Hello and welcome to the Becker's Healthcare Podcast. My name is Chanel Bunger and today I'm excited to speak with Dr. Anna Allen and we're going to talk about AI in the exam room, patient tools, clinical support and the strategy behind it. And Dr. Allen, thank you so much for joining me today. Thank you very much for having me. Please do feel free to call me Hannah. Perfect. Well, to get us started out, before we jump into anything, could you please introduce yourself and give us an overview of your work in healthcare? Thanks so much for having me. Really excited to be here. So I'm a clinician by background. So I'm a general practitioner based in London. I've been a GP for the past 14 years and I've worked in the digital health space alongside of practicing frontline healthcare in AI and telemedicine and digital health for the past 10 years as well. So loads of learnings, loads of challenges along the way. And I think we're at this real sort of pivotal moment in time where it's so exciting to see this novel technology coming to frontline healthcare and genuinely having an impact in how we practice as clinicians from general practice to hospital medicine, to communities, to emergency care, et cetera. And I guess I've been most excited about how we can kind of use data and use AI to do all of the stuff that either we aren't necessarily trained to do as doctors and the kind of admin burden associated with being a clinician, right? We spend a third of our working week now on admin, which is not patient facing and doesn't benefit the patient. And actually what we want to do is focus on the human side of healthcare. And that's what I think we've got. We've landed ourselves in a place where now we've got the tools and the AI and the technology to allow us to do that, which I'm really excited about. So my current role is as chief medical officer at Heidi for the UK and Europe, where we're seeing a significant kind of adoption and traction over there, but also around the world as well, over in the US, here and in Australia and various other places. Very exciting. Absolutely. And now that we have a sense of your background, there have been a wave of announcements about healthcare AI tools that provide health information and decision support. Can you break down the different categories of these tools and who are they really designed to serve? Yeah, of course. So if we kind of take a step back and we kind of look at the jobs that either clinicians are doing, patients are doing, or hospital systems are doing, there are sort of five or six-ish categories that you can kind of put these things into. So one which has been heavily covered lately is the transcription sort of functionality. So that is converting speech to a templated structured note for me as a clinician, and they're certainly clinician facing. You also have obviously kind of consumer facing patient products that we know of. And then you have things like clinical decision support tools. So, you know, one of the tools that we've recently brought out is an evidence-based tool that allows me to not have multiple tabs open, allows me to look up latest guidelines, latest formularies, latest ways to manage patients, because it's impossible to retain all of that information in our own heads as clinicians when, you know, medical knowledge is doubling every 73 days. So we have the transcription tools, the clinical decision support tools. Then we have the kind of operating system tools that focus on, you know, improving productivity, improving efficiency through the hospitals. Maybe it's, you know, how are we going to increase the throughput of patients through the emergency department, for example. And, you know, the executive boards of hospitals are interested in those kind of systems as well And then we have patient facing tools So they could be a whole host of different sort of layers of intelligence et cetera wearables internet of things that allow patients to have insights into their own health as well And we've seen a whole host of patients using things like chat GPT to support themselves of late. And then kind of lastly, the big bucket is around intelligence, right? Like how do we move from this reactive healthcare system where people get sick and they need to go to the emergency department or people get sick and they end up with me as a GP. But how do we actually stop that influx of patients coming in through that kind of top of funnel, through the front door? And in order to do that, we need to look at the data. And in order to do that, we need to have a system where we can see kind of population level data to provide us with intelligent ways to manage our patients and improve our systems ultimately. That's how I kind of think of those like five or six systems and who they're speaking to. Yeah. And you mentioned consumer facing tools, digging deeper a bit on that, where the quote unquote, Dr. AI tools aimed at patients, what guidance would you offer physicians on how to navigate conversations when patients bring those insights into the exam room? Yeah, that's a really good question. It's something I've been navigating myself in clinic. The one thing I would start with is that if we take a step back, this isn't new, right? It's a new way of bringing it to us. And it's a new sort of format that can be a bit intimidating for some clinicians, but it's not new that patients will bring accessory information with them or things that they've already validated. They already come with an idea of what they think is wrong with them, which is actually quite helpful when you embrace it. You know, since the dawn of time or the dawn of the internet, patients were always Googling their symptoms and bringing those in, which already frames how they're thinking about it, right? So the thing that's changed is the kind of confidence, the quality and the fluency of the information and data they're bringing in now. Now, what that means for us as clinicians is that our role has changed. So we're not necessarily the person who's coming up with the first idea. Perhaps the AI has got it right and it likely has, but how do we explain to the patient that we still need to look at them holistically because AI isn't doing that currently. They haven't got necessarily access to all of their medical information and their latest blood tests and their family history because I've known them for 20 years or whatever that part we still as a human to human need to explain to them and to be able to kind of relate to them on a human level I think once we can kind of explain the wider sort of accountability framework and also the holistic 360 viewpoint that we have of them as a patient as a person then actually you can have a really great conversation using the information as a clinician, but embracing that change in the role, right? Where we have to be able to relate to them on a human level, but often we can be deemed as kind of gatekeepers, right? How do we help them to navigate the information that they've got, which is probably quite accurate, but we need to, part of our role is to educate them in, you know, is that source a good source? Is it a high quality source? If it's not, why is it not? What should they be looking for? And how might that mean that they've got some misinformation there? So that's part of our role now is explaining the quality of the information that they've got already and what that means for them as a person. Absolutely. And on the flip side of that, for AI tools designed for clinicians, what are two or three key factors leaders should evaluate when vetting these tools? Gosh, I think there were quite a lot to be honest. To kind of break it down to two or three I think we need to stop thinking of these tools as standalone tools right I think we need to look at how they fit in with a system Just another tool in the toolbox Exactly Exactly And you end up then, you know, procuring 20 different tools for 20 different purposes. That's 20 different training programs. And what if the adoption's low, you know? So, which inevitably it probably will be. So I think we need to look at how does it fit into your overarching kind of thesis and goal? What does success look like for it? And how does it work on a kind of interoperability level? So what workflows do you have already and how will this fit in and improve those workflows? You know, it's not just a siloed product or siloed tool for individual clinicians. We have to think bigger on this level. Absolutely. And as providers begin adopting these tools, how should they think about them within their broader AI strategy and what does taking a big picture approach really mean in this practice? Yeah, great question. So I think kind of leading on from what we were saying before is not kind of siloed AI. We need to think about what does a world look like where we've got an AI enabled healthcare system, right? And I encourage people to kind of think about the P's. So it's easy to remember. Rather than individual products, think about problems that we're solving. And rather than individual products, let's think about pathways that we're wanting to improve. Because if we're constantly asking what problems are we having, we're thinking on a system level rather than in a siloed individual level. And that's really important when we're thinking about these, you know, AI, bringing AI into existing workflows. And I would say to look for an AI partner rather than a scribe, rather than a clinical decision support. It needs to be, you know, the way that we're approaching things is looking at what problems hospital systems are having and clinicians are having and coming up with innovative solutions that utilize certain product functionality and features that will ultimately benefit the overarching goal of the hospital as a whole. Got it, got it. And now looking ahead, we all know that AI is moving incredibly fast. And with that, how do you see AI reshaping clinical decision support over the next two years? And what's one thing leaders need to do today? Yeah, that's a really good question. I think up until today, it's been an accessory, sort of ancillary to how we're consulting and how we're practicing and how the systems are working. I think we're moving towards it becoming embedded into that system. So what does that embedded AI supported system and AI supported clinician look like that will enable us to practice at the top of our game, at the top of our practice, right? And I believe that in time, patients will only want to see clinicians who are AI augmented because they know they'll have a better, more accurate, more precise, more listened to experience. And so I think for leaders in this field, we need to then be looking at like, where is this going to have the most impact, right? What workflow, which part of my system should I look at to start augmenting with AI first? So a lot of people look at, say, general practice because the high footfall or the emergency department, for example, but look at that on a system and workflow basis. So how can I augment the pathway and improve the pathway for the clinicians, for the hospital system overarching and for the patients ultimately? Because that's how we're going to ultimately result in better patient care, which is what we're all here for. Well Hannah I want to thank you for your time today but before I let you go is there anything else that listeners should know I think the way that we are approaching things at Heidi is in a very agile and innovative way and so I think I would encourage people to look out for that next wave of innovation and in particular how it applies to your hospital system. So I'm thinking things like revenue cycle management for example, we've got some really innovative partnerships and really innovative ways of approaching this that are going to be ultimately benefiting the hospital systems and it's coming very, very soon in the near future. So watch out for that one. Well, that's a great note to end on. I want to thank you, Hannah, Dr. Allen for your time today. And we also want to thank our podcast sponsor, Heidi. You can tune into more podcasts from Becker's Healthcare by visiting our podcast page at beckershospitalreview.com. Hi everyone. This is Chanel with Becker's Healthcare. Thank you so much for tuning into the Becker's Healthcare podcast series. Today, we're going to talk about AI in the exam room, patient tools, clinical support, and the strategy behind it. Joining me for today's discussion is Dr. Hannah Allen, the Chief Medical Officer at Heidi. Hannah, Dr. Allen, thank you so much for being here today. Yeah, I got it. And you mentioned the consumer facing tools or that quote-unquote Dr. AI. Yeah, and you mentioned consumer-facing tools. Digging deeper a bit on that, were the quote-unquote Dr. AI tools aimed at patients? What guidance would you offer physicians on how to navigate conversations when patients bring those insights into the exam room. As providers begin adopting these tools, how do you think they should think about them within their broader AI strategy? Got it, got it. And as providers begin adopting these tools, how should they think about it? Got it, got it. And as providers begin adopting these tools, how should they think about them within their broader AI strategy? And what does taking a big picture approach really mean in practice? And what does, and then, and then what does taking a quote unquote big, and then what does taking a big picture approach really mean in practice? Really mean in practice? got it got it and as providers begin adopting these tools how should they think about them within their broader ai strategy and what does taking a big picture approach really mean in practice now looking ahead how do you see ai reshaping clinical decision support over the next few years and what's one thing leaders need to know now looking ahead how do you see ai reshaping clinical. Now looking ahead, how do you see AI reshaping clinical decision support over the next few years? And what's one thing leaders need to do today? Now that we have a sense of your background, let's jump into the conversation a bit. There have been a wave of announcements about healthcare AI tools that provide health information and decision support. Can you break down the different categories of these tools, and who are they really designed to serve?
