1
00:00:05,340 --> 00:00:06,460
Hello everyone.

2
00:00:06,460 --> 00:00:09,540
Thank you for tuning in to the
Becker's Healthcare podcast series.

3
00:00:09,870 --> 00:00:12,860
My name is Mariah Mohammed,
writer at Becker's Healthcare,

4
00:00:12,880 --> 00:00:17,700
and today we'll be discussing how alerts
in the E H R system may help bridge the

5
00:00:17,940 --> 00:00:22,260
17 year gap between publication of
evidence-based guidelines and consistent

6
00:00:22,260 --> 00:00:24,740
implementation and practice
by healthcare providers.

7
00:00:25,350 --> 00:00:29,820
We are thrilled to be joined by
Dr. Ralph Rlo and Dr. Nihar Desai.

8
00:00:29,970 --> 00:00:33,060
Before we get started,
Dr. Rlo and Dr. Desai,

9
00:00:33,160 --> 00:00:37,140
can you please share a bit more about
yourself, starting with you, Dr. Rlo?

10
00:00:37,990 --> 00:00:41,870
Certainly. Yeah. So my name is Ralph
Frio. I'm a clinical pharmacy specialist,

11
00:00:42,040 --> 00:00:46,350
uh, that's been working, uh, with Yale
Health System for over a decade. Um, and,

12
00:00:46,350 --> 00:00:47,910
and tr transitioned a few, uh,

13
00:00:47,910 --> 00:00:50,790
years ago over to the School of
Medicine to partner with, uh,

14
00:00:50,790 --> 00:00:53,070
some wonderful cardiologists like niha on,

15
00:00:53,070 --> 00:00:54,910
on some exciting trials like Prompt Lipid.

16
00:00:55,450 --> 00:00:58,400
Perfect. Thank you very
much. And then Dr. Desai?

17
00:00:58,870 --> 00:01:00,480
Sure. Uh, Nihar Desai.

18
00:01:00,480 --> 00:01:04,560
I'm a cardiologist here at the Yale
School of Medicine and Vice Chief of

19
00:01:04,560 --> 00:01:06,120
Cardiology here. And, um,

20
00:01:06,120 --> 00:01:09,520
also very interested in implementation
science and doing pragmatic clinical

21
00:01:09,520 --> 00:01:12,480
trials, which we're really
excited to talk about, um, today.

22
00:01:13,030 --> 00:01:14,200
Yeah, absolutely.

23
00:01:14,200 --> 00:01:18,880
Thank you again so much for joining us
today to the both of you and Dr. Desai.

24
00:01:18,880 --> 00:01:23,280
If I could ask you the first question
for those tuning in who may have not seen

25
00:01:23,280 --> 00:01:27,200
the featured presentation at the
American Heart Association Conference,

26
00:01:27,380 --> 00:01:32,200
can you please explain the objective
of the Prompt Lipid trial and summarize

27
00:01:32,200 --> 00:01:33,400
the key findings for us?

28
00:01:33,980 --> 00:01:35,870
Sure. Merh delighted too. So,

29
00:01:35,870 --> 00:01:40,550
prompt Lipid was really designed
to evaluate whether automated

30
00:01:40,550 --> 00:01:45,510
electronic alerts with guideline based
recommendations could help to improve

31
00:01:45,510 --> 00:01:50,110
the management of hyperlipidemia among
patients with established cardiovascular

32
00:01:50,110 --> 00:01:53,270
disease at very high risk
for cardiovascular events.

33
00:01:53,850 --> 00:01:58,510
And what we were very excited to see
in this trial of about 100 providers

34
00:01:58,690 --> 00:02:03,590
and over 2000 patients is
a significant increase in

35
00:02:03,590 --> 00:02:06,750
the use of evidence-based lipid
lowering therapies. In fact,

36
00:02:06,750 --> 00:02:11,390
we saw that those providers
who received this decision

37
00:02:11,390 --> 00:02:16,150
alert were about 40% more
likely to intensify lipid

38
00:02:16,150 --> 00:02:16,983
lowering therapy.

39
00:02:17,210 --> 00:02:21,230
And then amongst the subset of providers
who sort of more favorably interacted

40
00:02:21,230 --> 00:02:22,063
with the alert,

41
00:02:22,120 --> 00:02:26,990
we saw a more than twofold increase in
the use of evidence-based lipid lowering

42
00:02:26,990 --> 00:02:27,823
therapies.

43
00:02:28,400 --> 00:02:33,280
Wonderful. Yeah. Thank you so much for
that background information. Now, Dr.

44
00:02:33,420 --> 00:02:33,900
Rlo,

45
00:02:33,900 --> 00:02:38,600
the prompt lipid research team
felt strongly that patients with

46
00:02:38,600 --> 00:02:43,600
very high risk atherosclerotic
cardiovascular disease would make for a

47
00:02:43,600 --> 00:02:47,680
compelling population to
test an innovative approach
to quality improvement.

48
00:02:47,860 --> 00:02:49,240
Can you explain why that is?

49
00:02:50,180 --> 00:02:54,400
Yes, it certainly can, uh, Mariah,
and that patients with atheros, uh,

50
00:02:54,400 --> 00:02:58,840
sclerosis, uh, cardiovascular
D uh, disease or A S C B D, um,

51
00:02:58,840 --> 00:03:02,880
we know that their, their l LDL well,
um, L D L cholesterol L LDL C is,

52
00:03:02,880 --> 00:03:06,320
has been a critical modifiable risk
factor, um, for their disease that,

53
00:03:06,320 --> 00:03:09,640
that we as providers can control.
And in lipid lowering therapy,

54
00:03:09,640 --> 00:03:13,240
particularly intensive lipid lowering
therapy for this high risk patient

55
00:03:13,240 --> 00:03:15,200
population, you know,
it's been recommended by,

56
00:03:15,200 --> 00:03:18,760
by clinical practice guidelines
for years. Um, however,

57
00:03:19,140 --> 00:03:23,520
the evidence practice gap between what
the guidelines say we should be doing and

58
00:03:23,520 --> 00:03:27,520
the therapies our patients actually
receive in, in real world practice, um,

59
00:03:27,520 --> 00:03:28,720
is vast. Um,

60
00:03:28,720 --> 00:03:32,520
only about half of patients with A S C
V D are actually prescribed any statin

61
00:03:32,550 --> 00:03:35,080
whatsoever in the United States. Um,

62
00:03:35,080 --> 00:03:38,640
and fewer than a quarter of these
patients are, are even prescribed, uh,

63
00:03:38,640 --> 00:03:42,440
the appropriately recommended
high-intensity statin. So
as it currently stands,

64
00:03:42,440 --> 00:03:47,200
sort of passive diffusion
of these guidelines, um, to,
to our providers and, uh,

65
00:03:47,200 --> 00:03:51,120
to make sure that these therapies reach
patients, um, just really isn't enough.

66
00:03:51,120 --> 00:03:54,000
And, and clinical inertia
really, um, you know,

67
00:03:54,550 --> 00:03:58,760
prevents our providers from, from passing
on these therapies to patients. Um,

68
00:03:59,040 --> 00:04:02,520
so, so we were hoping that the C H R
alert might do something to change that.

69
00:04:03,520 --> 00:04:06,940
Yes. Thank you for those
insights. Dr. Rlo, Dr.

70
00:04:06,940 --> 00:04:11,780
Desai of the many E H R alerts that
fire every day for clinicians using

71
00:04:11,780 --> 00:04:13,860
the electronic health record system.

72
00:04:14,050 --> 00:04:18,180
Very few are ever subject to thorough
evaluation after they're deployed.

73
00:04:18,330 --> 00:04:21,660
What prompted your research team
to go through the extra effort?

74
00:04:22,210 --> 00:04:25,820
Yeah, it's a great question, and
I think as you heard from Ralph,

75
00:04:26,370 --> 00:04:30,220
that there was a real sense of urgency
that our team felt that, you know,

76
00:04:30,220 --> 00:04:34,100
patients with high risk atherosclerotic
cardiovascular disease really needed

77
00:04:34,520 --> 00:04:36,620
intensification of their
lipid lowering therapy.

78
00:04:36,990 --> 00:04:41,060
We thought that an alert
like this that was real time,

79
00:04:41,060 --> 00:04:44,860
that was targeted and tailored,
um, would be helpful to providers,

80
00:04:45,040 --> 00:04:49,140
but we were committed to actually testing
it in a very rigorous way to actually

81
00:04:49,140 --> 00:04:53,820
define and determine whether an alert
like this would actually do the things

82
00:04:53,820 --> 00:04:57,140
that we wanted it to do. And again,
we were very pleased with, um,

83
00:04:57,140 --> 00:04:59,940
the results of prompt lipid
that really again showed, um,

84
00:04:59,940 --> 00:05:04,100
the benefits of an alert like this to
improve the quality of care for our

85
00:05:04,100 --> 00:05:05,660
patients with very high risk A C B D.

86
00:05:06,240 --> 00:05:09,370
Yeah, and to, to dovetail off
Neer's point there as well,

87
00:05:09,560 --> 00:05:11,210
I think all too often when, uh,

88
00:05:11,210 --> 00:05:14,090
when health systems look to
implement alerts like these in the,

89
00:05:14,090 --> 00:05:18,090
in electronic health record, um, very
rarely are they ever actually, you know,

90
00:05:18,090 --> 00:05:20,130
formally studied in a,
in a prospective fashion,

91
00:05:20,130 --> 00:05:23,330
let alone in a randomized controlled
trial. And, and, you know,

92
00:05:23,330 --> 00:05:26,890
I think the consequence of that is, is,
you know, contributes to alert fatigue.

93
00:05:26,890 --> 00:05:30,970
Many of these alerts, you
know, as important as they
may be, tend to get ano uh,

94
00:05:30,970 --> 00:05:32,770
ignored. You know, for
instance, we're, we're,

95
00:05:32,770 --> 00:05:34,890
we're in deep flu season right now. Uh,

96
00:05:34,890 --> 00:05:37,370
I can't even tell you how many charts
I've opened up just to get pinged with

97
00:05:37,370 --> 00:05:39,770
these alerts that, that I know most
of our clinicians click through. So,

98
00:05:39,780 --> 00:05:41,370
so it's important to understand in a,

99
00:05:41,570 --> 00:05:44,850
in a well controlled setting how these
alerts actually perform and, and,

100
00:05:44,850 --> 00:05:46,290
you know, do what they're intended to do.

101
00:05:47,340 --> 00:05:50,230
Absolutely. Yeah, I completely agree.
And it seems that that actually,

102
00:05:50,230 --> 00:05:54,790
that really did pay off, thankfully. Um,
if I could come back to you actually,

103
00:05:54,790 --> 00:05:57,520
Dr. Rlo, the providers, uh,

104
00:05:57,520 --> 00:06:02,320
surveyed in both the Prompt HF and
prompt lipid trials seem to view the

105
00:06:02,320 --> 00:06:06,720
alert quite favorably and often accepted
the recommendations as suggested.

106
00:06:07,030 --> 00:06:11,080
What about your team's approach
enabled that user finding what you say.

107
00:06:12,300 --> 00:06:14,900
<Laugh>? Yeah, Mariah, that was
a really good point. Um, um,

108
00:06:14,900 --> 00:06:17,740
that might've been one of
the most surprising things
I think that we came away

109
00:06:17,740 --> 00:06:19,740
with. Um, uh, some of
the results from our,

110
00:06:19,740 --> 00:06:22,260
our various prompt studies
for 'em and promptly it's, uh,

111
00:06:22,260 --> 00:06:26,180
certainly notwithstanding, um, I I think
it has to do with our secret sauce and,

112
00:06:26,180 --> 00:06:28,300
and that's, that's engaging our, you know,

113
00:06:28,300 --> 00:06:32,220
highest volume cardiology stakeholders
upfront and early in that process.

114
00:06:32,320 --> 00:06:35,340
You know, give them a seat at the
table, listen to them, uh, you know,

115
00:06:35,340 --> 00:06:39,380
engage them, you know, recruit them for
feedback. And, and we actually had a,

116
00:06:39,380 --> 00:06:43,700
a couple separate focus groups with,
with, you know, the busiest, um, you know,

117
00:06:43,700 --> 00:06:46,060
lipid lowering therapy prescribers
to make sure, you know,

118
00:06:46,060 --> 00:06:48,940
these are the ones in the front lines
of healthcare doing the work that,

119
00:06:48,940 --> 00:06:52,380
that we think is important and, and
that we wanna see more of. So, you know,

120
00:06:52,380 --> 00:06:56,180
by asking them for their input on not
just the build of the alert, but you know,

121
00:06:56,180 --> 00:06:59,260
when it should fire what the
order set should look like,

122
00:06:59,410 --> 00:07:03,300
what labs should display, um, we
felt like, you know, their input was,

123
00:07:03,300 --> 00:07:06,220
was incredibly valuable to make
sure we designed something that was

124
00:07:06,220 --> 00:07:10,100
user-friendly for the people who would
actually use it and buy the people that

125
00:07:10,100 --> 00:07:12,820
used it. Because cuz sure, you
know, Nihar and I have, you know,

126
00:07:12,820 --> 00:07:15,140
spent our careers caring for
cardiovascular patients, but,

127
00:07:15,160 --> 00:07:18,420
but we want to ask what others think
about that too, to make sure that,

128
00:07:18,420 --> 00:07:19,260
you know, it, it,

129
00:07:19,260 --> 00:07:23,300
it helps facilitate their clinical
workflow rather than hinder it in any way.

130
00:07:24,420 --> 00:07:25,230
Yeah. Yeah,

131
00:07:25,230 --> 00:07:28,870
making sure everyone feels heard is
definitely essential for these types of

132
00:07:28,870 --> 00:07:31,670
trials. And Dr. Desai, if I could ask you,

133
00:07:31,990 --> 00:07:36,830
alert fatigue is a pressing concern for
many frontline clinicians and may have

134
00:07:36,830 --> 00:07:41,670
consequences for patient safety as
well as contribute to provider pa uh,

135
00:07:41,670 --> 00:07:46,390
burnout. How did the prompt lipid
trial help address this barrier?

136
00:07:47,250 --> 00:07:49,420
Yeah, it's a great question
and a great point and I'll,

137
00:07:49,420 --> 00:07:51,140
I'll build on much of
what Ralph, you know,

138
00:07:51,140 --> 00:07:54,020
just started to discuss about
the design of the alert.

139
00:07:54,020 --> 00:07:57,660
And so alert fatigue is a real
concern. I think all of us, you know,

140
00:07:57,660 --> 00:08:02,180
practicing on the front lines are,
are loathe to get additional alerts,

141
00:08:02,180 --> 00:08:06,020
you know, especially those that interrupt
or interfere, um, you know, with,

142
00:08:06,020 --> 00:08:09,740
with direct clinical workflow. And I
think one of the things that we did,

143
00:08:09,740 --> 00:08:14,260
you heard about from Ralph was we engaged
frontline providers and said, Hey,

144
00:08:14,260 --> 00:08:17,220
what, what do you wanna see? What
should the alert look like? Um,

145
00:08:17,220 --> 00:08:20,500
what time in your clinical
encounter should it fire?

146
00:08:20,810 --> 00:08:24,820
I think the other thing that we did was
very helpful was we brought and pulled

147
00:08:24,820 --> 00:08:29,700
information from the medical record
that might otherwise take 2, 3, 4

148
00:08:29,700 --> 00:08:33,300
other clicks and actually
brought that into the alert.

149
00:08:33,480 --> 00:08:37,780
And I think the final piece of it is
that we made the alert pretty easy to

150
00:08:37,900 --> 00:08:39,740
interact with that if you
wanted to do the right thing,

151
00:08:39,940 --> 00:08:41,460
if you wanted to open the order set,

152
00:08:41,460 --> 00:08:46,060
that the orders were actually embedded
within the alert itself. Um, and so that,

153
00:08:46,060 --> 00:08:50,300
that step from interacting with the
alert to then doing the right thing was

154
00:08:50,300 --> 00:08:53,660
about as simple and straightforward
as could be. And so I think, you know,

155
00:08:53,660 --> 00:08:57,880
many of the in secret ingredients
as, as Ralph calls them, um, <laugh>,

156
00:08:57,940 --> 00:09:00,640
you know, kind of, you know,
get, getting the engagement of,

157
00:09:00,640 --> 00:09:04,160
of clinicians upfront, um, really
thinking deep about, you know,

158
00:09:04,160 --> 00:09:05,760
what the alert itself looks like,

159
00:09:05,760 --> 00:09:08,560
pulling information into
it and then embedding,

160
00:09:08,560 --> 00:09:12,120
ordering capability within the alert
to really make it a very seamless

161
00:09:12,120 --> 00:09:14,440
experience. So that at
least I hope, you know,

162
00:09:14,440 --> 00:09:18,480
gives you a sense of what we were trying
to do to try and overcome that kind of

163
00:09:18,480 --> 00:09:21,960
intrinsic fear that everyone
has about alert fatigue.

164
00:09:22,900 --> 00:09:25,230
Yeah. And, and I think
another point to add on there,

165
00:09:25,230 --> 00:09:28,350
there ni at least to emphasize again
cause I I think it's so important,

166
00:09:28,350 --> 00:09:32,590
is that the, the alert only fires
when the patient has, you know,

167
00:09:32,590 --> 00:09:37,190
low hanging fruit, you know, guideline
recommended but not yet implemented,

168
00:09:37,600 --> 00:09:40,390
um, change with their
therapy. So you know, it,

169
00:09:40,390 --> 00:09:43,630
the alert's not gonna suggest them to do
something that would be duplicative or,

170
00:09:43,630 --> 00:09:45,190
you know, the patient's
already on a statin.

171
00:09:45,190 --> 00:09:47,830
Why is this telling me to
prescribe another one? Um, it,

172
00:09:47,830 --> 00:09:52,270
it's really just a meaningful easy
change at the point of care only when

173
00:09:52,420 --> 00:09:54,110
it's indicated for the patient.

174
00:09:55,490 --> 00:10:00,100
Yeah. Yeah. Thank you very much to, to
both of you for that information, Dr.

175
00:10:00,100 --> 00:10:02,060
Relo, before I let you both go,

176
00:10:02,060 --> 00:10:06,100
I must ask now that the prom
lipid trial is completed,

177
00:10:06,370 --> 00:10:10,700
what are the next steps for your research
team to further explore opportunities

178
00:10:10,870 --> 00:10:12,180
on these findings?

179
00:10:12,770 --> 00:10:14,220
Yeah, great question, right?

180
00:10:14,220 --> 00:10:17,660
We always need to look beyond
sort of our own house, um, to, to,

181
00:10:17,660 --> 00:10:20,180
to make sure that we can then
extrapolate these findings to,

182
00:10:20,180 --> 00:10:24,100
to other systems as well. And, and I
think the, the, the prompt research team,

183
00:10:24,350 --> 00:10:28,780
et cetera, definitely has some big plans
for that. Um, for, for one, this, uh,

184
00:10:28,780 --> 00:10:32,700
the prompt lipid trial was conducted
exclusively at Yale Health System and the

185
00:10:32,700 --> 00:10:36,260
outpatient, uh, clinic's, uh,
both cardiology and medicine. Um,

186
00:10:36,260 --> 00:10:38,980
but we would certainly like to
see, you know, how this, you know,

187
00:10:38,980 --> 00:10:43,140
study performs beyond our own, you know,
neighborhood beyond our community, um,

188
00:10:43,140 --> 00:10:47,380
and expand that to, to other
interested sites. Um, additionally the,

189
00:10:47,380 --> 00:10:51,900
the study only took place, um, on, on
Epic as our E H R platform across Yale.

190
00:10:52,270 --> 00:10:55,500
Um, but certainly wouldn't be nice if,
if we could expand that to other eh,

191
00:10:55,500 --> 00:10:56,500
h r platforms as well.

192
00:10:56,500 --> 00:10:59,820
It's not sort of pigeonhole
ourselves just to one particular eh,

193
00:10:59,820 --> 00:11:04,220
r platform and make it more interoperable
for other sites that that may be using

194
00:11:04,220 --> 00:11:07,700
different ones. Um, and, you know,
expand it to, to other clinical settings.

195
00:11:07,700 --> 00:11:09,900
You know, maybe focus on
primary care or endocrine,

196
00:11:09,900 --> 00:11:11,980
but we know a lot of these
patients, um, you know,

197
00:11:11,980 --> 00:11:13,700
may also frequently see providers as well.

198
00:11:14,440 --> 00:11:16,370
Yeah. And I, I I, I
might just add, you know,

199
00:11:16,370 --> 00:11:19,570
I think there's a lot of
important work ahead and,

200
00:11:19,570 --> 00:11:23,170
and I think we're very excited to
build on the results of prompt, um,

201
00:11:23,170 --> 00:11:27,210
and prompt lipid to then, you know, think
about what else can be done. I mean,

202
00:11:27,210 --> 00:11:30,770
I think Mariah, you very nicely kind of
said right at the front of, you know,

203
00:11:30,770 --> 00:11:32,930
how do we bridge the gap
between evidence and practice?

204
00:11:32,930 --> 00:11:37,490
How do we shorten that time from when
things go into the guidelines and they're

205
00:11:37,650 --> 00:11:41,610
actually received by patients? And so I
think there's a lot of work to do, um,

206
00:11:41,610 --> 00:11:43,290
and we're really excited
to build on, you know,

207
00:11:43,290 --> 00:11:44,810
the exciting results for Prompt Lipid.

208
00:11:45,510 --> 00:11:50,280
Absolutely. And I'm also excited
to hear back on what you finds. Um,

209
00:11:50,280 --> 00:11:52,840
and I wanna thank both
of you, Dr. Rlo and Dr.

210
00:11:52,920 --> 00:11:56,160
Desai for your time and thought
provoking conversation today.

211
00:11:56,450 --> 00:11:59,320
We also wanna thank our
podcast sponsor Amgen.

212
00:11:59,380 --> 00:12:03,560
You can tune into more podcasts
from Becker's Healthcare
by visiting our podcast

213
00:12:03,670 --> 00:12:07,400
page at becker's hospital
review.com/podcast.

214
00:12:07,680 --> 00:12:08,840
Thank you both again.