
Loading summary
A
Hello, everyone. This is Erica Spicer Mason with Becker's Healthcare. Thank you so much for tuning in to the Beckers Healthcare podcast series. Today we're going to talk about how organizations are thinking about connectivity, where AI is delivering practical value, why healthcare requires a different approach to AI, and how patient access continues to drive operational efficiency initiatives. And joining me for this conversation is David Edwards, the Chief Technology Officer at Relation. David, welcome to the podcast. Thank you so much for joining us today.
B
Thanks, Erica. It's good to be here.
A
It's good to have you here. And before we dive into the conversation, I wanted to ask you to share just a little bit more about yourself and the work that you're leading in the healthcare space.
B
Yeah, sure thing. So I am the Chief Technology Officer, as you mentioned, at Relation. I've been in the company about one year. We are essentially a patient access company that specializes in intelligence scheduling. My career started in 1991. That certainly dates myself with Cerner. And I started as a software engineer. I spent a couple decades there. Most of my time was in leadership roles. Eventually, at one point, I became an engineering fellow at Cerner. And then about 10 years ago, I stepped away from health care to really cut my teeth on my first CTO job. And I actually found that quite useful to really step away from health care after 25 years, do something different, focus on really improving operational skills, running a technology organization, learning more about the business, working with boards of directors and so on, and really stepping back into healthcare. Having all that sort of experience over the last 10 years has really helped me become a better CTO in the context of the healthcare industry.
A
Well, great to learn that about you, David. And certainly some time away can help bring new perspective, new skills to what you're doing currently. And I think that's so important in an industry as complex as healthcare. And we know this industry has long struggled with fragmented systems and siloed data. So I kind of wanted to start our conversation there. From your perspective, how is healthcare's progress on connectivity reshaping? What's possible right now, and where do you think the biggest gaps are? Still there.
B
Yeah. So when I, when I stepped away from healthcare in 2016, the FHIR specification, I think most of your listeners probably know what that means, but if not, it's the FAST Healthcare Interoperability Resources Specification that essentially was a standard that was starting to really gain a lot of traction. And at the time Cerner had implemented it, I think EPIC at the time was sort of dabbling in it, and then sort of fast Forward. I come, you know, back into healthcare a year ago and starting to look at where that specification has emerged, you're starting to see, I think, really a critical mass adoption, which is fantastic when it comes to improving the connectivity, interoperability of systems. So what's really exciting to me about the FHIR spec is that when you get critical mass adoption, we're not there yet, by the way, and I'll comment on that in a bit. When you get a critical mass there, it starts to open up all kinds of opportunities on the edge of these systems to innovate. I think that's really, for me, what's exciting. If you think about what's happening in the AI space, particularly in the agentic realm of AI, once you open these systems through open standards and liberate that data, there's all kinds of interesting things you can do on the edges. And we'll talk about a little bit about this. But voice AI is an area, I think that's pretty exciting and creates a lot of value there. But in terms of some of the gaps, I think despite having sort of critical mass around a standard like fhir, there is still an adoption issue. And just to illustrate my point, relation integrates with about 10 different EMR and practice management systems today. Only a few of them actually support the FHIR specification. So despite the fact that we've got a lot of momentum, the reality is broad adoption just isn't there. There's a variety of reasons for that. There's cost, time complexity and so on. I think that's sort of the anchor that's preventing sort of more and more, let's say, acceleration of innovation on the edges of these systems.
A
Yeah. So, David, it sounds like since you have come back into the healthcare space, you've seen some progress and connectivity, especially with the FIRE specs, as you mentioned, but still some gaps in adoption, room for growth. And it sounds like there's definitely some opportunities and excitement behind AI, whether we're talking agentic AI, voice AI. So I'd like to bring our conversation there next. There's certainly no shortage of AI hype in the industry, but leaders are being asked to make very real and heavily scrutinized decisions around their AI investments. So where are you seeing the technology deliver genuine and measurable value right now? And how do you distinguish that from the noise?
B
Yeah, there's certainly a lot of hype out there. I think I can certainly appreciate the C suite and the pressure that they're getting from their boards, their owners, to adopt AI. And look, there's despite the hype, there is a lot of very interesting problems that are being solved out there. So I think cutting through that noise is important. And some of the areas that we are seeing very interesting results in as, as I mentioned was the Voice AI. We can talk about that here in a little bit. But the key thing is, I think for all C suite leaders, the important thing to remember is go back to ask yourself this question. What business problem are we solving? What are the pain points that we feel as an organization? And can, can we reimagine solving those problems in a different way, using AI as sort of a first step in that process? I think that that's the really hard transition that I found. I mean, particularly, you know, I work in the software development side of the fence here. But there's a lot of sort of institutional fear and resistance of, you know, what does this mean and how does this work? There's just a lot of sort of like change management that has to happen. Don't make poor decisions around AI projects that don't actually return value to you. So again, I go back to sort of what business problem are we solving? Try to avoid the sort of fear of missing out that would lead to poor decision to run a project. But areas that we see in particular that are driving clear value is think of GNA and sales and marketing functions inside the company. We're seeing a lot of our, those organizations start to really lean in and start with first principles and reimagining how they can get their work done. So a lot of times it's an efficiency and productivity improvement. On the product side, again, Voice AI, we could talk about that a little bit more in detail, but I think there's, there's a really, really interesting, fantastic work that's going on there. On the support side of the organization. What we're finding is that the AI is helping us resolve issues a lot faster than we could in the past. So we're getting master productivity gains there as well.
A
Yeah, really interesting to hear where you're seeing some of these measurable gains happening, David, and also appreciate you highlighting some of the parameters that you'd encourage leaders to think about when assessing AI tools. So, you know, staying really grounded in the business problem they're solving, reimagining solving those problems a different way. But you, you highlighted Voice AI a few times, so I want to go a little bit deeper there. This is something that seems to be gaining traction in healthcare. So I'd love to know what you think is driving greater adoption of Voice AI and what should leaders realistically expect from it?
B
Yeah, and this is, this is a great question. And this is an example of something that is solving a real business problem. You know, tying it back to my earlier point. A lot of these, particularly the organizations that we sell into, they have call centers with very high turnovers, 30 to 40%, long wait times for patients, eight to five operations and so on. And it ultimately creates a less than ideal experience for the patient when they're trying to interact with the, the, the practice to, to reschedule or do any kind of activity and so on. And so the, what the voice AI is doing essentially is creating a new system of engagement. So it is, think of it as an agent that is 24, 7, never quits their job, always there to help. And there turns out that there's a large number of routine calls that happen. Things like verifying appointments, rescheduling appointments, canceling appointments. And there's actually a lot of, there's kind of a long tail of interesting use cases there that we're finding as well. All of these can actually be done through an intelligent agent. Now, what makes it really valuable is when the, the voice AI system or the agent itself is integrated with the system of record. So unless you do that, you're not really getting the full benefit of that AI experience. And what I mean by that is I can call an agent that is connected into our scheduling system that knows when I'm scheduled for the appointment and it knows when, for example, if I need to reschedule it, it knows what those open times are going to be. And so I can actually interact with this software agent entirely without a human and solve 50% of my calls can be handled by an agent. Now, and that's what we're finding is that many of our clients are realizing that they're getting incredible value, better patient experiences as a result of that. And look, the hard things go to a patient or they go to a human. When there's something that we can't solve, we route it to human. And that's actually when you want to talk to somebody. But if it's routine things, then let the software, let the machines do what they're really good at doing.
A
Yeah, it's a great example, David, and I appreciate you highlighting this particular use case. And I want to make sure too, that we touch a little bit more on, I guess, adoption and kind of the change management piece that you highlighted earlier. Whether we're talking about clinical applications or administrative applications, like the example you gave, when it comes to AI. Healthcare, of course, has higher stakes and concerns around things like safety compliance. It's fundamentally different from other industries that might be adopting the same tools. So I'd love to know from your perspective, what makes getting AI right in this environment uniquely difficult, but also imperative, and what does responsible adoption really look like?
B
Yeah, that's a great question. And healthcare has always been much more conservative in its adoption of technology, and rightly so. I mean, we're talking about patient safety, highly regulated environment, so on. And so we, we always tend to put a lot more controls and guardrails in place around the systems we build. So much more conservative. But I, I think it's when you look at AI, one thing that people need to realize, and I think we all know this, but we all tend to forget about this, is that these are probabilistic systems. Right. They make mistakes. You know, the AI industry calls it hallucination, but ultimately what it is that means is it's making mistakes. And. Which is different than many of the software systems that we've built in the past, which are deterministic. Right. It always works the same way. And so with AI, the way that you mitigate, you have to think about the risk associated with what you're using the AI to solve. And the way that you mitigate that is by putting controls and guardrails in place. And a lot of times that means human in the loop. Right. And in particular in healthcare, I think always having a human in the loop is a critical piece. But again, you have to weigh the risk associated with that use case. Let me give you an example of something where there's extremely low risk. One thing that we do is we actually use a secondary AI to evaluate, in the sentiment of a call that a patient has with a voice AI agent. Well, that's a case where it's low risk. It just provides a score for us that a human can go look at if they want to. But that's a case where, you know, it's something that we can tolerate. We don't need to put controls and guardrails in place around. If you look at something else, like ambient listening, these are systems where a software agent will listen to the conversation between a clinician and a patient and then transcribe that into a clinical note. Now, it's because the AI can make a mistake. It's extremely important for the clinician to be in the loop to verify that the AI actually produced something that was accurate. So again, I think at the end of the day, the complexity and the criticality of safety in the healthcare space just requires us to have strong guardrails and controls in place to mitigate that risk. But again, I tell everybody, don't be afraid of AI. Don't be afraid of what AI can do to help solve problems. Just make sure that you understand the risk and that you mitigate that appropriately.
A
Yeah, and I appreciate you sharing. David, how relation is approaching those extra checks and guardrails as you mentioned? Again, another helpful example for our listeners. And we've covered a lot of ground today already, but just wanted to check before we close, is there anything that we haven't covered that you think deserves more attention in this conversation?
B
Yeah, you know, when I think it ties into my last point about leaning into AI and being curious, I think it's really important for organizations to not fear the potential here. I would say lean into it, experiment. Do things that are safe, low risk. What I often find is that if you just start small and get a win, and when I say a win, I mean go prove out that AI can help you solve a problem, then it starts to open up the aperture of possibilities that an organization would say, okay, well, we got a whole new set of problems we could go solve. It starts to change the muscle memory of the organization and reduce the institutional resistance. So that's kind of my parting advice is just lean into it, be curious and experiment.
A
Great parting advice. David, I want to thank you again for the time that you've made for our conversation today. Thanks again for joining Beckers.
B
Thank you, Erica. I appreciate the time.
A
And of course, we'd also like to thank our podcast sponsor for this episode. Relation listeners, be sure to tune into more podcasts from Becker's Healthcare by visiting our podcast page at beckershospitalreview. Com.
Podcast: Becker’s Healthcare Podcast
Episode: The New Healthcare Access Layer: Connectivity, AI, and Real Operational Gains
Host: Erica Spicer Mason
Guest: David Edwards, Chief Technology Officer at Relation
Date: May 12, 2026
This episode explores the intersection of connectivity and artificial intelligence (AI) in healthcare, with a particular focus on how advancements in data standards and AI-driven workflow automation are enabling operational efficiencies and improving patient access. David Edwards, CTO at Relation, brings decades of experience to the discussion, offering insights into tangible ways AI—especially voice AI—is making measurable impacts and how healthcare organizations can responsibly and effectively adopt new technologies despite industry-specific challenges.
[02:01 – 05:05]
Advancement of FHIR Standard:
Current Gaps:
[05:05 – 08:19]
Pragmatic AI Adoption:
Operational Gains:
[08:19 – 11:27]
Why Voice AI is Taking Off:
Mechanics & Impact:
[11:27 – 15:06]
Risk, Compliance, and Human Oversight:
Responsible Adoption:
[15:06 – 16:20]
Start Small, Prove Value, Expand:
Leaning into Curiosity: