Optimizing UM with automation and AI

Hello, everyone. And thank you for joining

today's webinar

Interco Auto Review,

Optimizing um with automation

and A I.

My name is Heather Volmer with Optum and

I will be your host today.

Before we begin, please note the

following housekeeping items

at the bottom of your audience console

are multiple application widgets. You

can use to customize your viewing experience.

If you have any questions during the webcast,

you can click on the Q and A widget at the bottom

of your screen to submit a question.

If you experience any technical difficulty,

please click on the help widget. It covers

common technical issues.

You can expand your slide area by

clicking on the maximize icon on

the top right of the slide window

or by dragging the bottom right

corner of the slide window.

There is a survey widget which you can use

at the end of the webcast to provide us

with feedback on today's presentation.

Additionally, this presentation uses streaming

audio. You may listen to the audio

through your computer speakers or headphones

to ensure the best possible system

performance. Please be sure to shut

down any VPN connections

and connect directly to the internet

and with that. I will hand it over to

Laura mcintyre.

Thank you, Heather

and good afternoon everyone.

My name is Laura mcintyre. I

lead our strate strategic client

teams for inter qual. Um

I've worked for change health care for many

years, um which is now part of Optum.

I am a registered nurse with a Bachelor of Science

degree in nursing and a master's in health

care administration and I've been in the industry

for more years than I'd like to count.

Um But it's a pleasure to be here today

to share with you how Interpol auto

review can be used to help optimize

your um program. And to share a little

bit about how some of our customers

who are using auto review have realized

success in their um transformation

journey with this capability. So I'm

excited to be here.

So to start off, I wanted to share a little

bit about some of what we're all seeing and feeling

in the industry because regardless

of what type of organization you work

for, we're all facing challenging

times across the health care industry

globally, right? There's a global

nursing shortage. Um As

a result of the pandemic, we've seen

this thing that we're dubbing the great resignation

across health care, which really

further compounds this problem that's been

building for years. And honestly,

there's really no clear path to how,

how we're gonna really reduce that nursing

um shortage, how we're gonna really reduce

that gap in in nursing resources

to compound things. We've got an aging

population that brings with

it more complex care needs

and those numbers are just continuing to grow.

We also know that access to behavior,

health and mental health services is

incredibly challenging right now.

And compared to other industries,

we have um really

outdated technologies and a

tremendous amount of inefficiency in our

systems.

Our processes are all highly manual,

they're tedious, they're fraught with error

and authorizations are really

time consuming

to add on to that. The back and forth

between hospitals and and health

plans is incredibly burdensome

for anyone that's involved in that process.

And as we all know, costs are still continuing

to rise. So since the onset of COVID,

we're actually seeing denial rates rising,

they've gone up about 11% nationally.

And we know that the entire appeals process

is a challenge, a challenge that's resource

intensive and burden burdensome.

We also are seeing that success rates

of appeals are declining. So increasing

the lack of resources to prevent and

work those denials, which means we need to be

diligent

about getting the right level of care from

the start.

So we're really at a tipping point more

now than ever. And we're really needing

to look at what opportunities and capabilities

exist out there to help reduce the burden,

to help offset the impact of the nursing

shortage, to help ensure that we're

doing things accurately and appropriately

from the start. And that we're doing the best

we can to manage costs um

in the most efficient way possible.

So we talk a little bit about,

you know, how we can think about

automation. It's not all doom and gloom. Right.

We, we've got a lot that's going on out

there in the market. If you've been thinking about

how to transform your UN program.

Um There's a lot of things to be thinking

about and a lot of capabilities to look at

out there um at change healthcare.

Now, often we've been looking at technology

to help drive efficiencies

in the process as the workflow processes

of today for more than a decade. Um

It's incredibly hard complex

work and to be blunt, there's no

silver bullet out there, there's no silver

bullet solution that's complete or perfect

um capabilities like um artificial

intelligence and robotic process automation

are evolving. Uh We know

that the industry standards think fire,

you hear, you hear fire anywhere, any

in any article or any uh podcast

you're listening to related to health care, industry

standards like that are also continuing

to evolve. So we're not quite there yet

in terms of having an end to end

all encompassing perfect solution,

but we don't need to wait until technology

is perfect. There's still a tremendous amount

of value to be gained by

taking advantage of capabilities that are

in the market today. Transform in

your program with these automation capabilities

really can help reduce administrative

burden we know that they can help

drive accuracy and also

foster collaboration uh between

payers and providers really trying to build

trust because we all know that's lacking

today. Um So anything

we can do

to employ machines to

do what they're good at, can free

up reach sources, those precious clinical resources

to do what they're good at. Right.

So with staffing challenges that I think

we all have today, we can really

take some of those routine tasks

or basic tasks that are high volume

out of the equation, freeing up

clinical staff and allowing them to focus

on more complex patients and thus being

able to get to a state where we

can actually begin to manage

by exception.

So I think it's a, it's a place we all wanna be.

I'm gonna talk a lot today about the Interpol auto review

capability. I'll jump into some case

scenarios and do a couple of demonstrations

for you, but just to talk a little bit about

it. Um But before we jump

in and start, um at Interpol, we've

had a team of clinical consultants who have done thousands

of audits chart audits, case audits to

really help our customers opti optimize

the use of um Interpol

in our in our capability solutions. And

in that audit process, we have found

somewhere around one out of three cases,

have some sort of error or some sort

of inaccuracy or some

sort of um process that

they didn't package up the case in a

way that made clinical sense to a health plan

and therefore getting denied or, or they're challenged

with it. So whether it's the right subset

being selected, whether it's driving

to the right level of care or

whether it's finding criteria

that's validated by data

that exists in the E hr it could be

any of that.

And we know that auto review Interpol auto

review can help solve for a lot of that.

It really can drive to accuracy because

we're populating a medical necessity review

with data from the

source. The source of truth that E hr

and we're automatically creating that

ro review without having a human

even intervening it. So that

drives um trust and

sensibility

little bit about inter Q auto review and just

give you a little background if you're not familiar with

it. This is a capability, it's a cloud

based solution that automates

the medical necessity reviews

at the point of admission decision

using real time data from your

E hr

this robotic process automation

capability direct directly integrates

into your workflow which is critically

important. And I'll talk a little bit more about that as we

move forward.

But with the admission order serving

as the trigger to begin the automated

process, the process kicks

off when that um admission order

hits the record, it creates

a medical necessity review. It

populates that medical necessity review

with the data in the E hr

and completes that review. So it extracts

that data structured data and

unstructured data from the E

hr and transmits that review

with the embedded data back into

your um workflow.

OK. This is where your staff can then

access that review

for the ones that are fully completed. They

can validate that that's appropriate

and correct and send it off to the the

health plan the way they normally would

or if it's only partially completed,

they can edit it or finish the review

if you will. And it will also

provide a notification back to the E hr

So really exciting to think

about the possibilities here and to think about

the cases that we maybe can take off

the workload of the care manager today,

as I mentioned, it integrates directly into

the workflow and from a workflow standpoint,

just to give you a sense for where inter qual

auto review comes into play, you

may have a patient that presents to the er

or the EV uh they're triaged

and there's some sort of clinical assessment process

that occurs. And once the clinician

or physician makes a determination

that the patient needs to be admitted to the hospital,

regardless of whether or not it's observation

status or an inpatient status,

the auto review capability is going to trigger

and that auto review will trigger

and continue to run for the 1st

24 hours. It'll collect

data and it will populate that review

for the 1st 24 hours

um automating that um episode

day one review for you.

So really saving a lot of time

for the care manager and in trying to

gather that information and

do that work from the start.

So um to recap,

it's integrated the workflow, it automates

the review by triggering from the point

of admission order. It continues

to collect data for that case for the 1st

24 hours and populates the

criteria, the inter quality criteria,

not only with the checkmark indicating that

that criteria point is met or not

met, but it also embeds

the data, the EMR data and

its source date and time stamps

and brings it right into your um system

under the covers. What we've done is tag

the words within Interpol to

standard nomenclature databases.

And we have a rigorous process for mapping

the data to make it machine readable.

And we perform a technical and

clinical validation process to ensure

accuracy

and auto review is gonna select the appropriate

subset based on the admission diagnosis.

It's going to select the appropriate level

of care and the corresponding

criteria point based on that

patient severity, the interventions

that have been delivered and it's gonna

be defensible and clear because it's gonna

have that actual data coming from the

record embedded as that source

of truth

today. This cap

capability is available

in the market. We are able

to automate over 80% of

condition um focused

reviews for the um admission

day. So that first episode day,

while we're not yet able to auto

populate all data points, we are

able to fully complete about 15%

of reviews to a level of care.

And for those that were not able to fully complete,

we can populate about 70

75% of the criteria needed

to complete that review. So there's

a tremendous amount of value in those that

we can partially complete as well. Um

I don't wanna discount those because I think that piece

is incredibly important. So for example,

if you're doing a medical necessity review,

and there are four criteria that you need

to collect four data points that you need

to collect to complete that review,

we can capture and populate three

of those. So if we're not able to fully get

all of them, we can get at least at least three

of those four. So that means the

user simply needs to open up that automated

review, find the single criterion

we were not able to collect or not able

to verify or interpret

and they can simply check that record to see

if that criterion can be manually completed.

So the amount of time savings

can be substantial, there's always

gonna be some criteria that are challenging

uh to make machine readable such

as um things like the word greater

than baseline. Um These are things

that are gonna require a clinical interpretation.

They're a little bit harder to get at a little

bit more challenging for a machine to

read and accurately interpret

it. Because if we can't ensure accuracy

is there, we're not gonna have that auto

populate. We're gonna still rely on uh care

managers or utilization managers to go in and verify

those criteria points, but lots

of value still to be had, whether it's fully

met to a level of care or even just

partially met to that level of care.

And because auto review is a cloud based technology,

we also include along with that a really

robust reporting platform.

Um This is where uh you can monitor

how many auto reviews are generated, you can

track user adoption um

as well as productivity measures such

as the number of reviews that users are completing.

Um our reporting platform

has interactive drill downs.

It really allows you to focus on the data

that's pertinent to whatever issue

is at hand. So we have our level

data that allows you to look at

um um behaviors on a case by

case level. Um We have performance

visualizations that allow you at a glance

information. So you can see how the auto

review capability is performing.

We have productivity reports

by user by facility

by subset lot lots of ways

that you can drill down and

figure out, you know how things are going

or where you might have opportunities to improve.

We even have observation versus inpatient

data along with estimated roys.

Um And many, many many more things

um our platform is also somewhat customizable.

So if we don't have a report that you're looking for

information on, we can work with you to figure

that out.

But our ru customers uh today

are using these reports really

to advantage, trying to their advantage, trying

to, you know, keep their finger on the pulse

of what's happening across their department

and improving the results by taking

action based on them, whether it's you

individual user education, whether

it's process changes or process

improvements or maybe even system

updates. So I'll talk a little bit more about

that when we get into the results.

Ok. So let's jump in. Um

Before I jump into the demonstration, I let me walk

through and set the stage with a patient case

scenario. So I'll take you through

this and then I'm gonna, I'm gonna move over into

our, our, our application to be able to show

you what on review would look like.

Um So in this case scenario, we

have an 89 year old patient

that presents to the Ed with a three

day history of shortness of breath,

cough, weakness

and fever that started that morning.

Um In the Ed, he had

02 sacs that ranged from

85 to 87% on

room air. Uh looks like he was

placed on three liters of nasal cannula.

And then his 02 sacs came up to somewhere

around 93 to 97.

He had orders to be admitted for community

acquired pneumonia.

Um He started on antibiotics.

He got some IV fluids, he

got it. He had a chest X ray that showed

right lower lobe infiltrate and

had a bunch of uh standard

blood work urinalysis, cultures,

things of that nature.

You can see in his vitals that his blood pressure

was low on arrival. His temperature

was low grade.

He had a couple of abnormal um lab

values there. B un elevated

at 42 sodium was

1 32. His U A was negative

COVID was negative definitely has

a white count that's elevated and and

those blood cultures are still pending.

So let me move over. I'm gonna

uh move over to

the application,

share that screen with you.

Now when we integrate

inter qual auto review, we

integrate it into your

um workflow application

and and run against the E hr

so we don't have access to that. So what we

here have here is our uh

acne care manager fig tool

that we've created to be able to demonstrate and

show you the capability. So

when you, when you have auto review integrated

and you go in and you access

the case and you click on it just like

I did, it's gonna bring you right into that medical

necessity review and you can see that

the medical necessity review has already

been completed for me. I don't have

to actually do anything with this case

other than take a peek and say I'm gonna send

it off to the health plan

and there's a couple of things I'll point out to you,

you can see that the acute level of care has

been met

and you can see we have some gray text to the

left of it here with some arrows. E

hr updates complete. This

means that the auto review capability

has run for 24 hours or

more. Once it runs for 24

hours or more, you'll see that it's complete.

And if it was still running, uh

meaning it was less than 24 hours,

you would see that circle, these

arrows would be turning in a circle

and it would say E hr updates are,

are being gathered, right? So it's continuously

gathering data. So as additional

data comes into the E hr maybe it's

imaging results or lab tests

or, or other orders, it's

gonna continue to collect that information

until the 24 hour

I don't mark,

but initial of care was met and somebody

decided to complete the review and that's gonna stop

the auto review from running. And in this

case, acute level of care has been

met. When I click on acute

level of care and expand,

I can see that it says pneumonia confirmed

by imaging and it's a both rule.

And when I expand that I'll see the findings

and the intervention. So I'm just gonna expand

a little bit more so I can show you here,

what we're looking at for the obs level

of care.

And if you look at pneumonia confirmed by imaging,

you're gonna see a file folder next to it with

some um little dots in it that's

called an ellipsis. You'll also see

other uh criteria points have

folders next to it with a plus sign

that that is a visual cue or

an indicator to let you know that there is e

hr data that has been collected

relative to that criteria point.

And if I were to click on this one next

to pneumonia confirmed by imaging, it's gonna

pull up for me the imaging results.

So this is the imaging test, imaging

narrative. I can see that the chest CT

angiogram was done. I have

the ability to click on these little plus signs

next to it and you're gonna be able to see

the uh the narrative summary.

You're gonna see some text that might be highlighted

that the auto review capability is focused

on. In this case, it's saying bilateral

multifocal ground glass opacities

and nodules likely reflecting

multifocal pneumonia. If

there's multiple results, you're gonna see multiple

results in there. And so it

is confirming that there is likely

a pneumonia by imaging.

So that one was checked, you're gonna see

that some criteria are outlined

but they are not filled in. Uh That

means that the auto review capability found

data, but it did not, it either

did not support the criterion

or we were unable to make a

full interpretation and the user

needs to come in and take a look at that.

I can take a look at these two criterion

and see what's happening in those folders

if I want to. But I don't even

have to do that for this case because

this patient is meeting at the acute level

of care because they have a pneumonia confirmed by imaging.

They are meeting the curve 65

criterion and they are meeting the imaging

component. Um Sorry, the intervention

component, excuse me,

if I were to look at the curve 65 criteria,

we can see that the system auto selected

four components.

There was no confusion that we mentioned

in our clinical scenario,

but we did mention that he had a B UN of

42. So that meets this criterion.

And if I were to click on that file folder,

it's going to tell me uh what

the code name and description are.

It'll tell me what the recorded values

are. And if there's multiple values

those would be listed in here. Um

This is just a demonstration. So the numbers

might not be accurate here. We this is,

you know, just sort of pulled together for demonstration

purposes, but it will show you all

of the results for A B

UN. It would show you all of the

results for the respiratory rate. So any

recorded values date and time stamp

for what this individual's respiratory

rate was um you know what the value

was where it was found in the record

and the timing of it.

So we know that he was um tit

it, we know that the blood pressure was on

the low side and we know that this individual

was elderly. So from the facts

of the case, uh looking at what was found

in the E hr this is exactly how it

would work. It would auto select the criteria

points and we know that this patient was

also receiving anti-infective. And again,

we could pull this up, we could see that he's on

set Axon 1 g, um,

and 50 mL piggyback. Um

And how many times it was administered

at the time the data was collected, date and time

stamp. So it's gonna pull all the information,

whether it comes from the MA R, whether

it comes from vital signs, whether it's a diagnosis.

Um It's really gonna pull all the data that it can

to be able to satisfy that review.

So that's an example of a case

that's fully met. And the user would simply

take a look at this and say, yep, this all seems great.

I'm done. I don't need to do anything. I don't

need to hunt in the E hr I don't need to find

any data points to pull together.

Um For this review, I'm free to send

this off and, and move along to my next case.

So I'm gonna close this one.

I'm gonna come back into our, our

um homemade um application

here and I'm gonna open up another case that

was only partially met. So I wanna

show you what it would look like if the review

was only partially met.

Um In this case, you'll see it says partial.

So it did not fully meet the level of care

and I can open up any of these levels

and see what data it was able to

gather. So if the request is for the patient

to be inpatient, we'll start here

again and again, it's

got pneumonia confirmed by imaging. It

has some of the finding criteria

and it has that intervention for anti-infective.

So it did meet for the intervention,

it did meet for the uh uh curb

65 criterion. And

then there's a couple of other things that maybe couldn't

be confirmed.

We can look at the imaging study for pneumonia

and we can see that it is able to

pull some of that data and be able to read

what that imaging narrative is. And we

can see that it does say right, lower lobe pneumonia

infiltrate as with underlying pneumonia.

So it's telling us that it is believing

that this is pneumonia.

And for this case example, it wasn't

able to fully read that or read it

with enough accuracy and confidence to be

able to select criteria point. But

as the user, I can go ahead and select

that and say yes, I want

to select that criteria point because I am

verifying that that is accurate. And it is true

based on that imaging study. And I don't

even have to go out to the record to find that.

So it's, it's, it's able to pull some of

that same thing with this one. Here. You can see

for two lobes, there is a file folder

and I can pull that open and take a look

at what it says. And it doesn't say that

there's two lobes, it says it's a right lower lobe.

So this criteria point is not met.

Um But it is providing me that E

hr data in there to go along with

my case and help me paint that picture.

So that's how it would work. If it was a

partially met case, I would come in,

I would look at the case. I would see what was

met, see where we maybe weren't able

to find criteria or

maybe where we weren't able to satisfy

it enough that we could check that criteria point

and I can finish the review myself.

But I don't have to go hunt for the anti-infective.

I didn't have to go hunt for the curve 65.

I didn't have to go hunt for the imaging narrative.

It's all been pulled in here for me and I

just need to verify it and complete it on my

own.

So fairly straightforward, fairly simple

um that those are just two examples

of how the auto review capability works

and how it looks. So let me

close out of this

and I will come,

I will stop sharing that screen

and come back on over to

the review here. So those are

our two examples to walk you through around

how the auto review capability would work.

Um Depending on what system you're in.

Like I said, um we're able

to do some things um a little

bit more robustly and epic than we are in Cerner,

but we certainly can provide information

in Cerner that helps you to complete that review.

Um But we're still working on evolving

the capability.

As I mentioned earlier, there is

no perfect capability that exists

out there in the market. These technologies

with A I and robotic process automation

are continuing to evolve. So

we keep growing our road map and keep

building functionality and capability

to help evolve things. Um A as

time goes on

today, we've got um somewhere

around 100 and 50 hospitals or so using

auto review in the market. And

when it comes to um

program transformation with

automation, there's a tremendous

amount of lessons that

we've learned. Uh We've got lots of

hospitals in various stages of

implement implementation. And um as

we continue to grow, we're seeing over 25,000

automated reviews occurring on a monthly

basis and that number is just continuing

to tick up. But as we implement

and integrate our customers. We're

learning a tremendous amount about how

we can tweak and massage things

to make them work better, how we can help

our customers to do things different or better.

So we're learning from our users, we're continuously

improving this capability and

evolving our solution as the industry

standards evolve and change as well.

So, really exciting to think about

what we can do and and to see some of

the successes that we have within the market.

So when it comes to automating

with a capability like Interpol auto

review, um with those 150

or so hospitals that we have, we've

learned that when it comes to

um program transformation with

automation, it's really a three

legged stool to reach success

and value. And the, the three

legs of that stool are really, you

know, looking critically at your systems,

really looking at your processes

and looking at your people. And

we're finding that if you're not focusing

time and attention on any one of those

um stool legs, if you will,

um it's really hard to see the success

and the value that you want to get.

So let me talk a little bit about

that because when I say clinical systems,

I'm referring to your current

technology infrastructure.

And this really, this really has to be the starting point

because if your technology isn't ready for you

on transformation,

the processes in the people, those legs of the

stool really, really, really don't matter right, you, you've

got to have technology that is ready

and it's ripe for, for receiving

the auto review and being able to drive

those efficient. So some questions

to ask yourself is has your E hr

been upgraded in the last 18

months or so with fire servers

and web service availability,

those kinds of things, we really need to make

sure you're on um a current

E hr um version

or if you are thinking about upgrading,

keeping that in your mindset as you,

as you head up on this journey, in terms

of the phases and stages of getting to automation,

also your um or cm workflow

tool. Are you on the latest version? Um

Is it compatible for integration

and will it allow for for this type of automation

capability?

So lots to lots to think about

there for sure.

And then

when it comes to processes,

it's really important to step back and understand

the impact that the automation is gonna have

on your processes and your workflow.

Um You know, really thinking about what

you want your future state program to

look like because now is the time

to be able to make some

changes and maybe tweak some processes

to help optimize how the capability

is gonna work.

And then the la the last piece uh last but

not least is the people, right? Um

When it comes to people, they're really gonna be

your, your key to their success. Um

We've learned that uh change is hard

and change comes hard to some folks, especially

where we've got, you know, an aging

uh nursing workforce and

we often don't spend enough time

or in invest enough energy in

preparing our

for this type of change.

So I think it's critically important

um to ensure that you've got

a really good go live readiness plan

that you're thinking about allocating

your resources in the right way and

that you're really clear with your goals

and your, your, your key performance indicators

or metrics of measure what metrics matter

to you

and how you're going to measure those to define

your success and bring your people

along on some of that journey with you.

Um You can have the most advanced technology

on the market out there, but if you don't

have the people that are willing to embrace it,

um And really, you know, stepping

up to be champions to, you know,

envision this future state can really

make or break your process.

It's also a really good time to

be thinking about staff competency and what

else you can do to um bolster people

and get support um around moving

towards automation and technology.

Um People get afraid that when you bring in a technology

or capable a capability, it's,

it's gonna diminish their value or

potentially reduce um you

know, what is needed from them. And, and that's not

the case. We, we're just not finding that to be

the case we, we often talk

with our customers or, or our customers

that are thinking about the auto review capability

in terms of the efficiencies and

the time savings, which can translate

to a lot when it comes to um FTEs

or, or resources.

I can tell you though, we've never encountered

a customer that's decided to cut their

resources because they've been able to

employ a capability like this. I

think we all can appreciate

that, you know, with staffing shortages

and challenges, we don't have enough people

to do the things that we wanna do and it's

very infrequent that we encounter

a hospital that's able to do all of the medical

necessity, reviews or the due diligence

when it comes to um other functions

for um or CM or discharge

planning or other programs

that you're looking to do because you don't

have the resources to do it. So we

definitely uh are not seeing the there's

a resource attrition, but we're

seeing that people are able to reallocate resources

in places or to support

programs where they can really have an impact

and make a difference. So super

super exciting to um to

think about the potential, right? If

you, if you think about taking some

of the routine rote basic

cases off the plate, really the the

impact can be transformational to free

up your resources.

So let me share a little bit. Um

This is a compilation of some

of the real world results that our customers

are achieving and seeing today with

the Interpol

a auto review capability,

uh we have an organization that's seen

um a 76% decrease

in medical necessity denials over

one year. Now, is that 100%

related to Interpol auto review and what

that can deliver? No, probably

not. But it's a combination

of utilizing the Interpol auto review capability

and them in

some clinical and technical workflow

improvements. So bringing our capability

in really gave them some support

and guidance for really digging

in and rethinking some of their processes

in ways that they could do things different,

do it better and do it with more accuracy.

And those results are translating to real

real world dollars for them.

Uh We've seen an organization decrease

their uh condition code 44

by 56%.

Um And that the automation is really

what allowed them for, for them to be able

to do uh more reviews earlier

on in the process. Uh And therefore

patients, patients were being placed in the

right level of care the first time

around or cases were being

escalated to a physician advisor

with partial information earlier on

in the process. So if you're getting it right

from the outset, you don't have to go back

and make changes after the fact. So

reducing condition code 40 fours

by driving to a more accurate level of

care from the outset.

Uh We have an organization that has

um had 6.2

FTEs and it translates to

somewhere around a million dollars in annual

savings. And again, not that they were

cutting staff, but they are able

to increase their staff productivity

with review volume, enough volume

that it's equating to somewhere in the order

of 6.2 FTEs um

from their old manual processes. So

6.2 FTEs that they

don't have to hire, uh, doing

more with less, uh, it's translating

to over 40,000 reviews more

a year that they're able to do. Um

saving them some again, real dollars

and being able to take their staff

and stretch them further. So, really

exciting results we're seeing with that customer.

Uh We have another customer that um

has had 78% of reviews

being automated. Um This customer

is really exciting because when we say

automated, we mean automated, either

fully completed or partially completed.

But they, they achieve this result

by rolling out auto review, recognizing

that the results that they were getting are hoping

for were not coming to fruition.

And when we were able to dig in with them and

help them understand what the root

cause was around that

they needed to change a single process

that they had with their physician documentation

and help them optimize or really

ignite what was being able

to be automated. So they started out

with 7% of reviews being automated.

And when they made this change to their process,

they went from 7% to 78%.

So really impactful changes

just by digging in and looking at the data saying,

why the heck aren't we getting more automated

reviews here? So, really um

super excited to see where that customer

has been able to take um auto review

and really help them on their transformation journey.

We have a customer that has had uh somewhere

between 30 40% decrease

in the time to complete a medical necessity

review. Um This is really what we're

seeing on average with our customers. We

can really cut the time to do a review down.

So if you think about those partial reviews

that I I mentioned where we're able to get

some of the data but not all of it, they

were able to complete that review and significantly

less time than, than what they were doing previously.

And then the last statistic I have on

here um is that we have

a customer that has seen a 13%

overall decrease in the time it takes

to conduct that first review. And

this is really exciting because if you come in

to start your work day as a un nurse

and auto review has been running

and you're looking at your work list or

your work cue and many of those

reviews are already fully completed for

you and others are partially completed

for you.

Well, you can just see how your work day can be much

easier and how you can get to more reviews,

quicker, reduce that lag time

and really be able to increase your productivity.

So super exciting results and

these numbers are just gonna continue to grow as

we evolve our capability and, and

learn from our customers. But that

robust reporting capability that we have

is really what's helped us to help

our customers dig in and

really be thoughtful about making meaningful

changes, whether it's to their systems,

whether it's to their processes or whether

it's with their people. Um It could

be any one of those. And again, that's the three legged

stool that we've seen to really help

ensure and drive success.

OK.

So

just in summary, uh what I

shared with you today around auto

review, really auto review

is designed and intended to help

you reduce, you know, administrative

task burden by driving

efficiency. Uh really reducing

the burden of that manual review process

by integrating directly into the workflow,

right? So you're not going outside of a medical

necessity review um to be

able to find this information just pulling right

into your workflow. So really

um having that integration in there is really

gonna eliminate the need to toggle in and

out of disparate systems and bring

efficiencies.

We also are driving to accuracy.

Um Interpol

on review is not a black box capability.

There's many black box capabilities on

the market with, you know, promises of

A I uh and, and promises

of, of having you uh eliminate

the need to be able to do a medical assess

the review overall. Um We don't

believe that that's the the route to go,

you know, at the end of the day, Interpol

and auto review is founded on

the gold standard evidence-based

Interpol criteria that many of you have been

using for a long time. It really

creates transparency and

leaves no guess work for health plans

as to the validity um of

the need for care at whatever

level is being requested. So

when we compare results

with of auto review, with manual case

audits, we know that there, the automation

really brings um accuracy

in the form of getting to the right subset,

right? We're looking at the codified

admission diagnosis, the driver,

the reason for being there and we're getting

to the right subset. We're driving

to the right level of care

and the right criterion um

being selected and supported with the source

of truth really to ensure accuracy.

And then lastly, this capability

really um is also designed to help

foster collaboration and trust.

We know that there is still uh

a lot of challenge between payers

and providers in the market. We write

Interpol and sell Interpol to both

sides. Uh We don't write it for

one side or the other. We really try to

toe the line and write the criteria based

on the evidence. And our, our goal

and ideal state is to help foster

collaboration by providing a capability

that both sides of the equation can use

and can trust. And I know

that when we're out talking to health plans,

they have a AAA real

interest and intent to automate

their off decisions based on data

that comes from the source of truth

that they can trust

that's not been manipulated by a human.

So I think, I think, you know, our future

and where we're headed from a strategic

trajectory is really exciting

to think about those cases that

we can just take off the plate. I mean,

no one's arguing if a stemi

needs to be in a hospital, inpatient

level of care, whether you're the hospital

or the health plan, we're sort of all

in agreement with that. I mean, these patients

are sick and they're complex. Um

And yet we're still doing medical

necessity reviews on the hospital side, on the health

plan side, there's a lot of back and forth

and, you know, looking for clinical information

to validate and support it. If we can take those

things out of the equation, we're really

gonna be able to make some significant

strides. And um really

the automating on the front end has such a tremendous

capability to help transform

how we operate in the future. And it's

really just the tip of the iceberg when it

comes to meaningful

transformative change. So

I hope you found this helpful. Um

I think we're gonna, we're gonna pause and

maybe take a couple of questions here, but I hope

you found this helpful Interpol um

is here to

help you along that journey and

for our customers that do choose

to license Interpol auto review. We

don't integrate this capability and then leave

you on your own.

We do provide consultative services

to ensure that you're optimizing

the use of it and that we're really helping

you be thoughtful about your, your

p your processes

and your systems and making

small adjustments to make sure that these kinds

of capabilities can really help you

and really drive me

efficiency. So um would love

to see you uh consider um Interpol auto

review.

So, Heather with that, I think we're gonna

um pause

and, and take a few questions.

Yes. Thank you so much, Laura.

Before we get started with the Q and a portion

of today's webcast, we'd like you to take

a moment to remind you of the survey located

beneath the slide deck area within your audience

console. We'd love it.