1
00:00:00,480 --> 00:00:02,850
This is Laura Dedo with the
Becker's Healthcare Podcast.

2
00:00:03,470 --> 00:00:05,810
I'm thrilled today to be
joined by Dr. Chen, Kai Kao,

3
00:00:05,940 --> 00:00:09,850
who's the Chief Clinical
Informatics Innovation Officer
at Uch Chicago Medicine.

4
00:00:10,230 --> 00:00:12,090
Kai, it's a pleasure to have
you on the podcast today.

5
00:00:13,700 --> 00:00:16,210
Thank you so much for having me.
It's such a pleasure to be here.

6
00:00:17,070 --> 00:00:18,930
Now, I know we have a lot to talk about.

7
00:00:18,930 --> 00:00:21,290
There's so much happening in
the clinical informatics space,

8
00:00:21,310 --> 00:00:23,290
and certainly I'm intrigued by your title,

9
00:00:23,290 --> 00:00:26,050
having both the clinical
informatics and innovation in there.

10
00:00:26,150 --> 00:00:28,250
But before we dive into
my broader questions,

11
00:00:28,270 --> 00:00:30,370
can you tell us a little bit more
about yourself and your background?

12
00:00:32,020 --> 00:00:35,440
Of course. So, um, I'm a
hospitalist by training. Um,

13
00:00:35,440 --> 00:00:37,960
so I spend a lot of time
seeing inpatients, um,

14
00:00:37,960 --> 00:00:41,560
mostly general medicine patients,
and have done that for, you know,

15
00:00:41,560 --> 00:00:45,200
about 12 to 13 years by now.
And I also hold, you know,

16
00:00:45,200 --> 00:00:48,880
some administrative positions in
hospital, including, as you mentioned,

17
00:00:48,880 --> 00:00:53,320
the Chief Clinical Run Innovation
Officer, which I really focused a lot of,

18
00:00:53,460 --> 00:00:57,960
you know, um, informatics operations,
especially for inpatient, uh,

19
00:00:58,240 --> 00:01:01,280
capacity access and
qualities. And also, you know,

20
00:01:01,280 --> 00:01:03,720
thinking about on the innovative
side, you know, you know,

21
00:01:03,940 --> 00:01:08,640
all the interesting projects that really
help accelerate the care. You know,

22
00:01:08,640 --> 00:01:12,320
improving the quality and
also expanding our access, uh,

23
00:01:12,320 --> 00:01:13,520
sort of the things that I do as well.

24
00:01:15,900 --> 00:01:16,500
Excellent. Wow.

25
00:01:16,500 --> 00:01:19,660
So it sounds like you certainly have
your hands full at your Chicago medicine

26
00:01:19,720 --> 00:01:20,180
and you know,

27
00:01:20,180 --> 00:01:23,180
from your perspective and vantage
point with so many different roles,

28
00:01:23,270 --> 00:01:26,740
where do you see the big opportunities
for healthcare as well as the headwinds

29
00:01:26,740 --> 00:01:27,980
that you have your eyes on right now?

30
00:01:29,670 --> 00:01:32,080
Yeah, I think we have, you know,

31
00:01:32,110 --> 00:01:35,720
obviously we are in the post
covid era where, you know,

32
00:01:35,800 --> 00:01:40,720
a lot of things are still, you know,
being restored. Um, but in the meantime,

33
00:01:40,740 --> 00:01:44,960
if one thing accelerated through
covid, that will be the overall,

34
00:01:45,060 --> 00:01:48,960
the telehealth and, and overall
the informatics, um, you know, um,

35
00:01:48,960 --> 00:01:53,440
opportunities. Um, so we're excited
to see, you know, there's a lot of,

36
00:01:53,780 --> 00:01:57,800
um, booming, um, programs around
how do we spend, you know,

37
00:01:58,040 --> 00:02:00,360
outreach to the patients
through, for example,

38
00:02:00,360 --> 00:02:05,200
remote patient monitoring or Apache
Ho I spend a lot of time, um, um,

39
00:02:05,200 --> 00:02:08,680
it's called Hospital at home, uh,
ho essentially providing, you know,

40
00:02:08,690 --> 00:02:11,640
acute hospital, um, care
to the patients home.

41
00:02:12,100 --> 00:02:16,680
So I see a lot of opportunity where
home become a new alternative site

42
00:02:16,900 --> 00:02:21,680
of care in the future. And, you
know, a lot of these services, um,

43
00:02:21,680 --> 00:02:26,600
enabled by technology informatics,
um, solutions were, you know, um,

44
00:02:26,790 --> 00:02:30,640
helping us to be able to provide the
care in the patient's home, you know,

45
00:02:30,640 --> 00:02:32,280
where they really belongs. Um,

46
00:02:32,340 --> 00:02:36,080
so I think there's really
a lot of opportunity here
and that's why we have been,

47
00:02:36,220 --> 00:02:38,120
um, you know, working on
over the last three years.

48
00:02:39,670 --> 00:02:43,570
That's great to hear. And
you know, especially looking
at the hospital at home,

49
00:02:43,770 --> 00:02:47,050
I know that's a trend across the country
and some organizations are further

50
00:02:47,050 --> 00:02:50,370
along the journey than others.
And, you know, in your experience,

51
00:02:50,440 --> 00:02:51,810
what have you learned in, um,

52
00:02:51,810 --> 00:02:55,930
setting up a program like that and
what does it take to, to be successful?

53
00:02:57,840 --> 00:03:00,060
Yeah, we are just started, um,

54
00:03:00,160 --> 00:03:04,460
we have been planning this for almost
two years and we just saw our first

55
00:03:04,780 --> 00:03:05,660
patients, um,

56
00:03:05,660 --> 00:03:09,900
earlier this year and we are now planning
to launch hopefully in the next few

57
00:03:09,900 --> 00:03:14,220
months. Um, you know, officially,
so very exciting for us. Um, and,

58
00:03:14,560 --> 00:03:16,940
you know, you asked the
key question, you know,

59
00:03:16,940 --> 00:03:20,900
where it really takes a lot
of resources and planning, um,

60
00:03:20,900 --> 00:03:25,660
to kick up this such a program. Um, we,
we think there is a lot of value. Um,

61
00:03:25,680 --> 00:03:28,980
so I think recognizing what the
value is for the organization,

62
00:03:29,030 --> 00:03:33,140
cause the answer could
be slightly different for
organization to organization,

63
00:03:33,600 --> 00:03:38,140
but for us it would be improving
our capacity access. Uh,

64
00:03:38,200 --> 00:03:42,180
we constantly are full all the
time, um, in the hospital. Um,

65
00:03:42,180 --> 00:03:44,020
so there's not a way to quickly, you know,

66
00:03:44,220 --> 00:03:47,980
increase the bad numbers that
open us up for more, um, you know,

67
00:03:48,000 --> 00:03:52,700
sicker patients that can, um, you know,
um, be hospitalized. Um, so really,

68
00:03:52,800 --> 00:03:56,380
you know, hospital home becomes a
way to really expand the access, um,

69
00:03:56,380 --> 00:03:59,260
allow more patients to be
treated. In the meantime,

70
00:03:59,300 --> 00:04:02,180
I think there's a huge component
for patient experience. Um,

71
00:04:02,320 --> 00:04:07,020
so patients are up and mobile a lot more
often than they traditionally are in

72
00:04:07,020 --> 00:04:09,420
the hospital. In the first
patient that we just had,

73
00:04:10,060 --> 00:04:14,860
I observed the patient in the hospital
where he was mostly bedbound because he

74
00:04:15,120 --> 00:04:17,220
was deemed have to have
certain floor risk.

75
00:04:17,640 --> 00:04:20,220
And so even though he has his
cane just right next to him,

76
00:04:20,440 --> 00:04:21,380
it doesn't walk around.

77
00:04:21,400 --> 00:04:25,380
He literally just pretty much tied to
the bed and watch TV most of the time.

78
00:04:25,720 --> 00:04:29,580
But once we transfer him home, we
saw him actually get up, walk around,

79
00:04:30,050 --> 00:04:31,660
uses Cane to walk in the living room,

80
00:04:32,120 --> 00:04:34,060
he eating the meal that
we send over to him,

81
00:04:34,200 --> 00:04:38,940
and just quickly we see the smiley
face, the energy in him just, you know,

82
00:04:39,230 --> 00:04:41,100
after a few hours we return from him home.

83
00:04:41,400 --> 00:04:44,540
So we believe home is really the
optimal place for a lot of patients,

84
00:04:44,540 --> 00:04:48,940
especially elderly patients to recover.
And I think that's something that,

85
00:04:49,200 --> 00:04:49,540
you know,

86
00:04:49,540 --> 00:04:53,460
even though we know for sure we still
not be able to provide care at home

87
00:04:53,460 --> 00:04:57,820
because, you know, a lot of limitations
we had before. Um, but now with the,

88
00:04:58,200 --> 00:05:01,980
you know, with the covid, you know,
with the waiver program in front of cms,

89
00:05:02,280 --> 00:05:05,860
now we finally have the opportunity and
also reimbursement mechanism to actually

90
00:05:05,860 --> 00:05:09,260
make this work. So I think
recognizing those values, um,

91
00:05:09,580 --> 00:05:13,490
identifying if that's important for
your organization, then it's all about,

92
00:05:14,190 --> 00:05:18,650
you know, investments and, and software
planning. Um, I would just say this is,

93
00:05:18,830 --> 00:05:22,050
uh, a very, um, robust community, um,

94
00:05:22,060 --> 00:05:24,050
given the programs still
pretty young and new,

95
00:05:24,640 --> 00:05:29,010
most of the hospital home programs
across the country already supporting one

96
00:05:29,010 --> 00:05:32,610
another. Um, so if there's any
questions, you know, um, you know,

97
00:05:32,610 --> 00:05:34,850
there's always a lot of
people who were, you know,

98
00:05:34,850 --> 00:05:39,250
happy to answering those questions
voluntarily. And if anyone, you know,

99
00:05:39,270 --> 00:05:42,250
is listening are interested in knowing
more about Host Their Home, again,

100
00:05:42,290 --> 00:05:45,490
I would really love to return the favor
as well, and happy to get that resource,

101
00:05:45,790 --> 00:05:47,650
um, provide you if there's any questions.

102
00:05:49,030 --> 00:05:52,920
Fantastic. I, I think that's so helpful
to know and just really great results,

103
00:05:53,260 --> 00:05:57,860
um, even for a single person to be able
to have that opportunity to go home and

104
00:05:57,860 --> 00:06:01,340
really come to life more. I, I
think that's so important. Now,

105
00:06:01,570 --> 00:06:02,700
when you look more broadly,

106
00:06:02,920 --> 00:06:05,980
how are you thinking about growth and
adding value to the organization overall?

107
00:06:08,090 --> 00:06:11,540
Yeah, so, um, it's just like I
mentioned, I think, you know, um,

108
00:06:11,920 --> 00:06:15,700
we need to really improve the access
for both inpatient and outpatients.

109
00:06:15,840 --> 00:06:18,940
And I think we are doing a
lot more now through, um,

110
00:06:18,940 --> 00:06:23,740
also at home and telemedicine, um, type
of initiative, um, to expand the access.

111
00:06:24,080 --> 00:06:25,060
Um, but in the meantime,

112
00:06:25,060 --> 00:06:29,620
we also need to just make it easier for
patients to actually, um, contact us,

113
00:06:29,650 --> 00:06:32,660
make appointments, ask
questions. So I think, you know,

114
00:06:32,700 --> 00:06:35,940
a lot of the initiative
around digital front door,

115
00:06:36,060 --> 00:06:39,580
a lot of initiative around how do we
improve the patient engagement, you know,

116
00:06:39,580 --> 00:06:43,220
through the patient portal and how do we
actually direct the patient to make the

117
00:06:43,220 --> 00:06:45,660
appointment to the right providers.
I mean, I think there's a lot,

118
00:06:45,660 --> 00:06:48,340
still a lot of opportunities
where, you know, um,

119
00:06:48,340 --> 00:06:52,180
that we need to work on to improve the
growth and also add more value. And,

120
00:06:52,200 --> 00:06:56,340
and ultimately I think traditionally
healthcare has been a very, a episodic,

121
00:06:57,160 --> 00:06:59,740
um, services where patient gets sick,

122
00:06:59,930 --> 00:07:04,100
they either go to our urgent care clinic
or go to ED and be hospitalized even.

123
00:07:04,680 --> 00:07:08,260
Um, but all we know is like
per episode, like in between,

124
00:07:08,280 --> 00:07:13,180
we don't have a ton of, um, way
to engage and also, um, monitor,

125
00:07:13,640 --> 00:07:18,220
um, you know, their progress.
Um, and but now I think again,

126
00:07:18,220 --> 00:07:22,260
with remote patient monitoring with, you
know, some of the AI that's coming up,

127
00:07:22,520 --> 00:07:24,860
I'm sure it's on hot
topics everywhere, um,

128
00:07:24,860 --> 00:07:28,660
that there's some interesting opportunity
where we can provide more seamless

129
00:07:28,660 --> 00:07:32,980
service, um, by leveraging
these, um, new tools to, um,

130
00:07:33,280 --> 00:07:35,460
engage patients and
answer their questions.

131
00:07:37,620 --> 00:07:38,990
That makes a lot of sense. You know,

132
00:07:38,990 --> 00:07:42,750
and it's really fascinating to hear
about how the Paradigm healthcare is

133
00:07:42,830 --> 00:07:44,510
shifting and changing, you know, to,

134
00:07:44,570 --> 00:07:47,590
to bring in more patient
engagement as well as connection,

135
00:07:48,010 --> 00:07:52,830
the episodes of care and,
and really bridging that gap
from your perspective in,

136
00:07:52,850 --> 00:07:55,150
in thinking about how healthcare
is changing, evolving,

137
00:07:55,540 --> 00:07:59,510
what is an important risk or an investment
that is really worth making this

138
00:07:59,510 --> 00:08:00,343
year?

139
00:08:02,080 --> 00:08:06,920
I mean, on the news, obviously LGBT
is everywhere. Um, and you know, I,

140
00:08:06,960 --> 00:08:10,960
I think all in all, I think
we, it's worth to really, um,

141
00:08:11,980 --> 00:08:16,320
not necessarily invest, but
at least explore, um, you
know, all these, you know,

142
00:08:16,620 --> 00:08:18,440
AI solutions. It's ongoing.

143
00:08:18,510 --> 00:08:22,080
It's not just about large
language models such as J G B T,

144
00:08:22,080 --> 00:08:26,160
but there's also a couple of others, you
know, for imaging, for videos. I mean,

145
00:08:26,160 --> 00:08:30,680
we are in the era where there's a
revolution coming in, um, you know,

146
00:08:30,680 --> 00:08:34,440
through these ai. So how would
that impact our healthcare? Um,

147
00:08:34,460 --> 00:08:37,040
how do we make sure we put
that in the guard rail so you,

148
00:08:37,060 --> 00:08:40,520
you would benefit our patients, but in
the meantime we make sure, for example,

149
00:08:40,540 --> 00:08:42,960
you wouldn't provide any false
information to our patients.

150
00:08:43,320 --> 00:08:46,200
I think that's definitely one
thing worth exploring and,

151
00:08:46,220 --> 00:08:48,560
and really coming up with
the structure for that.

152
00:08:48,710 --> 00:08:52,800
Like how do we government that in the
meantime, being able to support and adapt,

153
00:08:53,060 --> 00:08:56,140
you know, all these newest solutions,
um, you know, from the market.

154
00:08:58,380 --> 00:09:00,020
Fascinating to hear about. And, and, um,

155
00:09:00,020 --> 00:09:04,060
I really love your point in terms of
looking at artificial intelligence and how

156
00:09:04,060 --> 00:09:07,620
that can incorporate in healthcare,
especially clinical care, effectively in,

157
00:09:07,760 --> 00:09:10,580
you know, when you talk about
the images and, and, um,

158
00:09:11,300 --> 00:09:13,460
I know there's a lot
going on there as well.

159
00:09:13,960 --> 00:09:17,900
Are there any technologies that you've
been particularly intrigued by or just in

160
00:09:17,900 --> 00:09:21,220
general looking at artificial
intelligence, large language
models and how that,

161
00:09:21,560 --> 00:09:22,860
um, some of those, you know,

162
00:09:22,860 --> 00:09:24,980
data-driven initiatives can
really make a difference?

163
00:09:26,380 --> 00:09:30,350
Yeah, we are looking at this in a
couple different fronts. Um, so one is,

164
00:09:30,410 --> 00:09:33,750
you know, you know, just on
the news there's, you know, um,

165
00:09:34,020 --> 00:09:38,310
partnership between Epic Open
AI and, and Microsoft as well.

166
00:09:38,850 --> 00:09:43,710
So we are also exploring, you know,
the, the implementation of, um,

167
00:09:43,810 --> 00:09:48,470
you know, piloting using essentially GD
like, um, language models in, you know,

168
00:09:48,470 --> 00:09:53,470
sort of the ehr, um, which can really
help improve the documentation,

169
00:09:53,780 --> 00:09:58,350
ease of practice in the meantime also
potentially can improve the patient, um,

170
00:09:58,350 --> 00:10:02,470
experience as well by having,
you know, a more, you know,

171
00:10:02,720 --> 00:10:07,110
human um, and model. They can talk to the
patient, provide some of the feedback.

172
00:10:07,610 --> 00:10:08,630
Um, but at minimum,

173
00:10:08,810 --> 00:10:13,230
at least you can help us draft some of
the messages and be more efficient, um,

174
00:10:13,230 --> 00:10:17,150
which just help solve a lot of
issues obviously around burn out. Um,

175
00:10:17,150 --> 00:10:19,390
cause you know, we, we keep talking about,

176
00:10:19,390 --> 00:10:21,510
so there's a lot of burn
out not just in providers,

177
00:10:21,530 --> 00:10:24,950
but in nurses and essentially
everyone in healthcare. Um,

178
00:10:24,970 --> 00:10:29,350
so this is just one example where
I think AI can help us be more, um,

179
00:10:30,010 --> 00:10:34,590
uh, more efficient, um, and in some way
more capable, um, because, you know,

180
00:10:34,650 --> 00:10:38,350
we usually wouldn't have that much
time to really come up with, uh,

181
00:10:38,510 --> 00:10:40,510
a long message thorough messages.

182
00:10:40,530 --> 00:10:44,670
But now AI can actually help put together
a lot of meaningful summaries in a

183
00:10:44,670 --> 00:10:47,510
relatively be short period
of time. And, um, you know,

184
00:10:47,510 --> 00:10:49,670
we still need people to
sign off and, you know,

185
00:10:49,670 --> 00:10:53,350
before this message was sent out,
but that obviously largely, um,

186
00:10:53,350 --> 00:10:57,030
reduced the time, um, um,
that we spent. You know,

187
00:10:57,110 --> 00:11:00,310
I also think there's just a lot of
opportunities in the future, you know,

188
00:11:00,310 --> 00:11:05,110
in this regard. Um, you know,
we are also exploring like, um,

189
00:11:05,960 --> 00:11:09,940
you know, can we apply for example,
some of the robots in healthcare?

190
00:11:10,160 --> 00:11:13,940
So we have, you know, we are actually
purchasing and not purchasing, um,

191
00:11:14,260 --> 00:11:19,220
contracting with some monthly robots
that can help to support, um, you know,

192
00:11:19,250 --> 00:11:22,980
some of the medications
or labs basements, um,

193
00:11:23,200 --> 00:11:27,340
to help us just be more efficient. Um,
you know, even including all of ours.

194
00:11:28,000 --> 00:11:30,780
Um, there's also a couple
of other like mbn um,

195
00:11:31,250 --> 00:11:35,300
technical services we are also looking
into to see if they can further, um,

196
00:11:35,300 --> 00:11:38,620
improve our, um, care efficiency
and, and patient experience.

197
00:11:39,400 --> 00:11:42,730
Fantastic. And, and that is
really great to hear. Now,

198
00:11:42,730 --> 00:11:45,290
before we wrap up our
conversation, I just wanted to ask,

199
00:11:45,350 --> 00:11:48,210
are there any other opportunities for
growth that you're seeing in the future?

200
00:11:49,850 --> 00:11:54,220
Um, I would just say, oh no, I think the
application of AI and the automation,

201
00:11:54,410 --> 00:11:58,460
like, you know, if we can take
off the burden, um, you know,

202
00:11:58,570 --> 00:12:03,500
from essentially any clinician in
the care processes by doing AI and

203
00:12:03,500 --> 00:12:07,300
automation, then they can really be
focusing on the more important stuff,

204
00:12:07,300 --> 00:12:11,780
which is talking to the patients. Because
I think really think nothing replaces,

205
00:12:12,320 --> 00:12:16,620
you know, the face-to-face interactions
and conversation. Um, so we,

206
00:12:16,640 --> 00:12:19,940
we sh share a lot about, you
know, the opportunities and,

207
00:12:19,940 --> 00:12:21,700
and applications of these technology,

208
00:12:21,760 --> 00:12:25,940
but I think fundamentally our goal is
to be able to make us more efficient so

209
00:12:25,940 --> 00:12:27,660
people can spend more
time with the patients,

210
00:12:28,030 --> 00:12:32,020
focus on more complex cases where
AI won't be able to help, um,

211
00:12:32,020 --> 00:12:33,940
fully and really, you know,

212
00:12:33,940 --> 00:12:37,980
improve the care processes and
experience in those regards. Um,

213
00:12:38,140 --> 00:12:41,020
I think in the future,
you know, um, everyone,

214
00:12:41,360 --> 00:12:46,020
not just healthcare people just need to
learn how to use these AI tools cuz the

215
00:12:46,020 --> 00:12:48,540
people who know how to use
them ultimately, you know,

216
00:12:48,540 --> 00:12:51,940
will be a lot more efficient
than those who don't. Um, and,

217
00:12:51,960 --> 00:12:56,900
and so this becomes sort of a homework
assignment or, or sort of things we, we,

218
00:12:57,280 --> 00:12:58,900
you know, we, we have to learn, you know,

219
00:12:58,900 --> 00:13:02,940
in the next few years about how we
actually apply that in a good way in

220
00:13:02,940 --> 00:13:03,773
healthcare.

221
00:13:05,570 --> 00:13:09,140
Fantastic. Well, Kai, thank you so much
for joining us on the podcast today.

222
00:13:09,220 --> 00:13:12,820
This has been such a fun and insightful
conversation and I'm really looking

223
00:13:12,820 --> 00:13:16,420
forward to your sessions as well at the
Health IT Digital Health and Revenue

224
00:13:16,430 --> 00:13:20,900
Cycle event in October. I think it'll
be amazing time in such, uh, great,

225
00:13:21,320 --> 00:13:25,780
um, information and data sharing and
uh, all sorts of conversations going on.

226
00:13:25,880 --> 00:13:27,420
So I'm looking forward
to seeing you there.

227
00:13:28,930 --> 00:13:30,400
Thank you so much, and again,

228
00:13:30,400 --> 00:13:34,000
thanks again for having me and very much
looking forward to the event as well.