.039 And I think we're starting to see that now all the way from the start of a patient encounter where they're making an appointment, making that easier doing referral referral management can be done through these new tools doing payment and all the other pieces of the administration of a practice or hospital will make care more efficient and will maybe make a small dent in the cost of care. 00:28:51.359 --> 00:28:54.720 So these are the things that we'll probably see in the next five years. 00:28:55.119 --> 00:29:11.680 I'm hoping that we rein in some of the excessive promises you know such as AI is going to replace doctors and I don't think it's going to do that, but I think it's going to make delivering care easier for physicians to do that. 00:29:11.839 --> 00:29:22.160 And I think it will help improve outcomes if the appropriate guardrails and and guidance is put forward for these tools. 00:29:23.680 --> 00:29:39.039 My nephew's a professor at Notre Dame and we are having this discussion and he was saying that there's the healthcare systems are now frequently hiring philosophers in from Notre Dame and you know one of the groups that he's been working with as part of this. 00:29:39.200 --> 00:29:55.680 In other words ethicists and philosophers because like you say that there has to be these guardrails there has to be we've got to think about this beyond but look how this what this can do and look at what this technology can achieve there's got to be ethics as part of it. 00:29:55.839 --> 00:30:11.599 There's got to be a bigger philosophy around it and I just thought that was fascinating that these healthcare systems are are in you know actively engaging in the discussion you just had with regard to making sure we have those proper guardrails and things like that. 00:30:11.839 --> 00:30:13.839 So there's there's some danger there. 00:30:13.920 --> 00:30:24.559 That's actually the next question I have and we've kind of you know skated around it a little bit what are any risks that you see that this could affect especially older adults in a negative way? 00:30:25.039 --> 00:30:25.519 Sure. 00:30:25.759 --> 00:30:34.079 I think you know while AI shows a lot of promise there are risks as you say and I think they fall into multiple categories. 00:30:34.240 --> 00:30:43.839 One of them is access and so certain to these tools that other populations may not. 00:30:44.160 --> 00:30:54.640 I think too that there's biases in these tools the LLMs the large language models learn on vast troves of data. 00:30:54.960 --> 00:31:12.640 If you are a smaller population or a population that does not have its data in these systems then there could be

what's called bias in the system and and when it gives answers to questions it will not consider certain populations. 00:31:13.119 --> 00:31:29.680 It also can overgeneralize and where that could be a problem for an older population is there's a difference between a 68 year old who's healthy and active versus a 92 year old who's frail and may have cognitive decline and so on. 00:31:29.920 --> 00:31:50.640 So in considering treatments for those unless that LLM that's providing that information was trained on enough people across that spectrum you're not going to necessarily get uh get the best answers there'll be biases inherent in that system. 00:31:50.880 --> 00:32:40.400 I think I think to you know we talk about two physicians becoming de-skilled that's something that we need to monitor and be cognizant of you know it could be very easy uh physicians use something called open evidence actually they use chat GPT too but using these LLMs there there are probably really good ways to use them in supporting the care that you give but I think if you become overreliant, you know, a lot of I know what I did as an emergency physician relied on my experience in the thousands of patients that I saw and we say oh my gut told me well it's more than your gut it's something that you've seen this is pattern recognition this ability to connect with the patient those things are lost in AI. 00:32:40.480 --> 00:32:44.480 The AI doesn't connect with a patient it doesn't know context. 00:32:44.640 --> 00:32:56.559 It may not understand that you know it may send up a signal that an older person is not taking their medication but it doesn't understand the context that maybe they lost their health insurance maybe they can't afford it. 00:32:56.799 --> 00:33:01.759 Maybe they're taking care of a family member and so they're neglecting their own care. 00:33:01.920 --> 00:33:05.039 It doesn't get all that context in there. 00:33:05.200 --> 00:33:10.720 And so I think we can't become too overly reliant on technology. 00:33:10.960 --> 00:33:41.279 Technology we don't live for the technology the technology is there to support to support us so I was I've been reading a book recently that kind of discussed this concept that and it's the position of the book is that some of the greatest innovations in the world were the result of either a mistake and then exploring that mistake or just thinking outside the box think not thinking within the protocols you know and that's the what AI can't do. 00:33:41.440 --> 00:33:52.720 AI can't you know the AI can create duplicate art but it can't create new art where you know no one would have thought of using this texture or this color or whatever. 00:33:52.880 --> 00:33:55.519 It's out of the norm it's out of the protocols. 00:33:55.599 --> 00:34:37.280 And some of the greatest innovations in the world are because of thinking outside of the way that is the standard way of thinking and so I thought that was a fascinating concept and I think

this is true you know for folks that are worried about AI and they're losing you know think they're going to lose their jobs over this I think the key on this is that there is it has its limitations that it can't it can't think outside of the those protocols that it it has been fed and some of the most important decisions in healthcare sometimes are are more nuanced than well you have a 62 year old that's experiencing these things. 00:34:37.440 --> 00:34:45.199 Well even you mentioned the difference between a 62 year old and 92 year old even two 62 year olds might be drastically different. 00:34:45.679 --> 00:34:58.000 It's a 62 year old that is running marathons and you can and and you know you can see a 62 year old that has not taken care of themselves and both of them have differ you know risks. 00:34:58.159 --> 00:35:01.679 And so this is one of those things that I think you're you're 100% right. 00:35:01.760 --> 00:35:10.079 The nuance is still very important in the medical field, in the legal field being able to you know apply appropriately. 00:35:10.480 --> 00:35:25.840 Right humans are so complex and there's so much very variability across humans that I think AI is really good at algorithms and it knows what it knows, but it may not go outside of what it knows. 00:35:25.920 --> 00:35:28.559 So and and I think you need humans to do that. 00:35:28.719 --> 00:35:33.360 So well I really appreciate you taking the time with us today. 00:35:33.519 --> 00:35:50.159 This has been fascinating do you have any key takeaways that you'd like to leave our audience with yeah sure I think I think people need to realize that their health is their own there isn't anybody more invested in that than you are. 00:35:50.320 --> 00:36:02.239 And I think it's really important to become a full participant in your care and you could do that by being prepared for any emergency and being prepared for your appointments. 00:36:03.440 --> 00:36:10.159 Know your health insurance coverage and what your needs are with regard to your healthcare coverage. 00:36:10.400 --> 00:36:26.000 I think it's also important to be prepared for older adults being prepared for the end of life I think doing that for your family and your caregivers is really the biggest gift that you could give them so those are some of the things that I would leave people with. 00:36:26.800 --> 00:36:28.159 Well we've been talking to Dr. 00:36:28.239 --> 00:36:29.679 Terry Adirum Dr. 00:36:29.760 --> 00:36:57.519 Adirum what's the best way to get more information about some of the research you've done some of the books that you've written I set up a website terryadrammd.com that has a lot of this information as well as some resources for people to use I think to my my book The Prepared Patient can also be helpful in that regard and it's available from any bookseller but you could get it from Amazon pretty easily. 00:36:57.920 --> 00:36:58.400 Great. 00:36:58.639 --> 00:37:09.360 Well thank you so much and folks if you've enjoyed this conversation don't forget to subscribe to advice from your advocates and you can find that anywhere that you listen to podcasts. 00:37:09.519 --> 00:37:12.159 So thanks for listening and we'll talk to you next time. 00:37:13.920 --> 00:37:25.360 Thanks for listening to learn more visit manner lawgroup.com thanks for listening to learn more visit manner lawgroup.com