Latest / Becker’s Healthcare Podcast / Paul Testa, MD, Chief Health Informatics Officer at NYU Langone Health

Transcript

Paul Testa, MD, Chief Health Informatics Officer at NYU Langone Health · Becker’s Healthcare Podcast

• 0:02
  - Every health system executive right now is being
• 0:04
  - asked to do something that has no useful
• 0:06
  - precedent,
• 0:07
  - make durable technology decisions in an environment that
• 0:10
  - looks meaningfully different every six months.
• 0:13
  - The pace isn't slowing,
• 0:14
  - the vendor landscape isn't getting simpler, and the
• 0:17
  - cost of missteps
• 0:18
  - keeps rising.
• 0:19
  - Becker's eleventh annual IT and Revenue Cycle Conference,
• 0:23
  - The Future of AI and Digital Health, exists
• 0:25
  - for exactly this challenge.
• 0:27
  - Here, you'll find more than 2,500 executive level
• 0:30
  - attendees,
• 0:31
  - 600 plus speakers, and four days of sessions
• 0:34
  - built around the questions leaders are actually wrestling
• 0:36
  - with.
• 0:37
  - How to scale AI responsibly,
• 0:39
  - how to protect increasingly complex systems,
• 0:42
  - how to bring more performance out of the
• 0:44
  - revenue cycle, and how to make tech investments
• 0:46
  - that hold up over time.
• 0:48
  - Join us September fourteenth through the seventeenth at
• 0:51
  - the Hilton Chicago.
• 0:53
  - You can register on our events page at
• 0:55
  - beckershospitalreview.com
• 0:57
  - by clicking on the events tab in the
• 0:59
  - upper right.
• 1:01
  - This is Laura Dierda with the Becker's Healthcare
• 1:04
  - podcast. I'm thrilled today to be joined by
• 1:06
  - Paul Testa, chief health informatics officer at NYU
• 1:09
  - Langone Health. Paul, it's a pleasure to have
• 1:11
  - you on the podcast today.
• 1:13
  - Thanks for having me. Always a pleasure.
• 1:15
  - Absolutely. Now I'm excited for our conversation. I
• 1:18
  - know we'll talk through some of the cool
• 1:20
  - things that you're doing as well as perspective
• 1:22
  - on the future. But before we dive in,
• 1:24
  - can you tell me a little bit more
• 1:25
  - about yourself as well as NYU Langone?
• 1:29
  - Happy to. Thank you.
• 1:31
  - So I'm a born and bred New Yorker,
• 1:34
  - which, you know, really identifies a lot of
• 1:36
  - who I am, I feel like. I've been
• 1:38
  - at NYU Langone now for twenty two years,
• 1:41
  - give or take. I did my residency in
• 1:43
  - emergency medicine there.
• 1:45
  - And,
• 1:46
  - it's it's been home for me for a
• 1:48
  - very long time, so it's been incredible to
• 1:49
  - watch it grow and,
• 1:52
  - and excel.
• 1:55
  - I,
• 1:55
  - right now, get to the privilege of leading
• 1:58
  - health informatics for the system,
• 2:00
  - which has been really, I think, a strategic
• 2:02
  - differentiator as we grow, as we expand, as
• 2:04
  - we bind
• 2:05
  - ourselves together as a large academic health system.
• 2:10
  - Informatics is the glue that holds us in
• 2:12
  - many ways, and it's it's been really incredible
• 2:14
  - to be able to watch
• 2:16
  - as we
• 2:18
  - continue to set quality
• 2:19
  - standards for ourselves
• 2:21
  - to leverage all the technologies. And then, you
• 2:24
  - know, let it let it mark the one
• 2:25
  - minute moment where I mentioned AI,
• 2:28
  - how we've sort of heavily leaned into AI
• 2:31
  - and what that's meant for us as a
• 2:32
  - health system.
• 2:34
  - So it's been a great venture, and,
• 2:36
  - I'm loving what I do.
• 2:39
  - That's fantastic to hear. And, you know, to
• 2:41
  - your point, just having that data information informatics
• 2:44
  - and and translating that into something that's beneficial
• 2:47
  - across the board,
• 2:48
  - is so important in health care right now
• 2:50
  - and something that is a huge unlock for
• 2:52
  - organizations once they once they can do it
• 2:54
  - well.
• 2:54
  - Now what are some of the powerful forces
• 2:57
  - that you see shaping digital health and IT
• 2:59
  - at NYU Langone, and how do you anticipate
• 3:01
  - it'll evolve over the next year or so?
• 3:06
  - An incredibly powerful force, I think,
• 3:09
  - has to be the economics of what we
• 3:11
  - do.
• 3:12
  - We need to be really conscious stewards
• 3:15
  - of the resources
• 3:16
  - that we have access to and deploy.
• 3:20
  - It's
• 3:21
  - it's
• 3:22
  - not as challenging to deploy 10 of something
• 3:25
  - or
• 3:26
  - to have a small innovation project that has
• 3:29
  - great
• 3:30
  - results in
• 3:32
  - one location or five locations.
• 3:34
  - But to steward our resources and to get
• 3:36
  - the the bang for the buck
• 3:38
  - as we
• 3:39
  - are in sort of every facet of the
• 3:41
  - way we deliver care, you have to do
• 3:43
  - it at scale. So real innovation in my
• 3:45
  - book is anything that we can do at
• 3:47
  - scale successfully.
• 3:49
  - And
• 3:51
  - the
• 3:52
  - the cost of
• 3:55
  - what it will take to
• 3:57
  - innovate,
• 3:58
  - to do AI at scale, to continue to
• 4:01
  - drive the quality and safety, you know, and
• 4:03
  - most importantly or or the binding is the
• 4:05
  - the digital patient experience partnered with the digital
• 4:08
  - clinician experience.
• 4:10
  - We have to do that understanding what our
• 4:11
  - success metrics are and what our return on
• 4:13
  - investment is. And there are multitude of ways
• 4:15
  - of measuring both those things.
• 4:17
  - So we're we're setting out a standard internally
• 4:20
  - at NYU Langone
• 4:21
  - to
• 4:22
  - to measure like for like. So when we
• 4:25
  - deploy a project, we look at what our
• 4:26
  - measures of success will be, and then we
• 4:28
  - try to use the same measures for the
• 4:29
  - next project where we can so we can
• 4:31
  - compare apples and apples
• 4:33
  - and look at the way we spend our
• 4:34
  - resources.
• 4:36
  - I think all of that is in service
• 4:38
  - of our patients and our clinicians
• 4:40
  - and ensuring that it's an absolute, just incredible,
• 4:43
  - joyful digital experience
• 4:45
  - in some of the trying times of our
• 4:46
  - patients and our clinicians' lives.
• 4:52
  - That makes a lot of sense. And, you
• 4:53
  - know, I I think it's so critical
• 4:55
  - to be able to have that great patient
• 4:58
  - experience because right now, it seems like there
• 5:01
  - is more and more of a need and
• 5:02
  - demand for things to be
• 5:05
  - seamless across the board. But especially in the
• 5:07
  - health care space, there could be a lot
• 5:08
  - of complexity. So how are you seeing, those
• 5:11
  - solutions being,
• 5:13
  - driven at NYU lingo and health?
• 5:16
  - What are the ways that you're seeing patient,
• 5:19
  - preferences change, and how are you meeting those,
• 5:22
  - expectations on the technology as well as the
• 5:25
  - clinical care side?
• 5:27
  - Oh, I love that question. Thank you. The
• 5:30
  - simultaneously,
• 5:31
  - patients expect, rightfully so,
• 5:34
  - the most cutting edge,
• 5:36
  - highest quality care. That's their given right and
• 5:39
  - what they show up at our front door
• 5:41
  - expecting, and we have to give that to
• 5:42
  - them each and every time.
• 5:44
  - But they also expect, and we want to
• 5:47
  - give,
• 5:48
  - the most empathic
• 5:49
  - connections and care.
• 5:51
  - And sometimes
• 5:52
  - when technology is deployed,
• 5:55
  - maybe in less than a thoughtful fashion or
• 5:57
  - without the patient and the clinician front and
• 5:59
  - center,
• 6:00
  - that empathy can suffer.
• 6:02
  - So we're finding so much, for example, with,
• 6:04
  - with leveraging AI to create patient friendly versions.
• 6:08
  - I love us calling it patient friendly. It's
• 6:09
  - all patient friendly. The fact of the matter
• 6:11
  - is we've been writing notes for decades.
• 6:14
  - We never and the patients weren't our primary
• 6:16
  - consumer, so we're shifting, and we're gonna figure
• 6:18
  - out what it means. But we created a
• 6:20
  - rubric
• 6:22
  - at NYU, went on clear rubric
• 6:25
  - of defining patient friendliness in five different domains.
• 6:28
  - So everything we put in front of patients
• 6:31
  - has to pass that rubric as well as
• 6:33
  - an AI safety checklist and make sure it's
• 6:35
  - ethically and safely deployed. We also have to
• 6:37
  - ensure that it's
• 6:38
  - ingested the way our patients want it.
• 6:42
  - Be it
• 6:43
  - in a you
• 6:44
  - know, they're finding the patient's report and tells
• 6:46
  - us quite quickly
• 6:47
  - that the AI generated summaries
• 6:50
  - can be,
• 6:51
  - or and AI generated
• 6:54
  - versions,
• 6:55
  - the patient friendly versions, everything from our
• 6:58
  - e e, our echo results
• 7:00
  - to our discharge summaries
• 7:02
  - to every day, giving discharge the the summaries
• 7:05
  - of a patient's stay in the ED in
• 7:07
  - the, inpatient setting.
• 7:09
  - Patients
• 7:10
  - are finding the
• 7:12
  - the AI generated versions to be more empathic
• 7:15
  - scoring. And I'm not gonna say that's a
• 7:18
  - bad thing from our docs, but we we
• 7:20
  - optimize for something different with the doctor's outputs
• 7:22
  - and the nurses' outputs,
• 7:24
  - and this gives an opportunity to tune in
• 7:26
  - a different direction.
• 7:27
  - So I think
• 7:29
  - meeting both their safety, quality expectations,
• 7:32
  - as well as the experience, that's our challenge
• 7:34
  - here.
• 7:35
  - That makes a ton of sense, and I
• 7:37
  - love the way you laid that out so
• 7:39
  - perfectly because it is, such an important thing
• 7:42
  - to understand what the patients are dealing with
• 7:44
  - and how they,
• 7:45
  - are are wanting to have the experience of
• 7:48
  - health care and then exactly what AI can
• 7:50
  - do and and how then the caregivers can
• 7:52
  - lean into that, relationship that they have and
• 7:55
  - what they can, do to most helpfully,
• 7:58
  - bring the patients into,
• 8:00
  - recovery.
• 8:01
  - Now from your perspective, could you tell us
• 8:03
  - about something you're doing at NYU Langone Health
• 8:06
  - that,
• 8:07
  - has been a big swing you took in
• 8:08
  - the last year that paid off? What did
• 8:10
  - you do, and what were the results?
• 8:13
  - You know, I think the biggest swing we're
• 8:15
  - taking right now, and I I don't have
• 8:17
  - I don't have all the outcomes yet, but
• 8:18
  - I know where we're going,
• 8:20
  - is,
• 8:22
  - you know, there's so much,
• 8:24
  - inches of headline and conversations around ambient documentation,
• 8:28
  - how important the ambient documentation is for the
• 8:30
  - experience of patients and the doctors.
• 8:32
  - But what is
• 8:34
  - trailing, and I think it'll be more revolutionary
• 8:37
  - to the experience aspect, is nursing ambient documentation.
• 8:41
  - So we're deeply involved both with our EHR
• 8:44
  - vendors as well as third parties and our
• 8:46
  - own efforts internally with our own AI expertise
• 8:49
  - of what it's gonna mean for nurses to
• 8:51
  - be able to practice out loud, to speak
• 8:53
  - their practice, which they've been doing for centuries.
• 8:57
  - But capturing that rather than having to run
• 8:59
  - to flow sheet rows to document.
• 9:01
  - I think, you know, the
• 9:03
  - the the space of an exam room
• 9:05
  - with a physician is a very different experience
• 9:07
  - than an inpatient setting in a room with
• 9:09
  - a nurse.
• 9:10
  - I think it's we we have seen
• 9:12
  - incredible gains
• 9:14
  - with ambient documentation
• 9:16
  - for physicians.
• 9:17
  - I can't wait to see what we're gonna
• 9:18
  - have with nurses using this tool at scale.
• 9:21
  - So we're we are early in that journey,
• 9:25
  - but I think it's gonna dramatically change the
• 9:28
  - practice
• 9:29
  - as well as how nurses are able to
• 9:31
  - connect
• 9:32
  - with their patients and manage their day.
• 9:35
  - So I I have incredibly high hopes. I
• 9:38
  - think the
• 9:39
  - the way putting it is, you know, the
• 9:41
  - the win for Ambien and the physicians, and
• 9:43
  - it was an easy one to see.
• 9:45
  - For nurses, it's gonna be a little bit
• 9:46
  - more it's gonna be a lot more effort,
• 9:48
  - but I think the, the incremental benefit is
• 9:50
  - gonna be incredible.
• 9:55
  - That makes a ton of sense. And, you
• 9:57
  - know, it is cool to have that opportunity
• 9:58
  - to then support nurses,
• 10:00
  - who are such a critical
• 10:02
  - part of the the patient care journey, to
• 10:05
  - then have that ambient support there too. So,
• 10:08
  - you know, when you look at that change
• 10:09
  - management aspect of it, as you mentioned, doing
• 10:12
  - nursing out loud and and being able to,
• 10:14
  - fit within the workflows,
• 10:17
  - how do you approach that? Obviously, like you
• 10:19
  - said, with physicians is one thing, but,
• 10:22
  - what is what do you do to make
• 10:24
  - sure that, you you know, you're working with
• 10:26
  - the teams and it's as seamless as it
• 10:28
  - can be?
• 10:31
  - Oh, it's a tricky one.
• 10:33
  - And and yet in some ways, it's incredibly
• 10:35
  - simple, and that is to meet the nurses
• 10:38
  - and patients where they live and where they
• 10:40
  - are.
• 10:41
  - And the very first thing we learned
• 10:43
  - is
• 10:44
  - nurses,
• 10:45
  - the units that we had been using, our
• 10:48
  - proof of concept for ambient
• 10:50
  - documentation
• 10:51
  - were quieter,
• 10:52
  - frankly. They were used to engaging in their
• 10:56
  - practice
• 10:57
  - in a less out loud way.
• 11:01
  - I'm not gonna take any hits on that
• 11:03
  - as an ER doc. I'm I'm kinda I
• 11:04
  - can get loud myself, but I think we're
• 11:07
  - more practiced, and there's an expectation
• 11:11
  - of
• 11:12
  - speaking as we go.
• 11:13
  - So,
• 11:14
  - with our CNO,
• 11:16
  - and our informatics
• 11:18
  - nurse leadership,
• 11:20
  - We engaged on a project essentially, practice out
• 11:22
  - loud.
• 11:23
  - And it it
• 11:25
  - while every one of our nurses has a
• 11:27
  - clinical mobile companion, an iPad with a suite
• 11:29
  - of apps,
• 11:30
  - right now,
• 11:32
  - it was a step we could pave the
• 11:34
  - way without any technology, and that was simply
• 11:37
  - peer to peer engaging
• 11:39
  - in what it meant to practice out loud
• 11:41
  - and having nurses
• 11:42
  - assess themselves
• 11:44
  - and what speaking their practice and as they
• 11:47
  - engage in their assessment,
• 11:49
  - means for them and getting used to it
• 11:51
  - because some took to it very quickly, and
• 11:53
  - some are still working on it. So it
• 11:55
  - it was we we engaged in this practice
• 11:58
  - out loud program
• 12:00
  - irrespective of technology, just just just to move
• 12:02
  - them towards,
• 12:04
  - reengaging their own voice,
• 12:07
  - as nurses at bedside.
• 12:09
  - And it's been incredibly,
• 12:11
  - enlightening to watch.
• 12:14
  - I don't wanna say
• 12:16
  - a a renewed voice because it I don't
• 12:18
  - think it went anywhere,
• 12:19
  - but it's stretching different muscles,
• 12:22
  - flexing different muscles for many of our nurse
• 12:24
  - colleagues
• 12:26
  - that are really leaning into it. And it's
• 12:29
  - it's incredibly transformative
• 12:31
  - to what we're hearing bedside.
• 12:36
  - That makes a ton of sense, and thank
• 12:38
  - you so much for sharing that with us.
• 12:40
  - I think it's so helpful,
• 12:41
  - for all health systems as they're going through
• 12:43
  - some of these changes to understand what's working
• 12:45
  - well and what can make a difference.
• 12:47
  - Now that, you know, you you've been working
• 12:50
  - on these projects and getting technology in the
• 12:52
  - door for both the patients as well as
• 12:53
  - clinicians, what's the most important problem that you're
• 12:56
  - trying to solve next with AI or technology?
• 13:02
  - The I I I won't say easy wins,
• 13:04
  - but the early wins,
• 13:06
  - with the application of artificial intelligence and generative
• 13:09
  - AI in health care has has had a
• 13:11
  - lot to do with what we've spoken about,
• 13:13
  - summarization,
• 13:14
  - extraction.
• 13:18
  - But,
• 13:19
  - you know, as an ED doc, the magic
• 13:21
  - is in the flow.
• 13:22
  - So we're beginning to inject models
• 13:25
  - not just at the extraction point of, why
• 13:28
  - is this patient in front of me?
• 13:30
  - But
• 13:31
  - or or what transpired in the last twenty
• 13:34
  - four hours to six months with this patient,
• 13:36
  - be it ambulatory or inpatient.
• 13:38
  - We're looking for the the wins to be
• 13:40
  - had in the flow of an ED, the
• 13:42
  - flow of a hospital, the flow of a
• 13:43
  - patient in the arc of their care. And
• 13:45
  - that is sort of I think, you know,
• 13:47
  - it's it's it's often referred to as longitudinal
• 13:49
  - decision support. I don't think that really captures
• 13:52
  - it,
• 13:53
  - But how do we
• 13:55
  - have models which are ever present or can
• 13:58
  - be ever present? And they can be
• 14:00
  - always,
• 14:03
  - vigilant
• 14:04
  - as a copilot, as we often hear the
• 14:07
  - term, for nurses and doctors
• 14:09
  - in their day to days and in the
• 14:11
  - week to weeks. So we've deployed models in
• 14:14
  - the emergency department that are looking for high
• 14:16
  - risk diagnoses that can get
• 14:19
  - missed or
• 14:20
  - anchored away from with an earlier diagnosis.
• 14:24
  - And we're reducing that model so it's ever
• 14:25
  - present watching the documentation,
• 14:27
  - watching
• 14:29
  - the inputs from lab results and for imaging,
• 14:32
  - and then
• 14:33
  - providing a nudge to say, hey. Maybe it's
• 14:35
  - time you know, we can see in your
• 14:37
  - documentation you're thinking the diagnosis is x.
• 14:40
  - The risk or the likelihood of it being
• 14:42
  - y has now hit a threshold. And I
• 14:44
  - wanna just drop you a nudge to say,
• 14:46
  - reassess.
• 14:47
  - Now that is that is bread and butter
• 14:49
  - ER practice.
• 14:50
  - But I think any any, you know, any
• 14:52
  - comfortable,
• 14:55
  - ER doc will tell you that we can
• 14:57
  - anchor bias. So
• 14:59
  - we welcome those nudges
• 15:00
  - when appropriate, when well tuned. And when when,
• 15:03
  - frankly, they aren't imperfect. We're not looking for
• 15:05
  - a diagnosis machine. But the opportunity to offer
• 15:09
  - a nudge away from anchoring bias is incredibly
• 15:11
  - powerful. And frankly, it's a it's a
• 15:14
  - blanket of security that the patients and the
• 15:16
  - nurse and the doctors are finding together in
• 15:18
  - the ERs.
• 15:22
  - That's incredible to hear about it. And and,
• 15:24
  - really, it seems like it'd be helpful,
• 15:27
  - across the board for folks in in the
• 15:30
  - clinicians working in the ER as well as,
• 15:32
  - you know, the the patients too. So, thank
• 15:35
  - you for sharing that example, and and we'll
• 15:37
  - be excited to hear more about how it
• 15:38
  - goes as you continue to roll things out.
• 15:41
  - And, you know, in in thinking about the
• 15:43
  - future, what are you most curious about or
• 15:45
  - or what rises to the top of some
• 15:46
  - of the things that you're keeping your eye
• 15:48
  - close on as time goes on?
• 15:54
  - Most curious about that. That's a that's a
• 15:55
  - great one. I
• 15:58
  - as the evidence mounts
• 16:01
  - that
• 16:03
  - having a clinician in the loop may degrade
• 16:06
  - performance,
• 16:08
  - Inevitable. It will it is inevitable. We will
• 16:10
  - find circumstances where the addition of a
• 16:13
  - clinician
• 16:14
  - degrades an AI model's performance.
• 16:18
  - And the ethical quandary of when we
• 16:21
  - maybe need to remove the person in the
• 16:23
  - loop to ensure higher quality.
• 16:26
  - We're getting really close to that moment.
• 16:28
  - Am I suggesting we should set these models
• 16:30
  - loose right now? Absolutely not. We have a
• 16:32
  - lot to learn and our patients expect for.
• 16:35
  - But I'm very curious as to how
• 16:38
  - the,
• 16:41
  - frankly,
• 16:42
  - disruptive
• 16:42
  - nature to our
• 16:44
  - definition of professionalism and our definition of who
• 16:47
  - we are as clinicians
• 16:49
  - is gonna be shaken
• 16:51
  - When there are moments we need to step
• 16:53
  - back and say,
• 16:54
  - the central thing for me to do is
• 16:56
  - to let this
• 16:58
  - process run and mean not to interfere, but
• 17:01
  - still have the courage to say, I need
• 17:03
  - to validate. I need to be there.
• 17:05
  - So
• 17:06
  - more complex way of things just saying, human
• 17:09
  - the loop, we're gonna learn isn't always always
• 17:11
  - the protection we think it is.
• 17:14
  - And as we're learning with these models,
• 17:17
  - I'm very curious as to what it means
• 17:19
  - to be a clinician and how we're gonna
• 17:21
  - define our role working with these models and
• 17:24
  - with these tools as is almost every profession.
• 17:27
  - I'm no word about being replaced.
• 17:29
  - And if it means to be the provision
• 17:32
  - of safer care to replace me in that,
• 17:33
  - then so be it. Replace me.
• 17:35
  - But I I think what I'm most curious
• 17:38
  - about in the next one to three years
• 17:41
  - is how we move into that space of
• 17:44
  - acknowledging, at times, the performance
• 17:47
  - of these models
• 17:49
  - outperform me,
• 17:51
  - outperform a team,
• 17:53
  - and therefore, we have to understand what its
• 17:55
  - role here would be moving forward.
• 17:59
  - Well, that's such an interesting question. And, you
• 18:02
  - know, it sounds like you see that horizon
• 18:04
  - not too far off in the next one
• 18:06
  - to three years as as the technology continues
• 18:08
  - to evolve. And, you know, when you're looking
• 18:10
  - at it,
• 18:12
  - what types of discussions do you have with
• 18:14
  - your colleagues around this, or how do you
• 18:15
  - really see,
• 18:17
  - the humans in the loop being able to
• 18:20
  - make that estimation and and figure out what's
• 18:23
  - gonna be most impactful for patients without putting
• 18:25
  - them too much at risk?
• 18:29
  - Well, here's the great thing.
• 18:31
  - As a profession as professions,
• 18:34
  - doctors and nurses have been assessing
• 18:36
  - what tools we bring to bear for the
• 18:38
  - betterment of our patients for century.
• 18:41
  - So I think we're we have the rubric,
• 18:44
  - if not
• 18:45
  - the line by line assessment tools.
• 18:48
  - We have a model of questioning,
• 18:51
  - proving,
• 18:53
  - obtaining trust,
• 18:55
  - and deploying.
• 18:57
  - And that's what I'm hearing from my colleagues
• 18:59
  - is we're still in the trust building phase,
• 19:01
  - and I think that's appropriate.
• 19:04
  - And we've learned a lot from basic CDS,
• 19:07
  - from EHR implementations,
• 19:09
  - from radiology
• 19:10
  - at this for thirty plus years, from
• 19:12
  - from the digitization of pathology,
• 19:15
  - in the last year, many of the most
• 19:16
  - recent years.
• 19:17
  - So we're we're we're we're reasonably good skeptics
• 19:20
  - with an ability to prove, and we have
• 19:22
  - a scientific process to do that.
• 19:24
  - I think we're definitely still in the in
• 19:27
  - the trust building phase for both our patients,
• 19:30
  - and our on our peer clinicians.
• 19:33
  - And that's what they want is an explanation
• 19:35
  - of how these things
• 19:37
  - are doing what they do
• 19:38
  - and when do they really feel when they
• 19:40
  - earn the level of trust that we can
• 19:42
  - hand over some of the control.
• 19:44
  - Again, we're not there yet, but I'm not
• 19:47
  - I don't think it's five to ten to
• 19:49
  - fifteen years away. I think it's it's much
• 19:51
  - closer than that.
• 19:53
  - So the reckoning is coming is coming, and
• 19:55
  - that reckoning will only be,
• 19:58
  - resolved or acknowledged or moved past once we
• 20:00
  - build the trust together.
• 20:04
  - That's fascinating to hear. Paul, thank you so
• 20:06
  - much for joining us on the podcast today.
• 20:08
  - This has been such a fascinating discussion. I
• 20:11
  - think you've shared some really unique examples of
• 20:13
  - ways technology and AI is is,
• 20:15
  - helping the organization and really making and and
• 20:18
  - transforming patient care and then looking into the
• 20:20
  - future, what the possibilities hold. So
• 20:23
  - thank you so much for your time, and
• 20:24
  - I look forward to connecting with you as
• 20:25
  - well in person at our health IT and
• 20:27
  - revenue cycle conference coming up in September. I
• 20:30
  - know you'll be speaking on a panel, and
• 20:32
  - it'll be great to kinda dig deeper into
• 20:34
  - some of these themes, and,
• 20:35
  - I'll enjoy seeing you there.
• 20:38
  - Thanks for the privilege, Mark.