26d82b3c-6769-49ed-94ea-40002ca91529/23-0
00:00:03.136 --> 00:00:07.145
We're here today with Doctor
Michael Kirchoff of Cooper

26d82b3c-6769-49ed-94ea-40002ca91529/23-1
00:00:07.145 --> 00:00:07.646
Health.

26d82b3c-6769-49ed-94ea-40002ca91529/34-0
00:00:08.016 --> 00:00:11.126
He is the chief innovation and
Patient safety officer.

26d82b3c-6769-49ed-94ea-40002ca91529/46-0
00:00:11.136 --> 00:00:12.266
Michael, thank you for joining
us today.

26d82b3c-6769-49ed-94ea-40002ca91529/52-0
00:00:12.726 --> 00:00:13.566
Thank you for having me.

26d82b3c-6769-49ed-94ea-40002ca91529/84-0
00:00:14.186 --> 00:00:18.176
So for those who don't know,
Cooper University Healthcare is

26d82b3c-6769-49ed-94ea-40002ca91529/84-1
00:00:18.176 --> 00:00:21.707
an academic health system,
pacing Camden, NJ with 900

26d82b3c-6769-49ed-94ea-40002ca91529/84-2
00:00:21.707 --> 00:00:23.276
physicians and 663 beds.

26d82b3c-6769-49ed-94ea-40002ca91529/119-0
00:00:23.646 --> 00:00:28.259
Today we have a dual topics that
I think will be of interest to

26d82b3c-6769-49ed-94ea-40002ca91529/119-1
00:00:28.259 --> 00:00:32.583
many of your peers across the
United States, and that those

26d82b3c-6769-49ed-94ea-40002ca91529/119-2
00:00:32.583 --> 00:00:35.466
topics are metadata and
technical debt.

26d82b3c-6769-49ed-94ea-40002ca91529/147-0
00:00:36.246 --> 00:00:40.660
Many organizations have a
laundry list of things that they

26d82b3c-6769-49ed-94ea-40002ca91529/147-1
00:00:40.660 --> 00:00:45.148
like to accomplish, and Dr
Kirchoff, you've been working to

26d82b3c-6769-49ed-94ea-40002ca91529/147-2
00:00:45.148 --> 00:00:47.466
ameliorate your technical debt.

26d82b3c-6769-49ed-94ea-40002ca91529/156-0
00:00:47.536 --> 00:00:50.166
Tell our lesson ners kind of
what?

26d82b3c-6769-49ed-94ea-40002ca91529/167-0
00:00:50.216 --> 00:00:53.166
You know what is the breadth of
your technical debt?

26d82b3c-6769-49ed-94ea-40002ca91529/172-0
00:00:53.176 --> 00:00:54.146
How do you prioritize?

26d82b3c-6769-49ed-94ea-40002ca91529/189-0
00:00:54.156 --> 00:00:56.724
How are you tackling this
problem and innovative way over

26d82b3c-6769-49ed-94ea-40002ca91529/189-1
00:00:56.724 --> 00:00:57.166
at Cooper?

26d82b3c-6769-49ed-94ea-40002ca91529/190-0
00:00:58.016 --> 00:00:58.486
Sure.

26d82b3c-6769-49ed-94ea-40002ca91529/235-0
00:00:58.496 --> 00:01:02.501
So, you know, in general, I
think a lot of organizations are

26d82b3c-6769-49ed-94ea-40002ca91529/235-1
00:01:02.501 --> 00:01:06.375
grappling with technical debt
and really the point of view

26d82b3c-6769-49ed-94ea-40002ca91529/221-0
00:01:04.946 --> 00:01:04.986
E.

26d82b3c-6769-49ed-94ea-40002ca91529/235-2
00:01:06.375 --> 00:01:10.446
that I bring from an innovation
of patient safety side is the

26d82b3c-6769-49ed-94ea-40002ca91529/235-3
00:01:10.446 --> 00:01:13.006
safety implications of technical
debt.

26d82b3c-6769-49ed-94ea-40002ca91529/237-0
00:01:10.586 --> 00:01:10.856
That.

26d82b3c-6769-49ed-94ea-40002ca91529/282-0
00:01:13.016 --> 00:01:16.592
So we have a great team on the
IT side and on the budgetary

26d82b3c-6769-49ed-94ea-40002ca91529/282-1
00:01:16.592 --> 00:01:19.691
side of really addressing
technical debt and really

26d82b3c-6769-49ed-94ea-40002ca91529/282-2
00:01:19.691 --> 00:01:23.564
focusing on making sure that our
life cycle plans make sense and

26d82b3c-6769-49ed-94ea-40002ca91529/282-3
00:01:23.564 --> 00:01:26.126
are keeping up with the latest
technology.

26d82b3c-6769-49ed-94ea-40002ca91529/293-0
00:01:26.136 --> 00:01:28.516
I have no complaints about our
team there.

26d82b3c-6769-49ed-94ea-40002ca91529/342-0
00:01:28.526 --> 00:01:32.144
They really do a fantastic job
and our Co CEOs, especially

26d82b3c-6769-49ed-94ea-40002ca91529/342-1
00:01:32.144 --> 00:01:35.945
Doctor Mazzarelli, is really
focused on making sure that life

26d82b3c-6769-49ed-94ea-40002ca91529/342-2
00:01:35.945 --> 00:01:39.134
cycling not only from our
applications but also our

26d82b3c-6769-49ed-94ea-40002ca91529/324-0
00:01:36.466 --> 00:01:36.746
Yes.

26d82b3c-6769-49ed-94ea-40002ca91529/342-3
00:01:39.134 --> 00:01:42.199
hardware that we're really
focusing on that as an

26d82b3c-6769-49ed-94ea-40002ca91529/344-0
00:01:40.846 --> 00:01:40.986
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/342-4
00:01:42.199 --> 00:01:42.996
organization.

26d82b3c-6769-49ed-94ea-40002ca91529/396-0
00:01:43.266 --> 00:01:47.685
But what we talked about earlier
was there's this other side of

26d82b3c-6769-49ed-94ea-40002ca91529/362-0
00:01:46.786 --> 00:01:47.016
That's.

26d82b3c-6769-49ed-94ea-40002ca91529/396-1
00:01:47.685 --> 00:01:51.689
technical debt that I see
looking at patient safety and I

26d82b3c-6769-49ed-94ea-40002ca91529/373-0
00:01:50.346 --> 00:01:50.416
E.

26d82b3c-6769-49ed-94ea-40002ca91529/396-2
00:01:51.689 --> 00:01:55.348
hear about from other
organizations when we talk not

26d82b3c-6769-49ed-94ea-40002ca91529/378-0
00:01:52.306 --> 00:01:52.536
It.

26d82b3c-6769-49ed-94ea-40002ca91529/396-3
00:01:55.348 --> 00:01:59.076
only about technology, but also
about patient safety.

26d82b3c-6769-49ed-94ea-40002ca91529/427-0
00:01:59.206 --> 00:02:02.415
And that's specifically there is
another impact of technical

26d82b3c-6769-49ed-94ea-40002ca91529/434-0
00:01:59.886 --> 00:02:00.476
Yeah, E.

26d82b3c-6769-49ed-94ea-40002ca91529/427-1
00:02:02.415 --> 00:02:05.834
depth other than the fact that I
just have an application that's

26d82b3c-6769-49ed-94ea-40002ca91529/427-2
00:02:05.834 --> 00:02:08.096
that's not being supported
anymore, right?

26d82b3c-6769-49ed-94ea-40002ca91529/464-0
00:02:08.106 --> 00:02:11.297
That's what everybody knows
about, or I have a piece of

26d82b3c-6769-49ed-94ea-40002ca91529/464-1
00:02:11.297 --> 00:02:14.488
hardware that is now out of
warranty and I can't get it

26d82b3c-6769-49ed-94ea-40002ca91529/452-0
00:02:12.186 --> 00:02:12.396
And.

26d82b3c-6769-49ed-94ea-40002ca91529/464-2
00:02:14.488 --> 00:02:17.736
serviced and now I suddenly have
a large capital outlay.

26d82b3c-6769-49ed-94ea-40002ca91529/503-0
00:02:17.746 --> 00:02:21.080
That's the technical debt most
people are familiar with, and I

26d82b3c-6769-49ed-94ea-40002ca91529/503-1
00:02:21.080 --> 00:02:23.778
think a lot of healthcare
organizations are really

26d82b3c-6769-49ed-94ea-40002ca91529/503-2
00:02:23.778 --> 00:02:27.006
understanding that they need to
focus on that and understand

26d82b3c-6769-49ed-94ea-40002ca91529/503-3
00:02:27.006 --> 00:02:28.646
that they need to address that.

26d82b3c-6769-49ed-94ea-40002ca91529/506-0
00:02:27.206 --> 00:02:27.396
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/526-0
00:02:28.866 --> 00:02:32.328
But there's other implications,
and that's really what I wanted

26d82b3c-6769-49ed-94ea-40002ca91529/526-1
00:02:32.328 --> 00:02:35.086
to talk to your listeners and
viewers today about.

26d82b3c-6769-49ed-94ea-40002ca91529/582-0
00:02:35.096 --> 00:02:39.376
But just you know, there are
other impacts besides the bottom

26d82b3c-6769-49ed-94ea-40002ca91529/582-1
00:02:39.376 --> 00:02:43.726
line besides the money we need
to spend to keep our systems up

26d82b3c-6769-49ed-94ea-40002ca91529/556-0
00:02:41.256 --> 00:02:41.476
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/582-2
00:02:43.726 --> 00:02:47.453
to date and our applications
running properly and our

26d82b3c-6769-49ed-94ea-40002ca91529/582-3
00:02:47.453 --> 00:02:51.526
hardware safe from cyber attack
and still within warranty.

26d82b3c-6769-49ed-94ea-40002ca91529/591-0
00:02:51.536 --> 00:02:55.946
And that's when I postpone doing
some work.

26d82b3c-6769-49ed-94ea-40002ca91529/617-0
00:02:52.196 --> 00:02:53.746
That's and.

26d82b3c-6769-49ed-94ea-40002ca91529/607-0
00:02:56.076 --> 00:03:00.205
What are the implications for
workflows and patient flows and

26d82b3c-6769-49ed-94ea-40002ca91529/607-1
00:03:00.205 --> 00:03:01.536
patient information?

26d82b3c-6769-49ed-94ea-40002ca91529/624-0
00:03:01.866 --> 00:03:04.478
And I think that's the thing
that a lot of people lose focus

26d82b3c-6769-49ed-94ea-40002ca91529/624-1
00:03:04.478 --> 00:03:04.606
of.

26d82b3c-6769-49ed-94ea-40002ca91529/622-0
00:03:05.596 --> 00:03:06.026
Right.

26d82b3c-6769-49ed-94ea-40002ca91529/648-0
00:03:06.036 --> 00:03:08.891
So the main mission of a
hospital or healthcare delivery

26d82b3c-6769-49ed-94ea-40002ca91529/648-1
00:03:08.891 --> 00:03:11.845
system is to provide care
increasingly with the transition

26d82b3c-6769-49ed-94ea-40002ca91529/648-2
00:03:11.845 --> 00:03:12.846
to value based care.

26d82b3c-6769-49ed-94ea-40002ca91529/674-0
00:03:12.976 --> 00:03:16.158
It's also to keep patient
populations well and out of

26d82b3c-6769-49ed-94ea-40002ca91529/674-1
00:03:16.158 --> 00:03:19.752
receiving at least duplicative
care or potentially avoidable

26d82b3c-6769-49ed-94ea-40002ca91529/674-2
00:03:19.752 --> 00:03:20.046
care.

26d82b3c-6769-49ed-94ea-40002ca91529/739-0
00:03:20.796 --> 00:03:25.230
So could you walk our listeners
through a concrete use case or

26d82b3c-6769-49ed-94ea-40002ca91529/739-1
00:03:25.230 --> 00:03:29.663
anecdote of the last year or two
when you've had conversations

26d82b3c-6769-49ed-94ea-40002ca91529/739-2
00:03:29.663 --> 00:03:33.534
with leadership and with
clinicians on your team about

26d82b3c-6769-49ed-94ea-40002ca91529/739-3
00:03:33.534 --> 00:03:38.108
the implications of the patient
safety implications of technical

26d82b3c-6769-49ed-94ea-40002ca91529/739-4
00:03:38.108 --> 00:03:41.979
debt and how that actually
played out and affected the

26d82b3c-6769-49ed-94ea-40002ca91529/739-5
00:03:41.979 --> 00:03:45.356
prioritization of reducing that
technical debt?

26d82b3c-6769-49ed-94ea-40002ca91529/737-0
00:03:45.896 --> 00:03:46.486
Sure.

26d82b3c-6769-49ed-94ea-40002ca91529/764-0
00:03:46.556 --> 00:03:49.703
So a lot of times what happens
is if you're gonna incur

26d82b3c-6769-49ed-94ea-40002ca91529/764-1
00:03:49.703 --> 00:03:53.356
technical debt, there's there's
the things we just talked about.

26d82b3c-6769-49ed-94ea-40002ca91529/774-0
00:03:53.366 --> 00:03:55.446
But there's also a latent errors
built into that.

26d82b3c-6769-49ed-94ea-40002ca91529/811-0
00:03:55.456 --> 00:03:59.226
So for example, you're expanding
as an organization and you're

26d82b3c-6769-49ed-94ea-40002ca91529/811-1
00:03:59.226 --> 00:04:02.757
electronic health record has a
facility structure and that

26d82b3c-6769-49ed-94ea-40002ca91529/811-2
00:04:02.757 --> 00:04:06.467
facility structure, there's a
lot of overhead associated with

26d82b3c-6769-49ed-94ea-40002ca91529/811-3
00:04:06.467 --> 00:04:06.766
that.

26d82b3c-6769-49ed-94ea-40002ca91529/839-0
00:04:06.776 --> 00:04:10.186
So for example, we need to build
that out and test it, but then

26d82b3c-6769-49ed-94ea-40002ca91529/839-1
00:04:10.186 --> 00:04:13.224
there's also all the other
dependencies on that facility

26d82b3c-6769-49ed-94ea-40002ca91529/839-2
00:04:13.224 --> 00:04:13.756
structure.

26d82b3c-6769-49ed-94ea-40002ca91529/856-0
00:04:13.766 --> 00:04:17.826
All the downstream systems,
right, I don't think anybody has

26d82b3c-6769-49ed-94ea-40002ca91529/856-1
00:04:17.826 --> 00:04:19.356
a single best of breed.

26d82b3c-6769-49ed-94ea-40002ca91529/893-0
00:04:19.366 --> 00:04:24.225
We all have legacy systems that
are emirs are interfacing with

26d82b3c-6769-49ed-94ea-40002ca91529/864-0
00:04:21.526 --> 00:04:21.676
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/893-1
00:04:24.225 --> 00:04:28.235
any changes in facility
structure and then metadata

26d82b3c-6769-49ed-94ea-40002ca91529/893-2
00:04:28.235 --> 00:04:32.862
sociated with what's going on
with patient flow or place of

26d82b3c-6769-49ed-94ea-40002ca91529/893-3
00:04:32.862 --> 00:04:34.096
care, et cetera.

26d82b3c-6769-49ed-94ea-40002ca91529/935-0
00:04:34.326 --> 00:04:38.949
All those require quite a bit of
overhead on as far as testing

26d82b3c-6769-49ed-94ea-40002ca91529/935-1
00:04:38.949 --> 00:04:43.277
and validation, and sometimes
when organizations grow they

26d82b3c-6769-49ed-94ea-40002ca91529/935-2
00:04:43.277 --> 00:04:47.899
want to grow and get those new
spaces up and running, and they

26d82b3c-6769-49ed-94ea-40002ca91529/928-0
00:04:43.496 --> 00:04:43.746
It.

26d82b3c-6769-49ed-94ea-40002ca91529/935-3
00:04:47.899 --> 00:04:48.486
may not.

26d82b3c-6769-49ed-94ea-40002ca91529/975-0
00:04:48.496 --> 00:04:51.786
They either may not do the
testing, which is a big mistake,

26d82b3c-6769-49ed-94ea-40002ca91529/975-1
00:04:51.786 --> 00:04:55.021
but more often than not they
say, well, let's use existing

26d82b3c-6769-49ed-94ea-40002ca91529/975-2
00:04:55.021 --> 00:04:57.926
structures because I can't
afford to do the testing.

26d82b3c-6769-49ed-94ea-40002ca91529/973-0
00:04:56.286 --> 00:04:56.426
That.

26d82b3c-6769-49ed-94ea-40002ca91529/982-0
00:04:57.936 --> 00:04:59.466
And I realized that testing is
important.

26d82b3c-6769-49ed-94ea-40002ca91529/993-0
00:04:59.476 --> 00:05:02.826
So now you've introduced latent
errors in the system.

26d82b3c-6769-49ed-94ea-40002ca91529/1023-0
00:05:03.476 --> 00:05:07.355
I've the technical that I've
taken on is not to building out

26d82b3c-6769-49ed-94ea-40002ca91529/1023-1
00:05:07.355 --> 00:05:10.980
the representation of my
organization appropriately, and

26d82b3c-6769-49ed-94ea-40002ca91529/1023-2
00:05:10.980 --> 00:05:14.286
then that may have potential
implications later on.

26d82b3c-6769-49ed-94ea-40002ca91529/1032-0
00:05:12.056 --> 00:05:12.086
E.

26d82b3c-6769-49ed-94ea-40002ca91529/1078-0
00:05:14.296 --> 00:05:17.605
So for example, if you're in an
organization and for many

26d82b3c-6769-49ed-94ea-40002ca91529/1038-0
00:05:15.176 --> 00:05:15.346
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1078-1
00:05:17.605 --> 00:05:20.857
organizations have disaster
plans, and in those disaster

26d82b3c-6769-49ed-94ea-40002ca91529/1078-2
00:05:20.857 --> 00:05:24.394
plans they've built out in their
facility structure, disaster

26d82b3c-6769-49ed-94ea-40002ca91529/1078-3
00:05:24.394 --> 00:05:27.246
beds and now we have a sudden
need for new space.

26d82b3c-6769-49ed-94ea-40002ca91529/1075-0
00:05:24.396 --> 00:05:24.626
Yes.

26d82b3c-6769-49ed-94ea-40002ca91529/1117-0
00:05:25.266 --> 00:05:28.816
Now if E.

26d82b3c-6769-49ed-94ea-40002ca91529/1107-0
00:05:27.376 --> 00:05:31.260
So an organization may consider
using some of those best

26d82b3c-6769-49ed-94ea-40002ca91529/1107-1
00:05:31.260 --> 00:05:35.006
designated for these Serge plans
as patient care area.

26d82b3c-6769-49ed-94ea-40002ca91529/1134-0
00:05:35.296 --> 00:05:39.487
But the point is there are
downstream dependencies related

26d82b3c-6769-49ed-94ea-40002ca91529/1134-1
00:05:39.487 --> 00:05:43.536
to that, so classic patient
safety example would be I've

26d82b3c-6769-49ed-94ea-40002ca91529/1143-0
00:05:39.506 --> 00:05:40.896
Well, uh. Uh.

26d82b3c-6769-49ed-94ea-40002ca91529/1134-2
00:05:43.536 --> 00:05:44.246
built out.

26d82b3c-6769-49ed-94ea-40002ca91529/1150-0
00:05:44.356 --> 00:05:48.696
I've used existing facility
structure that caused the bed.

26d82b3c-6769-49ed-94ea-40002ca91529/1149-0
00:05:47.706 --> 00:05:48.556
At all.

26d82b3c-6769-49ed-94ea-40002ca91529/1218-0
00:05:48.806 --> 00:05:52.840
You know, Serge, bed XYZ and now
there's a code blue or medical

26d82b3c-6769-49ed-94ea-40002ca91529/1157-0
00:05:48.846 --> 00:05:49.116
And.

26d82b3c-6769-49ed-94ea-40002ca91529/1187-0
00:05:51.096 --> 00:05:51.166
E.

26d82b3c-6769-49ed-94ea-40002ca91529/1218-1
00:05:52.840 --> 00:05:56.937
emergency in that and that goes
to our downstream legacy systems

26d82b3c-6769-49ed-94ea-40002ca91529/1218-2
00:05:56.937 --> 00:06:00.592
to our operators at call an
overhead page and now they're

26d82b3c-6769-49ed-94ea-40002ca91529/1204-0
00:05:59.266 --> 00:06:01.666
Yeah. Anything.

26d82b3c-6769-49ed-94ea-40002ca91529/1218-3
00:06:00.592 --> 00:06:04.437
calling overhead page to search
space XYZ as opposed to some

26d82b3c-6769-49ed-94ea-40002ca91529/1218-4
00:06:04.437 --> 00:06:07.336
physical location that
clinicians know about.

26d82b3c-6769-49ed-94ea-40002ca91529/1259-0
00:06:07.346 --> 00:06:10.920
So by kicking down the road,
even building out facilities

26d82b3c-6769-49ed-94ea-40002ca91529/1259-1
00:06:10.920 --> 00:06:14.370
structures and doing this
downstream testing, etcetera,

26d82b3c-6769-49ed-94ea-40002ca91529/1259-2
00:06:14.370 --> 00:06:17.821
you introduce latent errors in
the system that may have

26d82b3c-6769-49ed-94ea-40002ca91529/1254-0
00:06:17.576 --> 00:06:17.856
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1259-3
00:06:17.821 --> 00:06:19.176
implications later on.

26d82b3c-6769-49ed-94ea-40002ca91529/1288-0
00:06:19.356 --> 00:06:22.817
And in our owner organization,
as we identify those, you know

26d82b3c-6769-49ed-94ea-40002ca91529/1288-1
00:06:22.817 --> 00:06:26.500
we then go to leadership and go,
hey guys, you missed this and we

26d82b3c-6769-49ed-94ea-40002ca91529/1288-2
00:06:26.500 --> 00:06:27.616
go back and do that.

26d82b3c-6769-49ed-94ea-40002ca91529/1347-0
00:06:27.626 --> 00:06:30.694
But you know when you're
initially thinking about the

26d82b3c-6769-49ed-94ea-40002ca91529/1305-0
00:06:29.846 --> 00:06:31.506
Uh, OK.

26d82b3c-6769-49ed-94ea-40002ca91529/1347-1
00:06:30.694 --> 00:06:34.159
total cost of ownership around
expanding an organization and

26d82b3c-6769-49ed-94ea-40002ca91529/1312-0
00:06:33.596 --> 00:06:33.826
And.

26d82b3c-6769-49ed-94ea-40002ca91529/1347-2
00:06:34.159 --> 00:06:37.283
expanding the facility, you
really need to think about

26d82b3c-6769-49ed-94ea-40002ca91529/1347-3
00:06:37.283 --> 00:06:40.861
there's other costs associated
with that, not just the PCs I'm

26d82b3c-6769-49ed-94ea-40002ca91529/1347-4
00:06:40.861 --> 00:06:44.156
putting in that space or the
monitors I'm putting in that

26d82b3c-6769-49ed-94ea-40002ca91529/1341-0
00:06:41.886 --> 00:06:42.206
It's.

26d82b3c-6769-49ed-94ea-40002ca91529/1347-5
00:06:44.156 --> 00:06:44.496
space.

26d82b3c-6769-49ed-94ea-40002ca91529/1351-0
00:06:44.546 --> 00:06:44.806
Yes.

26d82b3c-6769-49ed-94ea-40002ca91529/1375-0
00:06:44.606 --> 00:06:48.401
But it's also that I really look
at the facility structure and

26d82b3c-6769-49ed-94ea-40002ca91529/1375-1
00:06:48.401 --> 00:06:51.955
how that, how that interface all
our downstream systems et

26d82b3c-6769-49ed-94ea-40002ca91529/1375-2
00:06:51.955 --> 00:06:52.376
cetera.

26d82b3c-6769-49ed-94ea-40002ca91529/1418-0
00:06:52.606 --> 00:06:56.011
And that's something that the
safety side, we've been drawing

26d82b3c-6769-49ed-94ea-40002ca91529/1418-1
00:06:56.011 --> 00:06:59.470
a lot of attention to so that we
don't fall into that risk and

26d82b3c-6769-49ed-94ea-40002ca91529/1421-0
00:06:58.996 --> 00:06:59.216
There's.

26d82b3c-6769-49ed-94ea-40002ca91529/1418-2
00:06:59.470 --> 00:07:02.874
introduce those latent errors
into our system and can protect

26d82b3c-6769-49ed-94ea-40002ca91529/1418-3
00:07:02.874 --> 00:07:04.356
our patients in the future.

26d82b3c-6769-49ed-94ea-40002ca91529/1433-0
00:07:04.626 --> 00:07:07.586
When that information then
flows, the other downstream

26d82b3c-6769-49ed-94ea-40002ca91529/1433-1
00:07:07.586 --> 00:07:08.016
systems.

26d82b3c-6769-49ed-94ea-40002ca91529/1429-0
00:07:08.896 --> 00:07:09.326
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1453-0
00:07:09.336 --> 00:07:12.817
And you also mentioned not only
physical the integration of

26d82b3c-6769-49ed-94ea-40002ca91529/1453-1
00:07:12.817 --> 00:07:16.066
physical and digital, but also
just within the digital.

26d82b3c-6769-49ed-94ea-40002ca91529/1467-0
00:07:16.176 --> 00:07:19.244
For example, when you're out of,
organizations are going through

26d82b3c-6769-49ed-94ea-40002ca91529/1467-1
00:07:19.244 --> 00:07:20.706
an application rationalization.

26d82b3c-6769-49ed-94ea-40002ca91529/1485-0
00:07:20.816 --> 00:07:24.656
If you're mapping old metadata
to new metadata, then the lookup

26d82b3c-6769-49ed-94ea-40002ca91529/1485-1
00:07:24.656 --> 00:07:26.216
tables don't work anymore.

26d82b3c-6769-49ed-94ea-40002ca91529/1505-0
00:07:26.226 --> 00:07:28.476
For example, in some cases, and.

26d82b3c-6769-49ed-94ea-40002ca91529/1489-0
00:07:26.646 --> 00:07:26.846
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1494-0
00:07:28.116 --> 00:07:28.336
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1497-0
00:07:28.346 --> 00:07:29.146
And that's another.

26d82b3c-6769-49ed-94ea-40002ca91529/1503-0
00:07:29.216 --> 00:07:30.136
That's a classic example.

26d82b3c-6769-49ed-94ea-40002ca91529/1517-0
00:07:30.146 --> 00:07:32.300
I know we were gonna talk both
about technical debt and

26d82b3c-6769-49ed-94ea-40002ca91529/1517-1
00:07:32.300 --> 00:07:32.646
metadata.

26d82b3c-6769-49ed-94ea-40002ca91529/1527-0
00:07:32.656 --> 00:07:34.496
This is where the two really
meet, right?

26d82b3c-6769-49ed-94ea-40002ca91529/1534-0
00:07:34.186 --> 00:07:34.446
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1554-0
00:07:34.506 --> 00:07:39.052
Because with all these
transactions is the metadata,

26d82b3c-6769-49ed-94ea-40002ca91529/1554-1
00:07:39.052 --> 00:07:44.455
the physical location and the
virtual location, and those have

26d82b3c-6769-49ed-94ea-40002ca91529/1551-0
00:07:43.036 --> 00:07:43.256
And.

26d82b3c-6769-49ed-94ea-40002ca91529/1554-2
00:07:44.455 --> 00:07:45.226
to align.

26d82b3c-6769-49ed-94ea-40002ca91529/1580-0
00:07:45.276 --> 00:07:48.434
And sometimes you know, for
example, we build a new tower

26d82b3c-6769-49ed-94ea-40002ca91529/1580-1
00:07:48.434 --> 00:07:52.026
and now we have, we've built out
all this new facility structure.

26d82b3c-6769-49ed-94ea-40002ca91529/1606-0
00:07:52.036 --> 00:07:55.370
We have old facility structure
and that metadata is someone has

26d82b3c-6769-49ed-94ea-40002ca91529/1606-1
00:07:55.370 --> 00:07:58.546
to go back and rationalize that
and make sure it all aligns.

26d82b3c-6769-49ed-94ea-40002ca91529/1639-0
00:07:58.896 --> 00:08:02.876
Because if you don't fix those,
you potentially introduce latent

26d82b3c-6769-49ed-94ea-40002ca91529/1639-1
00:08:02.876 --> 00:08:06.794
errors and those latent errors
might be as dangerous as sending

26d82b3c-6769-49ed-94ea-40002ca91529/1639-2
00:08:06.794 --> 00:08:09.426
somebody to the wrong location
for a code.

26d82b3c-6769-49ed-94ea-40002ca91529/1651-0
00:08:09.496 --> 00:08:11.116
But they might be inside it.

26d82b3c-6769-49ed-94ea-40002ca91529/1654-0
00:08:09.556 --> 00:08:09.836
But.

26d82b3c-6769-49ed-94ea-40002ca91529/1656-0
00:08:11.126 --> 00:08:11.686
Insidious.

26d82b3c-6769-49ed-94ea-40002ca91529/1688-0
00:08:11.696 --> 00:08:14.217
Like for example the
accommodation code that's

26d82b3c-6769-49ed-94ea-40002ca91529/1688-1
00:08:14.217 --> 00:08:17.488
associated with that facility
structure may come across as a

26d82b3c-6769-49ed-94ea-40002ca91529/1688-2
00:08:17.488 --> 00:08:20.545
Med surge when reality it's a
step down or critical care

26d82b3c-6769-49ed-94ea-40002ca91529/1688-3
00:08:20.545 --> 00:08:20.866
space.

26d82b3c-6769-49ed-94ea-40002ca91529/1720-0
00:08:20.996 --> 00:08:24.660
Now you're missing out on that
facility that facility charge

26d82b3c-6769-49ed-94ea-40002ca91529/1702-0
00:08:23.736 --> 00:08:23.926
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1720-1
00:08:24.660 --> 00:08:28.204
because the metadata on the, the
the accommodation code is

26d82b3c-6769-49ed-94ea-40002ca91529/1715-0
00:08:26.196 --> 00:08:27.866
Now that that.

26d82b3c-6769-49ed-94ea-40002ca91529/1720-2
00:08:28.204 --> 00:08:30.726
different in somebody forgot to
fix that.

26d82b3c-6769-49ed-94ea-40002ca91529/1724-0
00:08:30.736 --> 00:08:31.936
So there's a lot.

26d82b3c-6769-49ed-94ea-40002ca91529/1751-0
00:08:31.976 --> 00:08:35.423
There's a lot of cost to
rationalization and growing and

26d82b3c-6769-49ed-94ea-40002ca91529/1751-1
00:08:35.423 --> 00:08:38.688
part of that is the safety
component and the metadata

26d82b3c-6769-49ed-94ea-40002ca91529/1744-0
00:08:36.646 --> 00:08:36.966
And.

26d82b3c-6769-49ed-94ea-40002ca91529/1751-2
00:08:38.688 --> 00:08:39.836
associated with it.

26d82b3c-6769-49ed-94ea-40002ca91529/1818-0
00:08:39.846 --> 00:08:42.861
But it also has also
understanding the total cost of

26d82b3c-6769-49ed-94ea-40002ca91529/1818-1
00:08:42.861 --> 00:08:46.047
ownership of these migration
strategies and making sure

26d82b3c-6769-49ed-94ea-40002ca91529/1777-0
00:08:45.156 --> 00:08:45.396
Uh-huh.

26d82b3c-6769-49ed-94ea-40002ca91529/1818-2
00:08:46.047 --> 00:08:49.745
you're going back and having the
time and resources allocated to

26d82b3c-6769-49ed-94ea-40002ca91529/1818-3
00:08:49.745 --> 00:08:53.386
make sure you're looking at all
that legacy stuff and doing all

26d82b3c-6769-49ed-94ea-40002ca91529/1792-0
00:08:49.796 --> 00:08:50.066
And then.

26d82b3c-6769-49ed-94ea-40002ca91529/1818-4
00:08:53.386 --> 00:08:56.856
the appropriate stress testing
to make sure that information

26d82b3c-6769-49ed-94ea-40002ca91529/1814-0
00:08:55.916 --> 00:08:56.376
OK.

26d82b3c-6769-49ed-94ea-40002ca91529/1818-5
00:08:56.856 --> 00:08:58.676
flows to where it needs to flow.

26d82b3c-6769-49ed-94ea-40002ca91529/1843-0
00:08:59.166 --> 00:09:03.178
Billing is proper and that those
physical locations match up with

26d82b3c-6769-49ed-94ea-40002ca91529/1843-1
00:09:03.178 --> 00:09:05.366
the virtual location in the
system.

26d82b3c-6769-49ed-94ea-40002ca91529/1907-0
00:09:03.236 --> 00:09:07.694
Ah, so it sounds like if any of
our listeners are persuaded and

26d82b3c-6769-49ed-94ea-40002ca91529/1907-1
00:09:07.694 --> 00:09:12.082
interested by this application
rationalization, technical debt

26d82b3c-6769-49ed-94ea-40002ca91529/1907-2
00:09:12.082 --> 00:09:16.470
awareness about the implications
technical debt has on patient

26d82b3c-6769-49ed-94ea-40002ca91529/1907-3
00:09:16.470 --> 00:09:20.719
safety, it sounds like a lot of
governance and processes and

26d82b3c-6769-49ed-94ea-40002ca91529/1907-4
00:09:20.719 --> 00:09:24.828
testing or kind of what they
need to begin implementing in

26d82b3c-6769-49ed-94ea-40002ca91529/1907-5
00:09:24.828 --> 00:09:27.196
order to account for these
risks.

26d82b3c-6769-49ed-94ea-40002ca91529/1898-0
00:09:27.426 --> 00:09:27.616
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/1913-0
00:09:27.626 --> 00:09:30.316
What it comes down to is just
best industry practice, right?

26d82b3c-6769-49ed-94ea-40002ca91529/1929-0
00:09:30.426 --> 00:09:33.316
Your your information technology
department really needs to make

26d82b3c-6769-49ed-94ea-40002ca91529/1929-1
00:09:33.316 --> 00:09:34.916
you know good stakeholder
analysis.

26d82b3c-6769-49ed-94ea-40002ca91529/1937-0
00:09:34.926 --> 00:09:37.676
Good project management, good
communication skills.

26d82b3c-6769-49ed-94ea-40002ca91529/1941-0
00:09:37.686 --> 00:09:38.356
It's all stuff.

26d82b3c-6769-49ed-94ea-40002ca91529/2018-0
00:09:38.366 --> 00:09:42.501
Everybody who's listening just
knows about, but what a lot of

26d82b3c-6769-49ed-94ea-40002ca91529/2018-1
00:09:42.501 --> 00:09:46.169
teams fail to realize is the
extent of it and how many

26d82b3c-6769-49ed-94ea-40002ca91529/2018-2
00:09:46.169 --> 00:09:50.303
resources you actually need to
do a full stakeholder analysis

26d82b3c-6769-49ed-94ea-40002ca91529/2018-3
00:09:50.303 --> 00:09:53.771
and a full risk analysis of
these minute mitigation

26d82b3c-6769-49ed-94ea-40002ca91529/2018-4
00:09:53.771 --> 00:09:57.505
migrations to make sure that
you've looked at all those

26d82b3c-6769-49ed-94ea-40002ca91529/1998-0
00:09:55.436 --> 00:09:55.676
OK.

26d82b3c-6769-49ed-94ea-40002ca91529/2018-5
00:09:57.505 --> 00:10:01.506
downstream systems and not just
that the information flows.

26d82b3c-6769-49ed-94ea-40002ca91529/2010-0
00:09:58.196 --> 00:09:58.426
OK.

26d82b3c-6769-49ed-94ea-40002ca91529/2034-0
00:10:01.696 --> 00:10:05.126
But then when it pops up at the
other end, does that end user?

26d82b3c-6769-49ed-94ea-40002ca91529/2046-0
00:10:05.176 --> 00:10:06.956
Is that does that data make
sense to them?

26d82b3c-6769-49ed-94ea-40002ca91529/2055-0
00:10:06.966 --> 00:10:08.056
Do they know where to go?

26d82b3c-6769-49ed-94ea-40002ca91529/2101-0
00:10:08.146 --> 00:10:11.296
And if you forget to look at
your operators and now you're

26d82b3c-6769-49ed-94ea-40002ca91529/2062-0
00:10:08.226 --> 00:10:08.416
So.

26d82b3c-6769-49ed-94ea-40002ca91529/2101-1
00:10:11.296 --> 00:10:14.552
pushing them physical locations
that no longer make sense or

26d82b3c-6769-49ed-94ea-40002ca91529/2101-2
00:10:14.552 --> 00:10:17.968
you're pushing it to your coders
and you're billing department,

26d82b3c-6769-49ed-94ea-40002ca91529/2088-0
00:10:14.866 --> 00:10:15.026
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/2101-3
00:10:17.968 --> 00:10:19.836
and then those accommodation
code.

26d82b3c-6769-49ed-94ea-40002ca91529/2110-0
00:10:19.846 --> 00:10:21.666
So it makes sense you're going
to miss out.

26d82b3c-6769-49ed-94ea-40002ca91529/2135-0
00:10:22.486 --> 00:10:26.227
So I did mention just a moment
ago about metadata, which is the

26d82b3c-6769-49ed-94ea-40002ca91529/2135-1
00:10:26.227 --> 00:10:28.916
second topic that we're going to
cover today.

26d82b3c-6769-49ed-94ea-40002ca91529/2224-0
00:10:29.066 --> 00:10:33.125
And on the I mentioned kind of
screwing up mapping between

26d82b3c-6769-49ed-94ea-40002ca91529/2224-1
00:10:33.125 --> 00:10:36.977
lookup tables when you're
mapping from old metadata new

26d82b3c-6769-49ed-94ea-40002ca91529/2224-2
00:10:36.977 --> 00:10:41.449
metadata, the topic of metadata
itself is becoming of increasing

26d82b3c-6769-49ed-94ea-40002ca91529/2224-3
00:10:41.449 --> 00:10:45.645
interest to many healthcare
delivery systems as Gen AI rises

26d82b3c-6769-49ed-94ea-40002ca91529/2224-4
00:10:45.645 --> 00:10:49.635
in prominence as a topic that
everyone is focusing on, in

26d82b3c-6769-49ed-94ea-40002ca91529/2224-5
00:10:49.635 --> 00:10:53.693
particular when healthcare
deliveries are sons are looking

26d82b3c-6769-49ed-94ea-40002ca91529/2224-6
00:10:53.693 --> 00:10:58.096
to integrate Gen AI into their
healthcare delivery system, they

26d82b3c-6769-49ed-94ea-40002ca91529/2224-7
00:10:58.096 --> 00:11:00.916
often need to train models on
their own.

26d82b3c-6769-49ed-94ea-40002ca91529/2301-0
00:11:01.326 --> 00:11:05.359
Uh data of different kinds,
whether it be Phi or other sorts

26d82b3c-6769-49ed-94ea-40002ca91529/2301-1
00:11:05.359 --> 00:11:09.524
of metadata, and I think we were
today, we're going to discuss

26d82b3c-6769-49ed-94ea-40002ca91529/2301-2
00:11:09.524 --> 00:11:13.292
kind of what, what data
governance, what needs to happen

26d82b3c-6769-49ed-94ea-40002ca91529/2301-3
00:11:13.292 --> 00:11:17.325
to ensure that you have the
right metadata in order to train

26d82b3c-6769-49ed-94ea-40002ca91529/2301-4
00:11:17.325 --> 00:11:21.423
your, your, your Gen AI models
and kind of what is what we're

26d82b3c-6769-49ed-94ea-40002ca91529/2301-5
00:11:21.423 --> 00:11:25.522
best practices in the past and
how they how are they evolving

26d82b3c-6769-49ed-94ea-40002ca91529/2301-6
00:11:25.522 --> 00:11:25.786
now.

26d82b3c-6769-49ed-94ea-40002ca91529/2303-0
00:11:26.236 --> 00:11:26.666
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/2339-0
00:11:26.676 --> 00:11:30.319
So this is a hard question to
answer, and obviously there's

26d82b3c-6769-49ed-94ea-40002ca91529/2339-1
00:11:30.319 --> 00:11:34.023
the broader concepts of like
bias in the data, et cetera and

26d82b3c-6769-49ed-94ea-40002ca91529/2339-2
00:11:34.023 --> 00:11:37.726
healthcare or qualities when
you're deploying these systems.

26d82b3c-6769-49ed-94ea-40002ca91529/2345-0
00:11:37.736 --> 00:11:40.026
And that's where that data
really shines.

26d82b3c-6769-49ed-94ea-40002ca91529/2354-0
00:11:40.216 --> 00:11:43.526
But the other is the the tipple.

26d82b3c-6769-49ed-94ea-40002ca91529/2358-0
00:11:43.536 --> 00:11:44.806
Set the simple lasers.

26d82b3c-6769-49ed-94ea-40002ca91529/2414-0
00:11:45.016 --> 00:11:49.120
You don't know because when we
look at these systems and we're

26d82b3c-6769-49ed-94ea-40002ca91529/2414-1
00:11:49.120 --> 00:11:53.158
looking at sort of the emergent
characteristics of the hidden

26d82b3c-6769-49ed-94ea-40002ca91529/2414-2
00:11:53.158 --> 00:11:57.132
layers in these systems, we
don't really understand a lot of

26d82b3c-6769-49ed-94ea-40002ca91529/2414-3
00:11:57.132 --> 00:12:01.235
times what those features are,
that the system is keying on to

26d82b3c-6769-49ed-94ea-40002ca91529/2414-4
00:12:01.235 --> 00:12:04.166
understand to make that
generative decision.

26d82b3c-6769-49ed-94ea-40002ca91529/2424-0
00:12:04.356 --> 00:12:07.466
And for humans, a lot of times,
how many people?

26d82b3c-6769-49ed-94ea-40002ca91529/2431-0
00:12:07.476 --> 00:12:08.946
The diagnosis makes sense,
right?

26d82b3c-6769-49ed-94ea-40002ca91529/2449-0
00:12:08.956 --> 00:12:13.116
But it may turn out that
location may play a role there.

26d82b3c-6769-49ed-94ea-40002ca91529/2447-0
00:12:11.076 --> 00:12:11.376
At.

26d82b3c-6769-49ed-94ea-40002ca91529/2476-0
00:12:13.126 --> 00:12:16.613
So when we're thinking about the
data and the metadata, you're

26d82b3c-6769-49ed-94ea-40002ca91529/2473-0
00:12:14.366 --> 00:12:14.386
E.

26d82b3c-6769-49ed-94ea-40002ca91529/2476-1
00:12:16.613 --> 00:12:19.491
going to use in training
systems, the more data the

26d82b3c-6769-49ed-94ea-40002ca91529/2480-0
00:12:18.846 --> 00:12:19.216
Huh.

26d82b3c-6769-49ed-94ea-40002ca91529/2476-2
00:12:19.491 --> 00:12:20.266
better, right?

26d82b3c-6769-49ed-94ea-40002ca91529/2522-0
00:12:20.276 --> 00:12:23.844
We're seeing that across the
industry, bigger generative

26d82b3c-6769-49ed-94ea-40002ca91529/2522-1
00:12:23.844 --> 00:12:27.600
systems, the bigger the data
set, the better, the more rich

26d82b3c-6769-49ed-94ea-40002ca91529/2501-0
00:12:26.416 --> 00:12:26.586
Yep.

26d82b3c-6769-49ed-94ea-40002ca91529/2522-2
00:12:27.600 --> 00:12:31.481
the data system that we're rich
the data the better and other

26d82b3c-6769-49ed-94ea-40002ca91529/2522-3
00:12:31.481 --> 00:12:34.611
contexts of that data,
specifically the metadata,

26d82b3c-6769-49ed-94ea-40002ca91529/2518-0
00:12:31.766 --> 00:12:32.076
And.

26d82b3c-6769-49ed-94ea-40002ca91529/2522-4
00:12:34.611 --> 00:12:34.986
right.

26d82b3c-6769-49ed-94ea-40002ca91529/2559-0
00:12:34.996 --> 00:12:39.423
So maybe not just patient
demographics and diagnosis gets

26d82b3c-6769-49ed-94ea-40002ca91529/2559-1
00:12:39.423 --> 00:12:44.078
me to where I need to go, but
maybe the system at the hidden

26d82b3c-6769-49ed-94ea-40002ca91529/2574-0
00:12:41.166 --> 00:12:41.316
Well.

26d82b3c-6769-49ed-94ea-40002ca91529/2559-2
00:12:44.078 --> 00:12:48.352
layer is keying on location
data, time of day, data, et

26d82b3c-6769-49ed-94ea-40002ca91529/2559-3
00:12:48.352 --> 00:12:48.886
cetera.

26d82b3c-6769-49ed-94ea-40002ca91529/2637-0
00:12:48.996 --> 00:12:52.841
So if you're not saving that
data and then putting it into

26d82b3c-6769-49ed-94ea-40002ca91529/2637-1
00:12:52.841 --> 00:12:57.012
your datasets for your training,
you may be missing out on that

26d82b3c-6769-49ed-94ea-40002ca91529/2589-0
00:12:53.896 --> 00:12:54.106
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/2637-2
00:12:57.012 --> 00:13:01.248
and what further confounds that
is, you often don't know how the

26d82b3c-6769-49ed-94ea-40002ca91529/2598-0
00:12:57.226 --> 00:12:57.466
And.

26d82b3c-6769-49ed-94ea-40002ca91529/2637-3
00:13:01.248 --> 00:13:05.485
original basis that was trained
was the basis of trade with that

26d82b3c-6769-49ed-94ea-40002ca91529/2637-4
00:13:05.485 --> 00:13:08.613
metadata are not thus
introducing that metadata

26d82b3c-6769-49ed-94ea-40002ca91529/2637-5
00:13:08.613 --> 00:13:11.936
aligned with the previous
training regimen or not.

26d82b3c-6769-49ed-94ea-40002ca91529/2633-0
00:13:09.236 --> 00:13:09.426
That's.

26d82b3c-6769-49ed-94ea-40002ca91529/2692-0
00:13:12.306 --> 00:13:16.500
And then when there's a mismatch
between the data set of initial

26d82b3c-6769-49ed-94ea-40002ca91529/2650-0
00:13:13.106 --> 00:13:13.336
And.

26d82b3c-6769-49ed-94ea-40002ca91529/2692-1
00:13:16.500 --> 00:13:20.113
training and then fine tuning,
how does that affect the

26d82b3c-6769-49ed-94ea-40002ca91529/2692-2
00:13:20.113 --> 00:13:24.243
emergent features in the latent
space, and how does that affect

26d82b3c-6769-49ed-94ea-40002ca91529/2692-3
00:13:24.243 --> 00:13:28.436
the decision making the weights
of that and that's hard to know.

26d82b3c-6769-49ed-94ea-40002ca91529/2748-0
00:13:28.526 --> 00:13:32.625
So it's not only important to
have this data because it helps

26d82b3c-6769-49ed-94ea-40002ca91529/2748-1
00:13:32.625 --> 00:13:36.525
the AI key on items that the
human observer may not key on

26d82b3c-6769-49ed-94ea-40002ca91529/2748-2
00:13:36.525 --> 00:13:40.888
and get you more robust answers,
but it also has implications for

26d82b3c-6769-49ed-94ea-40002ca91529/2748-3
00:13:40.888 --> 00:13:44.722
off the shelf AI that now you're
trading and what was the

26d82b3c-6769-49ed-94ea-40002ca91529/2748-4
00:13:44.722 --> 00:13:46.506
training regimen initially?

26d82b3c-6769-49ed-94ea-40002ca91529/2752-0
00:13:46.516 --> 00:13:47.186
What was that data?

26d82b3c-6769-49ed-94ea-40002ca91529/2827-0
00:13:47.196 --> 00:13:50.916
What was the metadata and do
those things align and then the

26d82b3c-6769-49ed-94ea-40002ca91529/2765-0
00:13:48.376 --> 00:13:48.636
That.

26d82b3c-6769-49ed-94ea-40002ca91529/2827-1
00:13:50.916 --> 00:13:54.757
bigger 50,000 foot view in as
much as that, what this has been

26d82b3c-6769-49ed-94ea-40002ca91529/2775-0
00:13:51.276 --> 00:13:51.586
Great.

26d82b3c-6769-49ed-94ea-40002ca91529/2788-0
00:13:54.706 --> 00:13:54.966
That.

26d82b3c-6769-49ed-94ea-40002ca91529/2827-2
00:13:54.757 --> 00:13:58.598
a day to do with respect to bias
and all the other things that

26d82b3c-6769-49ed-94ea-40002ca91529/2827-3
00:13:58.598 --> 00:14:02.195
we're really focused on in
healthcare and and so much show

26d82b3c-6769-49ed-94ea-40002ca91529/2827-4
00:14:02.195 --> 00:14:05.549
that Joint Commission is
starting to focus on this and

26d82b3c-6769-49ed-94ea-40002ca91529/2827-5
00:14:05.549 --> 00:14:06.646
the FDA et cetera.

26d82b3c-6769-49ed-94ea-40002ca91529/2833-0
00:14:06.566 --> 00:14:06.716
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/2880-0
00:14:06.906 --> 00:14:10.896
You know this bias in these
general systems is a hot button

26d82b3c-6769-49ed-94ea-40002ca91529/2880-1
00:14:10.896 --> 00:14:14.953
topic, and the metadata and
understanding the context of the

26d82b3c-6769-49ed-94ea-40002ca91529/2880-2
00:14:14.953 --> 00:14:19.342
data and the person that is part
of that data in system is super,

26d82b3c-6769-49ed-94ea-40002ca91529/2880-3
00:14:19.342 --> 00:14:22.866
super important for
understanding bias and outcomes.

26d82b3c-6769-49ed-94ea-40002ca91529/2873-0
00:14:20.016 --> 00:14:20.306
Yes.

26d82b3c-6769-49ed-94ea-40002ca91529/2891-0
00:14:23.906 --> 00:14:28.157
You know, some of the examples
you provide remind me of classic

26d82b3c-6769-49ed-94ea-40002ca91529/2891-1
00:14:28.157 --> 00:14:29.086
public health.

26d82b3c-6769-49ed-94ea-40002ca91529/2900-0
00:14:29.656 --> 00:14:32.836
Uh case studies, for example.

26d82b3c-6769-49ed-94ea-40002ca91529/2904-0
00:14:32.916 --> 00:14:34.546
I think in the 1970s.

26d82b3c-6769-49ed-94ea-40002ca91529/2926-0
00:14:36.586 --> 00:14:41.038
There was not clear evidence at
the time that both systolic and

26d82b3c-6769-49ed-94ea-40002ca91529/2926-1
00:14:41.038 --> 00:14:44.516
diastolic blood pressure were
worth measuring it.

26d82b3c-6769-49ed-94ea-40002ca91529/2947-0
00:14:44.526 --> 00:14:47.712
There was meaning to the lower
number and so we didn't use to

26d82b3c-6769-49ed-94ea-40002ca91529/2947-1
00:14:47.712 --> 00:14:48.996
collect that information.

26d82b3c-6769-49ed-94ea-40002ca91529/2974-0
00:14:49.006 --> 00:14:52.227
And it wasn't only until later
that we incorporate that we

26d82b3c-6769-49ed-94ea-40002ca91529/2974-1
00:14:52.227 --> 00:14:55.556
found, I think Professor Mike
Clagg from Johns Hopkins found

26d82b3c-6769-49ed-94ea-40002ca91529/2974-2
00:14:55.556 --> 00:14:56.866
out, hey, you know what?

26d82b3c-6769-49ed-94ea-40002ca91529/3032-0
00:14:56.876 --> 00:15:00.423
Both numbers actually do have
value and meaning, and gosh, I

26d82b3c-6769-49ed-94ea-40002ca91529/3032-1
00:15:00.423 --> 00:15:03.969
wish we'd been collecting both
numbers for a long time so it

26d82b3c-6769-49ed-94ea-40002ca91529/3032-2
00:15:03.969 --> 00:15:07.748
could be for future discoveries
where another one that's kind of

26d82b3c-6769-49ed-94ea-40002ca91529/3032-3
00:15:07.748 --> 00:15:11.353
a fun one to to talk about is
before the Queen Indoor Air Act

26d82b3c-6769-49ed-94ea-40002ca91529/3032-4
00:15:11.353 --> 00:15:12.806
in the late 20th century.

26d82b3c-6769-49ed-94ea-40002ca91529/3047-0
00:15:13.026 --> 00:15:16.126
Coffee drinking was an indicator
of lung cancer because it was

26d82b3c-6769-49ed-94ea-40002ca91529/3047-1
00:15:16.126 --> 00:15:17.306
correlated with smoking.

26d82b3c-6769-49ed-94ea-40002ca91529/3074-0
00:15:17.316 --> 00:15:20.971
People used to have a cigarette
with their coffee, and so it may

26d82b3c-6769-49ed-94ea-40002ca91529/3074-1
00:15:20.971 --> 00:15:24.063
be helpful to predict lung
cancer by capturing who's a

26d82b3c-6769-49ed-94ea-40002ca91529/3074-2
00:15:24.063 --> 00:15:24.906
coffee drinker.

26d82b3c-6769-49ed-94ea-40002ca91529/3091-0
00:15:24.976 --> 00:15:27.465
But of course, those are the
sorts of things that just seems

26d82b3c-6769-49ed-94ea-40002ca91529/3084-0
00:15:25.026 --> 00:15:25.206
Yep.

26d82b3c-6769-49ed-94ea-40002ca91529/3091-1
00:15:27.465 --> 00:15:28.036
so irrelevant.

26d82b3c-6769-49ed-94ea-40002ca91529/3095-0
00:15:28.046 --> 00:15:29.656
There's no causal relation.

26d82b3c-6769-49ed-94ea-40002ca91529/3112-0
00:15:29.736 --> 00:15:32.726
Why would I care about coffee if
I want to study one cancer?

26d82b3c-6769-49ed-94ea-40002ca91529/3172-0
00:15:32.856 --> 00:15:36.872
So just take that as an analogy
for, you know, 60 years later

26d82b3c-6769-49ed-94ea-40002ca91529/3172-1
00:15:36.872 --> 00:15:40.046
with different kinds of
contextual data and data

26d82b3c-6769-49ed-94ea-40002ca91529/3172-2
00:15:40.046 --> 00:15:43.738
elements and data space about
what might be relevant and

26d82b3c-6769-49ed-94ea-40002ca91529/3172-3
00:15:43.738 --> 00:15:46.847
predictive for either a
patient's health or for

26d82b3c-6769-49ed-94ea-40002ca91529/3172-4
00:15:46.847 --> 00:15:49.956
something that could help train
a Gen AI model.

26d82b3c-6769-49ed-94ea-40002ca91529/3168-0
00:15:50.586 --> 00:15:50.886
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/3191-0
00:15:50.896 --> 00:15:56.486
And this is key in the emergence
the emergent characteristics of

26d82b3c-6769-49ed-94ea-40002ca91529/3191-1
00:15:56.486 --> 00:15:57.346
the model.

26d82b3c-6769-49ed-94ea-40002ca91529/3232-0
00:15:58.356 --> 00:16:02.506
The human again may think this
data point is irrelevant to the

26d82b3c-6769-49ed-94ea-40002ca91529/3199-0
00:15:59.996 --> 00:16:00.026
E.

26d82b3c-6769-49ed-94ea-40002ca91529/3232-1
00:16:02.506 --> 00:16:06.590
outcome from a face validity
perspective, but it turns out it

26d82b3c-6769-49ed-94ea-40002ca91529/3232-2
00:16:06.590 --> 00:16:10.740
actually is important for the
model, and that makes inter till

26d82b3c-6769-49ed-94ea-40002ca91529/3232-3
00:16:10.740 --> 00:16:13.506
interoperability research very
difficult.

26d82b3c-6769-49ed-94ea-40002ca91529/3249-0
00:16:13.516 --> 00:16:17.424
It makes explain ability very
difficult, so it's a fine

26d82b3c-6769-49ed-94ea-40002ca91529/3249-1
00:16:17.424 --> 00:16:19.586
balance, so more interpretable.

26d82b3c-6769-49ed-94ea-40002ca91529/3258-0
00:16:19.596 --> 00:16:21.196
I want something to be the more
explainable.

26d82b3c-6769-49ed-94ea-40002ca91529/3294-0
00:16:21.206 --> 00:16:24.815
I want to model to be the more
I'm going to give data that has

26d82b3c-6769-49ed-94ea-40002ca91529/3282-0
00:16:24.306 --> 00:16:24.596
That.

26d82b3c-6769-49ed-94ea-40002ca91529/3294-1
00:16:24.815 --> 00:16:28.365
facility to a human observer
that this makes sense that these

26d82b3c-6769-49ed-94ea-40002ca91529/3294-2
00:16:28.365 --> 00:16:30.026
are associated or correlated.

26d82b3c-6769-49ed-94ea-40002ca91529/3304-0
00:16:30.956 --> 00:16:34.226
But it turns out more data
density.

26d82b3c-6769-49ed-94ea-40002ca91529/3314-0
00:16:35.136 --> 00:16:39.146
Other metadata may actually help
the system become more accurate.

26d82b3c-6769-49ed-94ea-40002ca91529/3321-0
00:16:39.376 --> 00:16:40.476
The converse is also true.

26d82b3c-6769-49ed-94ea-40002ca91529/3345-0
00:16:40.776 --> 00:16:44.916
Sometimes it's Data Aronofsky
trains the system for outcomes,

26d82b3c-6769-49ed-94ea-40002ca91529/3345-1
00:16:44.916 --> 00:16:49.056
so again on the interpretability
side and face validity side.

26d82b3c-6769-49ed-94ea-40002ca91529/3357-0
00:16:45.336 --> 00:16:45.766
OK.

26d82b3c-6769-49ed-94ea-40002ca91529/3376-0
00:16:49.506 --> 00:16:53.884
Umm it it's it can cut both ways
and it just really speaks the

26d82b3c-6769-49ed-94ea-40002ca91529/3368-0
00:16:53.216 --> 00:16:53.656
Managed.

26d82b3c-6769-49ed-94ea-40002ca91529/3376-1
00:16:53.884 --> 00:16:56.246
need to do this type of
research.

26d82b3c-6769-49ed-94ea-40002ca91529/3389-0
00:16:56.726 --> 00:17:00.336
To metadata can provide a kind
of audit if you're trying to.

26d82b3c-6769-49ed-94ea-40002ca91529/3398-0
00:17:00.346 --> 00:17:02.936
If they've been problem problems
interpreting past data.

26d82b3c-6769-49ed-94ea-40002ca91529/3399-0
00:17:03.246 --> 00:17:03.686
Yep.

26d82b3c-6769-49ed-94ea-40002ca91529/3401-0
00:17:03.696 --> 00:17:04.386
Sometimes.

26d82b3c-6769-49ed-94ea-40002ca91529/3446-0
00:17:04.396 --> 00:17:08.326
Uh, but it also may provide a
more data rich environment from

26d82b3c-6769-49ed-94ea-40002ca91529/3446-1
00:17:08.326 --> 00:17:12.002
the model than to key on a
characteristic of the data set

26d82b3c-6769-49ed-94ea-40002ca91529/3446-2
00:17:12.002 --> 00:17:15.108
that a human wouldn't
necessarily pick up on and

26d82b3c-6769-49ed-94ea-40002ca91529/3446-3
00:17:15.108 --> 00:17:17.326
potentially make it more
accurate.

26d82b3c-6769-49ed-94ea-40002ca91529/3443-0
00:17:15.756 --> 00:17:15.956
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/3459-0
00:17:17.596 --> 00:17:19.186
It's hard to say.

26d82b3c-6769-49ed-94ea-40002ca91529/3476-0
00:17:18.006 --> 00:17:22.580
I before we wrap up, I I'd love
to hear about the contextualize

26d82b3c-6769-49ed-94ea-40002ca91529/3476-1
00:17:22.580 --> 00:17:26.796
this conversation about metadata
in your work with sepsis.

26d82b3c-6769-49ed-94ea-40002ca91529/3502-0
00:17:26.886 --> 00:17:30.606
I think that Cooper Healthcare
is a national leader in sepsis

26d82b3c-6769-49ed-94ea-40002ca91529/3502-1
00:17:30.606 --> 00:17:34.326
care and I think you have an
interesting story about metadata

26d82b3c-6769-49ed-94ea-40002ca91529/3502-2
00:17:34.326 --> 00:17:34.926
and stuff.

26d82b3c-6769-49ed-94ea-40002ca91529/3515-0
00:17:35.106 --> 00:17:35.606
Is that right?

26d82b3c-6769-49ed-94ea-40002ca91529/3510-0
00:17:35.856 --> 00:17:36.326
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/3531-0
00:17:36.336 --> 00:17:39.644
So early on we were doing a lot
of the the this early work and

26d82b3c-6769-49ed-94ea-40002ca91529/3531-1
00:17:39.644 --> 00:17:40.746
understanding sepsis.

26d82b3c-6769-49ed-94ea-40002ca91529/3536-0
00:17:41.876 --> 00:17:45.646
We were collecting what had face
validity.

26d82b3c-6769-49ed-94ea-40002ca91529/3550-0
00:17:45.996 --> 00:17:47.346
What's your heart rate this
hour?

26d82b3c-6769-49ed-94ea-40002ca91529/3555-0
00:17:47.356 --> 00:17:48.036
What's your heart rate?

26d82b3c-6769-49ed-94ea-40002ca91529/3564-0
00:17:48.046 --> 00:17:49.786
Next hour, what's your blood
pressure this hour?

26d82b3c-6769-49ed-94ea-40002ca91529/3572-0
00:17:49.936 --> 00:17:51.246
What your heart rate, your
heart.

26d82b3c-6769-49ed-94ea-40002ca91529/3589-0
00:17:51.256 --> 00:17:54.560
Blood pressure the next hour and
it turns out that that overall

26d82b3c-6769-49ed-94ea-40002ca91529/3589-1
00:17:54.560 --> 00:17:56.366
trend is really, really
important.

26d82b3c-6769-49ed-94ea-40002ca91529/3648-0
00:17:56.436 --> 00:17:59.585
But when you want to start to
create models that predict that

26d82b3c-6769-49ed-94ea-40002ca91529/3648-1
00:17:59.585 --> 00:18:02.836
earlier and earlier and earlier,
the amount of variation minute

26d82b3c-6769-49ed-94ea-40002ca91529/3648-2
00:18:02.836 --> 00:18:05.782
to minute, 2nd to 2nd is
actually a stronger predictor of

26d82b3c-6769-49ed-94ea-40002ca91529/3648-3
00:18:05.782 --> 00:18:08.981
patients that are going to then
deteriorate before the obvious

26d82b3c-6769-49ed-94ea-40002ca91529/3648-4
00:18:08.981 --> 00:18:11.876
deterioration of high heart rate
and low blood pressure.

26d82b3c-6769-49ed-94ea-40002ca91529/3656-0
00:18:12.076 --> 00:18:12.406
Uh-huh.

26d82b3c-6769-49ed-94ea-40002ca91529/3679-0
00:18:12.146 --> 00:18:15.772
Once you have high heart rate
and low blood pressure, you know

26d82b3c-6769-49ed-94ea-40002ca91529/3679-1
00:18:15.772 --> 00:18:18.994
you're really behind 8 ball
picking up that decrease in

26d82b3c-6769-49ed-94ea-40002ca91529/3677-0
00:18:18.666 --> 00:18:18.886
That.

26d82b3c-6769-49ed-94ea-40002ca91529/3679-2
00:18:18.994 --> 00:18:21.296
variation earlier on is very
important.

26d82b3c-6769-49ed-94ea-40002ca91529/3714-0
00:18:21.306 --> 00:18:24.454
So when we did a lot of our
initial early studies, we were

26d82b3c-6769-49ed-94ea-40002ca91529/3714-1
00:18:24.454 --> 00:18:27.549
capturing at the bedside heart
rate and blood pressure is

26d82b3c-6769-49ed-94ea-40002ca91529/3714-2
00:18:27.549 --> 00:18:29.096
emitted to minute 2nd to 2nd.

26d82b3c-6769-49ed-94ea-40002ca91529/3731-0
00:18:29.286 --> 00:18:32.816
But in the data sets that we
exported, we did it every hour.

26d82b3c-6769-49ed-94ea-40002ca91529/3798-0
00:18:32.966 --> 00:18:36.541
And then what happens is we did
some great research on that

26d82b3c-6769-49ed-94ea-40002ca91529/3798-1
00:18:36.541 --> 00:18:40.055
space and then we started to
learn that understanding this

26d82b3c-6769-49ed-94ea-40002ca91529/3798-2
00:18:40.055 --> 00:18:43.987
data at smaller and smaller time
intervals and understanding that

26d82b3c-6769-49ed-94ea-40002ca91529/3764-0
00:18:40.596 --> 00:18:40.816
And.

26d82b3c-6769-49ed-94ea-40002ca91529/3798-3
00:18:43.987 --> 00:18:47.681
variability could be predictive,
we couldn't go back and redo

26d82b3c-6769-49ed-94ea-40002ca91529/3798-4
00:18:47.681 --> 00:18:51.314
those studies because we just
jettisoned all that other data

26d82b3c-6769-49ed-94ea-40002ca91529/3798-5
00:18:51.314 --> 00:18:53.816
that would have been potentially
helpful.

26d82b3c-6769-49ed-94ea-40002ca91529/3836-0
00:18:53.966 --> 00:18:58.269
Now the issue there is back in
the day, is it very expensive to

26d82b3c-6769-49ed-94ea-40002ca91529/3836-1
00:18:58.269 --> 00:19:01.967
store all this data as that's
getting cheaper and more

26d82b3c-6769-49ed-94ea-40002ca91529/3826-0
00:18:59.896 --> 00:19:00.206
Happy.

26d82b3c-6769-49ed-94ea-40002ca91529/3836-2
00:19:01.967 --> 00:19:03.916
inexpensive and easier to do?

26d82b3c-6769-49ed-94ea-40002ca91529/3842-0
00:19:02.746 --> 00:19:02.896
It.

26d82b3c-6769-49ed-94ea-40002ca91529/3886-0
00:19:04.286 --> 00:19:07.672
I would always make the argument
that just collect as much data

26d82b3c-6769-49ed-94ea-40002ca91529/3853-0
00:19:06.206 --> 00:19:06.416
OK.

26d82b3c-6769-49ed-94ea-40002ca91529/3886-1
00:19:07.672 --> 00:19:10.899
with as much density as you can,
because you never know when

26d82b3c-6769-49ed-94ea-40002ca91529/3881-0
00:19:08.136 --> 00:19:08.466
That.

26d82b3c-6769-49ed-94ea-40002ca91529/3886-2
00:19:10.899 --> 00:19:14.338
that's going to be helpful when
it's economically feasible to do

26d82b3c-6769-49ed-94ea-40002ca91529/3889-0
00:19:12.266 --> 00:19:12.456
Well.

26d82b3c-6769-49ed-94ea-40002ca91529/3886-3
00:19:14.338 --> 00:19:14.496
it.

26d82b3c-6769-49ed-94ea-40002ca91529/3897-0
00:19:14.506 --> 00:19:16.676
I wish we had that second to
second.

26d82b3c-6769-49ed-94ea-40002ca91529/3910-0
00:19:16.966 --> 00:19:19.249
A heart rate data because we
could have done a lot more

26d82b3c-6769-49ed-94ea-40002ca91529/3910-1
00:19:19.249 --> 00:19:19.616
research.

26d82b3c-6769-49ed-94ea-40002ca91529/3957-0
00:19:19.886 --> 00:19:23.697
We ultimately went on to do some
of that work, but the initial

26d82b3c-6769-49ed-94ea-40002ca91529/3957-1
00:19:23.697 --> 00:19:27.629
work really the data set didn't
support some of the new findings

26d82b3c-6769-49ed-94ea-40002ca91529/3949-0
00:19:24.916 --> 00:19:24.976
E.

26d82b3c-6769-49ed-94ea-40002ca91529/3957-2
00:19:27.629 --> 00:19:31.500
people were getting out of that
data, which was that second and

26d82b3c-6769-49ed-94ea-40002ca91529/3957-3
00:19:31.500 --> 00:19:32.346
2nd variation.

26d82b3c-6769-49ed-94ea-40002ca91529/3983-0
00:19:32.836 --> 00:19:36.331
And when it comes to sepsis,
just even having like 20 minutes

26d82b3c-6769-49ed-94ea-40002ca91529/3983-1
00:19:36.331 --> 00:19:39.883
earlier notification or an hour
or notification can make a lot

26d82b3c-6769-49ed-94ea-40002ca91529/3983-2
00:19:39.883 --> 00:19:41.066
of difference, right?

26d82b3c-6769-49ed-94ea-40002ca91529/3988-0
00:19:40.766 --> 00:19:41.646
Umm yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/3996-0
00:19:41.516 --> 00:19:44.726
So how would you advise health
systems listening to this?

26d82b3c-6769-49ed-94ea-40002ca91529/4012-0
00:19:44.736 --> 00:19:48.846
How to balance a cost of just
collecting a bunch of data?

26d82b3c-6769-49ed-94ea-40002ca91529/4039-0
00:19:48.856 --> 00:19:52.471
And by the way, if it's a lot of
information that's not

26d82b3c-6769-49ed-94ea-40002ca91529/4039-1
00:19:52.471 --> 00:19:56.408
interpreted and it's not, and
there's just garbage, you know

26d82b3c-6769-49ed-94ea-40002ca91529/4039-2
00:19:56.408 --> 00:19:57.376
it needs to be.

26d82b3c-6769-49ed-94ea-40002ca91529/4049-0
00:19:57.436 --> 00:19:59.966
How do you advise an
organization to not?

26d82b3c-6769-49ed-94ea-40002ca91529/4067-0
00:20:00.006 --> 00:20:03.256
You can lose the meaning of the
data if you just have everything

26d82b3c-6769-49ed-94ea-40002ca91529/4067-1
00:20:03.256 --> 00:20:05.056
collecting everything all the
time.

26d82b3c-6769-49ed-94ea-40002ca91529/4072-0
00:20:05.116 --> 00:20:06.036
How do you advise them?

26d82b3c-6769-49ed-94ea-40002ca91529/4084-0
00:20:06.046 --> 00:20:11.427
How to the balance, cost and
organization and future value

26d82b3c-6769-49ed-94ea-40002ca91529/4084-1
00:20:11.427 --> 00:20:12.886
potential value?

26d82b3c-6769-49ed-94ea-40002ca91529/4110-0
00:20:12.936 --> 00:20:15.056
How you how would you do that?

26d82b3c-6769-49ed-94ea-40002ca91529/4092-0
00:20:15.496 --> 00:20:15.866
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/4166-0
00:20:15.876 --> 00:20:18.674
No, that's a really good point
because you know the classic

26d82b3c-6769-49ed-94ea-40002ca91529/4166-1
00:20:18.674 --> 00:20:21.379
counterexample here is I OT
right if I if I have my watch

26d82b3c-6769-49ed-94ea-40002ca91529/4166-2
00:20:21.379 --> 00:20:23.757
collecting a bunch of
accelerometer data and heart

26d82b3c-6769-49ed-94ea-40002ca91529/4166-3
00:20:23.757 --> 00:20:26.695
rate data, but it doesn't know
the difference between what I'm

26d82b3c-6769-49ed-94ea-40002ca91529/4166-4
00:20:26.695 --> 00:20:29.587
sleeping on, I'm awake or when
it's off of when it's on, it's

26d82b3c-6769-49ed-94ea-40002ca91529/4166-5
00:20:29.587 --> 00:20:30.006
etcetera.

26d82b3c-6769-49ed-94ea-40002ca91529/4181-0
00:20:30.126 --> 00:20:33.975
There's a lot of garbage in that
data, and despite having great

26d82b3c-6769-49ed-94ea-40002ca91529/4173-0
00:20:30.166 --> 00:20:30.306
Yeah.

26d82b3c-6769-49ed-94ea-40002ca91529/4181-1
00:20:33.975 --> 00:20:34.756
data density.

26d82b3c-6769-49ed-94ea-40002ca91529/4198-0
00:20:36.236 --> 00:20:40.424
Might not be helpful, so it's
the same as the technical depth

26d82b3c-6769-49ed-94ea-40002ca91529/4198-1
00:20:40.424 --> 00:20:41.166
discussion.

26d82b3c-6769-49ed-94ea-40002ca91529/4202-0
00:20:41.556 --> 00:20:43.056
Good stakeholder analysis.

26d82b3c-6769-49ed-94ea-40002ca91529/4217-0
00:20:43.066 --> 00:20:45.562
Getting the right experts and
subject matter experts in there

26d82b3c-6769-49ed-94ea-40002ca91529/4217-1
00:20:45.562 --> 00:20:46.246
to talk about it.

26d82b3c-6769-49ed-94ea-40002ca91529/4241-0
00:20:46.656 --> 00:20:50.150
Know what the latest literature
shows in terms of how these data

26d82b3c-6769-49ed-94ea-40002ca91529/4241-1
00:20:50.150 --> 00:20:53.106
points may relate to the outcome
you're interested in?

26d82b3c-6769-49ed-94ea-40002ca91529/4268-0
00:20:53.176 --> 00:20:57.070
And then balancing that with the
total cost of either curating

26d82b3c-6769-49ed-94ea-40002ca91529/4252-0
00:20:53.266 --> 00:20:54.376
And yes.

26d82b3c-6769-49ed-94ea-40002ca91529/4268-1
00:20:57.070 --> 00:21:00.346
the data to your point is, does
the data make sense?

26d82b3c-6769-49ed-94ea-40002ca91529/4285-0
00:21:00.956 --> 00:21:04.283
Is the data clean or is there
garbage right that there's a

26d82b3c-6769-49ed-94ea-40002ca91529/4285-1
00:21:04.283 --> 00:21:05.466
cost related to that?

26d82b3c-6769-49ed-94ea-40002ca91529/4373-0
00:21:05.476 --> 00:21:09.299
Either technical or human to
look and validate that data, and

26d82b3c-6769-49ed-94ea-40002ca91529/4373-1
00:21:09.299 --> 00:21:13.123
then as the ultimate cost of
that storage and manipulation of

26d82b3c-6769-49ed-94ea-40002ca91529/4373-2
00:21:13.123 --> 00:21:17.069
that data, so understanding the
need and the potential benefits

26d82b3c-6769-49ed-94ea-40002ca91529/4373-3
00:21:17.069 --> 00:21:20.645
of more metadata and data
density and match that with the

26d82b3c-6769-49ed-94ea-40002ca91529/4331-0
00:21:20.186 --> 00:21:20.406
The.

26d82b3c-6769-49ed-94ea-40002ca91529/4373-4
00:21:20.645 --> 00:21:24.530
total cost of doing that work
and then having just really good

26d82b3c-6769-49ed-94ea-40002ca91529/4373-5
00:21:24.530 --> 00:21:28.415
project management getting the
right stakeholders at the table

26d82b3c-6769-49ed-94ea-40002ca91529/4359-0
00:21:27.706 --> 00:21:27.976
It.

26d82b3c-6769-49ed-94ea-40002ca91529/4373-6
00:21:28.415 --> 00:21:31.930
and having that conversation
because in the past we just

26d82b3c-6769-49ed-94ea-40002ca91529/4373-7
00:21:31.930 --> 00:21:33.656
never had that conversation.

26d82b3c-6769-49ed-94ea-40002ca91529/4399-0
00:21:33.866 --> 00:21:37.683
So my recommendation would be
just have the conversation so at

26d82b3c-6769-49ed-94ea-40002ca91529/4380-0
00:21:34.146 --> 00:21:34.386
So.

26d82b3c-6769-49ed-94ea-40002ca91529/4399-1
00:21:37.683 --> 00:21:41.136
least you know what you risk by
not saving all the data.

26d82b3c-6769-49ed-94ea-40002ca91529/4408-0
00:21:42.186 --> 00:21:44.916
I appreciate your time with us
today, Michael.

26d82b3c-6769-49ed-94ea-40002ca91529/4423-0
00:21:44.926 --> 00:21:48.987
We covered a lot of information
from meta technical debt to

26d82b3c-6769-49ed-94ea-40002ca91529/4423-1
00:21:48.987 --> 00:21:49.596
metadata.

26d82b3c-6769-49ed-94ea-40002ca91529/4452-0
00:21:50.026 --> 00:21:53.902
You've kind of already addressed
this, but if there was something

26d82b3c-6769-49ed-94ea-40002ca91529/4452-1
00:21:53.902 --> 00:21:57.659
that you wish you knew, if you
could speak to Michael Kirchhoff

26d82b3c-6769-49ed-94ea-40002ca91529/4452-2
00:21:57.659 --> 00:21:58.716
of four years ago.

26d82b3c-6769-49ed-94ea-40002ca91529/4454-0
00:21:58.906 --> 00:21:59.516
What?

26d82b3c-6769-49ed-94ea-40002ca91529/4485-0
00:21:59.526 --> 00:22:03.211
What would you tell yourself and
engaging in with kind of

26d82b3c-6769-49ed-94ea-40002ca91529/4485-1
00:22:03.211 --> 00:22:07.150
considering patient safety with
regards to technical debt and

26d82b3c-6769-49ed-94ea-40002ca91529/4485-2
00:22:07.150 --> 00:22:10.326
reviewing metadata, what would
you tell yourself?

26d82b3c-6769-49ed-94ea-40002ca91529/4527-0
00:22:11.796 --> 00:22:16.549
To really think about the
implications of it, what I find

26d82b3c-6769-49ed-94ea-40002ca91529/4527-1
00:22:16.549 --> 00:22:20.728
myself doing a lot of now is
really looking at the

26d82b3c-6769-49ed-94ea-40002ca91529/4527-2
00:22:20.728 --> 00:22:25.808
implications of this data as it
pertains to bias and equality

26d82b3c-6769-49ed-94ea-40002ca91529/4527-3
00:22:25.808 --> 00:22:27.856
and access in healthcare.

26d82b3c-6769-49ed-94ea-40002ca91529/4583-0
00:22:28.036 --> 00:22:31.883
Because that's ultimately we're
charged with the good keeping

26d82b3c-6769-49ed-94ea-40002ca91529/4583-1
00:22:31.883 --> 00:22:34.985
and the health of our
communities and knowing how

26d82b3c-6769-49ed-94ea-40002ca91529/4583-2
00:22:34.985 --> 00:22:38.398
whatever tool we're applying
applies to our particular

26d82b3c-6769-49ed-94ea-40002ca91529/4583-3
00:22:38.398 --> 00:22:41.686
communities and their
demographics and their data is

26d82b3c-6769-49ed-94ea-40002ca91529/4583-4
00:22:41.686 --> 00:22:45.346
something that I think every
health system struggles with.

26d82b3c-6769-49ed-94ea-40002ca91529/4603-0
00:22:45.356 --> 00:22:49.442
And it's a big item for us to
have an eye on because we're,

26d82b3c-6769-49ed-94ea-40002ca91529/4603-1
00:22:49.442 --> 00:22:51.076
it's our mandate, right?

26d82b3c-6769-49ed-94ea-40002ca91529/4717-0
00:22:51.136 --> 00:22:54.602
We need to care for those in our
communities and those

26d82b3c-6769-49ed-94ea-40002ca91529/4717-1
00:22:54.602 --> 00:22:58.320
communities vary from place to
place and time to time, and

26d82b3c-6769-49ed-94ea-40002ca91529/4717-2
00:22:58.320 --> 00:23:01.787
knowing how systems may or may
not meet the needs if a

26d82b3c-6769-49ed-94ea-40002ca91529/4717-3
00:23:01.787 --> 00:23:05.379
particular community is
something that we're focusing on

26d82b3c-6769-49ed-94ea-40002ca91529/4717-4
00:23:05.379 --> 00:23:09.160
a lot now because in the past
many vendors did not focus on

26d82b3c-6769-49ed-94ea-40002ca91529/4663-0
00:23:07.396 --> 00:23:07.636
That's.

26d82b3c-6769-49ed-94ea-40002ca91529/4717-5
00:23:09.160 --> 00:23:13.130
that and we find we're playing a
little catch up on that right

26d82b3c-6769-49ed-94ea-40002ca91529/4717-6
00:23:13.130 --> 00:23:16.974
now and knowing that back in the
day and really asking those

26d82b3c-6769-49ed-94ea-40002ca91529/4693-0
00:23:15.006 --> 00:23:15.166
And.

26d82b3c-6769-49ed-94ea-40002ca91529/4717-7
00:23:16.974 --> 00:23:20.692
questions, especially of the
vendors who were creating the

26d82b3c-6769-49ed-94ea-40002ca91529/4717-8
00:23:20.692 --> 00:23:24.410
systems to make sure that they
have inclusive data that is

26d82b3c-6769-49ed-94ea-40002ca91529/4722-0
00:23:23.916 --> 00:23:24.046
The.

26d82b3c-6769-49ed-94ea-40002ca91529/4717-9
00:23:24.410 --> 00:23:25.166
represented.

26d82b3c-6769-49ed-94ea-40002ca91529/4751-0
00:23:25.446 --> 00:23:29.064
Of our patient population and
that that data is free of biases

26d82b3c-6769-49ed-94ea-40002ca91529/4739-0
00:23:27.746 --> 00:23:27.966
And.

26d82b3c-6769-49ed-94ea-40002ca91529/4751-1
00:23:29.064 --> 00:23:32.279
is something those were those
were not frequently asked

26d82b3c-6769-49ed-94ea-40002ca91529/4751-2
00:23:32.279 --> 00:23:34.116
questions three, five years ago.

26d82b3c-6769-49ed-94ea-40002ca91529/4763-0
00:23:34.126 --> 00:23:35.976
Those are questions we asked
every time now.

26d82b3c-6769-49ed-94ea-40002ca91529/4782-0
00:23:37.236 --> 00:23:40.386
For our listeners, this has been
doctor Michael Kirchhoff, the

26d82b3c-6769-49ed-94ea-40002ca91529/4782-1
00:23:40.386 --> 00:23:43.486
chief innovation and Patient
Safety officer at Cooper Health.

26d82b3c-6769-49ed-94ea-40002ca91529/4795-0
00:23:43.576 --> 00:23:45.106
Michael, thank you so much for
joining us today.