This transcription is provided by artificial intelligence. We believe in technology but understand that even the smartest robots can sometimes get speech recognition wrong.

Newsday: The AI Staffing Dilemma - The Productivity Paradox with This Week Health

I'm Bill Russell, creator of this week Health, where our mission is to transform healthcare one connection at a time. Welcome to Newsday, breaking Down the Health it headlines that matter most. Let's jump into the news.

Bill Russell: , it's Newsday and the team is so well prepared. We are running so fast right now. US Sarah's hair is actually on fire. Drex is not. Um, And I'm covering mine up 'cause I haven't had a haircut in six weeks because I've been on the road. So, hopefully that will happen this week, although it's supposed to happen on Thursday and I forgot that's not gonna happen on Thursday.

It's a holiday.

Sarah Richardson: Let best cut it for you. See what happens,

Bill Russell: the craziness of it all the

Drex DeFord: humanity. We have noticed though, from time to time that you have been combing your hair, which we're super impressed also with

I'm like, it's branding. She's. You're with me. You're branded like I know. Or you're branding. I know. I know who you are. I'm like, okay. All right. We're gonna talk healthcare. I want to talk AI and I want to talk about its impact on healthcare, impact on the economy. And you know, it's interesting 'cause earlier this year when we were doing summits and we were doing city tour dinners, people would apologize.

They'd say, oh, I'm apologize. You know, I don't want to be the one to bring up ai. Well, that, those apologies have gone away. I mean, it's just like, oh my gosh, it's every day in my life. It's shown up in all these tools. Epic's throwing it at me, ServiceNow's throwing it at me Google, Microsoft, they're all throwing it at me.

Mm-hmm. And so we're putting this together. The interesting thing happened on our way to the, To, to the play. And that was that. We started to recognize that individuals. So we have this quadrant thing that we do with our employees, with our staff, and it's the quadrant is challenged. It's, there's a the x axis is challenged, are how challenged are you?

And the y axis is supported, right? So if you're not supported and you're challenged, you're stressed and you're burned out. And if you are, you know, you get the idea. If you are challenged and you're supported, you're in the right quadrant, you're actually doing good work and you're supported by the organization and good culture, it actually works.

We have noticed that people's bubbles have started moving from the top right hand quadrant over closer to cozy. Because they are getting stuff done a lot quicker. They're getting more stuff done, and we're implementing a new system. And that new system has automated with AI a whole bunch of things that they weren't doing.

So they're getting cozier and as a manager as a business owner and as a manager. And that's where I'm gonna pick up this conversation with you all. Do we believe I believe, I mean. This is gonna be a challenge for us. We're gonna get, we're gonna, our staffs are gonna start to be able to do more with less.

I don't know. I did the Eric Larson with I was, was at the he's a consultant and investor and he was at the conversation, the Bridge conference, and one of the things he said is, you should close every non-administrative. Every nonclinical administrative role you have open right now. He goes, if you don't and you hire for those, you'll probably not need those jobs in 12 months.

He was adamant on that, and he's saying that to a whole room full of healthcare people. I'm curious where you guys are falling down on this and what you're seeing and what you're hearing and how you would be approaching this as a manager.

Bill Russell: No.

You would still fly around. You said that in a post, you would still fly around. Who are you kidding? You love being with these people.

Sarah Richardson: I do actually, and I'm so grateful that we capped off a whole year of city tour dinners in New Orleans, and not just because New Orleans is a city where you celebrate, it's because just that group of people was a, they made up that dinner on their own and wanted us to come.

And just the types of conversations and that commitment to community, everything that we build was really special. So I'm grateful, you know, be home for a bit and. When you think about what AI can do for your teams, they can do them. If the mindset of the organization is such that's also gonna be a place that wants to head.

You keep really moving your own box around until you find the place that lets you go do all the things that you love to do and want to be experimentally capable of doing. That is something we have created here, so can we continue to X ourselves? Absolutely. And I've done it in other organizations. There are aspects though, that if I was working in a more conservative organization that prides itself on never having to lay people off, then I would be really thinking about, and how do you get a whole lot of people to not, you know, to wanna move on and do their own thing.

Where the most risk of, do we embrace AI or does it become a bit of the enemy to our culture from a healthcare system perspective, for sure.

Bill Russell: But it is easier to close racks?

Sarah Richardson: Yes.

Bill Russell: That, I mean the, one of the rural health systems. Yeah, I think we'll call this rural, it was rural health system. They laid off 90 people.

It was front page news. Everybody's talking about it at the barbershop kind of thing. Oh. They ended up hiring 30 of the people back. But still That was significant in that Yeah, in that community. If they had just closed the wrecks. That's not a front page story though, right?

Sarah Richardson: No. No. And it allows you to step into it gracefully because having to go back and rehire people can sometimes be embarrassing.

So there's a lot of. Spokes that come out of that hub of, Hey, AI becomes a central figure in our organization. Is it up-skilling? Is it laying off? Is it re-skilling? Is it moving people to different roles and opportunities? Is it closing open recs? All those things have to be orchestrated really well with hr, and that messaging comes from more than just it.

Drex DeFord: I think it's also just the culture of the, how are we gonna manage the work. Yeah, the work is still gonna be there One way or another. We need to figure out, I mean, I'm gonna go to Toyota production because that's what I do all the time. How are we gonna standardize the work so that ultimately it can be automated?

And we can use those same people to continue to sort of grind the edges off of those work processes so that they can be automated and they can become. Something we can do with ai, that's a special skill. This isn't something you just say I don't think you just say close all the open recs. 'cause in 12 months we're not gonna be doing that work.

That's not. No, that's not right. Yeah, you're going to have a program, a process in place to be able to make that a reality. Once you get momentum, you've changed the culture, those things are running, then I think you can really look at things like, we're not gonna need those open recs. And then we can start talking about closing them and maybe never opening them

Bill Russell: All right. I'm using AI over here and it's interesting to me 'cause I just loaded something that would've taken all day, maybe even two days, maybe even a week. I loaded three years of financials into our new system.

Yeah, it came back it, and sure enough, about an hour later I came back and all the financials were in the system matched to the right partner and the ones that were sort of anomalies, it sort of noted, Hey, these were anomalies. What do you wanna do with them to work? And I'm sitting there going.

Crap. If I were at a health system, I remember the number of times I went to the team and said, Hey, you know, what's it gonna take to stand up this dashboard and do this thing? And they'd be like, oh my gosh, we need a team of 10 people for 17 weeks. And you may never get that dashboard to look the way you want to.

And they're saying, yeah, what can I do for you? What do you want me to do? Oh, you looking at, they're, they work

Drex DeFord: 24 hours a day. Like you can actually set off one of those projects and then you can go home and those AI agent interns work all night. Mm-hmm. And never complain and don't let a day off and they just do their thing.

And the for, in your particular case, you had really. Clear, well structured data to start with. And you probably had really clear, well structured prompts to tell it exactly how you wanted it to do the work. And then when it didn't do it right, you were able to make just a tiny adjustment and then it gave you exactly what you wanted.

Drex DeFord: Mm-hmm.

Bill Russell: The advances on these things are amazing. Like amazing and I've been using 'em all year and like each one is like an incremental.

And then Gemini came out, I'm like, oh gosh. And then the new Claude came out. I'm like, oh my gosh. So you say, you know, five one, you say clearly define prompts and that kind of stuff. I've started to like just be like, I'm talking to you sloppy. Like, Hey I need this data in here. Can you do this? And it like, it starts its reasoning process.

It's like, well he was talking about this, but it appears he's talking about this. Let me, you know, clarify this with a question and 'cause you can watch it think and you can watch it reason and you can watch it come back to you and you're like, oh my gosh. It like, it keeps me from doing something stupid.

Yeah, it can be a consult as long as you go back

Sarah Richardson: and validate.

Bill Russell: You still have

I'm like, Hey, we're doing bug fix. We're doing bug fix on something that significantly changes our world in 2026.

Bill Russell: Yeah. So here's the question I have for, oh, again, Toyota, Leann guy, let's go to you. Toyota, Leann guy. Okay. Like, we should come up with a nickname. Make like Uncle Drs. I thought

Sarah Richardson: it was Uncle Drs.

Yeah. Maybe not

Bill Russell: the,

Sarah Richardson: and I have the, oh wait, I have the Toyota. He doesn't. So there you go.

Drex DeFord: Oh, you are the Toyota lien.

Bill Russell: So as we

Sarah Richardson: driver,

Bill Russell: as we were building out this this application and, you know, we're, we're looking at our internal processes, the thing that amazed me was the amount of time we spent. With people like going over here to find this and going over here to find this And going here and I mean, it took a PhD level person just to know where all the repositories were.

Let's just let's confine it to just the health IT organization. Not within healthcare, but just the health IT organization. How much of that is going on?

Drex DeFord: I think there's a ton of it. I mean, if you look at, you know, Toyota talks about several different kinds of waste, but the waste of search is a huge.

Part of this in our case, it's which Airtable, it's in lots of different places, or which sheet, or which Google sheet, or which Google doc is it in? And I need to go and update it here, and then I need to replicate that and put it over here by taking that waste That waste of search out of the process, it saves me a ton of time and it significantly decreases my aggravation with just like doing basic work becomes easier.

You can use the feature of AI that's in the tool set that you have, depending on the tool that you have to do things like actually take action if it's this kind of an alert. And so these things we're able to sort of set up and automate. I think in the beginning, you, we've all talked about this, artificial intelligence is like the smartest, dumbest person that you've ever, like.

They know everything, but they don't have any context. So in the beginning, human in the loop is really necessary. 'cause you, I think you've gotta look and see at the, see what kind of work that intern's doing. When they get really good at it, you can kind of cut 'em loose and let 'em, now you know what to do when it looks like this.

That's actually what people should be saying. Right. So to do more with less, it's just like, well then just figure out how to work more hours. That's not really what you want. Do more by taking waste out, by removing waste, so that you can do that same work with less energy.

Bill Russell: I'm gonna put us all in three different categories, probably comfortable categories because I think if you look at us, we're three, three different, phenotypes of CIO, Drex, I'll put you in the category of strong leader, process oriented. You know, staff oriented. Sarah, I will put you in the strong, well, actually, you're probably closer to me in terms of you, you are not risk averse. You will, you'll take some risks, you'll build some stuff, but you're very people oriented, very customer oriented.

I'm very much push the envelope the hell with people. Kind of, I mean, that's as you guys have experienced,

thing for all of us.

Bill Russell: Yeah. And this is this is why I tell people I don't have the temperament to be a CIO anymore.

'cause it is like. It was like I had to suppress this. Can we go a little

faster?

There

Drex DeFord: was a lot of back rubbing and hand holding. Is it? Yeah, it was very difficult. But not literally Sarah, not literally.

Sarah Richardson: No, I know. I actually was thinking about, I just probably don't wanna see Bill's engagement scores from back in the day.

Bill Russell: . Actually my engagement scores were really good because, oh, there

Drex DeFord: you go. Oh, well that's good. 'cause you know, one,

Bill Russell: one of the things you learn how to do with the kind of leader I am is you set a vision and you get people excited about the, you know, we're building the future of healthcare and here's how we're going to do it.

So that was really early. And people were like, people early adopt, oh my gosh. We're, you know, we are on the cutting edge. And so when you, and the other thing that happens at that point is people who see that sort of rally around you and that's who you hire and they get really excited. So I didn't have, yeah,

Drex DeFord: this is good. Bar talk for the people who work in your department when they go and sit down and go, guess what we did today? Yeah. We went to the cloud, you know, and people were like, wow, how can I come and work with you guys? So it definitely,

Sarah Richardson: or the thriving factor of where that works well is what you've built here is here's our vision.

Go do it, and we all check in once in a while, but there's no one holding your hand. There's nobody like telling you what to do per se. It's like if you like that autonomous, you might do great. You might not do great, but your boss has your back either way. I mean, we all came here for a very specific reason and it's like, can we run as fast as we can with a vision with very little quote unquote.

Bill Russell: strive to give you less oversight every day.

Drex DeFord: It's Sarah's running around as fast as she can with two pairs of scissors in her hands. That's what's happening in heels. In

Bill Russell: heels. Heels. In heels. Yes. So what I wanna do is I wanna go around to the three of us.

I wanna talk about how we would be approaching sort of the AI wave that's coming from a staff standpoint, from a, I don't know from an innovation standpoint, from a leadership standpoint of your own leadership and your leadership in the health system. If you want, I can go first and sort of give you a framework or you guys could choose to go first 'cause you don't want to hear me pontificate.

Drex DeFord: I say pontificate,

Like-minded people around to say, okay, what do we think is going to happen? I'd have those people in it. I'd have those people on the clinical side. I'd have it as many departments as I can represented and having conversations. A lot of those conversations would be individual conversations. They wouldn't be, I wouldn't be like pulling big groups together just yet.

I would try to sort of gauge like, where are we at? What do we think could be possible? Who are the. Friendlies to this and who are the people who are not, so you're sort of trying to figure out the landscape. But all along you're sort of trying things out. You know, we had a group at St.

Joe's that was doing taking our images and running 'em through computer vision and coming back with secondary reads. We were building that, did I ever intend to build a product and go live with that? No. the purpose of that was for them to get the experience and to learn it. How did we apply it?

I would be educating the leadership team. One of the things I did was meet with the c. EO on a monthly basis, and I would give her articles, I would give her books and we would have lunch and we'd have conversations on, you know, where I thought this stuff was going. She would educate me on healthcare. I would educate her on where this technology was going.

I would try to get her buy-in on, Hey, this is where I think this is going. This is how I think she would. You know, she would sort of chisel away at what I was thinking as were other people until we got to something that was like, all right, let's start to get the groups together. Now, there's a reality that's going on, which is all this AI is coming at you, so you do need to form the governance group that's going to be evaluating and all you have to do the basic blocking and tackling of AI coming into your organization.

Drex DeFord: So I was just gonna say that's like a terrible framework, but I'll try to go next.

Bill Russell: All right. You go next. I appreciate your your critique.

Drex DeFord: Okay. So I, you know, a lot of this the framework that you have built here, that we have built here is a lot of the very similar framework I would use in a health system. And it's also, as I go back to Toyota production systems and organizations. Like when I was at Seattle Children's, this is one of those things that we did over and over again too.

Failed and where they'd had successes and what they had already tried and why they're excited about what they're gonna try next. And with a lot of people watching that, you can't help but create a culture that people. In those other seats, look at the person on the platform and say, oh man, I could totally use that.

What you need sometimes culturally, is somebody else who's doing something that's solving a problem. That's almost like, or it can be exactly like the problem that you have. And when you see somebody doing that, then you start to try to. Replicate that. If you've never used these tools before, if you've never been any in the new models of the tools, which as you sort of talked about earlier, significantly different than the models that we had just two weeks ago there's some pretty cool stuff in there.

And there's a bunch of things I would do, but I think if I had like a starting nugget, it would be, let's make sure that folks are doing stuff. That's cool, that's working. Give him a platform to talk about it so that everyone else can see it and learn from it.

Bill Russell: All right, Sarah. Tell him while he is wrong.

Do it. .

Sarah Richardson: He's not wrong because, oh,

Bill Russell: come on.

Sarah Richardson: I will. I will say that there are aspects of what both of you would do that I absolutely support and agree with, and I believe we'd all get there. We would all have the same. Outcome to a degree. Three different ways of getting there. That's also the power in sourcing the talent.

As much as you have the people who are curious about learning, I also. Find the empathy in those that are gonna take longer to get there or have that biggest concern about, oh my gosh, change is scary. I don't know if I'm even capable of learning these things. Because you get some of these institutionalized roles.

People have been there for 20 or 30 years. That can be a win or it can be an albatross and you find the medium ground to help these people understand how to get there. 'cause you're gonna have this huge bell curve. Yeah. Of how the organization gets here. You gotta work with executive leadership, the board, and even HR to understand here's.

The accelerators, here's the middle, and by the way, here's the people who are lagging and what is our strategy around that goes back to my people side of the job. And yet, how is it aligning with the strategy of the organization in terms of what? What problems are we solving with our ability to do this?

So let's just say I've got Epic and I have ServiceNow and I have other a Bridge as an example. How are they helping us get there too? And embedding our selves to a degree into their success and strategy. Because then you've got a mutually beneficial output that can be lauded more globally. So to Dr.

This point, people platform to share what they're doing. We're learning because I see my universe, they see 200 universes or more, and. The orchestration back to all these big pieces, creating a beautiful puzzle. I'd be spending most of my energy making sure that the circus tent was up and running and hiring some really good lion Tamers.

And who's the guy that runs the circus inside the tent.

Bill Russell: Master ceremonies. The ring. The ringmaster. The ringmaster. That's it.

That's kind of how I would look at it. 'cause it's still gonna be a bit of a circus for a while.

Bill Russell: Well, I'm sure everyone has determined that Sarah is the perfect balance between Drex and I. She's like, if you wanna hire a CIO, she's the one you want to hire. Yeah. Um, Please, because, but I'd

Sarah Richardson: be a great chief of staff for either of you.

FYI Drex and I joke about what if we like, after all these interviews we do for HCSP, 'cause we change up our titles every time. I'm like, you know, I bet we'd be really good at running a hospital. He'd be like. Yeah, but we're not going to, we did that once before. I

Bill Russell: love Drex is like been there, done that, got the tshirt

Sarah Richardson: and I'm not doing that yet.

But we think we'd be still be pretty good at it.

Bill Russell: Oh man. Well, you know, it's really interesting as we sort of look at this, we did close one of our open recs. I mean, we're not that big of a company. We're 17 people. We had an open rec and we closed it. When we identified that there was one person in our organization that was incredibly cozy.

Drex DeFord: Help everybody else get cozy.

Bill Russell: Yep.

Drex DeFord: The other interesting thing about the model that we have here is that I feel like. The easy and sloppy way out for a lot of organizations is just hire people. You wind up with a whole bunch of people on staff and the work is getting done, but you have no way to kind of get yourself outta that.

It's like a legacy application now. It's just sort of built like that. The way we've built it here is we've hired people and pushed the AI button on them pretty early on and really often, and in fact, we don't hire somebody until it's just like, I got no other way to like get this work done.

Bill Russell: one of my advisors asked me yesterday, it's like, what are you gonna do with all this excess capacity? I'm like, you know, are you gonna let people go?

I'm like, no, I'm gonna grow.

Drex DeFord: You're, we're growing the right way as opposed to, it's just, it's funny to watch the you know what the folks who kind of are stuck in this mode of, like, I have people ask me all the time how we've got like a hundred people working there and it's like, no, we don't have a hundred people working here.

You know, and you kind of tell 'em the story and they're like, what? You only have 17 people working there. How can that be so. It's fun to be inside of a machine like we've built here in the way that it runs.

Bill Russell: Well, we're not on the road for a little while, so this is gonna be fun.

We've got some time to read and well, we've got a lot of, we're doing a couple end of year things. We're doing some things for our partners end of year, but it's all zoom based, so, should be fun. I assume we're gonna do predictions as well. You guys up for predictions?

Bill Russell: When are we gonna do that?

Sarah Richardson: I believe it's next week and we've got some end of year thoughts, some gratitude, some predictions.

Bill Russell: we got off the road and our team's like

Sarah Richardson: they're off the road. Oh yeah. there's no breaks. There's like, no even five minute break. I had to move some people around so I could actually just get a snack and, you know, take care of life for 10 minutes at a time.

So. Yeah, well,

Bill Russell: I, well, it's great to be off the road and I will crack you guys up to close this out. I've gotta close this up because my bocce starts in 23 minutes.

Sarah Richardson: Bocce ball.

Bill Russell: I'm on a bocce team and I've made like one match out of the first six, and today I can finally make it. It's not a retirement community, but they schedule things at like four O'.

Sounds

Sarah Richardson: like it.

Bill Russell: Yeah, exactly.

Sarah Richardson: Shuffleboards next bill. I mean, shuffleboards, when you can't really throw, you can just kind of shove things. You've got lots of trajectory in here. We,

Bill Russell: we,

We don't, although I will say at the conversation, I don't know if you saw this, but at the conversation they brought in some really.

Sarah Richardson: dang.

Bill Russell: he was great. I mean, in terms of entertainment, I mean, that's his shtick, right? He was very entertaining. And at one point he goes he goes, you know, people ask me a lot of stupid questions. People ask me stupid questions all the time. Like somebody asked me like. What do you think of pickleball?

He goes, what a stupid question. Pickleball sucks. And I thought, and that was it. And I thought, here it's, that's John McEnroe right there. the quote I will leave you with is this on John McEnroe. So he he was getting fined by the American Tennis Association, like left and right for his outbursts and that kinda stuff.

And you are entertaining, you're bringing in fans don't change. And he goes, so who are you gonna listen to? Jack Nicholson or the American Tennis Association? He goes I chose Jack Nicholson. And it's the foundation for everything he's done since. So, you know, our team's gonna get mad at me 'cause this went over the allotted time. It's great to hang out with you guys.

Drex DeFord: Yeah, it's good to see you too. Good to see. You too, Sarah, even though we're on the road all the time.

Sarah Richardson: Yes. But it's also what makes work so fun.

Bill Russell: hold right there while we're, while we're recording. Hold one second. Hold please. Oh.

Sarah Richardson: Wait, what happened?

Gonna show us his pickleball racket. And tell us it doesn't suck.

Drex DeFord: Wait. On my bot. Oh.

Bill Russell: There you go.

Drex DeFord: Oh,

Bill Russell: it's the Bucky's

Sarah Richardson: hat. My,

Bill Russell: my gift from the road from you guys.

Sarah Richardson: I'm so happy right now. I buy you a 300 bottle of wine. I never hear about it. I get you a $10 hat and you're like wearing it on camera. I know.

This is so

Thanks for listening to the 2 29 podcast. The best conversations don't end when the event does. They continue here with our community of healthcare leaders. Join us by subscribing at this week health.com/subscribe.

If you have a conversation, that's too good not to share. Reach out. Also, check out our events on the 2 29 project.com website. Share this episode with a peer. It's how we grow our network, increase our collective knowledge and transform healthcare together. Thanks for listening. That's all for now.