1
00:00:00,640 --> 00:00:03,199
Philips is a health tech leader focused on

2
00:00:03,199 --> 00:00:06,000
innovation that improves the health and well-being of

3
00:00:06,000 --> 00:00:09,460
people. Our health care technology and informatics solutions

4
00:00:09,599 --> 00:00:12,559
help care teams diagnose, treat, and manage more

5
00:00:12,559 --> 00:00:16,244
patients with greater precision, speed, and confidence across

6
00:00:16,244 --> 00:00:19,225
the care journey. With Philips, clinicians are empowered

7
00:00:19,285 --> 00:00:21,765
with streamlined insights in the moments that matter

8
00:00:21,765 --> 00:00:24,904
for every patient. Better care for more people.

9
00:00:25,204 --> 00:00:25,704
Philips.

10
00:00:28,890 --> 00:00:30,969
This is Haley Recker with the Becker's Health

11
00:00:30,969 --> 00:00:33,450
Care Podcast. We are live at the 9th

12
00:00:33,450 --> 00:00:36,350
annual health IT Digital Health and RCM conference.

13
00:00:36,570 --> 00:00:39,149
And today, I am joined by Nabil Shahadi,

14
00:00:39,369 --> 00:00:41,850
who is the chief clinical transformation officer at

15
00:00:41,850 --> 00:00:42,350
MetroHealth.

16
00:00:42,835 --> 00:00:44,835
Nabil, thank you so much for speaking with

17
00:00:44,835 --> 00:00:45,494
me today.

18
00:00:46,195 --> 00:00:47,795
To get us started, can you please share

19
00:00:47,795 --> 00:00:49,715
a bit about yourself, your background, and your

20
00:00:49,715 --> 00:00:50,774
role at your organization?

21
00:00:51,554 --> 00:00:55,015
Thank you, Haley, for hosting this podcast today.

22
00:00:56,000 --> 00:00:59,460
I'm a a physician, a urologist by training.

23
00:01:00,079 --> 00:01:03,039
I am the chief clinical transformation officer at

24
00:01:03,039 --> 00:01:03,539
MetroHealth,

25
00:01:04,159 --> 00:01:07,619
and MetroHealth is a super safety net hospital

26
00:01:07,840 --> 00:01:08,979
in Cleveland, Ohio.

27
00:01:09,465 --> 00:01:10,524
We are about $2,000,000,000

28
00:01:11,224 --> 00:01:11,884
on revenue,

29
00:01:12,984 --> 00:01:13,804
5 hospitals,

30
00:01:15,704 --> 00:01:17,084
pre emergency room,

31
00:01:17,704 --> 00:01:19,884
and more than 30 outpatient,

32
00:01:20,504 --> 00:01:22,924
clinic. We serve more than

33
00:01:23,549 --> 00:01:26,670
60% of our patient are governmental patients, which

34
00:01:26,670 --> 00:01:29,810
means Medicaid, Medicare patients, and uninsured.

35
00:01:31,069 --> 00:01:33,149
Alright. Thank you so much for sharing. So

36
00:01:33,149 --> 00:01:35,325
for my first question, I wanted to ask

37
00:01:35,484 --> 00:01:37,965
about AI adoption, which is currently exploding in

38
00:01:37,965 --> 00:01:40,125
health care. So in your view, what's the

39
00:01:40,125 --> 00:01:42,944
most significant or promising application of this technology

40
00:01:43,004 --> 00:01:45,244
right now, and how is this informing your

41
00:01:45,244 --> 00:01:47,024
organization's innovation strategy?

42
00:01:48,159 --> 00:01:50,479
Thank you for this great question. There's no

43
00:01:50,479 --> 00:01:52,799
easy answer for it. There's a lot of,

44
00:01:53,200 --> 00:01:54,340
potential excitement,

45
00:01:55,679 --> 00:01:58,819
but potentially also so many potential disappointment,

46
00:01:59,599 --> 00:02:00,340
as well.

47
00:02:00,719 --> 00:02:02,500
So how do we tackle this?

48
00:02:03,215 --> 00:02:05,875
I think few things we we can,

49
00:02:06,494 --> 00:02:06,994
zoom

50
00:02:07,375 --> 00:02:09,634
into and and have a conversation about.

51
00:02:10,334 --> 00:02:11,074
1 is,

52
00:02:11,694 --> 00:02:12,194
into

53
00:02:12,574 --> 00:02:13,074
data

54
00:02:13,775 --> 00:02:14,094
and,

55
00:02:14,574 --> 00:02:16,034
predictive data analytics.

56
00:02:16,960 --> 00:02:19,219
How can you leverage AI

57
00:02:19,520 --> 00:02:20,740
into streamlining,

58
00:02:21,680 --> 00:02:25,199
that data and and predict a disease before

59
00:02:25,199 --> 00:02:25,939
it becomes,

60
00:02:26,319 --> 00:02:27,620
a much bigger problem?

61
00:02:28,080 --> 00:02:30,719
How do you use it to help manage

62
00:02:30,719 --> 00:02:31,699
chronic disease

63
00:02:32,224 --> 00:02:33,764
on an outpatient setting?

64
00:02:34,385 --> 00:02:36,165
How do you use it

65
00:02:36,944 --> 00:02:40,405
within the inpatient setting to predict a potential

66
00:02:41,344 --> 00:02:44,625
bad outcome before it happens? Like potential very

67
00:02:44,625 --> 00:02:45,284
well known,

68
00:02:46,340 --> 00:02:46,840
sepsis

69
00:02:47,219 --> 00:02:48,280
sepsis prediction

70
00:02:48,900 --> 00:02:49,799
or fall,

71
00:02:50,259 --> 00:02:52,259
prediction on the nursing unit. So how do

72
00:02:52,259 --> 00:02:54,359
you help your nurses and clinical staff

73
00:02:54,900 --> 00:02:55,639
to prevent

74
00:02:56,259 --> 00:02:57,000
a catastrophic

75
00:02:58,305 --> 00:03:01,264
event before it happens? And can you have

76
00:03:01,264 --> 00:03:04,385
early warning signs that can help you in

77
00:03:04,385 --> 00:03:05,044
the management,

78
00:03:05,585 --> 00:03:06,324
of those,

79
00:03:07,344 --> 00:03:08,564
potentially disastrous,

80
00:03:09,025 --> 00:03:10,724
health outcomes for your patient?

81
00:03:11,310 --> 00:03:12,530
So that's on the,

82
00:03:13,710 --> 00:03:14,210
structured

83
00:03:14,750 --> 00:03:15,250
data

84
00:03:15,710 --> 00:03:16,210
side.

85
00:03:16,830 --> 00:03:17,330
However,

86
00:03:17,950 --> 00:03:18,770
on imaging,

87
00:03:19,550 --> 00:03:20,530
and diagnostic,

88
00:03:21,310 --> 00:03:22,370
there is a

89
00:03:23,069 --> 00:03:23,969
lot to

90
00:03:24,745 --> 00:03:25,965
be wishing for,

91
00:03:26,665 --> 00:03:30,044
with AI. Can AI help us into,

92
00:03:31,305 --> 00:03:33,245
some of the images in radiology,

93
00:03:33,544 --> 00:03:36,044
MRI, x-ray, CT scans, etcetera?

94
00:03:36,905 --> 00:03:37,405
And

95
00:03:37,705 --> 00:03:38,205
can

96
00:03:38,585 --> 00:03:39,085
AI

97
00:03:40,419 --> 00:03:41,560
accelerate the efficiency

98
00:03:41,860 --> 00:03:44,760
of the provider reading those x rays?

99
00:03:45,139 --> 00:03:45,620
Can it,

100
00:03:46,099 --> 00:03:46,599
unravel

101
00:03:47,379 --> 00:03:48,199
some subtle

102
00:03:49,139 --> 00:03:51,860
differences in the images, etcetera, that might lead

103
00:03:51,860 --> 00:03:52,919
to a

104
00:03:53,925 --> 00:03:54,745
different diagnostic,

105
00:03:55,685 --> 00:03:56,185
outcome.

106
00:03:56,645 --> 00:03:58,824
But the same can apply for pathology,

107
00:03:59,444 --> 00:04:01,044
reading slides for,

108
00:04:01,604 --> 00:04:02,104
cardiologists

109
00:04:02,485 --> 00:04:03,784
reading their own images,

110
00:04:04,965 --> 00:04:06,280
even dermatologists lesion.

111
00:04:06,819 --> 00:04:06,894
And can can can you look at patterns

112
00:04:06,894 --> 00:04:08,659
that the naked eye cannot, truly see? And

113
00:04:08,659 --> 00:04:10,739
can you really help AI with those diagnostic

114
00:04:10,739 --> 00:04:11,959
tools? Absolutely.

115
00:04:18,814 --> 00:04:21,214
You mentioned the intersection of AI and data

116
00:04:21,214 --> 00:04:23,055
in the very beginning there, and so I

117
00:04:23,055 --> 00:04:25,634
kinda wanna pivot the conversation to data specifically.

118
00:04:26,334 --> 00:04:28,574
On a daily basis, health care leaders are

119
00:04:28,574 --> 00:04:31,074
managing greater volumes of data and more devices

120
00:04:31,214 --> 00:04:33,375
across a growing number of care settings and

121
00:04:33,375 --> 00:04:33,875
populations.

122
00:04:34,620 --> 00:04:37,819
In this complex environment, what clinical data integration

123
00:04:37,819 --> 00:04:40,479
tools or practices are you seeing drive improvements

124
00:04:40,779 --> 00:04:43,019
in patient outcomes and operations? And then can

125
00:04:43,019 --> 00:04:44,000
you share an example?

126
00:04:44,699 --> 00:04:46,079
Sure. That's probably

127
00:04:47,194 --> 00:04:47,855
the biggest,

128
00:04:48,555 --> 00:04:50,875
conundrum and issue that we have to solve

129
00:04:50,875 --> 00:04:52,975
too. All AI tools

130
00:04:53,514 --> 00:04:55,995
are based on what kind of data we

131
00:04:55,995 --> 00:04:56,495
have,

132
00:04:57,194 --> 00:05:01,134
and bad data will lead to bad outcome.

133
00:05:01,919 --> 00:05:02,899
Biased data

134
00:05:03,360 --> 00:05:04,899
will lead to biased

135
00:05:05,600 --> 00:05:07,060
decision support tool,

136
00:05:07,839 --> 00:05:09,300
that we have today.

137
00:05:09,919 --> 00:05:10,899
So the opportunity,

138
00:05:12,479 --> 00:05:14,399
as I stated before, to get a lot

139
00:05:14,399 --> 00:05:16,740
of, predictive analytics, etcetera,

140
00:05:17,725 --> 00:05:18,225
only

141
00:05:19,245 --> 00:05:20,144
if your

142
00:05:20,524 --> 00:05:21,024
basic

143
00:05:21,644 --> 00:05:23,665
data gathering is clean.

144
00:05:24,285 --> 00:05:25,824
And what we struggle today

145
00:05:26,685 --> 00:05:27,824
as a health system,

146
00:05:28,444 --> 00:05:29,584
not just integrating

147
00:05:30,125 --> 00:05:30,625
data

148
00:05:31,670 --> 00:05:33,990
that come to us from all kind of

149
00:05:33,990 --> 00:05:36,490
different sources from outside of our organization.

150
00:05:37,990 --> 00:05:39,129
Because on an average,

151
00:05:39,509 --> 00:05:42,069
a patient that sees us at MetroHealth will

152
00:05:42,069 --> 00:05:44,410
only get 50% of their care at MetroHealth,

153
00:05:44,629 --> 00:05:47,455
but that the other 50% is somewhere else.

154
00:05:48,235 --> 00:05:50,415
So how can you account for that data?

155
00:05:50,875 --> 00:05:53,435
Also, the payers so we're the health system.

156
00:05:53,435 --> 00:05:55,115
There's the payer, and the payer have the

157
00:05:55,274 --> 00:05:56,735
those data. And can you

158
00:05:57,370 --> 00:05:57,870
bring,

159
00:05:59,050 --> 00:06:01,050
that data and integrate it with your,

160
00:06:01,689 --> 00:06:04,350
electronic medical records so you have that nice

161
00:06:04,649 --> 00:06:07,550
base of data that that you can

162
00:06:07,930 --> 00:06:11,470
use to build your your AI tools on?

163
00:06:12,064 --> 00:06:13,824
But that's only one part of the equation.

164
00:06:13,824 --> 00:06:15,204
The other part of the equation

165
00:06:15,985 --> 00:06:18,084
is within your own health system.

166
00:06:18,544 --> 00:06:20,485
There are all those peripheral

167
00:06:20,865 --> 00:06:21,524
and devices

168
00:06:22,544 --> 00:06:23,764
just like EKGs

169
00:06:25,310 --> 00:06:27,410
or or monitors on the inpatients,

170
00:06:27,870 --> 00:06:29,889
that are really collecting data?

171
00:06:30,349 --> 00:06:32,269
How do you interface all of those and

172
00:06:32,269 --> 00:06:35,310
bring them into a a a nice,

173
00:06:35,629 --> 00:06:37,250
data warehouse? And finally,

174
00:06:37,709 --> 00:06:38,529
are the wearable,

175
00:06:39,069 --> 00:06:39,889
with the consumers?

176
00:06:40,644 --> 00:06:42,584
And this is one of the biggest challenge.

177
00:06:42,644 --> 00:06:43,464
Every consumer

178
00:06:44,404 --> 00:06:45,144
is collecting

179
00:06:45,764 --> 00:06:46,264
data

180
00:06:46,805 --> 00:06:48,404
and wants to share that data with their

181
00:06:48,404 --> 00:06:50,324
provider, but that's gonna bring a lot of

182
00:06:50,324 --> 00:06:51,464
noise about,

183
00:06:52,485 --> 00:06:54,819
useful data that you can act on it

184
00:06:54,819 --> 00:06:56,759
or not, but that's probably where

185
00:06:57,459 --> 00:07:00,660
we're innovating the most, trying to curate and

186
00:07:00,660 --> 00:07:02,360
try to figure out what is

187
00:07:03,060 --> 00:07:04,120
the most important

188
00:07:04,580 --> 00:07:05,300
data element,

189
00:07:05,699 --> 00:07:08,915
that you're receiving from your patient wearable that

190
00:07:08,915 --> 00:07:09,735
a provider

191
00:07:10,035 --> 00:07:12,915
a a team of provider and their nurses

192
00:07:12,915 --> 00:07:15,074
and assistants can act upon and make a

193
00:07:15,074 --> 00:07:15,574
difference.

194
00:07:16,834 --> 00:07:18,935
In your view, how can health care organizations

195
00:07:19,314 --> 00:07:21,555
better support IT and clinical teams as they

196
00:07:21,555 --> 00:07:23,580
carry out innovation efforts, and what are the

197
00:07:23,580 --> 00:07:25,439
common pitfalls that you see here?

198
00:07:26,300 --> 00:07:27,520
Another good question.

199
00:07:28,379 --> 00:07:29,920
It's all about governance.

200
00:07:30,939 --> 00:07:32,240
I truly believe

201
00:07:32,540 --> 00:07:33,839
you have to have a strong,

202
00:07:34,459 --> 00:07:34,959
governance

203
00:07:35,259 --> 00:07:36,959
internally to your health system

204
00:07:37,935 --> 00:07:39,395
in order to streamline

205
00:07:41,455 --> 00:07:42,335
what kind of,

206
00:07:42,975 --> 00:07:44,835
application you're bringing in.

207
00:07:45,455 --> 00:07:46,754
A lot of those application

208
00:07:47,214 --> 00:07:47,714
promises

209
00:07:48,254 --> 00:07:49,634
they promise a lot,

210
00:07:50,290 --> 00:07:53,110
but most of the time, they've not proven

211
00:07:53,330 --> 00:07:54,230
their outcome.

212
00:07:55,170 --> 00:07:56,709
And without a

213
00:07:57,009 --> 00:07:59,990
true governance in place, how you can streamline,

214
00:08:00,449 --> 00:08:03,830
how you can allow small pilot to happen

215
00:08:04,285 --> 00:08:07,485
before you commit your organization into 6, 7

216
00:08:07,485 --> 00:08:07,985
figure

217
00:08:08,524 --> 00:08:09,024
expenses

218
00:08:09,805 --> 00:08:10,865
without any

219
00:08:11,644 --> 00:08:14,285
data to prove that the health outcome is

220
00:08:14,285 --> 00:08:14,785
better

221
00:08:15,165 --> 00:08:17,725
or actually your cost has improved and your

222
00:08:17,725 --> 00:08:19,024
efficiency have improved.

223
00:08:19,404 --> 00:08:20,810
I've seen it so many

224
00:08:21,350 --> 00:08:23,990
time that there is a rush to bring

225
00:08:23,990 --> 00:08:25,050
in the excitement,

226
00:08:26,230 --> 00:08:26,970
but then

227
00:08:27,350 --> 00:08:29,129
you spend a lot of

228
00:08:30,069 --> 00:08:30,569
resources

229
00:08:31,350 --> 00:08:34,070
and the outcome is not better and sometimes

230
00:08:34,070 --> 00:08:36,554
it's worse. So you gotta do baby steps.

231
00:08:37,575 --> 00:08:40,215
So that's that's one of my biggest thing.

232
00:08:40,215 --> 00:08:43,014
The other one is you don't wanna over

233
00:08:43,014 --> 00:08:46,154
govern and not allow innovation to happen.

234
00:08:46,774 --> 00:08:47,434
You wanna

235
00:08:47,990 --> 00:08:49,850
be very mindful that

236
00:08:50,230 --> 00:08:52,329
if you have a good governance structure,

237
00:08:52,870 --> 00:08:53,370
if

238
00:08:54,149 --> 00:08:56,549
you allow them to innovate, but take those

239
00:08:56,549 --> 00:08:59,269
baby steps, you are more likely to figure

240
00:08:59,269 --> 00:09:01,514
out what is the best that you're gonna

241
00:09:01,735 --> 00:09:05,434
ultimately invest and and and deploy system wide.

242
00:09:06,774 --> 00:09:08,615
Alright. Thank you so much for sharing. And

243
00:09:08,615 --> 00:09:10,934
to wrap up our conversation today, what's your

244
00:09:10,934 --> 00:09:12,774
top piece of advice for health care leaders

245
00:09:12,774 --> 00:09:15,195
as they prepare for further advancements in technology

246
00:09:15,450 --> 00:09:17,070
and greater demands for care?

247
00:09:18,009 --> 00:09:19,149
Don't be afraid

248
00:09:19,690 --> 00:09:20,590
of the technology,

249
00:09:21,370 --> 00:09:21,870
but

250
00:09:22,570 --> 00:09:25,290
you gotta also be cognizant that within your

251
00:09:25,290 --> 00:09:26,110
health system,

252
00:09:27,290 --> 00:09:30,330
people are gonna use technology whether you allow

253
00:09:30,330 --> 00:09:31,924
it or not. So

254
00:09:33,105 --> 00:09:33,924
embrace it,

255
00:09:34,384 --> 00:09:37,125
but embrace it in a very calculated way.

256
00:09:37,264 --> 00:09:38,964
Have a very strong governance,

257
00:09:39,904 --> 00:09:41,825
not just of what you adopt and what

258
00:09:41,825 --> 00:09:44,544
you don't adopt and what you invest in

259
00:09:44,544 --> 00:09:47,589
and not invest in, but also be very

260
00:09:47,589 --> 00:09:49,450
cognizant about your data infrastructure

261
00:09:49,830 --> 00:09:51,529
and invest into a

262
00:09:51,830 --> 00:09:53,450
good data warehouse. So

263
00:09:53,990 --> 00:09:54,889
it is not,

264
00:09:55,750 --> 00:09:57,289
bad data in

265
00:09:57,669 --> 00:10:00,855
bad outcome, bad decision. You want good data,

266
00:10:00,855 --> 00:10:03,654
and you want data that is unbiased so

267
00:10:03,654 --> 00:10:06,394
you don't create a health disparity while you're,

268
00:10:06,934 --> 00:10:08,075
relying on AI.

269
00:10:08,695 --> 00:10:10,455
Thank you so much for sharing your insight

270
00:10:10,455 --> 00:10:12,134
with us. Once again, we are at the

271
00:10:12,134 --> 00:10:14,794
9th annual health IT digital health and RCM

272
00:10:14,855 --> 00:10:15,355
conference.

273
00:10:15,740 --> 00:10:17,179
Nabil, thank you so much for being here.

274
00:10:17,179 --> 00:10:18,799
Thank you, Hailey. Thank you.