Latest / Becker’s Healthcare Digital Health + Health IT / Joseph Izzo, MD, Chief Medical Information Officer at San Joaquin General Hospital

Transcript

Joseph Izzo, MD, Chief Medical Information Officer at San Joaquin General Hospital · Becker’s Healthcare Digital Health + Health IT

• 0:00
  Philips is a health tech leader focused on
• 0:03
  innovation that improves the health and well-being of
• 0:06
  people. Our health care technology and informatics solutions
• 0:09
  help care teams diagnose, treat, and manage more
• 0:12
  patients with greater precision, speed, and confidence across
• 0:16
  the care journey. With Philips, clinicians are empowered
• 0:19
  with streamlined insights in the moments that matter
• 0:21
  for every patient. Better care for more people.
• 0:25
  Philips.
• 0:28
  This is Gracelyn Keller with the Becker's Healthcare
• 0:30
  podcast, and I am excited now to be
• 0:33
  joined by doctor Joseph Izzo, who is the
• 0:35
  chief medical information officer at San Joaquin General
• 0:38
  Hospital. So, doctor Izzo, thank you for joining
• 0:40
  me today.
• 0:41
  Let's get started by having you please share
• 0:43
  a bit about yourself and your background and
• 0:45
  your role in your organization.
• 0:47
  Yeah. Thank you for having me. So as
• 0:50
  you said, I'm the chief medical information officer
• 0:52
  for San Joaquin.
• 0:53
  We're a level 2 trauma center and a
• 0:55
  public hospital in the Central Valley of California.
• 0:58
  I'm also a practicing physician. I've worked in
• 1:00
  the
• 1:01
  emergency department for over 6 years. And I
• 1:03
  actually got my start, in computer science before
• 1:05
  I transitioned into medicine. And since that, I've
• 1:07
  been really working at the intersection between health
• 1:10
  care and technology.
• 1:11
  But particularly to my role, I oversee the
• 1:13
  health informatics and the clinical reporting teams. And
• 1:16
  you might say that our efforts are mostly
• 1:18
  directed at optimizing how clinicians engage with technology,
• 1:22
  fostering innovation, and and really ensuring clinical data
• 1:25
  stewardship. So I I do hope with that
• 1:27
  that my insights are valuable to the listeners.
• 1:29
  And again, I I appreciate you having me.
• 1:31
  Absolutely. Well, thank you for taking the time.
• 1:33
  And let's start our conversation today with AI
• 1:36
  adoption. So as we know, this is exploding
• 1:38
  in health care right now. So in your
• 1:40
  view, what is the most significant or promising
• 1:42
  application of this technology,
• 1:44
  and how is this informing your organization's innovation
• 1:47
  strategy?
• 1:49
  Right. You know, AI to me is excelling
• 1:51
  in spaces
• 1:52
  that are traditionally considered, I think, the administrative
• 1:55
  burden of medicine, especially,
• 1:57
  these areas that are very important to clinical
• 1:59
  medicine but are far enough,
• 2:01
  away from patient care where I think the
• 2:03
  margin for error is a little bit more
• 2:04
  forgiving. You know, I I don't think the
• 2:06
  tools are quite where they need to be
• 2:07
  right now for direct patient care. And, of
• 2:09
  course, not everyone may feel that way, and
• 2:11
  that's certainly an institutional decision. But these areas
• 2:14
  are traditionally burdensome to providers too, so it's
• 2:16
  it's a bit of a win win. I
• 2:18
  think the most pressing example that everyone's familiar
• 2:20
  with is AI scribes, but I will say
• 2:22
  that that really is the case. You know?
• 2:24
  In 2 of the depart emergency departments
• 2:27
  I work in, and
• 2:29
  even in our pilots at the hospital, the
• 2:29
  feedback is is just tremendous. You know, I've
• 2:31
  received paragraphs of gratitude from some of the
• 2:33
  physicians who have piloted the products, and it's
• 2:35
  more or less a day and night diff
• 2:36
  difference, for how their their charting changes and
• 2:39
  and how much administrative burden is on them.
• 2:42
  But I've also been really impressed with some
• 2:44
  of the recent work in in, medical record
• 2:46
  space, revenue cycle,
• 2:48
  clinical documentation improving, even automating nursing processes, and
• 2:52
  we've been exploring that at our hospital. And
• 2:54
  separately too from my role at San Joaquin,
• 2:56
  I've done some fractional work with large language
• 2:58
  models and agentic AI, and I've seen tremendous
• 3:01
  success too in areas of quality and safety
• 3:03
  auditing. And I've what I've noticed in all
• 3:05
  of these spaces, what it's doing is freeing
• 3:07
  up, staff and clinicians to do more creative
• 3:09
  and patient focused tasks.
• 3:11
  But, to your other question, for us to
• 3:13
  stay strategic, we really first had to start
• 3:15
  with, a robust executive governance structure and policy,
• 3:18
  and we really do vet any of these
• 3:20
  AI tools, for the reasons I mentioned. And
• 3:22
  I I personally authored the policy for a
• 3:24
  hospital, and I've been very proactive in getting
• 3:26
  clinician leaders involved in the dialogue,
• 3:29
  particularly with a physician innovation group who vets
• 3:31
  the products, try to attend many of the
• 3:33
  vendor conferences, including Becker's.
• 3:36
  Take attend local meetups and societies, and really
• 3:38
  stay informed so that we can make the
• 3:39
  best decisions on this technology. But either way,
• 3:42
  our strategy is that we are moving forward,
• 3:44
  and we wanna take advantage of these tools,
• 3:46
  that are promising.
• 3:48
  Absolutely.
• 3:49
  And kind of going off of this AI
• 3:51
  AI adoption,
• 3:52
  technological advancements within health care, we see that
• 3:55
  leaders are managing greater volumes of data and
• 3:58
  more devices on a growing number of care
• 4:00
  settings and populations. And this is a complex
• 4:03
  environment. So what clinical data integration tools or
• 4:06
  practices are you seeing drive improvements
• 4:09
  in patient outcomes? And can you share some
• 4:12
  examples of this?
• 4:14
  Yeah. Certainly. The truth is, I I think
• 4:16
  our organization is is actually limited in this
• 4:18
  space with very small teams of analysts. You
• 4:20
  know, we're a public hospital, like I mentioned,
• 4:21
  and that's not too uncommon from talking to
• 4:24
  other individuals from those spaces.
• 4:26
  And I end up, helping out with a
• 4:28
  fair number of the coding, the report writing
• 4:30
  myself.
• 4:31
  But on the plus side, I think that
• 4:32
  means that our teams are really close knit
• 4:34
  and really understand the data and are able
• 4:35
  to make changes so,
• 4:37
  that we do drive these improvements you mentioned.
• 4:40
  And in that context, what's what's really worked
• 4:41
  for us is being
• 4:43
  quite intentional,
• 4:44
  about reconciling all of these sources of data
• 4:47
  that you mentioned
• 4:48
  and kind of ingesting them
• 4:50
  into really only 1 or 2 sources of
• 4:52
  of truth. You know, if there's a census
• 4:54
  that says a 170 patients
• 4:56
  were at in the hospital at any given
• 4:59
  moment, it should be true across any data
• 5:00
  we query. And,
• 5:02
  for us, we do use our our EMR's
• 5:04
  cloud enterprise data warehouse for that purpose. But
• 5:07
  I've seen other successful strategies where,
• 5:10
  some of our clinics have done their own
• 5:11
  homegrown versions and similar efforts like that. But
• 5:14
  like you mentioned, there there's so many,
• 5:16
  different areas to pull data from and and,
• 5:18
  sources of data. And another thing in that
• 5:20
  in that vein is is actually being really
• 5:22
  realistic with workflows
• 5:24
  because I I think, everyone's familiar that most
• 5:26
  EMRs have a number of ways to maybe
• 5:28
  document or perform the same task. And so
• 5:30
  if there's, say, 5 different ways to document
• 5:32
  a fall risk, then that's effectively 5 potentially
• 5:35
  different locations on the back end to find
• 5:37
  that data. And for us, standardizing these workflows
• 5:40
  has been really important.
• 5:42
  Now, of course, there's a delicate balance between
• 5:44
  forcing folks to do something that is laborious,
• 5:46
  but produces the perfect data, and then optimizing
• 5:49
  their workflows. And I think we've learned really
• 5:51
  to walk that line.
• 5:53
  The example that really comes to mind in
• 5:55
  in this space is actually about using interpreter
• 5:57
  service,
• 5:58
  for our patient. And, you know, we know
• 6:00
  we're doing it, but a problem is really
• 6:02
  being able to demonstrate that we are, I
• 6:04
  say, for the case of auditing or for
• 6:06
  quality improvement. Because like you mentioned, this is
• 6:08
  so much disparate data. There's data from the
• 6:10
  health plan. There's data from interpreter and service
• 6:13
  networks, from registration, from nursing notes, from physician,
• 6:17
  progress notes, and so on. And however, by
• 6:19
  ingesting this all into one place, and then
• 6:21
  coming up with standard,
• 6:23
  reasonable, but expected workflows, we've actually been able
• 6:26
  to address this problem
• 6:28
  and make very good visualizations
• 6:30
  for the units to see how their staff
• 6:32
  are at documenting this data. And we've seen
• 6:34
  tremendous progress. Some departments actually went from ostensibly
• 6:38
  very low compliance in in in documenting that
• 6:40
  they were using interpreter service to near a
• 6:43
  100% just by able to have visibility
• 6:45
  and by having continuous quality improvement. And the
• 6:47
  truth is a lot of our efforts are
• 6:49
  structured just like this.
• 6:51
  Absolutely.
• 6:52
  And kind of going off of both of
• 6:54
  those points,
• 6:56
  I would love to know how you think
• 6:57
  that health care organizations can better support
• 7:00
  IT and clinical teams as we carry out
• 7:02
  innovation
• 7:03
  efforts. And then what are some of the
• 7:05
  common pitfalls that you're seeing here?
• 7:08
  Sure. I think, like I mentioned, we found
• 7:10
  the the most important first step is really
• 7:12
  to have that governance structure in place to
• 7:14
  protect these teams and gives them that space
• 7:16
  to to really explore. You want to be
• 7:19
  on the same page as an institution as
• 7:20
  you've got new technology.
• 7:22
  And I think it's also important to really
• 7:23
  foster communication
• 7:25
  between these teams.
• 7:27
  It really isn't helpful if these teams are
• 7:29
  working in silos and pursuing technology that may
• 7:31
  not be best for the whole organization.
• 7:34
  As a example, the clinician innovation group I
• 7:36
  mentioned regularly communicates with IT and the executive
• 7:39
  teams to make sure we're in agreement.
• 7:42
  On the flip side, actually, though, there is
• 7:44
  a tendency I've seen, maybe being a little
• 7:46
  bit too bureaucratic or overprotective,
• 7:48
  and I think it's something to be aware
• 7:49
  of. It's probably the most common pitfall I've
• 7:51
  seen, in my experiences.
• 7:53
  And to be fair, you know, a lot
• 7:55
  of these tools, particularly the AI ones, are
• 7:57
  a little intimidating at first glance, and that
• 7:59
  probably is what motivates the desire to be
• 8:01
  protective.
• 8:02
  But what we realize is even if we
• 8:03
  think we're being protected by being too strict,
• 8:05
  that will potentially limit innovation. And so as
• 8:08
  we've grown in this journey, we've learned to
• 8:10
  better strike that balance. But ultimately, I think
• 8:12
  it's important really to trust these teams. And
• 8:14
  more importantly, I I would say listen to
• 8:15
  them. You know? Give them that leeway to
• 8:17
  explore, but also that infrastructure to ensure the
• 8:19
  success.
• 8:21
  Absolutely. And as we wrap our conversation up
• 8:24
  today, I'd love to end on your top
• 8:26
  piece of advice for healthcare leaders as they
• 8:28
  prepare to
• 8:29
  take on further advancements in technology and greater
• 8:32
  demands for care in the future.
• 8:35
  Sure. I think it's really important,
• 8:37
  to stay informed of the technology and these
• 8:39
  use cases and the applications,
• 8:42
  and really be aware of the limitations. You
• 8:43
  know, it's something I try to work into
• 8:45
  my role daily, because it's really easy to
• 8:47
  fall into a trap maybe of pursuing this
• 8:49
  technology,
• 8:50
  like AI for the sake of AI, which
• 8:53
  is probably another common pitfall to your previous
• 8:55
  question, as well as maybe products that overpromise
• 8:57
  and don't necessarily deliver. And I think I
• 9:00
  found the best way for me and my
• 9:01
  team to insulate against that is to really
• 9:03
  be as informed as possible. And I think
• 9:05
  the time is now
• 9:07
  to do that, especially if you're listening to
• 9:08
  this podcast.
• 9:09
  And I feel like the barrier to entry
• 9:11
  is a lot lower than it's ever been
• 9:12
  with some of the general available tools such
• 9:14
  as large language models. But, just like your
• 9:16
  last question, ultimately, I I keep coming back
• 9:18
  to the goal should really be to build
• 9:20
  up these clinical and technology leaders, listen to
• 9:22
  their guidance, and give them that space to
• 9:24
  to work with these tools within the proper
• 9:26
  guardrails and governance to feel protected.
• 9:29
  Well, doctor Izzo, thank you so much for
• 9:31
  taking the time today to join me on
• 9:33
  the Becker's HealthCare podcast and share your insights
• 9:35
  with our listeners.
• 9:37
  Yeah. Thank you again for having me. I
• 9:39
  really appreciate it.