Latest / Becker’s Healthcare Podcast / AI Costs, Vendor Strategy and the Future of Hospital Capacity with Laura Dyrda

Transcript

AI Costs, Vendor Strategy and the Future of Hospital Capacity with Laura Dyrda · Becker’s Healthcare Podcast

• 0:02
  - There's a version of AI in health care
• 0:03
  - that lives mostly in press releases,
• 0:06
  - pilot programs, partnership announcements,
• 0:08
  - theoretical use cases.
• 0:10
  - And then there's the version that lives in
• 0:12
  - the work, the CIO trying to get a
• 0:15
  - model into production
• 0:16
  - without breaking a dozen workflows,
• 0:18
  - the revenue cycle leader weighing automation against accuracy,
• 0:22
  - the CISO who knows that every new integration
• 0:26
  - attack service.
• 0:27
  - Becker's eleventh annual IT and revenue cycle conference
• 0:31
  - is built for the people doing that work.
• 0:33
  - More than 2,500
• 0:35
  - executive level attendees,
• 0:37
  - 600 plus speakers,
• 0:38
  - and four days of sessions designed around the
• 0:41
  - real operational questions
• 0:43
  - behind AI, cybersecurity,
• 0:45
  - revenue cycle performance, and digital health.
• 0:48
  - Join us September 14 through the seventeenth.
• 0:51
  - For the agenda and event details, visit beckershospitalreview.com
• 0:55
  - and click on the events tab in the
• 0:57
  - upper right.
• 0:59
  - This is Scott Becker with the Becker's health
• 1:02
  - care podcast.
• 1:03
  - I'm thrilled today to be joined by Laura
• 1:05
  - Dirda,
• 1:06
  - editor in chief and just a fantastic leader.
• 1:09
  - She joins us regularly to talk to us
• 1:11
  - about a few of the stories that she's
• 1:12
  - following in health care. Laura, let me turn
• 1:15
  - it to you. What are a couple of
• 1:16
  - stories that you're following currently?
• 1:19
  - Well, thank you so much, Ted. And, you
• 1:21
  - know, right now in health care, there's a
• 1:23
  - a lot of focus and attention being placed
• 1:25
  - on how hospitals, health systems, and organizations
• 1:29
  - can really most efficiently and effectively work with
• 1:31
  - their teams,
• 1:32
  - galvanize the leadership as well as those on
• 1:34
  - the front lines around making sure they're adding
• 1:38
  - technology
• 1:39
  - in a smart and right way, incorporating AI
• 1:41
  - and being smart about how they're using it
• 1:43
  - in the way in which it's kind of
• 1:45
  - evolving within the space, which is fascinating.
• 1:48
  - What we've heard from CIOs
• 1:50
  - and IT leaders in addition to the CEOs,
• 1:52
  - CFOs, and c suite who are paying close
• 1:54
  - attention to that consumption is just as they
• 1:57
  - have continued to add technology,
• 1:59
  - point solutions, and those kinds of things, and
• 2:00
  - digital
• 2:01
  - transformation has accelerated
• 2:03
  - more vendors coming into the scene. Now they're
• 2:05
  - paying attention to how vendors are
• 2:08
  - are using AI and including that within their
• 2:11
  - larger governance structures. So, for example, sometimes they
• 2:14
  - would have had platforms or agreements with companies
• 2:18
  - before
• 2:19
  - the company incorporated AI into,
• 2:22
  - what they're doing. And, you know, once the
• 2:24
  - company does turn on that AI, they still
• 2:26
  - need to have an additional process to make
• 2:28
  - sure that the technology, the data, and the
• 2:31
  - information is used in a way that complies
• 2:33
  - with what the organization is trying to do
• 2:36
  - and,
• 2:36
  - will be healthy and helpful,
• 2:38
  - to just protect the organization from risk,
• 2:41
  - both on the
• 2:42
  - patient data side as well as on the
• 2:44
  - cost side because I've heard from many CIOs
• 2:47
  - that they're looking at some of the vendor
• 2:50
  - repricing
• 2:51
  - because of the AI is now outpacing inflation,
• 2:53
  - and they're looking at, AI licensing costs that
• 2:56
  - haven't yet proven some of their returns. And
• 2:59
  - so thinking through the technical debt they have,
• 3:01
  - it's eating up a lot of their budget
• 3:03
  - just to keep some of those old system
• 3:04
  - running. So going through and then cleaning that
• 3:06
  - up is a a priority. But at the
• 3:08
  - same time, platform consolidations
• 3:11
  - are pushing CIOs and vendor conversations earlier than
• 3:14
  - they would have expected in the past.
• 3:16
  - And then the second,
• 3:17
  - issue that has been coming up quite a
• 3:19
  - bit is that
• 3:20
  - AI,
• 3:21
  - features within some of these technology platforms are
• 3:23
  - being repriced at higher tiers.
• 3:26
  - Companies are seeing the value of them, plus
• 3:28
  - looking at, you know, token based and consumption
• 3:30
  - based pricing versus,
• 3:32
  - some of the original contracts that they've had.
• 3:35
  - And so, you know, I think a lot
• 3:37
  - of the major platforms are are working with
• 3:39
  - health systems to try to figure out what's
• 3:40
  - gonna be sustainable for them as the prices
• 3:43
  - of consumption continue to go up.
• 3:46
  - And it's just really fascinating to hear how
• 3:48
  - some of those things are factoring into the
• 3:51
  - broader conversations and about that. Health systems. Yeah.
• 3:55
  - No. Thank you. Talk about that because you
• 3:56
  - hit a nerve that people are talking about
• 3:58
  - every place
• 4:00
  - is this consumption based pricing.
• 4:02
  - It's almost like the old days of buying
• 4:04
  - minutes on your phone where the cost could
• 4:07
  - get out of control until at some point
• 4:08
  - they move to these unlimited packages.
• 4:11
  - And that seems like what we're hearing from
• 4:13
  - a whole cadre of users of AI
• 4:16
  - that the consumption based pricing makes it cost
• 4:19
  - prohibitive
• 4:20
  - unless you're, you know,
• 4:22
  - can really afford it. Talk about that a
• 4:24
  - little bit because I think you've had something
• 4:25
  - that's very people are watching very close.
• 4:28
  - Yeah. Absolutely. That is such a great point,
• 4:31
  - Scott. And certainly,
• 4:32
  - from the health system level, from the executives
• 4:34
  - and the c suite that we're hearing from,
• 4:37
  - you know, it would be really tough to
• 4:39
  - cut those contracts because their organizations
• 4:41
  - are now so ingrained,
• 4:43
  - with the technology.
• 4:44
  - Their teams are depending on it and relying
• 4:47
  - on it to do their jobs at a
• 4:48
  - very high level,
• 4:50
  - not only, obviously, the ambient technology, but how
• 4:53
  - AI is integrating into their data analysis,
• 4:57
  - everything about it. Not only that, using
• 5:00
  - AI enterprise subscriptions for things like, Anthrope's
• 5:04
  - cloud or or OpenAI's Chad GPT are are
• 5:07
  - areas that hospitals and health systems really see
• 5:10
  - as being space that their teams,
• 5:13
  - expect and wanna be able to build within.
• 5:16
  - And so,
• 5:17
  - having those conversations is incredibly important. I've heard
• 5:20
  - from multiple executives that, to your point, the
• 5:23
  - consumption based model just isn't sustainable for hospitals
• 5:26
  - and health systems. Their budgets won't allow for
• 5:29
  - continued increase in the same way that, for
• 5:31
  - example, four to 500 companies would. Now
• 5:35
  - on the flip side, I have also heard
• 5:37
  - a few,
• 5:38
  - executives
• 5:39
  - in those in high level positions at large
• 5:42
  - health systems across the country say they're not
• 5:45
  - super worried about that because when you look
• 5:47
  - at across the board, a lot of the
• 5:49
  - token usage and those kinds of things that
• 5:51
  - get truly, truly out of control
• 5:53
  - often are coming from building from within, and
• 5:56
  - and the developers who are trying to create
• 5:59
  - a a tech platform
• 6:00
  - for companies like Uber or other tech companies.
• 6:03
  - And so,
• 6:05
  - you know, they more sought pragmatically to say,
• 6:07
  - look. My team isn't gonna reach the levels
• 6:09
  - that those large companies will, and so I
• 6:11
  - think we could still keep it under control.
• 6:14
  - But to that point, I I think a
• 6:15
  - lot of health systems should in the meantime,
• 6:18
  - while they're trying to figure out what the
• 6:20
  - contract model is going to be with the
• 6:21
  - large companies like Anthropic or OpenAI.
• 6:25
  - Seeing a lot of the the smaller companies
• 6:28
  - and vendors that they're working with tick up
• 6:30
  - AI usage too, it all adds up. So,
• 6:33
  - one of the solutions that they've come up
• 6:34
  - with is really being thoughtful and mindful about
• 6:37
  - the types of subscriptions that they provide to
• 6:39
  - their,
• 6:41
  - employees and teams so that they aren't giving
• 6:43
  - unlimited access to everybody, but really truly finding
• 6:46
  - the, you know, select,
• 6:49
  - group or slice of the organization that might
• 6:51
  - need more power and more tokens and having
• 6:53
  - a very clear justification and monitoring that very
• 6:56
  - closely,
• 6:57
  - as time goes on just to keep those
• 7:00
  - costs under control as much as possible and
• 7:02
  - totally get into a more sustainable model for
• 7:04
  - using the tokens and AI and those kinds
• 7:07
  - of things. And then secondarily,
• 7:09
  - we have heard from CIOs and IT executives,
• 7:13
  - at small as well as large organizations,
• 7:17
  - that
• 7:17
  - they are
• 7:19
  - building into their vendor contracts
• 7:21
  - specific,
• 7:23
  - requirements
• 7:24
  - for those new contracts or re upping contracts
• 7:27
  - by adding AI that they
• 7:30
  - want to see the evidence that these things
• 7:32
  - are actually gonna work the way that they're
• 7:34
  - supposed to work. They wanna see, the data
• 7:36
  - behind it. They wanna look past some of
• 7:38
  - those claims. They're like, hey. This should provide
• 7:40
  - you a return on investment, but,
• 7:43
  - becoming more and more,
• 7:45
  - critically examining those claims and and wanting to
• 7:48
  - see the actual data behind it, wanting to
• 7:51
  - have those companies take on to some degree
• 7:54
  - risk with them so that the outcomes actually
• 7:56
  - do,
• 7:57
  - come that they are promising. So I think
• 7:59
  - those are some of the ways at the
• 8:00
  - moment that hospitals and systems are trying to
• 8:02
  - get around this uncertainty
• 8:04
  - in kind of the messy middle of of
• 8:06
  - these price consumption and,
• 8:08
  - fluctuations with AI. It's not a perfect solution,
• 8:11
  - but I think one that a lot of
• 8:12
  - organizations are trying to get around and and
• 8:15
  - and share some of those,
• 8:17
  - the ways that they're phrasing those contracts and
• 8:19
  - working with their partners so that it's not
• 8:21
  - all up on them, and and they won't
• 8:23
  - have to lose the technology as the prices
• 8:25
  - go up. But really, truly kind of working
• 8:27
  - with the vendors to say, look. Here's what
• 8:30
  - health care organizations look like, what we're able
• 8:31
  - to do, what we're able to spend on
• 8:33
  - technology, and how can we make this work.
• 8:36
  - It's fascinating, Laura. So that is fascinating to
• 8:38
  - see this evolution of AI spending and how
• 8:41
  - people have to allocate and how they're looking
• 8:42
  - at it and where they're gonna allocate their
• 8:45
  - their their spending towards these e a AI
• 8:48
  - tools,
• 8:49
  - and platforms.
• 8:50
  - What else are you watching closely?
• 8:52
  - That is fascinating. We're seeing that we're hearing
• 8:54
  - that every place in every organization
• 8:57
  - about the cost of consumption pricing and how
• 8:59
  - it's killing some budgets.
• 9:00
  - What else are you watching closely currently?
• 9:03
  - Anything else? Yeah. Absolutely. No. I think one
• 9:06
  - of the other things that we're really looking
• 9:08
  - at right now is, of course, seeing how
• 9:10
  - hospitals and systems are
• 9:12
  - are are thinking about their patient volume, planning
• 9:14
  - for the future based
• 9:16
  - on inpatient, outpatient,
• 9:18
  - and,
• 9:19
  - hospital at home, all of the things that
• 9:21
  - are coming together,
• 9:23
  - some technology powered, others, you know, just really
• 9:26
  - trying to figure out what is gonna be,
• 9:28
  - the best resource and the best ways to
• 9:30
  - to use and expand access to care to
• 9:32
  - their patient populations. And one of the things
• 9:34
  - that,
• 9:36
  - earlier this year, s g two came out
• 9:37
  - with a report that basically showed,
• 9:40
  - even with a lot of the move for
• 9:42
  - hospital at home and and outpatient,
• 9:44
  - the inpatient discharges are expected to increase around
• 9:47
  - ten percent through,
• 9:49
  - 2035.
• 9:50
  - And so a lot of health systems
• 9:53
  - anticipating
• 9:54
  - overall,
• 9:55
  - health care could see a 176,000,000
• 9:57
  - inpatient days annually,
• 9:59
  - which would strain even the health resource systems,
• 10:02
  - as that grows. And so,
• 10:05
  - one of the things we've been focused on,
• 10:07
  - I did an interview with doctor Abraham Barkuda,
• 10:10
  - who is the inaugural chief of the division
• 10:12
  - of hospital medicine at UT South Southwestern Medical
• 10:15
  - Center in Dallas. He talked through how he's
• 10:17
  - thinking through that, trying to figure out how
• 10:19
  - do we make sure to leverage the, obviously,
• 10:22
  - outpatient and hospital home models post acute as
• 10:25
  - much as possible. But then, two, as if
• 10:27
  - thinking about redesigning the system, not to forget
• 10:30
  - about the inpatient side,
• 10:31
  - what's gonna be needed, both on the infrastructure
• 10:34
  - as well as technology to efficiently get the
• 10:37
  - patients through, to make sure that clinicians are
• 10:42
  - equipped with the right education about how technology
• 10:45
  - should work so that they can design the
• 10:47
  - tools that they need and be able to
• 10:48
  - incorporate them easily into their workflows,
• 10:51
  - make sure clinical teams are able to identify
• 10:54
  - and be more predictive in the future,
• 10:56
  - on the inpatient settings so that they can
• 10:58
  - triage patients and,
• 11:00
  - you know, identify issues as quickly as possible
• 11:03
  - so that they can do what they need
• 11:05
  - to do in order to deal with this
• 11:07
  - projected expanded capacity
• 11:09
  - as volume likely will continue to go up.
• 11:11
  - But I think there's a variety of factors
• 11:13
  - that contribute to that, whether it's,
• 11:15
  - a a continued
• 11:16
  - increase in the aging population that will need
• 11:19
  - more inpatient care or whether it's hospitals closing
• 11:22
  - in some regions and and regional systems needing
• 11:25
  - to,
• 11:26
  - absorb some of that,
• 11:27
  - volume. I I think there's a lot of
• 11:29
  - reasons for that, but at the end of
• 11:30
  - the day, if hospitals
• 11:32
  - aren't remembering to think through the inpatient needs
• 11:35
  - that they'll have and make sure they're ready
• 11:37
  - for it, is gonna be a real challenge
• 11:39
  - over the next ten years or so. And
• 11:41
  - so that was one of the things that
• 11:42
  - was really interesting coming out of that conversation,
• 11:45
  - and we've seen it, you know,
• 11:47
  - with health systems recently continuing to invest in
• 11:50
  - additional
• 11:51
  - inpatient rooms, towers,
• 11:54
  - smart hospitals that are truly designed
• 11:57
  - in order to have efficient patient flow and
• 11:59
  - and thinking through what that inpatient experience is
• 12:02
  - gonna be like and how they can quickly
• 12:04
  - connect with clinical teams and,
• 12:06
  - keep an eye on those things. And so
• 12:08
  - it's just been very interesting to to look
• 12:10
  - at that and see how health system leaders
• 12:12
  - are navigating some of these, issues that likely
• 12:16
  - will come in the future.
• 12:18
  - No. It's literally fascinating. And there's so many
• 12:20
  - hospitals that are
• 12:21
  - under bedded where it seems like years ago,
• 12:23
  - everybody was over bedded.
• 12:25
  - Now it seems like everybody's full and trying
• 12:27
  - to figure it out and often full
• 12:30
  - with not necessarily high margin patients,
• 12:33
  - but but still still full and so critical
• 12:36
  - to America's
• 12:37
  - ecosystem in taking care of our patient population.
• 12:40
  - So fascinating to see this this effort to
• 12:42
  - make these health systems smarter and work better
• 12:45
  - and all that kind of stuff. Absolutely fascinating.
• 12:47
  - Yep. Definitely. I think, you know, the next
• 12:49
  - frontier additionally is the emergency department and and
• 12:52
  - trying to,
• 12:53
  - continue to optimize what that'll look like in
• 12:56
  - in getting the right patients that they're and,
• 12:59
  - you know, having that be more of an
• 13:01
  - efficient thing. It's it's fascinating to see how
• 13:03
  - some health systems are navigating through that and
• 13:06
  - trying to move patients into habit changes in
• 13:09
  - terms of, you know, instead of coming into
• 13:10
  - the emergency room, having them, you know, connected
• 13:13
  - virtually first and and more. And so a
• 13:16
  - lot on that forefront, a lot will continue
• 13:18
  - to follow and keep reporting on, but,
• 13:20
  - just really interesting to think about what health
• 13:23
  - care and and what that inpatient experience will
• 13:26
  - be like as well.
• 13:27
  - No. And I think that's fascinating. And those
• 13:28
  - things touch together too because I think many
• 13:31
  - else's feel like they don't have enough beds
• 13:33
  - to really see our patients to as well
• 13:35
  - or places to discharge patients to. So the
• 13:37
  - bed issue is a real issue too. And,
• 13:40
  - yes, absolutely trying to, on the front end,
• 13:43
  - keep people out of the ER if you
• 13:44
  - can do so through virtual means and other
• 13:46
  - things. Absolutely.
• 13:48
  - But at the end of the day, so
• 13:49
  - much of our population, including us, rebounced back
• 13:53
  - to going to the emergency room when we
• 13:55
  - can't get access elsewhere elsewhere.
• 13:58
  - And it just seems like we've returned to
• 14:00
  - that past where we're using the ERs a
• 14:02
  - lot unless you got these wait times, plus
• 14:04
  - the behavioral health issues are also keeping ERs
• 14:07
  - very full. Fastening, Laura, as always to visit
• 14:10
  - with you. Thank you for joining us on
• 14:12
  - the Becker's Healthcare Podcast.
• 14:14
  - What a great chance to visit with you.
• 14:15
  - Thank you very much. Thank you, Scott. It's
• 14:18
  - always a pleasure.