6a771959-1ff8-4201-8681-f53067b434e9/5-0
00:00:03.470 --> 00:00:05.811
Khan,
the Director of Digital Experience at

6a771959-1ff8-4201-8681-f53067b434e9/5-1
00:00:05.811 --> 00:00:09.270
Mount Sinai Health System. Neda,
thank you for joining us today.

6a771959-1ff8-4201-8681-f53067b434e9/6-0
00:00:09.750 --> 00:00:11.910
Thank you for having me.
It's great to be here.

6a771959-1ff8-4201-8681-f53067b434e9/8-0
00:00:12.630 --> 00:00:15.735
So for our listeners,
Mount Sinai Health System is

6a771959-1ff8-4201-8681-f53067b434e9/8-1
00:00:15.735 --> 00:00:18.474
headquartered in New York, New York,
with 3,

6a771959-1ff8-4201-8681-f53067b434e9/8-2
00:00:18.474 --> 00:00:22.736
200 beds across 7 hospitals supported by
9,000 physicians. Now, Neda,

6a771959-1ff8-4201-8681-f53067b434e9/8-3
00:00:22.736 --> 00:00:26.997
as a director of digital experience,
I think you have a lot of, well,

6a771959-1ff8-4201-8681-f53067b434e9/8-4
00:00:26.997 --> 00:00:31.502
interaction with or a mandate over
patient access and patient engagement.

6a771959-1ff8-4201-8681-f53067b434e9/8-5
00:00:31.502 --> 00:00:34.790
So I think I'd like to talk about those
topics today.

6a771959-1ff8-4201-8681-f53067b434e9/9-0
00:00:35.270 --> 00:00:39.294
Let's start with patient access.
I know that you've been running through a

6a771959-1ff8-4201-8681-f53067b434e9/9-1
00:00:39.294 --> 00:00:43.801
variety of change management initiatives.
Would you like to provide some context on

6a771959-1ff8-4201-8681-f53067b434e9/9-2
00:00:43.801 --> 00:00:47.826
what you've been working on regarding
patient access and change management

6a771959-1ff8-4201-8681-f53067b434e9/9-3
00:00:47.826 --> 00:00:48.470
initiatives?

6a771959-1ff8-4201-8681-f53067b434e9/10-0
00:00:49.670 --> 00:00:54.828
Yeah, absolutely. So, you know,
when we think about change management and

6a771959-1ff8-4201-8681-f53067b434e9/10-1
00:00:54.828 --> 00:00:58.662
patient access,
we want to think about like how we can

6a771959-1ff8-4201-8681-f53067b434e9/10-2
00:00:58.662 --> 00:01:03.959
reduce patient friction and how we can
reduce the complication to patients.

6a771959-1ff8-4201-8681-f53067b434e9/10-3
00:01:03.959 --> 00:01:08.560
And sometimes when we think about
technology and implementing new

6a771959-1ff8-4201-8681-f53067b434e9/10-4
00:01:08.560 --> 00:01:12.881
technology, you know,
we're always thinking about all the new

6a771959-1ff8-4201-8681-f53067b434e9/10-5
00:01:12.881 --> 00:01:13.230
bells

6a771959-1ff8-4201-8681-f53067b434e9/12-0
00:01:13.310 --> 00:01:16.863
and whistles,
how we can add to the workflows,

6a771959-1ff8-4201-8681-f53067b434e9/12-1
00:01:16.863 --> 00:01:20.946
add technology.
And a lot of times we're not thinking

6a771959-1ff8-4201-8681-f53067b434e9/12-2
00:01:20.946 --> 00:01:24.651
about how we can reduce, you know,
the workflow,

6a771959-1ff8-4201-8681-f53067b434e9/12-3
00:01:24.651 --> 00:01:30.851
reduce the amount of clicks for patients.
And so my role in thinking about access

6a771959-1ff8-4201-8681-f53067b434e9/12-4
00:01:30.851 --> 00:01:31.910
to patients is

6a771959-1ff8-4201-8681-f53067b434e9/13-0
00:01:32.230 --> 00:01:36.605
First, you know, what do patients,
you know, have access to?

6a771959-1ff8-4201-8681-f53067b434e9/13-1
00:01:36.605 --> 00:01:40.837
Not every patient is going to have access
to a smartphone.

6a771959-1ff8-4201-8681-f53067b434e9/13-2
00:01:40.837 --> 00:01:46.575
Not every patient has access to reliable
signal. And so when we have, you know,

6a771959-1ff8-4201-8681-f53067b434e9/13-3
00:01:46.575 --> 00:01:50.950
patients, you know,
in the worst case scenario, what kind of

6a771959-1ff8-4201-8681-f53067b434e9/14-0
00:01:51.270 --> 00:01:55.704
you know, programming,
what kind of technology can we provide to

6a771959-1ff8-4201-8681-f53067b434e9/14-1
00:01:55.704 --> 00:01:59.048
them that's going to be accessible to
them 24-7.

6a771959-1ff8-4201-8681-f53067b434e9/14-2
00:01:59.048 --> 00:02:02.732
And so oftentimes we think about like
text messaging.

6a771959-1ff8-4201-8681-f53067b434e9/14-3
00:02:02.732 --> 00:02:05.939
That's something that everybody has
access to.

6a771959-1ff8-4201-8681-f53067b434e9/14-4
00:02:05.939 --> 00:02:09.759
Patients are always going to be able to
access texting.

6a771959-1ff8-4201-8681-f53067b434e9/14-5
00:02:09.759 --> 00:02:11.670
They're doing it constantly.

6a771959-1ff8-4201-8681-f53067b434e9/16-0
00:02:12.030 --> 00:02:15.799
And so how can we use texting as a way to
change behavior,

6a771959-1ff8-4201-8681-f53067b434e9/16-1
00:02:15.799 --> 00:02:21.038
to get information to them about their
care, about, you know, their appointments?

6a771959-1ff8-4201-8681-f53067b434e9/16-2
00:02:21.038 --> 00:02:24.998
That's just a really easy way to get
information to patients.

6a771959-1ff8-4201-8681-f53067b434e9/16-3
00:02:24.998 --> 00:02:30.365
When we're changing, you know, their care,
texting is usually the easiest way to do

6a771959-1ff8-4201-8681-f53067b434e9/15-0
00:02:28.790 --> 00:02:29.270
Mhm.

6a771959-1ff8-4201-8681-f53067b434e9/16-4
00:02:30.365 --> 00:02:33.304
it.
Instead of trying to add an app that they

6a771959-1ff8-4201-8681-f53067b434e9/16-5
00:02:33.304 --> 00:02:34.390
have to download,

6a771959-1ff8-4201-8681-f53067b434e9/17-0
00:02:34.670 --> 00:02:38.690
That always adds friction.
You have to provide instructions to that

6a771959-1ff8-4201-8681-f53067b434e9/17-1
00:02:38.690 --> 00:02:42.710
workflow, things like that.
Adding friction always just reduces the

6a771959-1ff8-4201-8681-f53067b434e9/17-2
00:02:42.710 --> 00:02:45.430
amount of adoption that you're going to
have.

6a771959-1ff8-4201-8681-f53067b434e9/18-0
00:02:45.990 --> 00:02:49.885
And so what are some of the business and
clinical drivers that are that that

6a771959-1ff8-4201-8681-f53067b434e9/18-1
00:02:49.885 --> 00:02:52.870
you're trying seeking to address with
this text messaging?

6a771959-1ff8-4201-8681-f53067b434e9/19-0
00:02:53.830 --> 00:02:57.308
Yeah,
so some of the business drivers there is

6a771959-1ff8-4201-8681-f53067b434e9/19-1
00:02:57.308 --> 00:03:02.859
one cost. That's the number one.
When we're thinking about new technology,

6a771959-1ff8-4201-8681-f53067b434e9/19-2
00:03:02.859 --> 00:03:06.412
text messaging costs almost nothing.
It's like.

6a771959-1ff8-4201-8681-f53067b434e9/19-3
00:03:06.412 --> 00:03:11.889
005 cents to deliver a text message to
patients. And then on top of that,

6a771959-1ff8-4201-8681-f53067b434e9/19-4
00:03:11.889 --> 00:03:15.590
when we think about, you know,
the clinical side,

6a771959-1ff8-4201-8681-f53067b434e9/20-0
00:03:15.830 --> 00:03:21.368
You can actually do a lot with text
messaging when you're thinking about

6a771959-1ff8-4201-8681-f53067b434e9/20-1
00:03:21.368 --> 00:03:27.666
remote monitoring. You can do, you know,
collect patient reported outcomes through

6a771959-1ff8-4201-8681-f53067b434e9/20-2
00:03:27.666 --> 00:03:32.901
text messaging, conduct, you know,
symptom reports, symptom surveys.

6a771959-1ff8-4201-8681-f53067b434e9/20-3
00:03:32.901 --> 00:03:37.681
Tons of information can be collected and
sent through texting.

6a771959-1ff8-4201-8681-f53067b434e9/20-4
00:03:37.681 --> 00:03:39.350
And so we can actually

6a771959-1ff8-4201-8681-f53067b434e9/21-0
00:03:40.230 --> 00:03:43.804
you know,
move a lot of the needle through text

6a771959-1ff8-4201-8681-f53067b434e9/21-1
00:03:43.804 --> 00:03:49.091
messaging to help improve patient
behavior and patient health outcomes

6a771959-1ff8-4201-8681-f53067b434e9/21-2
00:03:49.091 --> 00:03:49.910
downstream.

6a771959-1ff8-4201-8681-f53067b434e9/22-0
00:03:50.630 --> 00:03:55.346
I know that you've been working on
prioritizing digital and AI solutions,

6a771959-1ff8-4201-8681-f53067b434e9/22-1
00:03:55.346 --> 00:04:00.381
as has everyone across the country,
to improve patient access and utilization.

6a771959-1ff8-4201-8681-f53067b434e9/22-2
00:04:00.381 --> 00:04:04.014
How are, I guess,
how is text messaging supporting this,

6a771959-1ff8-4201-8681-f53067b434e9/22-3
00:04:04.014 --> 00:04:08.156
or what other initiatives are you
pursuing to improve access and

6a771959-1ff8-4201-8681-f53067b434e9/22-4
00:04:08.156 --> 00:04:09.750
utilization? And I guess,

6a771959-1ff8-4201-8681-f53067b434e9/23-0
00:04:09.990 --> 00:04:12.714
I'd love if you could ground it.
For example,

6a771959-1ff8-4201-8681-f53067b434e9/23-1
00:04:12.714 --> 00:04:17.511
you said remote monitoring or patient
reported outcomes, kind of like, you know,

6a771959-1ff8-4201-8681-f53067b434e9/23-2
00:04:17.511 --> 00:04:21.479
what's very specifically,
what's one application where you've been

6a771959-1ff8-4201-8681-f53067b434e9/23-3
00:04:21.479 --> 00:04:25.270
seeking to leverage in order to improve
access and utilization?

6a771959-1ff8-4201-8681-f53067b434e9/24-0
00:04:26.070 --> 00:04:29.649
Yeah,
so one thing that we've been prioritizing

6a771959-1ff8-4201-8681-f53067b434e9/24-1
00:04:29.649 --> 00:04:32.706
is,
and it's actually this functionality

6a771959-1ff8-4201-8681-f53067b434e9/24-2
00:04:32.706 --> 00:04:37.329
through Epic because we are an Epic shop
here at Mount Sinai,

6a771959-1ff8-4201-8681-f53067b434e9/24-3
00:04:37.329 --> 00:04:42.847
which is called Cheers Campaigns.
And it's these text messaging campaigns

6a771959-1ff8-4201-8681-f53067b434e9/24-4
00:04:42.847 --> 00:04:45.830
that you can actually build based off of

6a771959-1ff8-4201-8681-f53067b434e9/25-0
00:04:46.190 --> 00:04:53.660
demographic information of the patient
based off of clinical information.

6a771959-1ff8-4201-8681-f53067b434e9/25-1
00:04:53.660 --> 00:05:01.635
And it basically builds these like text
messaging campaigns to promote either,

6a771959-1ff8-4201-8681-f53067b434e9/25-2
00:05:01.635 --> 00:05:05.270
you know, patient behaviors, promote

6a771959-1ff8-4201-8681-f53067b434e9/27-0
00:05:06.310 --> 00:05:11.996
you know, appointment reminders.
So we have a few different ones that are

6a771959-1ff8-4201-8681-f53067b434e9/27-1
00:05:11.996 --> 00:05:16.990
coming up. We have ones around no show,
no show recommendations.

6a771959-1ff8-4201-8681-f53067b434e9/27-2
00:05:16.990 --> 00:05:20.448
We have ones around breast cancer
screening,

6a771959-1ff8-4201-8681-f53067b434e9/27-3
00:05:20.448 --> 00:05:25.750
other different types of cancer
screenings. We also have ones around

6a771959-1ff8-4201-8681-f53067b434e9/26-0
00:05:22.790 --> 00:05:23.110
Yeah.

6a771959-1ff8-4201-8681-f53067b434e9/28-0
00:05:27.110 --> 00:05:27.270
Uh...

6a771959-1ff8-4201-8681-f53067b434e9/29-0
00:05:29.190 --> 00:05:32.776
So I think, Neda,
you work with the nudge unit to implement

6a771959-1ff8-4201-8681-f53067b434e9/29-1
00:05:32.776 --> 00:05:35.944
these kinds of nudges to improve patient
engagement.

6a771959-1ff8-4201-8681-f53067b434e9/29-2
00:05:35.944 --> 00:05:40.427
Does the nudge unit kind of work with
breast cancer screenings and no-show

6a771959-1ff8-4201-8681-f53067b434e9/29-3
00:05:40.427 --> 00:05:43.774
recommendations?
Can you explain a little bit about how

6a771959-1ff8-4201-8681-f53067b434e9/29-4
00:05:43.774 --> 00:05:45.030
the nudge unit works?

6a771959-1ff8-4201-8681-f53067b434e9/30-0
00:05:46.230 --> 00:05:49.659
Yeah,
so the nudge unit also worked with those

6a771959-1ff8-4201-8681-f53067b434e9/30-1
00:05:49.659 --> 00:05:55.496
types of screenings. And essentially,
the goal for that was to really implement

6a771959-1ff8-4201-8681-f53067b434e9/30-2
00:05:55.496 --> 00:06:01.114
small changes in the patient workflow to
really nudge them towards the right

6a771959-1ff8-4201-8681-f53067b434e9/30-3
00:06:01.114 --> 00:06:04.470
direction.
So whether that be sending a small

6a771959-1ff8-4201-8681-f53067b434e9/31-0
00:06:05.270 --> 00:06:09.773
you know,
message to remind them to do something or,

6a771959-1ff8-4201-8681-f53067b434e9/31-1
00:06:09.773 --> 00:06:14.022
you know,
sending them an educational reminder or

6a771959-1ff8-4201-8681-f53067b434e9/31-2
00:06:14.022 --> 00:06:20.820
changing a default in the workflow to
nudge them towards the right decision and

6a771959-1ff8-4201-8681-f53067b434e9/31-3
00:06:20.820 --> 00:06:23.030
making a different change.

6a771959-1ff8-4201-8681-f53067b434e9/32-0
00:06:23.750 --> 00:06:29.154
So how are you supporting these kind of
programs in the back end? Are these,

6a771959-1ff8-4201-8681-f53067b434e9/32-1
00:06:29.154 --> 00:06:34.347
what sort of infrastructure is supporting
this text message or nudge unit

6a771959-1ff8-4201-8681-f53067b434e9/32-2
00:06:34.347 --> 00:06:35.190
programming?

6a771959-1ff8-4201-8681-f53067b434e9/33-0
00:06:37.350 --> 00:06:42.366
There are a couple different
infrastructures that we have in place. So,

6a771959-1ff8-4201-8681-f53067b434e9/33-1
00:06:42.366 --> 00:06:46.616
you know, as I mentioned,
the EHR is the big one where we're

6a771959-1ff8-4201-8681-f53067b434e9/33-2
00:06:46.616 --> 00:06:51.353
pulling all this patient data,
but then we're connected to our text

6a771959-1ff8-4201-8681-f53067b434e9/33-3
00:06:51.353 --> 00:06:55.394
messaging vendors.
We also have an analytics team that we

6a771959-1ff8-4201-8681-f53067b434e9/33-4
00:06:55.394 --> 00:06:56.230
work with to

6a771959-1ff8-4201-8681-f53067b434e9/34-0
00:06:56.550 --> 00:07:02.475
help drive a lot of the, you know,
a lot of the decision-based like logic to,

6a771959-1ff8-4201-8681-f53067b434e9/34-1
00:07:02.475 --> 00:07:06.349
you know,
send these messages to the right patient

6a771959-1ff8-4201-8681-f53067b434e9/34-2
00:07:06.349 --> 00:07:12.350
populations based off of the different
criteria that we have for the different

6a771959-1ff8-4201-8681-f53067b434e9/34-3
00:07:12.350 --> 00:07:13.110
campaigns.

6a771959-1ff8-4201-8681-f53067b434e9/35-0
00:07:14.230 --> 00:07:18.319
So you're using,
you're pulling clinical data from the EHR,

6a771959-1ff8-4201-8681-f53067b434e9/35-1
00:07:18.319 --> 00:07:23.840
you have text message vendors that that
data is getting sent to in order to push

6a771959-1ff8-4201-8681-f53067b434e9/35-2
00:07:23.840 --> 00:07:27.521
to the right patients,
and you have an analytics team

6a771959-1ff8-4201-8681-f53067b434e9/35-3
00:07:27.521 --> 00:07:31.270
identifying which patients you want to be
reaching to.

6a771959-1ff8-4201-8681-f53067b434e9/36-0
00:07:31.590 --> 00:07:36.583
Has there been have there been any
technical challenges or challenges in

6a771959-1ff8-4201-8681-f53067b434e9/36-1
00:07:36.583 --> 00:07:42.193
staffing to manage all these applications
and the servers and infrastructure uses

6a771959-1ff8-4201-8681-f53067b434e9/36-2
00:07:42.193 --> 00:07:45.408
supported,
or is it hosted in a hyperscaler or

6a771959-1ff8-4201-8681-f53067b434e9/36-3
00:07:45.408 --> 00:07:48.691
on-prem?
Kind of what are some of the technical

6a771959-1ff8-4201-8681-f53067b434e9/36-4
00:07:48.691 --> 00:07:50.470
challenges of maintaining?

6a771959-1ff8-4201-8681-f53067b434e9/37-0
00:07:51.750 --> 00:07:53.670
kind of a viable program.

6a771959-1ff8-4201-8681-f53067b434e9/38-0
00:07:54.630 --> 00:07:58.234
Yeah,
so a lot of this is hosted on the cloud,

6a771959-1ff8-4201-8681-f53067b434e9/38-1
00:07:58.234 --> 00:08:04.521
but I think the biggest challenge that we
face is with the scalability sometimes,

6a771959-1ff8-4201-8681-f53067b434e9/38-2
00:08:04.521 --> 00:08:08.279
you know, our,
actually it's the texting vendor.

6a771959-1ff8-4201-8681-f53067b434e9/38-3
00:08:08.279 --> 00:08:11.806
Typically,
if we're pushing out a lot of text

6a771959-1ff8-4201-8681-f53067b434e9/38-4
00:08:11.806 --> 00:08:13.110
messages at once,

6a771959-1ff8-4201-8681-f53067b434e9/40-0
00:08:13.830 --> 00:08:17.801
Our texting,
or actually our cell carriers can

6a771959-1ff8-4201-8681-f53067b434e9/40-1
00:08:17.801 --> 00:08:24.308
sometimes read mass messages as spam.
So sometimes messages can get filtered

6a771959-1ff8-4201-8681-f53067b434e9/40-2
00:08:24.308 --> 00:08:27.857
out.
And so that is a risk for when we're

6a771959-1ff8-4201-8681-f53067b434e9/40-3
00:08:27.857 --> 00:08:33.350
doing bulk messaging to patients through
these texting programs.

6a771959-1ff8-4201-8681-f53067b434e9/39-0
00:08:31.750 --> 00:08:32.070
Mm.

6a771959-1ff8-4201-8681-f53067b434e9/41-0
00:08:34.150 --> 00:08:37.323
Other issues are, you know, cell service.
You know,

6a771959-1ff8-4201-8681-f53067b434e9/41-1
00:08:37.323 --> 00:08:41.350
if a patient's not near cell service,
they won't get the message.

6a771959-1ff8-4201-8681-f53067b434e9/41-2
00:08:41.350 --> 00:08:44.096
It's not that they,
it gets delivered later.

6a771959-1ff8-4201-8681-f53067b434e9/41-3
00:08:44.096 --> 00:08:48.002
They just don't get it.
So that's also an issue to be concerned

6a771959-1ff8-4201-8681-f53067b434e9/41-4
00:08:48.002 --> 00:08:50.870
about too,
when you're relying solely on text.

6a771959-1ff8-4201-8681-f53067b434e9/42-0
00:08:51.590 --> 00:08:54.556
Right.
And so how are you leveraging AI to drive

6a771959-1ff8-4201-8681-f53067b434e9/42-1
00:08:54.556 --> 00:08:56.070
these outreach campaigns?

6a771959-1ff8-4201-8681-f53067b434e9/43-0
00:08:56.870 --> 00:09:00.992
We are, yes, because AI,
a lot of this is natural language

6a771959-1ff8-4201-8681-f53067b434e9/43-1
00:09:00.992 --> 00:09:04.556
processing.
So when we're having the conversations

6a771959-1ff8-4201-8681-f53067b434e9/43-2
00:09:04.556 --> 00:09:07.770
with the patients,
when they respond back in,

6a771959-1ff8-4201-8681-f53067b434e9/43-3
00:09:07.770 --> 00:09:11.264
we're using that AI to help drive what,
you know,

6a771959-1ff8-4201-8681-f53067b434e9/43-4
00:09:11.264 --> 00:09:14.408
what the response is and what the,
you know,

6a771959-1ff8-4201-8681-f53067b434e9/43-5
00:09:14.408 --> 00:09:18.950
response back should be based on that
patient incoming response.

6a771959-1ff8-4201-8681-f53067b434e9/44-0
00:09:19.830 --> 00:09:25.800
So you're saying that a message gets sent
out to the patient, the patient replies,

6a771959-1ff8-4201-8681-f53067b434e9/44-1
00:09:25.800 --> 00:09:31.051
and then you leverage some Gen.
AI capabilities to automatically respond

6a771959-1ff8-4201-8681-f53067b434e9/44-2
00:09:31.051 --> 00:09:36.230
to the patient's reply and kind of go
back and forth in a conversation?

6a771959-1ff8-4201-8681-f53067b434e9/46-0
00:09:36.790 --> 00:09:40.716
Right.
And it's still contained to a certain

6a771959-1ff8-4201-8681-f53067b434e9/46-1
00:09:40.716 --> 00:09:45.950
number of responses,
so we can follow the right tree of our

6a771959-1ff8-4201-8681-f53067b434e9/46-2
00:09:45.950 --> 00:09:50.661
programming to get to the end result.
But it's still,

6a771959-1ff8-4201-8681-f53067b434e9/45-0
00:09:48.790 --> 00:09:49.270
Mhm.

6a771959-1ff8-4201-8681-f53067b434e9/46-3
00:09:50.661 --> 00:09:57.990
it's smart enough to know that like a yes
is the same as yeah, or the same as okay.

6a771959-1ff8-4201-8681-f53067b434e9/47-0
00:09:58.190 --> 00:10:02.150
so that it doesn't get stuck along the
conversation.

6a771959-1ff8-4201-8681-f53067b434e9/48-0
00:10:02.510 --> 00:10:07.087
At what point does a person read the
messages or respond to the patient

6a771959-1ff8-4201-8681-f53067b434e9/48-1
00:10:07.087 --> 00:10:08.550
instead of the Gen. AI?

6a771959-1ff8-4201-8681-f53067b434e9/50-0
00:10:09.990 --> 00:10:16.580
There would be some trigger points along
the way. And sometimes it's when,

6a771959-1ff8-4201-8681-f53067b434e9/50-1
00:10:16.580 --> 00:10:21.061
you know,
the patient has responded with a certain

6a771959-1ff8-4201-8681-f53067b434e9/50-2
00:10:21.061 --> 00:10:25.455
keyword like, you know, stuck or,
you know, stop.

6a771959-1ff8-4201-8681-f53067b434e9/50-3
00:10:25.455 --> 00:10:31.430
Those are certain keywords built into the
system that would trigger

6a771959-1ff8-4201-8681-f53067b434e9/49-0
00:10:30.670 --> 00:10:31.150
Mhm.

6a771959-1ff8-4201-8681-f53067b434e9/52-0
00:10:31.830 --> 00:10:36.814
a human to step in.
We also have some failsafes in there too,

6a771959-1ff8-4201-8681-f53067b434e9/52-1
00:10:36.814 --> 00:10:41.396
where, you know,
if at any time a patient needs to speak

6a771959-1ff8-4201-8681-f53067b434e9/52-2
00:10:41.396 --> 00:10:45.737
to a human,
they can indicate it by texting humans so

6a771959-1ff8-4201-8681-f53067b434e9/52-3
00:10:45.737 --> 00:10:48.470
that it can trigger that failsafe.

6a771959-1ff8-4201-8681-f53067b434e9/53-0
00:10:48.790 --> 00:10:53.593
And what about instead of just a human
text message response, a human call or a,

6a771959-1ff8-4201-8681-f53067b434e9/53-1
00:10:53.593 --> 00:10:57.328
yeah, I guess a call,
is there ever a time when somebody would

6a771959-1ff8-4201-8681-f53067b434e9/53-2
00:10:57.328 --> 00:10:58.870
actually call the patient?

6a771959-1ff8-4201-8681-f53067b434e9/55-0
00:10:59.830 --> 00:11:03.623
That is mostly for our like remote
monitoring programs.

6a771959-1ff8-4201-8681-f53067b434e9/55-1
00:11:03.623 --> 00:11:08.432
And it's typically for, you know,
if a patient's indicating, you know,

6a771959-1ff8-4201-8681-f53067b434e9/55-2
00:11:08.432 --> 00:11:13.376
they're feeling worse or, you know,
if they've triggered a response that

6a771959-1ff8-4201-8681-f53067b434e9/55-3
00:11:13.376 --> 00:11:18.524
requires that human call. So, for example,
we, you know, at the nudge unit,

6a771959-1ff8-4201-8681-f53067b434e9/54-0
00:11:13.430 --> 00:11:13.510
The.

6a771959-1ff8-4201-8681-f53067b434e9/55-4
00:11:18.524 --> 00:11:20.150
we had implemented a few

6a771959-1ff8-4201-8681-f53067b434e9/57-0
00:11:20.390 --> 00:11:25.692
blood pressure programs.
And if a patient's blood pressure seemed

6a771959-1ff8-4201-8681-f53067b434e9/57-1
00:11:25.692 --> 00:11:29.548
to be very,
very low or a very concerning blood

6a771959-1ff8-4201-8681-f53067b434e9/56-0
00:11:27.990 --> 00:11:28.550
Mhm.

6a771959-1ff8-4201-8681-f53067b434e9/57-2
00:11:29.548 --> 00:11:33.565
pressure,
that would trigger an alert in the Epic

6a771959-1ff8-4201-8681-f53067b434e9/57-3
00:11:33.565 --> 00:11:39.590
system to a provider that was on call to
call the patient within the hour.

6a771959-1ff8-4201-8681-f53067b434e9/59-0
00:11:40.150 --> 00:11:42.910
Got it.
And that's all from data that patient

6a771959-1ff8-4201-8681-f53067b434e9/59-1
00:11:42.910 --> 00:11:47.830
input into the text messaging program.
Do patients need to opt into this program?

6a771959-1ff8-4201-8681-f53067b434e9/58-0
00:11:44.390 --> 00:11:45.190
Yes.

6a771959-1ff8-4201-8681-f53067b434e9/61-0
00:11:48.710 --> 00:11:52.628
So it depended on the program and what,
you know,

6a771959-1ff8-4201-8681-f53067b434e9/61-1
00:11:52.628 --> 00:11:58.819
what we were able to get approval from
our privacy team on, because it really,

6a771959-1ff8-4201-8681-f53067b434e9/61-2
00:11:58.819 --> 00:12:04.070
you know, some programming,
you can have a default for an opt out,

6a771959-1ff8-4201-8681-f53067b434e9/61-3
00:12:04.070 --> 00:12:07.753
opt out enrollment for some low risk
programs,

6a771959-1ff8-4201-8681-f53067b434e9/60-0
00:12:04.470 --> 00:12:04.950
Mhm.

6a771959-1ff8-4201-8681-f53067b434e9/61-4
00:12:07.753 --> 00:12:11.750
but then for higher risk programs,
you did require

6a771959-1ff8-4201-8681-f53067b434e9/62-0
00:12:12.070 --> 00:12:13.350
opt-in enrollment.

6a771959-1ff8-4201-8681-f53067b434e9/63-0
00:12:14.630 --> 00:12:20.630
And then what have been patient reactions
to this text messaging engagement program?

6a771959-1ff8-4201-8681-f53067b434e9/64-0
00:12:21.750 --> 00:12:25.331
We've actually had really high feedback,
you know,

6a771959-1ff8-4201-8681-f53067b434e9/64-1
00:12:25.331 --> 00:12:31.018
NPS is in the high 90s on a lot of these
programs because it is, as I mentioned,

6a771959-1ff8-4201-8681-f53067b434e9/64-2
00:12:31.018 --> 00:12:36.495
it's just part of the regular workflow
for a patient to like be responding to

6a771959-1ff8-4201-8681-f53067b434e9/64-3
00:12:36.495 --> 00:12:41.270
text messages versus having to log in to
their MyChart app and then

6a771959-1ff8-4201-8681-f53067b434e9/65-0
00:12:41.350 --> 00:12:45.074
respond to a MyChart message from a
provider, you know,

6a771959-1ff8-4201-8681-f53067b434e9/65-1
00:12:45.074 --> 00:12:48.865
you're having to constantly log in,
remember your login.

6a771959-1ff8-4201-8681-f53067b434e9/65-2
00:12:48.865 --> 00:12:52.324
That workflow is just really annoying to
a patient.

6a771959-1ff8-4201-8681-f53067b434e9/65-3
00:12:52.324 --> 00:12:57.844
And so being able to use a text messaging
system to do the same thing is much more

6a771959-1ff8-4201-8681-f53067b434e9/65-4
00:12:57.844 --> 00:12:58.310
easier.

6a771959-1ff8-4201-8681-f53067b434e9/66-0
00:12:59.030 --> 00:13:02.381
Did you have any controlled group of
patients who did,

6a771959-1ff8-4201-8681-f53067b434e9/66-1
00:13:02.381 --> 00:13:06.828
who are compared to the text messaging
patients but did not receive text

6a771959-1ff8-4201-8681-f53067b434e9/66-2
00:13:06.828 --> 00:13:11.824
messages to evaluate the extent to which
this intervention with the text messages

6a771959-1ff8-4201-8681-f53067b434e9/66-3
00:13:11.824 --> 00:13:16.150
has had any positive impact on patient
adherence and patient outcomes?

6a771959-1ff8-4201-8681-f53067b434e9/67-0
00:13:17.150 --> 00:13:22.435
Yes, at the nudge unit,
we had done several RCTs to test,

6a771959-1ff8-4201-8681-f53067b434e9/67-1
00:13:22.435 --> 00:13:26.627
you know,
how the text messages compared to a

6a771959-1ff8-4201-8681-f53067b434e9/67-2
00:13:26.627 --> 00:13:33.280
survey or even sending messages in the
MyChart system and texting always

6a771959-1ff8-4201-8681-f53067b434e9/67-3
00:13:33.280 --> 00:13:36.470
prevailed over other usage systems.

6a771959-1ff8-4201-8681-f53067b434e9/68-0
00:13:36.870 --> 00:13:39.334
So,
texts prevailed and meaning that there

6a771959-1ff8-4201-8681-f53067b434e9/68-1
00:13:39.334 --> 00:13:40.710
was more responsiveness.

6a771959-1ff8-4201-8681-f53067b434e9/69-0
00:13:41.110 --> 00:13:45.110
Yes, adherence,
satisfaction was always higher.

6a771959-1ff8-4201-8681-f53067b434e9/70-0
00:13:46.550 --> 00:13:50.491
Okay, and okay,
so you've been finding a success with

6a771959-1ff8-4201-8681-f53067b434e9/70-1
00:13:50.491 --> 00:13:54.797
this program and success with leveraging
AI. What kind of,

6a771959-1ff8-4201-8681-f53067b434e9/70-2
00:13:54.797 --> 00:14:00.197
have there been any AI governance
challenges that you've run into or what

6a771959-1ff8-4201-8681-f53067b434e9/70-3
00:14:00.197 --> 00:14:05.963
has the process been like to bring on
vendors or to roll this out from a pilot

6a771959-1ff8-4201-8681-f53067b434e9/70-4
00:14:05.963 --> 00:14:07.350
to enterprise-wide?

6a771959-1ff8-4201-8681-f53067b434e9/71-0
00:14:08.150 --> 00:14:10.390
How have you gone about those processes?

6a771959-1ff8-4201-8681-f53067b434e9/72-0
00:14:11.350 --> 00:14:17.306
Yeah, so I would say at Mount Sinai,
it's been a different process than at the

6a771959-1ff8-4201-8681-f53067b434e9/72-1
00:14:17.306 --> 00:14:22.131
Nudge unit. At Mount Sinai,
we actually have a really robust AI

6a771959-1ff8-4201-8681-f53067b434e9/72-2
00:14:22.131 --> 00:14:25.900
governance process where we run an
assurance lab.

6a771959-1ff8-4201-8681-f53067b434e9/72-3
00:14:25.900 --> 00:14:29.670
So that is essentially a lab that like
pilots out

6a771959-1ff8-4201-8681-f53067b434e9/73-0
00:14:29.790 --> 00:14:36.602
these use cases at a really small scale
prior to us launching with patients to do

6a771959-1ff8-4201-8681-f53067b434e9/73-1
00:14:36.602 --> 00:14:41.587
some testing with the vendors,
with the LLMs, in, you know,

6a771959-1ff8-4201-8681-f53067b434e9/73-2
00:14:41.587 --> 00:14:46.654
a controlled test to ensure that it's
going to be, you know,

6a771959-1ff8-4201-8681-f53067b434e9/73-3
00:14:46.654 --> 00:14:51.390
viable and successful when we launch this
with patients.

6a771959-1ff8-4201-8681-f53067b434e9/74-0
00:14:51.990 --> 00:14:55.822
How large of the team, like what's the,
I guess,

6a771959-1ff8-4201-8681-f53067b434e9/74-1
00:14:55.822 --> 00:15:02.312
is this a program that pays for itself or
did you need to have leadership allocate

6a771959-1ff8-4201-8681-f53067b434e9/74-2
00:15:02.312 --> 00:15:08.412
FTEs to manage the texting vendor and to
manage the public cloud instances in

6a771959-1ff8-4201-8681-f53067b434e9/74-3
00:15:08.412 --> 00:15:12.870
which this program is hosted or to manage
the APIs with?

6a771959-1ff8-4201-8681-f53067b434e9/75-0
00:15:12.950 --> 00:15:15.990
Epic, kind of what, how has that worked?

6a771959-1ff8-4201-8681-f53067b434e9/76-0
00:15:16.870 --> 00:15:20.791
Yeah,
so we have a digital team already and

6a771959-1ff8-4201-8681-f53067b434e9/76-1
00:15:20.791 --> 00:15:24.712
they manage our other digital
applications.

6a771959-1ff8-4201-8681-f53067b434e9/76-2
00:15:24.712 --> 00:15:30.326
And so it didn't take additional FTE to
manage these programs.

6a771959-1ff8-4201-8681-f53067b434e9/76-3
00:15:30.326 --> 00:15:37.366
I would say where it does take some
additional FTE is on the operational side.

6a771959-1ff8-4201-8681-f53067b434e9/76-4
00:15:37.366 --> 00:15:37.990
So when

6a771959-1ff8-4201-8681-f53067b434e9/77-0
00:15:38.110 --> 00:15:41.623
you know,
we have alerts coming in for that low BP

6a771959-1ff8-4201-8681-f53067b434e9/77-1
00:15:41.623 --> 00:15:45.137
or, you know,
a concerning message from a patient.

6a771959-1ff8-4201-8681-f53067b434e9/77-2
00:15:45.137 --> 00:15:49.133
We need that, you know,
provider on the other end or some

6a771959-1ff8-4201-8681-f53067b434e9/77-3
00:15:49.133 --> 00:15:52.578
operational partner to respond to those
messages.

6a771959-1ff8-4201-8681-f53067b434e9/77-4
00:15:52.578 --> 00:15:58.227
And so we did have to ensure that there
were FTEs on the other side that would be

6a771959-1ff8-4201-8681-f53067b434e9/77-5
00:15:58.227 --> 00:15:58.710
able to

6a771959-1ff8-4201-8681-f53067b434e9/78-0
00:15:59.350 --> 00:16:01.190
You know, staff these programs.

6a771959-1ff8-4201-8681-f53067b434e9/80-0
00:16:01.670 --> 00:16:04.652
Was there an executive champion of this
program?

6a771959-1ff8-4201-8681-f53067b434e9/79-0
00:16:02.590 --> 00:16:02.790
And.

6a771959-1ff8-4201-8681-f53067b434e9/80-1
00:16:04.652 --> 00:16:08.304
I have to imagine there needed to be
clinician buy-in here.

6a771959-1ff8-4201-8681-f53067b434e9/80-2
00:16:08.304 --> 00:16:13.173
Was there the CMIO involved or anyone of
that nature in order to ensure that if

6a771959-1ff8-4201-8681-f53067b434e9/80-3
00:16:13.173 --> 00:16:18.042
physicians or nurses or NPs or PAs were
required to respond in the instance of,

6a771959-1ff8-4201-8681-f53067b434e9/80-4
00:16:18.042 --> 00:16:19.990
for example, low blood pressure,

6a771959-1ff8-4201-8681-f53067b434e9/81-0
00:16:20.510 --> 00:16:23.030
that they actually had availability and
interest in doing so.

6a771959-1ff8-4201-8681-f53067b434e9/82-0
00:16:24.070 --> 00:16:28.397
Yeah, at Sinai,
we have our Chief Digital Transformation

6a771959-1ff8-4201-8681-f53067b434e9/82-1
00:16:28.397 --> 00:16:34.775
Officer, who's our key executive sponsor.
And then we also work really closely with

6a771959-1ff8-4201-8681-f53067b434e9/82-2
00:16:34.775 --> 00:16:40.470
our Chief Population Officer on a lot of
these remote monitoring programs.

6a771959-1ff8-4201-8681-f53067b434e9/83-0
00:16:41.030 --> 00:16:46.214
And I suppose that a lot of these patient
reported outcome measures kind of feed

6a771959-1ff8-4201-8681-f53067b434e9/83-1
00:16:46.214 --> 00:16:50.886
into HEDIS measures or other public
reported quality measures to CMS and

6a771959-1ff8-4201-8681-f53067b434e9/83-2
00:16:50.886 --> 00:16:55.430
effects reimbursement rates.
Have you had any challenges kind of tying

6a771959-1ff8-4201-8681-f53067b434e9/83-3
00:16:55.430 --> 00:16:59.910
your program success to improved revenues
in order to secure funding?

6a771959-1ff8-4201-8681-f53067b434e9/83-4
00:16:59.910 --> 00:17:05.030
Or were there any other hoops you had to
jump through to secure initial funding

6a771959-1ff8-4201-8681-f53067b434e9/84-0
00:17:05.310 --> 00:17:07.430
to try out this program when you are
getting started.

6a771959-1ff8-4201-8681-f53067b434e9/85-0
00:17:09.270 --> 00:17:12.572
Yes,
I would say that there definitely have

6a771959-1ff8-4201-8681-f53067b434e9/85-1
00:17:12.572 --> 00:17:17.000
been some challenges initially,
especially with, you know,

6a771959-1ff8-4201-8681-f53067b434e9/85-2
00:17:17.000 --> 00:17:22.930
some of the CMS funded programs. You know,
for the blood pressure programming,

6a771959-1ff8-4201-8681-f53067b434e9/85-3
00:17:22.930 --> 00:17:28.409
some of them where we were actually
providing cuffs to patients required

6a771959-1ff8-4201-8681-f53067b434e9/85-4
00:17:28.409 --> 00:17:29.910
initial funding from

6a771959-1ff8-4201-8681-f53067b434e9/86-0
00:17:30.630 --> 00:17:34.864
the health system to, you know,
be able to provide cuffs for every

6a771959-1ff8-4201-8681-f53067b434e9/86-1
00:17:34.864 --> 00:17:39.983
patients because we knew they didn't have
cuffs at home to be able to take their

6a771959-1ff8-4201-8681-f53067b434e9/86-2
00:17:39.983 --> 00:17:44.596
blood pressure. So that took initial,
you know, business case providing,

6a771959-1ff8-4201-8681-f53067b434e9/86-3
00:17:44.596 --> 00:17:47.567
you know,
showing that the outcomes on the end

6a771959-1ff8-4201-8681-f53067b434e9/86-4
00:17:47.567 --> 00:17:50.790
would have that change and then be able
to provide

6a771959-1ff8-4201-8681-f53067b434e9/87-0
00:17:51.270 --> 00:17:57.270
ROI at the end, preventing patients from,
you know, requiring extra hospitalization,

6a771959-1ff8-4201-8681-f53067b434e9/87-1
00:17:57.270 --> 00:18:02.281
you know, reducing readmissions,
reducing visits that would eventually

6a771959-1ff8-4201-8681-f53067b434e9/87-2
00:18:02.281 --> 00:18:04.470
lead to, you know, saved costs.

6a771959-1ff8-4201-8681-f53067b434e9/88-0
00:18:05.350 --> 00:18:08.928
So how long, I don't know,
did you do a retrospective analysis to

6a771959-1ff8-4201-8681-f53067b434e9/88-1
00:18:08.928 --> 00:18:13.048
see how long it took to prove your ROI
that you originally projected to the

6a771959-1ff8-4201-8681-f53067b434e9/88-2
00:18:13.048 --> 00:18:16.571
executive sponsors so that they would pay
for those blood cuffs,

6a771959-1ff8-4201-8681-f53067b434e9/88-3
00:18:16.571 --> 00:18:17.710
blood pressure cuffs?

6a771959-1ff8-4201-8681-f53067b434e9/90-0
00:18:18.310 --> 00:18:21.680
Yeah,
so we were still in that process because

6a771959-1ff8-4201-8681-f53067b434e9/90-1
00:18:21.680 --> 00:18:25.840
these programs are still very early on.
I would say they,

6a771959-1ff8-4201-8681-f53067b434e9/90-2
00:18:25.840 --> 00:18:29.425
we've been live with some of them for six
months.

6a771959-1ff8-4201-8681-f53067b434e9/90-3
00:18:29.425 --> 00:18:35.234
I would say that we had more data at the
nudge unit to prove our ROI for some of

6a771959-1ff8-4201-8681-f53067b434e9/90-4
00:18:35.234 --> 00:18:36.310
those programs.

6a771959-1ff8-4201-8681-f53067b434e9/89-0
00:18:36.390 --> 00:18:38.310
And how did that go with the nudge unit?

6a771959-1ff8-4201-8681-f53067b434e9/91-0
00:18:38.950 --> 00:18:43.035
It went well.
We were able to prove we had, I think,

6a771959-1ff8-4201-8681-f53067b434e9/91-1
00:18:43.035 --> 00:18:46.504
saved,
I think it was like a two to one or a

6a771959-1ff8-4201-8681-f53067b434e9/91-2
00:18:46.504 --> 00:18:49.510
three to one savings for every patient.

6a771959-1ff8-4201-8681-f53067b434e9/94-0
00:18:49.990 --> 00:18:52.954
That's $3,
for every dollar spent you save,

6a771959-1ff8-4201-8681-f53067b434e9/94-1
00:18:52.954 --> 00:18:56.659
you generate,
save $3 or generated $3 and okay, saved,

6a771959-1ff8-4201-8681-f53067b434e9/92-0
00:18:53.110 --> 00:18:53.510
Okay.

6a771959-1ff8-4201-8681-f53067b434e9/93-0
00:18:55.430 --> 00:18:56.230
Saved.

6a771959-1ff8-4201-8681-f53067b434e9/94-2
00:18:56.659 --> 00:18:59.893
not generated. Okay, thank you. So again,
Neda,

6a771959-1ff8-4201-8681-f53067b434e9/94-3
00:18:59.893 --> 00:19:03.396
we're coming up to the end of this
podcast episode.

6a771959-1ff8-4201-8681-f53067b434e9/94-4
00:19:03.396 --> 00:19:08.449
And just to remind our listeners,
we've been discussing patient access and

6a771959-1ff8-4201-8681-f53067b434e9/94-5
00:19:08.449 --> 00:19:11.750
patient engagement at Mount Sinai Health
System.

6a771959-1ff8-4201-8681-f53067b434e9/95-0
00:19:12.070 --> 00:19:16.821
talking about text messaging and talking
about prioritizing AI solutions to

6a771959-1ff8-4201-8681-f53067b434e9/95-1
00:19:16.821 --> 00:19:21.884
improve patient access and utilization.
Any advice that you'd give to yourself a

6a771959-1ff8-4201-8681-f53067b434e9/95-2
00:19:21.884 --> 00:19:24.698
year ago,
maybe when you began some of these

6a771959-1ff8-4201-8681-f53067b434e9/95-3
00:19:24.698 --> 00:19:29.511
programs that you'd like to share with
people listening right now who may be

6a771959-1ff8-4201-8681-f53067b434e9/95-4
00:19:29.511 --> 00:19:33.950
interested in replicating your
experiences at their own health system?

6a771959-1ff8-4201-8681-f53067b434e9/96-0
00:19:35.350 --> 00:19:38.816
Yeah, I mean,
I think a couple things that I've already

6a771959-1ff8-4201-8681-f53067b434e9/96-1
00:19:38.816 --> 00:19:42.777
mentioned are like, you know,
thinking about how you can reduce

6a771959-1ff8-4201-8681-f53067b434e9/96-2
00:19:42.777 --> 00:19:47.543
friction instead of adding friction to
the patient workflow, thinking about,

6a771959-1ff8-4201-8681-f53067b434e9/96-3
00:19:47.543 --> 00:19:51.999
you know, as much as possible,
how you can build in defaults instead of

6a771959-1ff8-4201-8681-f53067b434e9/96-4
00:19:51.999 --> 00:19:56.270
thinking about how you add in reminders
or education. I think that's

6a771959-1ff8-4201-8681-f53067b434e9/97-0
00:19:56.390 --> 00:20:01.167
typically what people think about when
they're thinking about how to change

6a771959-1ff8-4201-8681-f53067b434e9/97-1
00:20:01.167 --> 00:20:06.321
patient behavior or increase engagement.
And then I think one thing that is a new

6a771959-1ff8-4201-8681-f53067b434e9/97-2
00:20:06.321 --> 00:20:10.218
thing from, you know,
the behavioral economics perspective is

6a771959-1ff8-4201-8681-f53067b434e9/97-3
00:20:10.218 --> 00:20:14.618
how you add in gamification into the way
that you build in, you know,

6a771959-1ff8-4201-8681-f53067b434e9/97-4
00:20:14.618 --> 00:20:17.510
patient engagement in some of these
programs.

6a771959-1ff8-4201-8681-f53067b434e9/98-0
00:20:17.590 --> 00:20:23.312
How can you add incentives to the
patients? You know, add in like streaks,

6a771959-1ff8-4201-8681-f53067b434e9/98-1
00:20:23.312 --> 00:20:29.110
games to like get them, you know,
thinking and promoting positive behavior.

6a771959-1ff8-4201-8681-f53067b434e9/99-0
00:20:29.990 --> 00:20:33.673
Fun. Gamification, streaks and games.
That's a great way to end it.

6a771959-1ff8-4201-8681-f53067b434e9/99-1
00:20:33.673 --> 00:20:36.219
So for our listeners,
this has been Neda Khan,

6a771959-1ff8-4201-8681-f53067b434e9/99-2
00:20:36.219 --> 00:20:39.957
a director of Digital Experience at Mount
Sinai Health System. Neda,

6a771959-1ff8-4201-8681-f53067b434e9/99-3
00:20:39.957 --> 00:20:42.070
thank you so much for joining us today.

6a771959-1ff8-4201-8681-f53067b434e9/100-0
00:20:42.550 --> 00:20:44.230
Thank you so much. It's been great.