Episode Overview
In this episode of Value-Based Care Insights, host Daniel Marino is joined by Don Woodlock, President of InterSystems, to explore how healthcare organizations can take a more strategic and disciplined approach to AI adoption. Drawing on InterSystems’ deep expertise in interoperability, data platforms, and AI-enabled solutions, the conversation focuses on where AI can truly enhance clinical workflows, reduce administrative burden, improve revenue cycle performance, and support better decision-making.
From ambient clinical documentation to smarter data connectivity and actionable AI insights, this discussion cuts through the hype to highlight how healthcare leaders can build an AI foundation that drives real value for clinicians, patients, and health systems, all while advancing the goals of value-based care.
LISTEN TO THE EPISODE:
Daniel Marino:
Welcome to Value-Based Care Insights. I am your host, Daniel Marino. Reflecting back on 2025, an issue that many of the healthcare leaders worked through was, of course, how to really make sense of artificial intelligence, and really the practical approach related to AI. And we've spent quite a bit of time talking about different solutions on the program, but, you know, as we start to think about moving into 2026, many IT leaders, many healthcare leaders are still thinking about what that practical approach is on incorporating artificial intelligence into both their business workflow and their clinical workflow. And it's created some real challenges. And the challenge is really around the potential distraction that artificial intelligence can create within the organization. So when I… when I've talked to many leaders within hospitals and with health systems across the country, they're asking quite a few questions around how do they begin to really evaluate what is that practical approach around artificial intelligence, and frankly, where do they start as they think about integrating it into their clinical operations, into their workflows, and so forth. And some of it has just been incorporated, right, into the technology that they have, you know, incorporated in their organization. For instance, EHRs have started to introduce ambient technology that has started to create some nice value for many of our physicians and our providers. Also, on the billing side, we've also seen a lot of opportunities around appeals and pre-authorizations and things that have been integrated into the practice management solution and have helped, really, our revenue cycle. But I think we've got a lot… a lot more… a farther place to go, a lot more to really consider as we think about how to practically integrate AI into our operations.
Well, I am really excited today to have a guest that has a lot of experience with artificial intelligence. Don Woodlock is president of InterSystems. InterSystems, as many of you have known, it's a long-standing technology solution. It's been involved in many hospitals and health systems over the years. We've had the experience of working with InterSystems as well as some other solutions, and InterSystems has always done a nice job of creating interoperability between disparate systems and creating that connection of data and technology. Don, welcome to the program.
Don Woodlock:
Daniel, great to be here. Looking forward to this.
Daniel Marino:
So, Don, maybe we could start a little bit with, where InterSystems is at right now. As we talk a little bit about the practical approach of integrating artificial intelligence as a solution, both into clinical operations, into business operations, into thinking about what that… that value is. Talk a little bit around where InterSystems is with… Trying to provide some of those solutions.
Don Woodlock:
Sure. If you don't mind, let me just start just with a few minutes on what InterSystems is before I kind of layer AI into that equation. I appreciate the kind words, we've been in the market for a long time. One of the centerpieces of what we do is interoperability solutions in healthcare and beyond, so making sure data can move around, can come together, can be reformatted, and that an organization can really have the data be where it's needed for decision-making and all that, good stuff. So we've been a long-standing interoperability player. We're also a data platform where people build applications on top of us. You know, most famously, Epic is built on our technology, so many… we have many customers, indirectly that use our technology in that way. And then we also have health solutions, most notably an outside the U.S. electronic health record, which has been a great, great area for us to introduce AI as well, but a number of solutions inside, inside the U.S. as well, targeted at kind of interoperability use cases, Master Patient Index, or ePrior authorization solutions, you know, things like… things like that.
Daniel Marino:
Yeah, so you're really touching all areas of the technology and integrated with the different technology, which puts you in a fantastic position to really start to take advantage of artificial intelligence and probably provide a lot of that value to the hospitals and health systems.
Don Woodlock:
Yeah, it's actually kind of an amazing time, because. Gen AI, in particular, can have a very positive impact on a lot of the areas that we work, so we've been quite enthusiastic adopters of the technology throughout our products.
Daniel Marino:
So, when you're… when you're working with some of these hospitals and some of the health systems right now, given the platforming, given the integration of your platform with the different technologies, where are you seeing most of the practical value or the practical, let's say, implementation, realization of AI coming from?
Don Woodlock:
Sure. So, I think we all, are enjoying our first win in Gen AI for providers, and that is ambient. As well, which you… which you mentioned. I was at… you might have been there as well. I was at the CHIME conference a few months ago, and it was really just a high-fiving each other around… around Ambient. Everybody was feeling good. about what we've collectively done as an industry. And this is lots of vendors, you know, in that, in that frame, but it's a great use case of Gen AI, it's a great kind of burden, or a large burden on physicians and clinicians, the documentation of what occurred, and people have done a great job, you know, of putting in place solutions, both building them and adopting them in health systems. And we have our own ambient solution that we built natively in our own EHR, and that's been a huge, huge hit.
Daniel Marino:
That is a quick win. And I… and I agree with you, you know, I've had some opportunities to talk with, many, many physician leaders who've taken the lead on really driving a lot of that, both designing it for their clinical workflows, but then also experiencing it in real time in terms of how it's impacted their delivery of care. And all of them said the same thing. I mean, it's saving them time, the information that they're getting is much more robust. And it's like the wow effect, right?
Don Woodlock:
Yeah, yeah.
Daniel Marino:
To experience it, we're seeing this value, and now folks are saying, okay, well, what's next?
Don Woodlock:
Yeah, sure, sure. And I think that, in terms of questions about what next, the other… what's next? The other area that we're doing a lot of work on is around, kind of, clinical summarization and a chat experience for a clinician to learn about the patient. Okay? So, you know, tell me about this patient. Have they been in here before for this? Do they have any history of cardiac disease? Is there anything from any other health system, you know, that I need to know? Sort of check the national networks, bring in relevant data? You know, that whole process of learning about the patient, do they have any risk factors for the surgery that I'm thinking of giving, to them. I mean, it just saves an enormous amount of time hunting and pecking their way through the chart, you know, for information that may, may be relevant for that.
Daniel Marino:
And I would even imagine, building on that. And this is where I think InterSystems has such value. Being able to aggregate all that data, but really synthesizing the data into critical information that could drive performance, right? Really help with diagnoses, help with the prognosis, help with the care plan in such a way that, you know, in such a way that we've never had that level of synthesized information like we've had now. I can only imagine that's just got to be a huge value driver for a lot of our physicians.
Don Woodlock:
I mean, absolutely. So, let me give an example. So, we, last year launched a product called our Health Gateway, Inter Systems Health Gateway, and it allows a provider to connect to national networks and bring that data in. So we keep track of a lot of stats, including our largest patient. So our largest patient, somebody asked for the patient record, and 4,800 documents came back. 4,859 documents. Now, what are you going to do with that.
Daniel Marino:
Exactly. Nothing. Yes. Nothing.
Don Woodlock:
I mean, nothing by itself. So our technology, of course, parses it all, so we can kind of, you know, have a single problem list, med list, allergy list, so that's good… that's good, too. But then Gen AI… Gen AI's happy to look through 4,800 documents, you know, synthesize it for you, or kind of answer, you know, dig through to find a question like, you know, is this a veteran that's had burn pit exposure, and just kind of find that and answer your questions. So, sometimes I think about it as… as the more success we have with interoperability, like bringing data together, like… like that, the more we create these giant charts. In a sense, and the more we create more information that can be digested easily through the old-fashioned user experience of clicking around.
Daniel Marino:
Yeah, absolutely.
Don Woodlock:
Gen AI is a really great… I mean, it's a well-timed solution. As we get better with interoperability, applying GenAI from a user experience point of view is really necessary, actually, for the Commission to get value from all this data that we brought together.
Daniel Marino:
Well, absolutely. I think it just helps define the path, right? Helps really define how we can provide more robust care plans and more, let's say, focused care delivery to patients who may have some challenging cases. Let me ask a little bit, though, around workflow. Are you seeing a lot of the clinical workflows, even the business workflows having to change because of the new AI technology and the information that we're getting? Or are you seeing that the workflows are actually pretty good, the information that we're giving right now is just allowing us to move faster through the workflows?
Don Woodlock:
Well, I would say that when the workflow doesn't need to change too much, it's gonna help with your AI adoption. So I think some of the early cases are partially because the workflow doesn't need to change too much. So let me… let me just… let me just, elaborate on Ambient a little bit part of the reason it's successful is just the wow of the technology, like you mentioned. It's just kind of amazing. It takes an audio and now gives me a nice soap note and what have you. But part of it, the reason that it's been adopted, is we already had a workflow a long time ago, you know, using dictation firms, and then using voice recognition, and then using scribes. You know, we've had, you know, Gen AI Ambient kind of fits into Kind of a multi-decade process that we've had of helping physicians with this task. You know, so we kind of slotted it into a workflow that people understood. There's a good human in the loop, right?
Daniel Marino:
Right, absolutely.
Don Woodlock:
It's the human that was just there with the patient. So it's the right human, they remember what happened, they sign the note, it's all good from a human-in-the-loop point of view, and everything's been worked out because of these prior things like scribes and dictation firms around the legal risks, who's responsible for inaccuracies, all that stuff's been worked out with prior versions of the technology, so Ambient really, in addition to having a lot of technical merits, it had a lot of workflow and legal things going for it. It was a familiar territory that it kind of slotted into.
Daniel Marino:
Slotted right in, because there was a lot of, to your point, there was a lot of pre-work and foundational elements that were in place that helped to guide it. If you're just tuning in, I'm Daniel Marino, and I'm here today talking with Don Woodlock, president of InterSystems, and we're talking about the practical deployment of artificial intelligence within hospitals, within the provider community. Don, I want to touch on an important topic that I think is guiding a lot of the decisions right now of many of the hospitals and healthcare leaders. Are you seeing the technology being deployed as a next generation of the technology, or is governance of AI within hospitals and health systems informing the decisions on what technology… what AI technology or AI-attributed technology should be implemented. Is it the technology, or is it the governance that's guiding it, or should govern… should guide it?
Don Woodlock:
I mean, well, it's… I think the governance should guide it, and since it's cost money, you know, some of these, solutions, I think it's getting…you know, visibility at the top. I do think having a really good governance process that you're happy with, that works, is an important part of AI adoption.
Daniel Marino:
Yeah, because I'll tell you, this is new, this is… this is new area for many hospitals and health systems. A lot of them do not have AI governance, and, you know, we've… a lot of them don't even under, you know, they have data governance, they have governance in, you know, managing, you know, medical groups, you know, they have different levels of governance, board governance, all that. But AI governance has not been something that they've tackled right now, and many of them are trying to figure out, well. What does it mean? How does it work? How do we start to integrate it?
Don Woodlock:
Yeah, and some of our earliest AI projects, which predated Gen AI, but just doing, you know, old-fashioned machine learning type of things, they would get slowed down for months while the health system figured out, well, who needs to approve what, and who do I need to involve, and, you know, without… you think of governance as slowing you down, but it's not correct, actually. Yeah. You know, getting governance in place, having it work, having a process, having the right stakeholders involved, having the right subject matter expertise on AI, that kind of… that kind of thing. It allows everything to move fast once you…to get that set up. And until you get that set up, you're gonna move slowly. And so I would really encourage every organization to sort of get their act together on that front.
Daniel Marino:
Yeah, I agree with you. I think having strong governance, you know, it's sort of, you know, it helps you define the North Star, right? Where you want to go, and to your point, it helps you move faster. So, you know, one of the vendors that I've been working with, technology vendors that we've been working with is doing a lot of work with aggregating data and putting some probability models together. That is informing both, the let's say, probability of a patient having an issue that would be exacerbated versus also defining what the right course of treatment should be. We're seeing a lot of that within cancer, right? So cancer, through a lot of these trials, are starting to aggregate all this information and put some of these models together. I guess two questions for you there. You know, how do you see some of that integrating with generative AI and also with the care models? And then second to that, how should hospitals and health systems start to take advantage of that as an opportunity.
Don Woodlock:
Yeah. Well, on the… on the sort of probability of an event happening, kind of the prediction aspect of models, you know, I… I… I personally, in our company, we were a big fan of… of this. Re-admission, you know, no-show appointment, you know, length of stay, you know, calculate, you know, those kinds of things, and I feel like none of that… that… none of it, but it really didn't take off in healthcare. I mean, Gen AI…was really the, I think, the eye-opener for AI in, in healthcare, and these predictive models only had, you know, so much impact. I think they'll actually be used now more, that sort of the AI door is open, the governance groups are there, people are more comfortable with it, and get a feel for how it how it might work, and so… so I think it'll actually, grow to some extent now. They can go together, which is kind of the essence of your… of your question, where you could have specialized models that you kind of fine-tune on your own data, your own situation. That sort of thing, but plug them into a larger agentic, you know, and Gen AI.
Daniel Marino:
Yeah, and I think, you know, you bring up a good point. I think first is trying to really understand the capabilities of Gen AI, right? And how do we… you know, and for hospitals and health systems to really start taking advantage of that, obviously the ambient technology. But there's so much more that it can begin to do, and especially if you start to incorporate that into your predictive probability modeling. And, you know, we talk a lot on the program about value-based care, and to move into more prospective monitoring of patients and using generative AI using some of these probability models to identify, more focused approaches to that high-risk cohort, or the rising risk cohort, for instance. The value drivers coming out of that, I mean, to me, just seems like it's enormous.
Don Woodlock:
Yeah, and just let me riff off what you're saying. You can imagine the following combination of technologies. Like, you use a predictive model to sort of have your highest risk patients bubble to the top. Whatever that might mean. Assigning case managers, putting them on the top of work list, whatever. And then you use Gen AI and, like, Agentic frameworks to help you get your work done. What's going on with this patient? What would some good next steps be? Can you arrange this? Can you help me arrange that? That kind of… that kind of thing, where Gen AI can really help, and Agentic AI can really help get… get some of the work done. But you can use some of your more experienced models to sort of bubble up those highest risk cases to the top.
Daniel Marino:
Well, I tell you, that's such an important point. So, really, what you drew out here is two value props that really should be measured, right? It's how you're using generative AI to identify, you know, that real high-risk cohort, or what have you. In your example, but also then creating the right level of information to increase efficiency of your team. So it really, if you're doing this the right way and focusing on both, those two value propositions are really what's going to drive a lot of phenomenal outcomes for you, and probably wrap into your ROI, I would assume.
Don Woodlock:
Yeah, I mean, to be very generic about what you just said, the first is a use case around where to focus. Right? Which patients should I focus on, or which claims are about to get rejected, and I should focus… you know, whatever. So where to focus is a good use of AI in a general way, and then speeding up your work tasks. Right. Some of your work tasks is, like, what's going on in this… with this patient, or this patient situation, and… and GenAI, like I said, they're happy to read those 4,800 documents, give you a summary, that's all… that's all great to a computer, not… not so friendly to a human. So, kind of helping you understand what's going on, and then get some of your work done.
Daniel Marino:
Yeah, yeah, I agree. So, Don, given where we're at right now, and, you know, you kind of touched on this earlier, right? The governance is important because these technologies, you know, they're not free, there's a cost to it. Are we starting to see the ROI come through, or are we still a little ways away from that?
Don Woodlock:
I think generally speaking, we're a little a ways from that, and part of it is our first, like I said, our first home run, we…many health systems did not focus on a hard ROI for ambient. They focused on physician, burnout, wellness, satisfaction and I applaud them for that. I mean, you know, it reminds… it reminds us, it reminds me that healthcare is not strictly a dollars and cents business. It's about something more important, and I think we've demonstrated that with AI so far. I think it… I think it's gonna be hard to ROI soon enough, you know, when we get to the impacting the billing, when we speed up work, and get more productive, those kinds of things. I think it's gonna be, easier to measure the ROI, and we're gonna demand it. And I think it's… I think it needs to be part of the conversation, so I'm not just a, let's just use AI, it's a great tool, it's a lot of fun, you know, I'm not that kind of… I wouldn't say that that's a good idea, but I think it's… I think it can be hard to measure the ROI in the early days of something.
Daniel Marino:
Yeah, I agree with you. I think there's that intrinsic value component, but I do feel like, you know, given where the continued strain on margins are, we have to keep our focus on the investment, and then what the potential opportunity is gonna… going to come out of that. Well, Don, this has been… this has been great, and I know, you know, a lot of our listeners valued a lot of your comments and your advice, and thoughts around generative AI, and certainly that practical approach. Any piece of advice that you might give to some of our listeners, those that are really thinking about advancing their AI strategy, or even integrating a lot of the capabilities that they're starting to come across?
Don Woodlock:
Yeah, I would, I'd say two things, maybe. One is, I think it's a good practice to align with your major vendors. So, they're gonna be rolling, and I'm thinking of the EHR vendors, maybe.
Daniel Marino:
Sure.
Don Woodlock:
They're going to be rolling out AI capabilities, you're going to be part of a community of other systems that are using it as well. I think that's a nice, good place to start. And a good center of gravity for your AI initiatives is to do it that way. And the second thing I'd say is we, you know, as you deploy AI, there's an impact… there's an impact of the data and the connection to your other systems that make AI really work. So if your data foundation is not so good, it's scattered, your data's kind of messy, your AI is going to be equally, you know, inaccurate, not as,you know, not as capable. And I would say, from an interoperability point, you know, Agentic AI is all about connecting… allowing AI to connect to the systems that you have and get some things done. And so having an interoperability framework that makes sense, that brings your organization together, is also going to empower your AI to really work well for your employees.
Daniel Marino:
Oh, great points. I agree, Don. I really appreciate that. Well, I want to thank you for coming on. This has been a great conversation, very enlightening for me. You obviously, you know, have had some great success, and I wish you and InterSystems a lot of success as you go forward, and work with a lot of hospitals and health systems. I mean, you are at the forefront of change. Frankly, we all are, and it's quite exciting.
Don Woodlock:
Great, thank you, Dan.
Daniel Marino:
If any of our listeners are interested in connecting with you, or maybe learning a little bit more about what the solutions are that InterSystems brings to the table, can you share your, contact information, or maybe direct them to your website or something?
Don Woodlock:
Yeah, two things. I mean, come to our website. We have all of our intersystems.com, we have all of our solutions outlined there, and certainly contact us if you're interested in any of that. And then feel free to connect with me directly on LinkedIn. I also have a video series called Code to Care, which is on LinkedIn and associated places where I explain these AI concepts. I love teaching, so it's a bit of a side job, but… Yeah, that's great. If you'd like to learn more about AI, you can connect with me on LinkedIn.
Daniel Marino:
Well, that's wonderful. Well, thanks again, Don. I really appreciate your time. Great conversation. And if any of our listeners are learning a little bit, are interested in learning more about this topic or any of our topics that we talk about on Value-Based Care Insights, please contact me at dmarino@luminaHP.com, or feel free to visit our website at luminaHP.com. I want to thank everyone for tuning in and listening. Until our next Insight, I am Daniel Marino, bringing you 30 minutes of value to your day. Take care.