Today in health, it we're going to take a look at generative AI and the patient experience. My name is bill Russell. I'm a former CIO for a 16 hospital system. And create, or this week health set, a channel set events dedicated to transform healthcare. One connection at a time. We want to thank our show sponsors who are investing in developing the next generation of health leaders. Notable service now, enterprise health parlance, certified health and Panda health.

Check them out at this week. health.com/today. All right, this and every new story we talked about, you can find on this week health.com/news, which is our website. We put it out there. If you want to become a contributor, you can check that out at the top of the page as well. All right. One last thing, let's see, share this podcast with a friend or colleague, use it as a foundation for daily or weekly discussions on the topics that are relevant to you and the industry. Little thing, we call mentoring.

They can subscribe wherever you listen to podcasts. All right.

Today, we're looking at a concept, the concept team to my, the forefront of my mind, because there's a new England journal of medicine. Article out there patient portal when patients take AI into their own hands. And so I was looking at that. And then there's also a couple of other articles.

And I tell the story all the time. I become CIO at St. Joe's. I get a phone call from one of the leading cloud providers and they say, Hey, congratulations. You're one of the largest users of our cloud platform. And so I went to the team. I said, Hey, do we have a contract for this? Sure enough, we had no contract for it.

But our physicians, our nurses, our administrators, they were all using this cloud platform. And in the absence of leadership and direction. The, these things will get adopted. And so I think that's going to be my biggest takeaway for this is get in front of this. The patients are already using generative AI. If you can somehow. I mean beyond education, integrated into their experience. It's going to be better if they can go to a trusted source like your health system and have you put the models in place that they can trust with regard to this. Okay. So we know the medical community remains cautious about the implementation of AI in clinical settings. And the author of this article for the new England journal of medicine, Carrie Goldberg argues that these patient led do it yourself approaches to using AI in healthcare should not only be monitored, but integrated into formal medical advice to ensure safe and effective practices.

Okay. So that's along the lines of what I'm saying. Let's get in front of this. Let's figure out how we can talk to our patients that are. I don't know our chronic patients and how they're going to be using it because they are, again, they are using it. I use it. You probably use it with regard to healthcare.

So there's an awful lot of relevance here. Let me go to the world. Economic forum. This is three ways. Generative AI is reshaping the patient experience. And they give three ways. Let's see general AI for patient use. Why now? And oh, here it is. How generative AI will revolutionize patient care health education assistance. Patient triage and co-pilots and disease management. And they they make the case for each one of these, obviously with health education. It's already being used for that.

Hey, I have stomach pain. I have these symptoms and it is providing them feedback. Patient triage and copilots, a given healthcare worker shortage health systems. Need help shifting resources away from non-urgent patients to those who require timely high touch care. Early virtual care assistants have failed. To scale due to lack of open source models for more than 7,000 languages spoken globally and predictive AIS limited grasp. Of nuances across cultural contexts and health system archetypes. With specific and culturally diverse training, large language models can help healthcare providers safety. Safely stratify patients. Entering the healthcare system and route their care accordingly.

So there we go. Patient triage is one of those things. Then in disease management let's see what has have say in their first year of treatment, up to 60% of chronically ill patients. Miss doses. Take the wrong dosage or abandoned treatment altogether costing health systems, hundreds of billions of dollars per year. While traditional algorithms can predict. When a patient is likely to drop off treatment, they struggled to suggest let alone execute effective interventions. I think it's going to be important to put these things in place.

How is AI going to interact with the patient? Because today it is going to be through your website, through your portal. In the future, it will be through voice assistance. And we're seeing that. RO and proliferate, and it could be a voice assistant that a health system puts in the patient's home. I could see that being a very real use case. And then, beyond that, W, we're looking at the confluence of robotics and this. And I think in a decade that's probably reality. And we're looking at. Robots interacting. Using generative AI for its language and its interaction with people and whatever else it uses for navigating. The robotics portion of that.

That's not what this episode's about. So I think we need to. Think as an industry, how are we going to do this? How are we going to integrate this? At these basic levels at the website and portal level, interacting with our patients. He goes on and talks about the need for strong collaboration. He gives four points, build trust through empathy and domain expertise.

We must prioritize fine tuning models on healthcare specific data. And having doctors test and rate responses to improve outputs. Absolutely. Mitigate against bias. It's important to correct for bias inherent in the models. Absolutely. Keep humans in the loop. This is one of the most critical items.

There needs to be escalation points. There needs to be points where it does. Get to a clinician who can assist the patient. And then finally plan to scale across contexts. Given resource constraints, developing more cost-effective means to run large language models. We'll be critical stakeholders must also work to create flexible deployment models. That recognize varying needs across regions. Excuse me. So that's essentially what I wanted to cover.

There's two articles. You can find them out on our website, three ways, gender vis reshaping, the patient experience. And patient portal when patients take AI into their own hands and they are taking it into their own hands and they will continue to take it into their own hands. I think it is really incumbent upon us to be the trusted source for information.

We keep coming back to this. How do we become the trusted source? We abandoned this a long time ago, we allowed Google to become doctor Google. And then we said, Hey, web MD, go out there. This is, good stuff and whatnot. I think it needs to be the local house system that is inserting themselves into the process so that, because that's where we're going to get care that as we are seeking information, we are essentially getting that information from our local health system. All right.

That's all for today. Don't forget. Share this podcast with a friend or colleague. You said it's the foundation for mentoring. We want to thank our channel sponsors who are investing in our mission to develop the next generation of health leaders. service now. Enterprise health parlance, certified health and 📍 Panda health.

Check them out at this week. Dot com slash today. Thanks for listening. That's all for now.