August 12, 2025 • 24 mins
This episode was recorded live at the 2025 Joy & Wholeness Summit in Asheville, North Carolina.
Dr. Xi (Sisi) Hu is a Co-Founder and the Chief Wellbeing Economist of Atalan Tech, a mission-oriented startup focused on using machine learning to predict and prevent clinician burnout and turnover. She is also a Fellow of the Center for Labor and a Just Economy at Harvard Law School and Research Economist at the National Bureau of Economic Research, with a research focus on labor issues. She specializes in modeling disruption and risks in labor markets, and is passionate about protecting the wellbeing of workers in the healthcare industry. Dr. Hu received a National Science Foundation grant to study COVID’s impact on healthcare workers. Her expertise is in risk science where her work has received media attention from around the world including the World Economic Forum, Sky News, the UN, among others.
Dr. Heather Schmidt is a family medicine physician in Fond du Lac, WI, who trained at Midwestern University and completed her residency at Exempla/St.Joseph’s Hospital. She became the Medical Director of Health and Wellness at Agnesian HealthCare in 2013 and transitioned to the System Medical Director of Healthy Work and Well-being for SSM Health in 2020. Dr. Schmidt has collaborated with various partners to integrate well-being into initiatives such as leadership development and employee safety and has been instrumental in implementing the SSM Health Care for Caregivers peer support system and Schwartz Rounds. She continues her medical practice with a focus on weight management and mental health and enjoys spending time with her husband and daughters, skiing and boating in Wisconsin.
Thanks for tuning in! Check out more episodes of The Well-Being Connector at www.bethejoy.org/podcast.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Introduction (00:00):
🎵 🎵
Voiceover (00:13):
Welcome to another episode of The Well-Being
Connector podcast hosted by RoyReid and sponsored by the
Coalition for Physician and APPWell-Being.
This episode is one of amulti-part series recorded live
at the 2025 Joy and WholenessSummit.
Thank you for listening.
Roy Reid (00:31):
Welcome to the Well-Being Connector podcast.
We are recording live from theJoy and Wholeness Summit here in
Asheville, North Carolina,sponsored by the Coalition for
Physician and APP Wellbeing.
I am joined now by Sisi Hu andHeather Schmidt.
They're getting ready to dotheir presentation here tomorrow
(00:53):
at the conference, and they'regoing to do a little bit of
conversation with me today tooutline the program they're
working on, the results they'regetting, and some of the things
that they see in the future.
So welcome.
Heather Schmidt (01:04):
Well, thank you for having us.
Roy Reid (01:06):
Absolutely. So we like to start with the journey that
brought you to wellness andwholeness and what motivated and
inspired you to do the worktoday.
So Heather, why don't you startout?
Heather Schmidt (01:17):
Okay.
Well, I'm a family medicinephysician.
And so I was actually trainedas a DO and that was
intentional.
I really am just very focusedon prevention and just that
holistic healing, kind of reallythinking about all of those
things that are innately with inus that can help us be as
healthy as we can be and thrive.
And so many years ago, Iactually was already starting to
(01:37):
get kind of frustrated withmedicine.
I wouldn't say burnt out, butfrustrated with how we did
things, you know, just alwaystaking care of the disease, but
not really thinking about thatwhole person.
And so shifted the way I did mymedical practice.
And that then grew into a rolewith my smaller healthcare
system at the time that wasfocusing on wellness with other
(01:59):
employer groups, our ownemployee and providers and then
also patients and schools andkind of just doing some fun
things in the community.
And that really grew intoreally thinking about what does
the work that we do look likeand how do we actually change
how we do our work to promotehealth versus just relying on
those kind of things that we doas individuals to keep ourselves
(02:22):
healthy, but recognizing thatimpact of the environment.
And we joined as part of abigger healthcare system.
So SSM Health And just hadreally gotten to this idea of
this changing people's mindsetabout wellness and opening
people's minds that it's not anindividual problem, that it's a
(02:42):
system problem.
This was 10, 12 years ago.
So it was the common practiceat that point was to rely on the
individual.
This was back in the day whenit was yoga and meditation,
right?
You just take care of yourselfand it's all up to you.
But so I feel like it's been along journey.
It's, you know, I think Istarted this maybe in 2010 or
(03:03):
so.
But just this whole idea of howdo we help people thrive and
help people stay so close totheir own meaning and purpose?
Because that's how I'vecontinued to do the work I'm
doing.
Well, how I stayed in medicine,really, right?
And so wellness is just one ofthose things to me that I think
is fundamental for ourhealthcare system.
(03:24):
And if we don't help our ownhealers, how are we going to
continue to take care of ourpatients and communities?
Wow.
Roy Reid (03:30):
Sisi Hu.
Sisi Hu (03:31):
Yes.
So I come from a slightlyunconventional background.
So I'm a trained economist.
When we first started thiswork, I was working as a
researcher at Harvard, lookingat different types of
disruptions.
So I actually think of myselfas a disruption scientist.
So I was looking at climatechange, how climate change
(03:54):
affects the labor market, theeconomy, and then I moved into
technological disruption, whichis when I started looking into
AI.
But anyway after as a sideproject we put in a proposal for
a National Science Foundationfunded grant looking at COVID
impact on healthcare workersthat's when we actually started
really getting into the burnoutand thinking through it from a
(04:15):
labour lens like what does thatmean for the wellbeing of our
clinicians and then we startedinterviewing health system
leaders like Heather and Heatherwas one of our first
respondents thank you very muchso I called him out Heather and
we started this work togetherand then I co-founded a company
with my co-founder Tiffany atLance.
(04:37):
We developed a workforceintelligence platform.
That's kind of how we gotstarted into this and we used
machine learning to predict thepeople who might be struggling
on their own silently, thinkingthrough how can we prove the ROI
of investing in our clinicians'well-being.
Awesome.
Well, and the funny thing wasthat she says she emailed me.
(04:59):
Somehow I caught the laborworkforce department and
thought, well, this is exactlywhat we need.
We need to look at thiseconomic impact.
We've got to figure this out.
And so we had this wonderfulconversation that just opened
this door for this partnershipthat really, I think, was divine
intervention because it reallycame at the time where as a
(05:20):
leader in a larger healthcaresystem across multiple states
how do you know how people trulyare doing and so this idea of
this dashboard and reallylooking at kind of different
types of data objective data incombination hopefully at some
point with subjective data thenwe truly know how people are
doing so we use data foreverything else for population
(05:40):
health so why are we not usingit for our own clinicians?
Roy Reid (05:45):
Unpack the program that you brought to share with
the group at the conferencetoday.
Tell me a little bit about whatthat looks like, how it's
working and what you're strivingto achieve with it.
Heather Schmidt (05:57):
Do you want me to take it?
Okay.
(06:21):
a little bit about, you know,well, what about interventions
we've done?
And how do we know that they'retruly effective?
And how do we invest in thoseones that are actually working?
And how do we scale them?
Because there are bestpractices all around our system.
(06:43):
So how do we identify those andbe able to get our other
regions being able to benefitfrom those same interventions?
And so that's really what we'representing tomorrow is one
particular or two particularprojects that came to us not
actually through the projectsthemselves weren't to improve
well-being.
They were to improve patientcare.
But the people that wererunning the project were very
(07:05):
interested in how did thatimpact well-being?
How did that impact retention?
And is it making a differenceor not?
Yeah, I think just kind of toadd, I think what excites me the
most is the potential of thismethodology that we developed
together to basically evaluateany intervention, any program
that you have that looks at, youknow, when it's a quality
(07:26):
project, safety project, or itcould be a well-being project.
But I think part of the storythat's missing right now is it's
really hard to quantify whatare the ROI, what is the return
of investment in your programs?
How do you actually prove thatyour program works?
So this methodology is designedto be actually retrospectively
create an RCP, like a randomizedcontrol trial through the data
(07:49):
infrastructure we set up.
So you can basically look atany program and then
retrospectively and on along-term basis say, hey, this
program is working on thisparticular way.
These are the metrics that weactually observe changes in and
they're statisticallysignificant and these are the
things that don't work.
We don't see any effects andthese are some of the unintended
consequences.
(08:09):
So I was really shocked, notshocked, but surprised, very
happy to see that the clinicalpharmacist program that SSM
Health had, had impact on notjust the quality metrics but
also the well-being.
So we saw actually turnoverrisk for that group of people
who received the intervention,we call it, reduced by
(08:32):
substantially amount, and thatequated to $1.6 million of cost
savings that you never thoughtabout quantifying before.
So being able to visibly showsome of this evidence and metric
changes, I think it's whatreally excites me.
Roy Reid (08:46):
So unpack the details of that work that you did and
describe for the audience theinterventions and then what that
did and then where it tookthem.
Heather Schmidt (08:55):
Sure, absolutely.
So our One of our ambulatorypharmacy leaders, her name is
Drea Meyer, we have a pharmacyresidency project.
And each resident has to do aproject.
A pharmacy residency, sorry.
And each pharmacy resident hasto do a project.
And so this particularresident, Brooke Watson is her
name, she came to me and justhappened to say, hey, do you
(09:17):
have any HR data on retentionfrom providers?
And I said, well, maybe wecould take this one step further
because she was looking at theeconomic impact of this
particular intervention.
And so we were able to, some ofour clinics have an embedded
clinical ambulatory pharmacist.
Some don't.
The project themselves actuallystarted because we have very
medically complicated patients.
(09:37):
And we know that a lot ofliterature will show that
pharmacists as part of thehealthcare team can help offload
some of that cognitive burden,can help manage those medical,
the medicines that go along withthese medical issues.
And so, but we don't alwayshave buy-in from some of the
clinicians, but we also don'tknow that we have funding to
make this possible across everysingle primary care site, right?
(10:00):
This is one of these, is thisgoing to make a difference?
And so she actually haddesigned her intervention with
subjective feedback, which Ithink is also very, very
important.
But we were trying to look atthe objective data variables.
So I took it to Cece and herteam and they said, well, yeah,
I think we can figure this out.
And so that's where it camefrom.
And so she could speak to thescience behind the predictive
(10:23):
analytics and machine learningpiece much more effectively than
I can.
But yeah, no, I think it wasreally cool to be actually able
to do that study because we gotthe data from Drian Burke and
then really found anothercontrol group.
Like, you know, how you woulddo a normal randomized control
(10:43):
trial, you would find anothergroup of people who are similar.
So we developed an algorithm tobasically match the treatment
group, the people who had theclinical pharmacists with
another control group.
And then the match was reallythe score it was so high.
I was like, oh, wow, you haveso many similar providers.
But anyway, and then we wereable to test the differences of
(11:05):
the people who actually gotthese help.
And we saw a lot ofimprovements around
documentation time, actuallyless pajama time as well, as
well as per encounter, fiveminutes increase that they were
able to spend more with theirpatients, which is what they're
coming to medicine for.
So being able to, I think Ijust, I was just so excited to
see the numbers and then to seethe graphs, which we'll present
(11:26):
tomorrow.
There are slightly confusing,but we'll make it work.
I know we've presented this,we've got really good feedback,
but I think we've got to a pointwhere it's easily explainable.
But super interesting resultsand hopefully we're also going
to look at DAX as well.
Yeah, so as you probably haveheard, the ambient documentation
(11:49):
is all rage, right?
So it's helping to use the AItechnology for good.
And personally, I got my DAXjust a couple weeks ago and And
it's absolutely amazing.
It's like a game changer.
But we've been trying to reallylook at how do we implement
this in the best way?
Because it's very expensive andit's a huge investment.
Everybody wants to do it, butit's, you know, we have to be
(12:12):
smart and good stewards offinancial resources as well.
And if people aren't going toreally use the tool, then why
would we invest in it?
And so we have been workingwith our IHT team and they have
been kind of providing us theearly adoption.
Like a list of those earlyadopters and then kind of adding
to that and then already seeingsome benefits.
(12:33):
And, you know, that it'shopefully going to help people
understand that this can be veryeffective and this is worth a
larger investment of both theprovider's time to learn how to
use the tool, but then alsofinancially, of course.
Roy Reid (12:51):
And describe a little more detail of the intervention
itself.
What happens and what'sAbsolutely.
Heather Schmidt (13:00):
So DAX is actually a tool.
And so people, if they get thelicense, so usually the way
we've been trying to identifythose are, we had our early
adopters, but then also ourleaders are recognizing who
really could use some help.
And then going to them, askingthem if they're interested in
using this tool, because someare not, you know, some are very
(13:21):
hesitant to use the tool.
And then being able to, oncethose licenses are secured, then
we actually do that.
a training and there's like asmall group, like a cohort of
people that are kind of meetingand learning from each other.
And then they get some time touse the tool and then they can
work with our IH team to dofurther training and kind of
more customization of the tool.
(13:41):
But I think what we've learnedis that it's not plug and play.
You know, it is going to be,you need to invest that time up
front and you also need to makesure that you are engaging
accepting of maybe your notelooks a little different than
what it would if you were todictate it or type it yourself.
So that's kind of theintervention in and of itself.
(14:01):
We're just on the back endwatching all of these kind of
objective data variables thathave been identified to be
drivers of workplace well-beingand burnout and turnover.
And so some things we're seeingare maybe expected and other
things we're like, oh, we didn'treally expect to see that as an
outcome, right?
And so it's this idea ofreal-time And being able to
(14:23):
course correct as needed is whatmy goal is, is to be able to
say, okay, we're doing all theseinterventions.
What's working, what's not?
What's working if we tweaked ithere, right?
But it's almost this objectivevoice of those participants as
we're also then trying to getthat subjective feedback from
them along the way.
Roy Reid (14:43):
And what are they doing now differently with the
tool?
What are the actions that arehappening on the unit within the
team that then creates themeasurement?
So
Heather Schmidt (14:54):
I'll speak to Dax mostly because, and I can
only really speak to my ownpractice, but when I go in with
a patient, I basically say, hey,I have this new documentation
assistance.
Are you okay with me using it?
Of course, you always have togive verbal consent.
They've all been veryreceptive.
I hit the button and I sit andtalk to the patient.
(15:14):
And so I can be present and Ican really truly hear, right?
I can see that patient.
I can like really make themfeel valued and that their time
matters as much as, you know,like then I'm spending that
quality time with them.
And then they, thedocumentation tool will tee up
my note for me.
And then I can correct it.
Sometimes if I have a littleextra time, I'll actually show
(15:35):
it to the patient.
We correct it together.
And then we, they feel reallycomfortable with the plan.
And then boom, I hit sign, noteis done.
No longer am I going back to myoffice and spending hours doing
my notes or after work andclosing my notes.
And so that's ultimately thegoal.
And this ambient documentationhas been kind of all the rage
(15:56):
and well-being for a while.
Yeah.
It's people, we're notnecessarily seeing like that
change in some of the EHRmetrics.
So there's more to it thanthat, right?
So it's really changingculture, I think, too.
Some of us are used to goinghome and logging back in and
we're used to doing our inboxafter hours.
And so we might have to changea little bit, you know, just
(16:17):
coaching along the way.
Sisi Hu (16:19):
Yeah.
Roy Reid (16:19):
So layman's terms, you're giving people their life
back.
Absolutely.
By automating the process andcreating the discipline around
that then gives that freedom towork on the other things that
are important to them.
Yes, absolutely.
And so Cece, from a numberstandpoint, from an outcome
standpoint, tell us about theresults your team takes and then
(16:43):
looks at for what the outcomesare going to be.
Sisi Hu (16:45):
For DAX or for the whole things that we've looked
at,
Roy Reid (16:50):
Start with whatever you think is most important.
Sisi Hu (16:53):
Okay.
So I think for just AI, likedocs in general, I think what we
found that was reallyinteresting is that the effect
really takes place fordifferently, for different
specialties.
And it depends on thedemographic.
So we saw younger docs are muchmore, actually.
most of what I'm looking for,willing to adopt new technology.
(17:16):
I think there is a piece aroundunderstanding, okay, out of
like a population of 4,000providers, who should you get
the technology first?
So there's a lot of work thatHeather and I were sort of
trying to figure out, well,should we be targeting maybe
like providers of a specificgroup that are really
struggling, for example, withthat augmentation time?
(17:37):
So going back to some of thedata modeling work that we do,
so we basically, for everybody,we make a prediction using
machine learning, right?
We take data that healthsystems already have, the
electronic and health recordsand HR data, but essentially
we're able to say, okay, all ofthese providers are struggling
in this particular issue, andthen they're really not doing
(17:57):
very well.
And those are the people, so wedeveloped actually a
prioritization card for rollingout some of these technological
advances, or even I think forother interventions, it's really
important to think about whowould benefit the most from a
particular intervention.
But being able to say, okay,this is the, you know, out of
4,000, the top 10, 100 who arereally struggling and they're
(18:17):
struggling specifically fordocumentation.
That's the subgroup of peopleyou should, you know, roll out
first and then help them, youknow, and then sort of like do a
hierarchical approach, right?
And I think that what I'mreally excited about with the
work that I'm doing with Heatheris we can sort of almost
customize this approach to othertypes of interventions because
we track 100 plus risk factors.
(18:39):
So if you're developing anintervention that targets not
necessarily documentationburden, but something else, we
can have a measure that reallysort of roll out that way.
So you can help really toidentify the people who will be
struggling the most.
Roy Reid (18:52):
And from all of that, what are the downstream benefits
that occur in their lives?
What are the changes thatyou're seeing happen?
And what's the transformationthat occurs with that team?
Heather Schmidt (19:04):
Yeah, so what we're hearing back from
subjective feedback fromtestimonials, right, people
using this tool is this isactually going to prolong If I
didn't have this, I might haveretired, right?
Things like this is a gamechanger.
This is giving me time back athome.
So we know that subjectivefeedback has been very positive.
(19:25):
When we want to be data-drivenand really understand, is this
going to be a bigger system-widechange we should make, then
let's actually continue to getthe objective feedback too,
right?
So that's kind of our goal isto really have that
comprehensive evaluation.
Yeah.
Okay.
(20:13):
we do and so with that soleintention of giving time back to
(20:47):
our clinicians and so our chiefdigital health officer Saad
Chaudhry is his name this is hisgoal is to give time back and
so we are working with that IHTstrategy team to look at
different interventions and whatis the timeline of that
implementation who would be thataudience that will receive this
intervention and how do weproactively track and so that's
(21:07):
kind of next phase is wherewe're trying to go
Roy Reid (21:11):
Outstanding.
So as we reach the end of ourtime together, what do you
envision being some of the otheroutcomes that you're hoping for
in terms of the overall impacton wellness, on wholeness with
people?
Heather Schmidt (21:27):
Are we just talking about...
You go first?
(21:57):
...believe nobody is trying tocause harm.
But everybody has a job to do,right?
So how do we actually know thatsometimes there are happy
accidents, but sometimes thereare unintended consequences and
these downstream effects of adecision that we make that we
don't right now have a great wayto measure that more
proactively.
(22:17):
And so in a perfect world,that's how I would see us using
this.
We're working with Adelante todevelop a nursing dashboard as
well.
And in our process of doingthis, we're working with our
bedside nurses.
We're working with our bedsidenursing leaders to really design
a tool.
They have been incrediblyamazing to work with, to design
a tool that's going to be mostmeaningful for how they do their
(22:38):
work.
And so how do we continue tostay as kind of keeping the
pulse on in a way that's goingto be scalable across the whole
system?
Roy Reid (22:48):
Sure.
Sisi anything?
Sisi Hu (22:50):
Oh, I want to talk about what we talked about
earlier, like that was your ideaaround connections, but being
able to measure connections in aquantified way through data.
We were thinking about comingup with some models to think
about, okay, well, can wemeasure maybe the ways that our
providers are connected withnurses, inbox messages, for
example, but being able to say,hey, if you affect, you know,
(23:14):
that contagion effect, but beingable to actually quantify it
and say, oh, well, if Dr.
Heather Schmidt here is notdoing well, who else in her
arena is going to be affected?
That would be so cool.
I think it's doable.
We have some literature thathas network science and has done
some work around that.
That's probably one of ourinnovation projects for next
(23:35):
year.
Stay tuned.
Other than that, I thinkquality and all the other stuff
like nursing for otherpopulations, residents, would be
really cool to look into too.
Roy Reid (23:46):
I look forward to our next podcast where you talk
about those things and how theimpact is having.
So thank you first for the workthat you're doing.
Thank you for sharing it withour audience today.
I know they're going to benefitfrom everything it is and look
forward to hearing how tomorrowgoes with your presentation.
Heather Schmidt (24:03):
Thank you so much.
Roy Reid (24:05):
Absolutely.
Voiceover (24:09):
Thank you for tuning in to the Well-Being Connector
podcast, brought to you by theCoalition for Physician and APP
Well-Being.
The Well-Being Connector offersinsightful conversations with
healthcare professionals devotedto fostering wholeness within
their organizations.
Each episode delves deep intothe holistic approach to
well-being, underscoring theimportance of physical, mental,
(24:30):
social, and spiritual health.
For more episodes, visit ourwebsite at www.bethejoy.org
/podcast.
🎵 🎵
Voiceover (00:13):
Welcome to another episode of The Well-Being
Connector podcast hosted by RoyReid and sponsored by the
Coalition for Physician and APPWell-Being.
This episode is one of amulti-part series recorded live
at the 2025 Joy and WholenessSummit.
Thank you for listening.
Roy Reid (00:31):
Welcome to the Well-Being Connector podcast.
We are recording live from theJoy and Wholeness Summit here in
Asheville, North Carolina,sponsored by the Coalition for
Physician and APP Wellbeing.
I am joined now by Sisi Hu andHeather Schmidt.
They're getting ready to dotheir presentation here tomorrow
(00:53):
at the conference, and they'regoing to do a little bit of
conversation with me today tooutline the program they're
working on, the results they'regetting, and some of the things
that they see in the future.
So welcome.
Heather Schmidt (01:04):
Well, thank you for having us.
Roy Reid (01:06):
Absolutely. So we like to start with the journey that
brought you to wellness andwholeness and what motivated and
inspired you to do the worktoday.
So Heather, why don't you startout?
Heather Schmidt (01:17):
Okay.
Well, I'm a family medicinephysician.
And so I was actually trainedas a DO and that was
intentional.
I really am just very focusedon prevention and just that
holistic healing, kind of reallythinking about all of those
things that are innately with inus that can help us be as
healthy as we can be and thrive.
And so many years ago, Iactually was already starting to
(01:37):
get kind of frustrated withmedicine.
I wouldn't say burnt out, butfrustrated with how we did
things, you know, just alwaystaking care of the disease, but
not really thinking about thatwhole person.
And so shifted the way I did mymedical practice.
And that then grew into a rolewith my smaller healthcare
system at the time that wasfocusing on wellness with other
(01:59):
employer groups, our ownemployee and providers and then
also patients and schools andkind of just doing some fun
things in the community.
And that really grew intoreally thinking about what does
the work that we do look likeand how do we actually change
how we do our work to promotehealth versus just relying on
those kind of things that we doas individuals to keep ourselves
(02:22):
healthy, but recognizing thatimpact of the environment.
And we joined as part of abigger healthcare system.
So SSM Health And just hadreally gotten to this idea of
this changing people's mindsetabout wellness and opening
people's minds that it's not anindividual problem, that it's a
(02:42):
system problem.
This was 10, 12 years ago.
So it was the common practiceat that point was to rely on the
individual.
This was back in the day whenit was yoga and meditation,
right?
You just take care of yourselfand it's all up to you.
But so I feel like it's been along journey.
It's, you know, I think Istarted this maybe in 2010 or
(03:03):
so.
But just this whole idea of howdo we help people thrive and
help people stay so close totheir own meaning and purpose?
Because that's how I'vecontinued to do the work I'm
doing.
Well, how I stayed in medicine,really, right?
And so wellness is just one ofthose things to me that I think
is fundamental for ourhealthcare system.
(03:24):
And if we don't help our ownhealers, how are we going to
continue to take care of ourpatients and communities?
Wow.
Roy Reid (03:30):
Sisi Hu.
Sisi Hu (03:31):
Yes.
So I come from a slightlyunconventional background.
So I'm a trained economist.
When we first started thiswork, I was working as a
researcher at Harvard, lookingat different types of
disruptions.
So I actually think of myselfas a disruption scientist.
So I was looking at climatechange, how climate change
(03:54):
affects the labor market, theeconomy, and then I moved into
technological disruption, whichis when I started looking into
AI.
But anyway after as a sideproject we put in a proposal for
a National Science Foundationfunded grant looking at COVID
impact on healthcare workersthat's when we actually started
really getting into the burnoutand thinking through it from a
(04:15):
labour lens like what does thatmean for the wellbeing of our
clinicians and then we startedinterviewing health system
leaders like Heather and Heatherwas one of our first
respondents thank you very muchso I called him out Heather and
we started this work togetherand then I co-founded a company
with my co-founder Tiffany atLance.
(04:37):
We developed a workforceintelligence platform.
That's kind of how we gotstarted into this and we used
machine learning to predict thepeople who might be struggling
on their own silently, thinkingthrough how can we prove the ROI
of investing in our clinicians'well-being.
Awesome.
Well, and the funny thing wasthat she says she emailed me.
(04:59):
Somehow I caught the laborworkforce department and
thought, well, this is exactlywhat we need.
We need to look at thiseconomic impact.
We've got to figure this out.
And so we had this wonderfulconversation that just opened
this door for this partnershipthat really, I think, was divine
intervention because it reallycame at the time where as a
(05:20):
leader in a larger healthcaresystem across multiple states
how do you know how people trulyare doing and so this idea of
this dashboard and reallylooking at kind of different
types of data objective data incombination hopefully at some
point with subjective data thenwe truly know how people are
doing so we use data foreverything else for population
(05:40):
health so why are we not usingit for our own clinicians?
Roy Reid (05:45):
Unpack the program that you brought to share with
the group at the conferencetoday.
Tell me a little bit about whatthat looks like, how it's
working and what you're strivingto achieve with it.
Heather Schmidt (05:57):
Do you want me to take it?
Okay.
(06:21):
a little bit about, you know,well, what about interventions
we've done?
And how do we know that they'retruly effective?
And how do we invest in thoseones that are actually working?
And how do we scale them?
Because there are bestpractices all around our system.
(06:43):
So how do we identify those andbe able to get our other
regions being able to benefitfrom those same interventions?
And so that's really what we'representing tomorrow is one
particular or two particularprojects that came to us not
actually through the projectsthemselves weren't to improve
well-being.
They were to improve patientcare.
But the people that wererunning the project were very
(07:05):
interested in how did thatimpact well-being?
How did that impact retention?
And is it making a differenceor not?
Yeah, I think just kind of toadd, I think what excites me the
most is the potential of thismethodology that we developed
together to basically evaluateany intervention, any program
that you have that looks at, youknow, when it's a quality
(07:26):
project, safety project, or itcould be a well-being project.
But I think part of the storythat's missing right now is it's
really hard to quantify whatare the ROI, what is the return
of investment in your programs?
How do you actually prove thatyour program works?
So this methodology is designedto be actually retrospectively
create an RCP, like a randomizedcontrol trial through the data
(07:49):
infrastructure we set up.
So you can basically look atany program and then
retrospectively and on along-term basis say, hey, this
program is working on thisparticular way.
These are the metrics that weactually observe changes in and
they're statisticallysignificant and these are the
things that don't work.
We don't see any effects andthese are some of the unintended
consequences.
(08:09):
So I was really shocked, notshocked, but surprised, very
happy to see that the clinicalpharmacist program that SSM
Health had, had impact on notjust the quality metrics but
also the well-being.
So we saw actually turnoverrisk for that group of people
who received the intervention,we call it, reduced by
(08:32):
substantially amount, and thatequated to $1.6 million of cost
savings that you never thoughtabout quantifying before.
So being able to visibly showsome of this evidence and metric
changes, I think it's whatreally excites me.
Roy Reid (08:46):
So unpack the details of that work that you did and
describe for the audience theinterventions and then what that
did and then where it tookthem.
Heather Schmidt (08:55):
Sure, absolutely.
So our One of our ambulatorypharmacy leaders, her name is
Drea Meyer, we have a pharmacyresidency project.
And each resident has to do aproject.
A pharmacy residency, sorry.
And each pharmacy resident hasto do a project.
And so this particularresident, Brooke Watson is her
name, she came to me and justhappened to say, hey, do you
(09:17):
have any HR data on retentionfrom providers?
And I said, well, maybe wecould take this one step further
because she was looking at theeconomic impact of this
particular intervention.
And so we were able to, some ofour clinics have an embedded
clinical ambulatory pharmacist.
Some don't.
The project themselves actuallystarted because we have very
medically complicated patients.
(09:37):
And we know that a lot ofliterature will show that
pharmacists as part of thehealthcare team can help offload
some of that cognitive burden,can help manage those medical,
the medicines that go along withthese medical issues.
And so, but we don't alwayshave buy-in from some of the
clinicians, but we also don'tknow that we have funding to
make this possible across everysingle primary care site, right?
(10:00):
This is one of these, is thisgoing to make a difference?
And so she actually haddesigned her intervention with
subjective feedback, which Ithink is also very, very
important.
But we were trying to look atthe objective data variables.
So I took it to Cece and herteam and they said, well, yeah,
I think we can figure this out.
And so that's where it camefrom.
And so she could speak to thescience behind the predictive
(10:23):
analytics and machine learningpiece much more effectively than
I can.
But yeah, no, I think it wasreally cool to be actually able
to do that study because we gotthe data from Drian Burke and
then really found anothercontrol group.
Like, you know, how you woulddo a normal randomized control
(10:43):
trial, you would find anothergroup of people who are similar.
So we developed an algorithm tobasically match the treatment
group, the people who had theclinical pharmacists with
another control group.
And then the match was reallythe score it was so high.
I was like, oh, wow, you haveso many similar providers.
But anyway, and then we wereable to test the differences of
(11:05):
the people who actually gotthese help.
And we saw a lot ofimprovements around
documentation time, actuallyless pajama time as well, as
well as per encounter, fiveminutes increase that they were
able to spend more with theirpatients, which is what they're
coming to medicine for.
So being able to, I think Ijust, I was just so excited to
see the numbers and then to seethe graphs, which we'll present
(11:26):
tomorrow.
There are slightly confusing,but we'll make it work.
I know we've presented this,we've got really good feedback,
but I think we've got to a pointwhere it's easily explainable.
But super interesting resultsand hopefully we're also going
to look at DAX as well.
Yeah, so as you probably haveheard, the ambient documentation
(11:49):
is all rage, right?
So it's helping to use the AItechnology for good.
And personally, I got my DAXjust a couple weeks ago and And
it's absolutely amazing.
It's like a game changer.
But we've been trying to reallylook at how do we implement
this in the best way?
Because it's very expensive andit's a huge investment.
Everybody wants to do it, butit's, you know, we have to be
(12:12):
smart and good stewards offinancial resources as well.
And if people aren't going toreally use the tool, then why
would we invest in it?
And so we have been workingwith our IHT team and they have
been kind of providing us theearly adoption.
Like a list of those earlyadopters and then kind of adding
to that and then already seeingsome benefits.
(12:33):
And, you know, that it'shopefully going to help people
understand that this can be veryeffective and this is worth a
larger investment of both theprovider's time to learn how to
use the tool, but then alsofinancially, of course.
Roy Reid (12:51):
And describe a little more detail of the intervention
itself.
What happens and what'sAbsolutely.
Heather Schmidt (13:00):
So DAX is actually a tool.
And so people, if they get thelicense, so usually the way
we've been trying to identifythose are, we had our early
adopters, but then also ourleaders are recognizing who
really could use some help.
And then going to them, askingthem if they're interested in
using this tool, because someare not, you know, some are very
(13:21):
hesitant to use the tool.
And then being able to, oncethose licenses are secured, then
we actually do that.
a training and there's like asmall group, like a cohort of
people that are kind of meetingand learning from each other.
And then they get some time touse the tool and then they can
work with our IH team to dofurther training and kind of
more customization of the tool.
(13:41):
But I think what we've learnedis that it's not plug and play.
You know, it is going to be,you need to invest that time up
front and you also need to makesure that you are engaging
accepting of maybe your notelooks a little different than
what it would if you were todictate it or type it yourself.
So that's kind of theintervention in and of itself.
(14:01):
We're just on the back endwatching all of these kind of
objective data variables thathave been identified to be
drivers of workplace well-beingand burnout and turnover.
And so some things we're seeingare maybe expected and other
things we're like, oh, we didn'treally expect to see that as an
outcome, right?
And so it's this idea ofreal-time And being able to
(14:23):
course correct as needed is whatmy goal is, is to be able to
say, okay, we're doing all theseinterventions.
What's working, what's not?
What's working if we tweaked ithere, right?
But it's almost this objectivevoice of those participants as
we're also then trying to getthat subjective feedback from
them along the way.
Roy Reid (14:43):
And what are they doing now differently with the
tool?
What are the actions that arehappening on the unit within the
team that then creates themeasurement?
So
Heather Schmidt (14:54):
I'll speak to Dax mostly because, and I can
only really speak to my ownpractice, but when I go in with
a patient, I basically say, hey,I have this new documentation
assistance.
Are you okay with me using it?
Of course, you always have togive verbal consent.
They've all been veryreceptive.
I hit the button and I sit andtalk to the patient.
(15:14):
And so I can be present and Ican really truly hear, right?
I can see that patient.
I can like really make themfeel valued and that their time
matters as much as, you know,like then I'm spending that
quality time with them.
And then they, thedocumentation tool will tee up
my note for me.
And then I can correct it.
Sometimes if I have a littleextra time, I'll actually show
(15:35):
it to the patient.
We correct it together.
And then we, they feel reallycomfortable with the plan.
And then boom, I hit sign, noteis done.
No longer am I going back to myoffice and spending hours doing
my notes or after work andclosing my notes.
And so that's ultimately thegoal.
And this ambient documentationhas been kind of all the rage
(15:56):
and well-being for a while.
Yeah.
It's people, we're notnecessarily seeing like that
change in some of the EHRmetrics.
So there's more to it thanthat, right?
So it's really changingculture, I think, too.
Some of us are used to goinghome and logging back in and
we're used to doing our inboxafter hours.
And so we might have to changea little bit, you know, just
(16:17):
coaching along the way.
Sisi Hu (16:19):
Yeah.
Roy Reid (16:19):
So layman's terms, you're giving people their life
back.
Absolutely.
By automating the process andcreating the discipline around
that then gives that freedom towork on the other things that
are important to them.
Yes, absolutely.
And so Cece, from a numberstandpoint, from an outcome
standpoint, tell us about theresults your team takes and then
(16:43):
looks at for what the outcomesare going to be.
Sisi Hu (16:45):
For DAX or for the whole things that we've looked
at,
Roy Reid (16:50):
Start with whatever you think is most important.
Sisi Hu (16:53):
Okay.
So I think for just AI, likedocs in general, I think what we
found that was reallyinteresting is that the effect
really takes place fordifferently, for different
specialties.
And it depends on thedemographic.
So we saw younger docs are muchmore, actually.
most of what I'm looking for,willing to adopt new technology.
(17:16):
I think there is a piece aroundunderstanding, okay, out of
like a population of 4,000providers, who should you get
the technology first?
So there's a lot of work thatHeather and I were sort of
trying to figure out, well,should we be targeting maybe
like providers of a specificgroup that are really
struggling, for example, withthat augmentation time?
(17:37):
So going back to some of thedata modeling work that we do,
so we basically, for everybody,we make a prediction using
machine learning, right?
We take data that healthsystems already have, the
electronic and health recordsand HR data, but essentially
we're able to say, okay, all ofthese providers are struggling
in this particular issue, andthen they're really not doing
(17:57):
very well.
And those are the people, so wedeveloped actually a
prioritization card for rollingout some of these technological
advances, or even I think forother interventions, it's really
important to think about whowould benefit the most from a
particular intervention.
But being able to say, okay,this is the, you know, out of
4,000, the top 10, 100 who arereally struggling and they're
(18:17):
struggling specifically fordocumentation.
That's the subgroup of peopleyou should, you know, roll out
first and then help them, youknow, and then sort of like do a
hierarchical approach, right?
And I think that what I'mreally excited about with the
work that I'm doing with Heatheris we can sort of almost
customize this approach to othertypes of interventions because
we track 100 plus risk factors.
(18:39):
So if you're developing anintervention that targets not
necessarily documentationburden, but something else, we
can have a measure that reallysort of roll out that way.
So you can help really toidentify the people who will be
struggling the most.
Roy Reid (18:52):
And from all of that, what are the downstream benefits
that occur in their lives?
What are the changes thatyou're seeing happen?
And what's the transformationthat occurs with that team?
Heather Schmidt (19:04):
Yeah, so what we're hearing back from
subjective feedback fromtestimonials, right, people
using this tool is this isactually going to prolong If I
didn't have this, I might haveretired, right?
Things like this is a gamechanger.
This is giving me time back athome.
So we know that subjectivefeedback has been very positive.
(19:25):
When we want to be data-drivenand really understand, is this
going to be a bigger system-widechange we should make, then
let's actually continue to getthe objective feedback too,
right?
So that's kind of our goal isto really have that
comprehensive evaluation.
Yeah.
Okay.
(20:13):
we do and so with that soleintention of giving time back to
(20:47):
our clinicians and so our chiefdigital health officer Saad
Chaudhry is his name this is hisgoal is to give time back and
so we are working with that IHTstrategy team to look at
different interventions and whatis the timeline of that
implementation who would be thataudience that will receive this
intervention and how do weproactively track and so that's
(21:07):
kind of next phase is wherewe're trying to go
Roy Reid (21:11):
Outstanding.
So as we reach the end of ourtime together, what do you
envision being some of the otheroutcomes that you're hoping for
in terms of the overall impacton wellness, on wholeness with
people?
Heather Schmidt (21:27):
Are we just talking about...
You go first?
(21:57):
...believe nobody is trying tocause harm.
But everybody has a job to do,right?
So how do we actually know thatsometimes there are happy
accidents, but sometimes thereare unintended consequences and
these downstream effects of adecision that we make that we
don't right now have a great wayto measure that more
proactively.
(22:17):
And so in a perfect world,that's how I would see us using
this.
We're working with Adelante todevelop a nursing dashboard as
well.
And in our process of doingthis, we're working with our
bedside nurses.
We're working with our bedsidenursing leaders to really design
a tool.
They have been incrediblyamazing to work with, to design
a tool that's going to be mostmeaningful for how they do their
(22:38):
work.
And so how do we continue tostay as kind of keeping the
pulse on in a way that's goingto be scalable across the whole
system?
Roy Reid (22:48):
Sure.
Sisi anything?
Sisi Hu (22:50):
Oh, I want to talk about what we talked about
earlier, like that was your ideaaround connections, but being
able to measure connections in aquantified way through data.
We were thinking about comingup with some models to think
about, okay, well, can wemeasure maybe the ways that our
providers are connected withnurses, inbox messages, for
example, but being able to say,hey, if you affect, you know,
(23:14):
that contagion effect, but beingable to actually quantify it
and say, oh, well, if Dr.
Heather Schmidt here is notdoing well, who else in her
arena is going to be affected?
That would be so cool.
I think it's doable.
We have some literature thathas network science and has done
some work around that.
That's probably one of ourinnovation projects for next
(23:35):
year.
Stay tuned.
Other than that, I thinkquality and all the other stuff
like nursing for otherpopulations, residents, would be
really cool to look into too.
Roy Reid (23:46):
I look forward to our next podcast where you talk
about those things and how theimpact is having.
So thank you first for the workthat you're doing.
Thank you for sharing it withour audience today.
I know they're going to benefitfrom everything it is and look
forward to hearing how tomorrowgoes with your presentation.
Heather Schmidt (24:03):
Thank you so much.
Roy Reid (24:05):
Absolutely.
Voiceover (24:09):
Thank you for tuning in to the Well-Being Connector
podcast, brought to you by theCoalition for Physician and APP
Well-Being.
The Well-Being Connector offersinsightful conversations with
healthcare professionals devotedto fostering wholeness within
their organizations.
Each episode delves deep intothe holistic approach to
well-being, underscoring theimportance of physical, mental,
(24:30):
social, and spiritual health.
For more episodes, visit ourwebsite at www.bethejoy.org
/podcast.
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
New Heights with Jason & Travis Kelce
Football’s funniest family duo — Jason Kelce of the Philadelphia Eagles and Travis Kelce of the Kansas City Chiefs — team up to provide next-level access to life in the league as it unfolds. The two brothers and Super Bowl champions drop weekly insights about the weekly slate of games and share their INSIDE perspectives on trending NFL news and sports headlines. They also endlessly rag on each other as brothers do, chat the latest in pop culture and welcome some very popular and well-known friends to chat with them. Check out new episodes every Wednesday. Follow New Heights on the Wondery App, YouTube or wherever you get your podcasts. You can listen to new episodes early and ad-free, and get exclusive content on Wondery+. Join Wondery+ in the Wondery App, Apple Podcasts or Spotify. And join our new membership for a unique fan experience by going to the New Heights YouTube channel now!
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.