Latest / Becker’s Healthcare Digital Health + Health IT / Using AI and Predictive Analytics to Turn Healthcare Data Into Action with Teresa Freeman

Transcript

Using AI and Predictive Analytics to Turn Healthcare Data Into Action with Teresa Freeman · Becker’s Healthcare Digital Health + Health IT

• 0:00
  - What we wanna do with AI is help
• 0:01
  - clinicians go from being reactive
• 0:04
  - to proactive.
• 0:05
  - At Philips, we can recognize patterns in monitoring
• 0:08
  - data across populations
• 0:10
  - and apply those insights to specific patients.
• 0:14
  - Safe and responsible AI can support sharper insights,
• 0:18
  - pave the way for faster diagnosis,
• 0:21
  - and help streamline care.
• 0:23
  - That can benefit clinicians, it can benefit health
• 0:26
  - systems. And most importantly, it can benefit patients
• 0:29
  - and their families.
• 0:33
  - Phillips.
• 0:34
  - Hi, everyone. This is Scott King with Becker's
• 0:35
  - Healthcare. Thank you so much for tuning in
• 0:37
  - to the Becker's Healthcare podcast series. Joining me
• 0:40
  - for today's discussion
• 0:41
  - is Theresa Freeman,
• 0:43
  - chief nursing information officer and vice president, West
• 0:47
  - Tennessee Health Care. Doctor Freeman, thanks so much
• 0:49
  - for joining us. It's great to have you
• 0:50
  - on the podcast. Good to be here.
• 0:52
  - I appreciate it. And, you know, before we
• 0:53
  - dive into these questions, I definitely wanna get
• 0:55
  - you taken a lot of things going on
• 0:56
  - in health care right now. But before I
• 0:57
  - do, I was wondering if you can just
• 0:58
  - kinda share a little about your career journey
• 1:00
  - and your background. Sure.
• 1:02
  - I am a nurse. I have been a
• 1:04
  - nurse,
• 1:04
  - for thirty nine years
• 1:06
  - and have really enjoyed it. Been a great,
• 1:09
  - great journey.
• 1:10
  - But I also went back to school and
• 1:11
  - got an MBA, and that led me in
• 1:13
  - a few different directions. I even worked in
• 1:15
  - accounting for a while, because they needed somebody
• 1:17
  - with some clinical experience. I also... In addition
• 1:20
  - to my, day job, again, of being the
• 1:22
  - nursing information officer for West Tennessee Health Care,
• 1:24
  - I also teach for one of our local
• 1:26
  - colleges of nursing adjunct faculty.
• 1:28
  - So, yeah, stay busy. That's great. You've worn
• 1:31
  - a lot of hats. Yeah. That's great. We're
• 1:32
  - gonna we're gonna need some of those hats
• 1:33
  - today here on the podcast.
• 1:35
  - The first topic I wanna get to is
• 1:37
  - health systems are receiving
• 1:39
  - a growing stream of patient data
• 1:41
  - from bedside monitoring, ambulatory devices, and care delivered
• 1:44
  - at home. Where is that data creating actionable
• 1:47
  - insights for clinicians,
• 1:49
  - and where does it still create more work
• 1:51
  - than insight? Yeah.
• 1:53
  - This is a question right up my alley.
• 1:55
  - I am an extremely analytical person by nature.
• 1:58
  - And if you've ever heard the term paralysis
• 2:00
  - by analysis,
• 2:02
  - that describes me to a t. I'm one
• 2:04
  - of those people that you can put a
• 2:05
  - multi tab spreadsheet in front of, and I
• 2:07
  - will just nerd out for hours.
• 2:10
  - Therein lies the problem. Sometimes we get so
• 2:12
  - much data,
• 2:14
  - that we can get paralyzed by looking at
• 2:16
  - it and trying to define it.
• 2:18
  - The EMR that we use actually has a
• 2:21
  - feature called slicer dicer, and I probably just
• 2:23
  - gave it away, the EMR that we use
• 2:25
  - to Yeah. To most folks out there.
• 2:27
  - But I joke with people that I'm always
• 2:30
  - one click away from brilliance or have an
• 2:32
  - absolutely nothing,
• 2:34
  - and sometimes can spend a lot of time
• 2:36
  - trying to get there.
• 2:38
  - So,
• 2:39
  - you know, we now have predictive analytics, we
• 2:41
  - have AI. Those are the tools that we
• 2:44
  - need to start using that will kind of
• 2:46
  - get rid of some of that noise that
• 2:47
  - we see when we have too much data
• 2:49
  - available,
• 2:50
  - and it will kinda start bringing those things
• 2:52
  - that we're really looking forward to the top.
• 2:54
  - But one piece of advice that I have,
• 2:55
  - especially for clinicians,
• 2:57
  - work with the people, those techie people that
• 2:59
  - are building the reports,
• 3:01
  - that are doing the mapping that's gonna get
• 3:04
  - us that information that we need. Tell them
• 3:06
  - upfront
• 3:07
  - what we want. And that way, if we
• 3:09
  - will put in that work upfront,
• 3:11
  - then it will make more actionable data for
• 3:14
  - us on the back end.
• 3:15
  - Yeah. And just kinda sticking with clinicians. You
• 3:17
  - know, clinicians are are trying to reach a
• 3:19
  - diagnosis.
• 3:20
  - When they do so, they must often piece
• 3:22
  - together information from radiology, cardiology, pathology, urology, and
• 3:25
  - other systems. Where does that fragmentation
• 3:28
  - create the most friction in your organization?
• 3:31
  - And how do you bring together data sources
• 3:33
  - and address this fragmentation or close the gaps?
• 3:36
  - Sure. I can tell you that, again, with
• 3:39
  - a nursing background, I hear patient comments
• 3:42
  - and complaints a a lot,
• 3:44
  - and and still read them on a regular
• 3:46
  - basis. And one of the most consistent complaints
• 3:48
  - that we get is that our physicians are
• 3:50
  - not talking to one another. Our providers are
• 3:52
  - not talking. I'm not gonna blame it on
• 3:53
  - physicians, but our our caregivers overall, they're not
• 3:55
  - talking to each other. They're not communicating,
• 3:57
  - and it doesn't always seem like one knows
• 3:59
  - what the other one is doing.
• 4:01
  - But even though I know that this whole
• 4:03
  - conference is about technology and AI,
• 4:06
  - what this is actually describing is not necessarily
• 4:08
  - a breakdown in technology. It's just a breakdown
• 4:10
  - in human communication,
• 4:12
  - and we kinda need to get back to
• 4:13
  - that. How often have you heard the phrase,
• 4:15
  - well, I talked to so and so today,
• 4:17
  - but what they really mean is I sent
• 4:18
  - them an email or I sent them a
• 4:19
  - text. Right. Right. There was no talking.
• 4:22
  - So, one great thing that technology is bringing
• 4:25
  - us, and again, the EMR that we use
• 4:27
  - uses something called insights which provides
• 4:30
  - a patient
• 4:31
  - chart summary.
• 4:33
  - So it kinda stops all that looking around
• 4:35
  - that people have to be doing and kinda
• 4:37
  - brings everything. Lab results,
• 4:40
  - x-ray, you know, all of that stuff is
• 4:42
  - in one spot. So every clinician, if they
• 4:45
  - will take the time, can look at that
• 4:47
  - and get a summary of what's going on
• 4:49
  - with the patient. But bottom line, it's still
• 4:51
  - never gonna replace two people talking to one
• 4:54
  - another.
• 4:55
  - Yeah. Absolutely. It's like I talked to so
• 4:56
  - and so. It was a one sentence text.
• 4:58
  - Exactly. Yeah. That. No. I know exactly what
• 5:00
  - you mean. You you meant... You were mentioning
• 5:02
  - patients. You know, patients increasingly move among the
• 5:05
  - hospital Mhmm. Outpatient settings and home. Where does
• 5:08
  - clinical context most often get lost during those
• 5:11
  - transitions,
• 5:12
  - and what would it take to give care
• 5:13
  - teams a more continuous view of the patient?
• 5:16
  - Sure. I mean, absolutely correct.
• 5:19
  - We do have a very mobile society, and
• 5:21
  - so they do see a lot of physicians.
• 5:23
  - And we also... You know, we live in
• 5:25
  - a social media and, Google and chat GPT.
• 5:28
  - Every patient before they ever go to a
• 5:30
  - doctor has already self diagnosed themselves. Right. And
• 5:33
  - if they don't get the diagnosis they want
• 5:35
  - from one, they could go to another.
• 5:37
  - So, yeah, it's it's an absolutely it's absolutely
• 5:39
  - no problem.
• 5:40
  - But I think it is becoming less common
• 5:42
  - than it than it used to be. One
• 5:44
  - thing is a lot of EMS
• 5:47
  - and,
• 5:48
  - EMRs,
• 5:49
  - they talk to one another
• 5:51
  - and share information. And that's the same with
• 5:53
  - the vendors. A lot of information shared. Again,
• 5:55
  - our EMR,
• 5:57
  - we actually have something that is called care
• 5:59
  - everywhere.
• 6:00
  - And so if a patient has ever been
• 6:02
  - in another facility that uses that, say, on
• 6:05
  - my EMR, we can pull it up right
• 6:07
  - there, and see exactly what happened. So a
• 6:09
  - lot a lot better than it used to
• 6:11
  - be.
• 6:12
  - With so many AI AI use cases competing
• 6:14
  - for attention, how do you distinguish a compelling
• 6:17
  - pilot from something ready to become part of
• 6:19
  - everyday care,
• 6:21
  - and what must be in place to scale
• 6:22
  - it responsibly?
• 6:23
  - Sure. Well, again, I work for West Tennessee
• 6:26
  - Health Care. And when you compare us to
• 6:28
  - some of the giant health systems out there,
• 6:29
  - we're relatively small, eight acute care hospitals.
• 6:33
  - So we are not typically what you would
• 6:34
  - call an early adopter.
• 6:36
  - We kinda wait for the big guys to
• 6:38
  - do a lot of the pilots and and
• 6:40
  - work the bugs out. And so after that,
• 6:43
  - then we come in and and take those.
• 6:45
  - So we don't, in our our system, we
• 6:48
  - don't typically end up with that problem most
• 6:50
  - of the time. Sure. Sure.
• 6:52
  - As technology relationships move beyond one time purchases,
• 6:55
  - what should a partner remain accountable for after
• 6:58
  - go live to make their relationship genuinely valuable
• 7:01
  - over time?
• 7:02
  - So the specific question you asked talks about
• 7:04
  - after go live, but I'm gonna kinda back
• 7:06
  - that up a little bit.
• 7:08
  - Primarily, they have to be a good partner
• 7:10
  - while you're working on the project.
• 7:12
  - They need to listen to you and try
• 7:14
  - to work with you to provide the services
• 7:16
  - that you need. Once that relationship has been
• 7:18
  - established,
• 7:19
  - then post to go live, they still need
• 7:21
  - to be present. You know, routine check ins
• 7:23
  - with you to see... Make sure that everything
• 7:25
  - is going well. And then again, if it's
• 7:27
  - not, work with you to to make sure
• 7:29
  - that it is. And one other thing, again,
• 7:31
  - kinda personally very important to me is
• 7:33
  - please don't blindside me. Keep me updated on
• 7:36
  - upgrades, changes that you're gonna be having, things
• 7:38
  - that are gonna sunset. You know, don't blindside
• 7:41
  - me in an email to tell me you're
• 7:42
  - fixing to shut my system off, and then
• 7:44
  - I'm gonna have to buy something new. So
• 7:45
  - that's what I look for in a vendor.
• 7:46
  - Yeah. No bad surprises. Exactly. And the last
• 7:49
  - question I have for you, doctor Freeman,
• 7:51
  - what's the future of integrated diagnostics or just
• 7:53
  - diagnostic services,
• 7:55
  - and where do you see yourself in the
• 7:56
  - next five years?
• 7:58
  - Well, you call me doctor Freeman, but my
• 8:00
  - my doctorate is in nursing. So,
• 8:02
  - I'm not a diagnostician, so to speak. That's
• 8:05
  - not that's not the scope of a nurse.
• 8:07
  - But,
• 8:08
  - you know, AI
• 8:09
  - is giving us so much these days.
• 8:12
  - It does allow, you know, again, our EMRs
• 8:14
  - to go out, search the chart, look at
• 8:16
  - what's going on with patients, and then make
• 8:17
  - suggestions. But I think that's the key point
• 8:19
  - is there that we need to understand that
• 8:22
  - these are suggestions
• 8:24
  - and that the clinician that has that knowledge
• 8:26
  - and that medical degree, that background is ultimately
• 8:29
  - the one that has to got... Has got
• 8:30
  - to make that final decision.
• 8:32
  - Thank you, doctor Freeman, for your time and
• 8:34
  - insights today. We also wanna thank our podcast
• 8:36
  - sponsor, Philips. You can tune in to more
• 8:39
  - podcasts from Becker's Healthcare by visiting our podcast
• 8:41
  - page at beckershospitalreview.com.