Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Rob Lott (00:00):
Hello, and welcome to
A Health Podicy. I'm your host,
(00:04):
Rob Lott. It's hard out therefor a rural hospital in America.
They serve a patient populationthat experiences
disproportionately high burdensof chronic disease and
mortality. And yet despiteplaying such a vital role in the
(00:26):
health of their communities,rural hospitals continue to
struggle.
It's hard to hire clinicians andother staff. Operational costs
are high and patient volumes arelow. And when it comes to the
most needed kinds of care,chronic disease management,
maternity care, geriatric care,the margins are incredibly
(00:48):
tight, if there's any margin atall. And so rural hospitals
continue to close. Too often,the finances and the math
underpinning them, they justdon't work out in favor of
continued operation.
And so what can we do about it?That's the subject of today's
health policy. I'm here withDoctor. Paula Chatterjee, an
(01:11):
assistant professor of medicineat the Perelman School of
Medicine and director of healthequity research at the Leonard
Davis Institute of HealthEconomics. She and her coauthors
have a paper in the July issueof Health Affairs titled, quote,
mixed evidence that ruralhospitals' finances improved
with participation in thePennsylvania Rural Health Model.
(01:36):
I can't wait to hear all aboutits findings and to learn all
about the work's potentiallessons for rural communities
nationwide. Doctor. PaulaChatterjee, welcome to the
podcast.
Paula Chatterjee (01:48):
Thank you so
much for having me.
Rob Lott (01:51):
So let's just jump
right in here, maybe with some
context. Can you tell us alittle bit about sort of the
state of rural healthcare inPennsylvania before this program
started? I know I kind ofprovided a really broad overview
of rural hospital financesgenerally. How's the experience
(02:11):
in Pennsylvania compared to thatbigger picture?
Paula Chatterjee (02:14):
Sure. So let's
jump in our time machine back to
the Commonwealth ofPennsylvania. In, let's say, the
early twenty tens. That's reallysort of the eve of the global
budget demonstration in thestate. And so at that time, I
think what we can say is thatabout a third of Pennsylvania's
population lives in rural areas.
(02:35):
That amounts to around three anda half million people who live
across around 8,000,000 acres.These are calculations that come
from the Pennsylvania Office ofRural Health. So it's a big
space for this three and a halfmillion people. In terms of
their demographics, in a lot ofways, mirrored what else was
going on in rural America at thetime. So about 20% of the rural
(02:58):
population in Pennsylvania was18, about 30% were baby boomers,
another 20% were in the 65group.
So we sort of see the typicaldemographics that we'd expect.
This is a population that isdisproportionately older. One of
the things that, you know, Ithink is always salient to me is
(03:18):
that since around 1970, ruralPennsylvania's per capita income
gap, when you compare it tourban areas, has doubled. And so
sort of, to me, that reallyserves as testament to some of
the entrenchment of poverty inthese areas and how long that's
been an issue. I'll say half ofrural Pennsylvanians around that
(03:39):
time were employed in themanufacturing industry,
wholesale, retail, health careand social services were really
emerging as sort of growingparts of economy there.
And so, that's really sort of asnapshot of the people. If you
want to hone in a little bit onthe healthcare infrastructure,
what I'll say is that some ofthe major health related issues,
you know, on the eve of theglobal budget demonstration were
(04:01):
really similar to what they aretoday. They included, you know,
poor access to mental healthservices. They include alcohol
related automobile deaths. Theyinclude an unfortunately high
prevalence of abuse and neglectof older adults.
And again, those are notnecessarily unique features
across rural populations ingeneral, but but are important
(04:21):
in the Pennsylvania context. Ifwe hone in on health care
delivery, I'll say that aroundthis time, there were estimates,
that rural Pennsylvania had onedoctor for every about 600 rural
residents. And they weresuffering from some of the same
challenges to retaining a viablehealth care workforce and
(04:42):
infrastructure that continue toplague us all today. You know?
It's it was the cost oftechnology, high wages, the cost
of obtaining highly educatedworkers.
All of that was related to sortof this workforce challenge. And
hospitals at that time, theyare, you know, today in a lot of
ways, were particularly relianton public payers. Right? They
were reliant on Medicare,reliant on medical assistance,
(05:03):
which is our Medicaid program.And that, combined with the high
costs of an older adultpopulation, was enough to put a
lot of financial strain onproviders.
And so if you want to look atsort of hard numbers on hospital
closures in rural Pennsylvaniain the pre global budget model
time, there are about threesalient closures or conversions
(05:23):
that happened. There was one in02/2006, another one in 02/2012,
and then a a conversion in 2014.Those are data from the
wonderful Schepp Center that Iknow so many of us are grateful
to for for collecting thesedata. So I I hope that gives at
least a little bit of a snapshotof what we're talking about
here.
Rob Lott (05:38):
That's great. And one
point of clarification when you
describe a conversion, what'shappening there?
Paula Chatterjee (05:43):
A conversion
can be defined kind of widely.
So sometimes conversions meanthat the hospital is shutting
down inpatient operations andreally sort of operating as a
freestanding ED. Sometimesthey're preserving other service
lines. It usually implies somedowngrade of service capability,
but it can differ sort ofdepending on the place or the
time.
Rob Lott (06:03):
Okay. Great. So
against this backdrop, you
studied, the Pennsylvania ruralhealth model, which I know
you've already described brieflyas a global budget. Can you tell
us a little bit more about howthat works and how specifically
this model was implemented?
Paula Chatterjee (06:20):
Sure. So the
Pennsylvania Rural Health Model
or, p a r h m, I'm gonna call itPARM, like, you know, chicken
parm, for the for the purpose ofour conversation.
Rob Lott (06:29):
I love a good chicken
parm. So
Paula Chatterjee (06:31):
Who does that?
Who does that? So, parm was
introduced by CMMI back in 2017.We're we're talking that's a
time when we're a couple yearsinto the ACA, but really
noticing that rural hospitalsare struggling in unique ways
outside of of coveragechallenges. And the way I like
to think about Parm is that itwas trying to test a theory,
(06:51):
right?
The theory was that if you couldprovide rural hospitals with a
predictable source of financing,I. E, a global budget, then
maybe you would give them thestability that they need to
transform their care deliverymodels to increase access,
improve quality, improve health,and, you know, ultimately, as
part of the program, to reducegrowth in expenditures across
(07:13):
payers. And really, the sort ofcombined, the hope was that this
global budget would be alinchpin in the strategy for
sort of keeping ruralPennsylvania hospitals open. And
so, again, a lot of what I seeParm as being about was relying
on this financial predictabilitypiece because it's otherwise
just really hard to ask ruralhospitals to surmount, like,
(07:36):
these herculean challenges infront of them, right? We're
asking these hospitals to meetan increasingly complex and
expensive, you know, nature ofcare for an aging population
while relying on public payers.
Though, of course, as we allknow, Medicare Advantage was
growing over the same time thatthis program was being
introduced. We ask thesehospitals to, like, meet this
(07:57):
challenge in a world where,like, single digit differences
in their censuses can havetremendous volume based
consequences for their finances.And so this, like, pesky
denominator issue really, like,rears its head in every
dimension, and that makes ithard for them to do big things.
It's hard to do big things in anunpredictable world for these
hospitals. And so PharmMike CSwas trying to solve that part of
(08:21):
the problem with the globalbudget.
It was, you know, trying to togive them this predictable
source of financing. I'm happyto go into some of the details
if that's helpful, but I'm alsohappy to to pause.
Rob Lott (08:31):
Yeah. Thank you. Maybe
give us, like, the the two
minute version of how a globalbudget, typically works. My
understanding is that they youknow, exactly as you described
it, there's a predictablepayment. And I guess some
questions are, what is thatpayment based on, and, you know,
how might it vary from year toyear?
(08:52):
How is that implemented?
Paula Chatterjee (08:54):
For each
hospital that voluntarily wanted
to participate in Parm, theywould have a global revenue
budget that was specific to eachpayer. So that includes
Medicare, Medicaid, andcommercial insurers that were
participating in the state, ofwhich there were a good number.
The global budgets were indexedto historic revenue, and then
(09:15):
they would make adjustments toaccount for different market
dynamics from year to year. Sothat's one way in which you
could expect some changes fromyear to year. The purpose of the
budget was really to cover allhospital based inpatient and
outpatient care.
It did not include things likephysician payment, for example.
And if you were a criticalaccess hospital, for example,
(09:35):
which already sort of operatesunder its own financing
mechanisms, but let's say youalso wanted to participate in
Parm, you certainly could. Andin those cases, the cost based
reimbursement continues, butthey sort of try and make it a
little bit more predictable bykeeping payment accounts equal
and consistent throughout theyear. Again, trying to enable
(09:56):
this world of improved cash flowand and trying to avoid some of
these periodic cash shortfalls.So again, all in the service of
that predictability.
Rob Lott (10:04):
Okay. Great. Well,
then let's just dive in. Can you
tell us a little bit about theoutcomes you studied in your
research paper and what youfound?
Paula Chatterjee (10:14):
For sure. So
our paper was trying to
understand what we like to callthe quote unquote first stage of
Parm. Right? We're trying to seeif hospitals that chose to
participate in the program hadfinancial improvements because
if they had those improvements,then that would allow them to be
successful in sort of what weconsider to be the second stage,
which is then allowing them totransform their care delivery to
(10:36):
meet the needs of patients inreally innovative ways that sort
of, you know, fire on allcylinders. So that's why we
focus on finances in this paper.
We chose four financial measuresto try and get a different
aspects of a hospital'sfinancial well-being. So we
chose two measures of margins,which again, are trying to give
us like a global financialpicture. One of those margin
(10:58):
measures was focused on revenuethat's specific to patient care.
The other measure of margins wasfocusing on sort of more global,
like broader sources of revenue.We also looked at a measure of
uncompensated care provided by ahospital.
And the reason we looked at thatis because it's a measure of
financial strain for a lot ofhospitals. They have strong
incentives to keep this numberlow. And then we looked at a
(11:22):
measure of something called thecurrent ratio, which is kind of
like a measure of hospitalliquidity. You know, the nitty
gritty is that it's taking sortof a measure of assets to
liabilities over the course of ayear, and it's trying to give us
a little bit of informationabout trajectory, about
financial trajectory for thathospital. And so combined, we
were hoping that these fourmeasures might tell us a little
(11:43):
bit about the global financialpicture for these hospitals that
participated in PARM.
So that's what we chose tomeasure. It's hard one one of
the things I'll say is thatwhenever you're talking about
small groups, it can be hard to,like, draw statistical
conclusions. And so here, we'retalking about 17 hospitals that
(12:03):
were participating in PARMcompared to 40 nonparticipating
hospitals in Pennsylvania forwhich we had full data in our
period. It's just hard. But butwe didn't want those small
numbers to stop us from tryingto pursue a rigorous analysis
here.
And what I'll say, what I wasreally excited about in this
paper is that we are lucky thatnew relatively new statistical
(12:25):
approaches have, like, caught upto this, like, analytical
challenge. And so I promise Iwon't, you know, torture our
listeners with too many details.But what I will say is I'll say
that in 2022, some, you know,really smart people came up with
a new estimator that reallyallows us to apply a difference
in differences framework tosmall sample sizes and still
(12:48):
uphold the validity of all thestatistical assumptions you need
to make to, you know, to ensurethat this type of analysis is
responsible. So again, not goinggo into the nitty gritty. I fear
I've already done so too much,but I think it's important to
say oh, go ahead.
Rob Lott (13:02):
No. Was going to say
this is super helpful context,
especially when we're looking atsort of such a, I don't want to
say narrow, but small startingpoint. And so glad you kind of
had those tools at yourdisposal. What did you find?
Paula Chatterjee (13:21):
So what we
found is that participation in
the Pennsylvania Rural HealthModel was associated with a 4.5
percentage point differentialincrease in operating margins
and a 4.7 percentage pointincrease in total margins.
However, what I will say is thatthose results became non
statistically significant onceyou started adjusting for some
(13:43):
other components. And when welooked at some of the other
measures of hospital finances,things like uncompensated care,
that measure of liquidity, weactually didn't find any
significant differences betweenthe hospitals that participated
in Parm and those that didn't.And so for that reason, we sort
of conclude by saying this is alittle bit of mixed evidence. In
some of the models, we seeimprovements.
(14:05):
In other models, we don't. Andthat's really trying to apply
some of the state of the artapproaches that we have to this
hard question.
Rob Lott (14:12):
Wonderful. I wanna
look under the hood of those
results with you in just aminute. But first, we'll take a
quick break. And we're back. I'mhere with doctor Paula
(14:39):
Chatterjee talking about thePennsylvania Rural Health Model.
So just a minute ago, you sharedthat the results were mixed.
Were you able to kind of lookbelow that surface level and
maybe identify some findingsthat that you want to look into
further or that might sort of bethe starting point for policy
(15:01):
discussions?
Paula Chatterjee (15:02):
For sure. We
tried to look at this question
in a bunch of different ways andreally tried to kick the tires
through a lot of differentstatistical approaches. You
know, we can get into thedetails, but, know, this was
obviously a murky time to beevaluating a payment program.
You know, there was a a the thesmall issue of the pandemic,
which happened to be smack dabright in the middle of the the
(15:24):
implementation of this program.And so we try to look a little
bit into how different featuresof the pandemic might have been
at play with this paymentprogram.
And I think there are a lot ofinteresting questions there
because, you know, we're we'relucky to be able to talk to some
of the folks who areparticipating in this program.
And it's just it's it'sfascinating what implementation
has looked like on the groundfor a lot of these hospitals.
(15:44):
You know, some hospitals beganparticipating in the global
budget and were able to set upcommunity health worker programs
to try and meet patients indifferent areas. Others have
been trying to invest indifferent telemedicine
capabilities. And so I think ourpaper sort of gives us that
30,000 foot view a little bit,and we're really excited to sort
of dive in a little bit furtherand figure out what exactly has
(16:07):
been going on on the ground interms of of deploying these
resources to meet the needs ofpatients.
Rob Lott (16:13):
Great. I'm curious
about those conversations for
the folks on the ground. Did,what's the most surprising thing
you heard from from them inthose conversations?
Paula Chatterjee (16:22):
You know, I I
perhaps it shouldn't be
surprising to me at all, but Ithink that, you know, in in
discussing some of thesefindings with the folks on the
ground, they you know, maybeI'll take a step back for a sec.
CMS has mentioned that the firstcouple years of Parm, when they
were deciding the allocations ofthe budget, there were concerns
that some of them may may haveactually been overestimates of
(16:42):
budgets, and that may have beenjust due to some details in
accounting. And what I'll say isthat even upon receiving, you
know, what some might considerwere overestimates, a lot of
these rural hospital leaderswere saying, it's gonna take a
lot for us to feel financiallystable. Right? These are folks
who have been dealing withfinancial volatility for decades
in many cases.
And while there's a lot whilewhile I think that the principle
(17:05):
of the global budget is reallyto achieve that stability on a
shorter term, you're askinghospital leaders to view this
program with a much longer lensand history. And so, I think
that, you know, for me, thatjust gives more field of the
fire to understand what are thespecific challenges, what which
of those challenges can be metwith a global budget, which
(17:26):
can't. And so for me, again, itperhaps shouldn't have been
surprising, but I think viewingit in the lens of some of these
long standing challenges was areally informative thing for me
to learn.
Rob Lott (17:34):
So that's a great
segue, I think, for my next
question, which is to ask abouthow this piece or the piece that
is this program fits in withmany other programs and payments
aimed at supporting ruralhealthcare, things like the
critical access hospitaldesignation, rural emergency
hospital program, for example,as well as all those programs
(17:56):
that aren't necessarily targetedat rural communities, but which
rural communities depend on340B, the higher Medicare rates
for hospital outpatientprocedures. All of those things
that have helped in a way overthe years boost revenue for
rural hospitals. How do you seeParm kind of fitting in that
(18:19):
ecosystem?
Paula Chatterjee (18:20):
I think I see
Parm as part of that suite of
programs that is trying to getat different drivers of these
financial challenges for ruralhospitals, right? Like I
mentioned, Parm is trying to getat this issue of volatility
through prospective, predictablefunding. I see the other
programs as trying to addresssimilar but still a little bit
(18:40):
distinct aspects of thefinancial challenge, right? I
see the critical access hospitalas an effort to increase
resources for smallgeographically isolated
hospitals based on, you know,how it was originally motivated
in 1997 Balanced Budget Act. Andit does this through cost based
reimbursement, but it remains avolume dependent program and is
(19:02):
not perspective like a globalbudget.
There's other programs, right?There's the Low Volume
Adjustment Hospital Program,which is from CMS as well. And
that's trying to get at thevolume related challenge. And it
gives you a payment adjustmentthat is commensurate to the
volume related challenges. I seethe rural emergency hospital
(19:23):
program, which is trying toaddress the issue of high fixed
costs for a lot of these ruralhospitals.
We're saying, Hey, it's okay foryou to power down some of those
inpatient operations. Weunderstand that that's expensive
for you, but let's keep otherservices robust and available.
And so, think of 340B and DISHand some of these other
(19:43):
financial sources in a in alittle bit of a separate bucket.
Again, addressing financialchallenges, but perhaps a little
bit more broadly generalizedacross the you know, outside of
the rural context, but, ofcourse, very much instrumental
for these hospitals. But Ithink, you know, after I I go
through this alphabet soup, I'llsay that I think all of this
(20:03):
together sometimes points me ina particular direction, which is
that we've come up with a lot ofapproaches to try and get at the
challenges of rural hospitals,but they're still struggling.
And I think that that has mademe want to think a little bit
more about upstream policyintervention and what that might
look like beyond the world ofhospital payment.
Rob Lott (20:22):
So that makes me think
of a paper that we published in,
the May issue of Health Affairsby, Caitlin Carroll at the
University of Minnesota and hercolleagues that looked at the
prices, at rural hospitals whenother nearby hospitals close. No
surprise, prices go up when thathappens for the surviving
(20:44):
hospitals. And we had Doctor.Carroll on this podcast a few
weeks ago. We talked about sortof that alphabet soup, the crazy
quilt of policies.
And I asked her if she felt likethey were working and she, in a
classic academic response said,yes and no, no shade there. But
(21:08):
she said basically they'reworking because many rural
hospitals, they're keeping manyrural hospitals from closing,
but they're not working becausemany rural hospitals are still
closing. And one of the theoriesshe offered for that disconnect
was that the policies justaren't targeted well enough. And
so I guess my question for you,do you agree with that
(21:29):
sentiment? How would you sort ofrespond to this this dynamic?
And then two, how would youassess parm in terms of how well
targeted it is?
Paula Chatterjee (21:42):
So for
starters, just good life advice
to agree with doctor CaitlinCarroll, who knows whose work I
love and admire, in this area. Ithink let me be a little
specific about that. I I think Iagree with Caitlin in that these
policies work for some hospitalsand they don't for others. And I
I think there's a couple ofreasons for this. Right?
Part of this is just becauserural America is beautifully
(22:03):
heterogeneous and has differentchallenges, and the existing
policy solutions we have addresssome but not others. You know,
another point I think thatKaelin makes beautifully is this
issue about fit. Right? Like, weI find this line of argument to
actually be consistent with whatfolks in community health
centers have told me recently.Right?
In their case, what's happeningwith them is they're being
(22:25):
presented with, like, an arrayof value based payment programs,
but they have little idea on howto choose between them. It's
hard for them to know ex antewhich ones are going to be
advantageous and which ones aregonna introduce undue risk or
complications of participating.They agree with the general
sentiment of what those policiesare trying to achieve, but it's
(22:45):
really hard to pick what thebest fit is. For rural
hospitals, I think the fit isnot clear because the financial
landscape of rural health careis just changing really fast
under their feet, right? Whetherit's acquisitions by bigger
health systems, opportunitiesto, you know, engage with
capital through private equityor growth in Medicare Advantage,
(23:07):
right?
It's hard to pick a financialstrategy when the crystal ball
is so murky. More broadly, I'llsay that I think asking rural
hospitals to assume upside ordownside risk is just gonna be a
big ask, right? They alreadyhave an absurd set of challenges
they're trying to navigate, andadditional uncertainty is just
not desirable. Right? Even if itoffers the opportunity for
(23:28):
financial stability, like inParm.
Right? Hospital leaders werestill concerned about the
uncertain. What if payers changetheir mind about budget
calculations? What if there arenew benchmarks? What if a new
financial threat presents itselflike a pandemic?
Right? These are reasonableconcerns. And I think that's why
we saw sort of, you know, asmall number of participants in
Parm despite the intention totry and get as many rural
(23:51):
hospitals involved. Whether thatmeans it's a good fit for them
or not, I'm not quite sure. Wedo know that some hospitals that
participated did feel like theywere able to weather the
pandemic a little bit betterthan others, but that's just one
dimension of fit.
I think we'd hear a differentstory if we talk to every
hospital.
Rob Lott (24:08):
Sure. Before we wrap
up, let let's take a step back
and maybe, I'd love to hear yourthoughts for our listeners
thinking about where we go fromhere. What do you think is the
next step on this front when itcomes to policy? And if your
paper were to maybe inspire anysort of concrete or specific
(24:29):
policy changes or improvements,what would you hope that to be?
Paula Chatterjee (24:35):
So can I give
two answers? Absolutely. The
reason there's two is one isthat, like, the 3,000 foot
level, and the other one is at30,000 feet. So my Go
Rob Lott (24:45):
for it.
Paula Chatterjee (24:45):
My 3,000 foot
takeaway is that global budgets
might still be a very usefultool for rural hospitals, right?
There's another way to frame ourresults is that despite the
tumultuous period in which Parmwas introduced, participating
hospital did not take afinancial hit, right? And that
is valuable and important,especially for a lot of rural
(25:06):
hospitals that see financiallosses from year after year,
right? I think that tryingglobal budgets in a state
without rate setting likePennsylvania in contrast to
someplace like Maryland isvaluable because you can see how
a global budget works in thewild. Right?
And and as we know, globalbudgets are only, you know,
going to become more prominent.They're included in the AHEAD
(25:27):
model and sort of continuing tofollow the trajectory of this
program, I think, is going to beimportant and interesting. My
30,000 foot takeaway goes back alittle bit to the earlier
discussion, which is thatthere's actually been quite a
bit of policy attention on ruralhospitals for the past several
decades, right? There's been alot of programming designed to
support them and address thechallenges that they find
(25:48):
themselves in. But despite that,they're still struggling, right?
Rural health outcomes are stillsuffering. And I think that
juxtaposition of facts demandsus to just, like, pause for a
second and ask why. Why are allof these well intentioned
policies not making big dents inthis problem? And I think the
way I find myself thinking aboutthis these days is that when you
(26:09):
talk to rural hospital leaders,they'll tell you about their
financial challenges, butthey'll actually spend most of
their time talking about theirpatients and their workers.
Right?
They'll tell you about how theburden of disease has just grown
so dramatically and how it'sharder and harder to keep highly
educated labor in in rural areasto sustain the hospital and how
their communities have beenstruggling for so long that
(26:30):
people leave, and that reducesthe tax base that's available to
try and keep the hospital open.And so I find myself trying to
take those observations and fitthem into what we know about
payment. And and I think wherethat leads me these days is I
think just focusing on paymentis going to continue to be an
incomplete solution. It's a veryattractive place to focus as a,
(26:52):
you know, admitted healthservices researcher because we
can control parts of it, right?And because the stakeholders are
confined to the healthcareindustry, we can try and frame
this as a healthcare problem.
But I think improving health inrural areas is going to require
investing in rural communitiesin a much broader way, right?
Like, only then do you have apopulation for the hospital to
serve. Only then can you keepclinicians there to support
(27:15):
these hospitals. By justfocusing on payment, I worry
that we're using a bucket toempty the ocean. And so my
30,000 foot view these days isthat I'm wondering if there's a
way for us to think about ruralhealth policies that are coupled
investments for rural people.
Right? I think we need tosustain economic opportunities
in these areas, expandbroadband, you know, attract
(27:35):
businesses and infrastructure tothese communities for payment
policy to have a chance at beingfully effective and realizing
the potential that we want forit. And so I think that that
that's sort of the directionthat I I I find myself wanting
to start thinking about thisproblem.
Rob Lott (27:51):
Great. Well, that's
probably a good spot for us to
wrap up. A wonderful way to tothink about it looking forward.
Doctor. Apala Chatterjee, thankyou so much for joining us on
the podcast.
Paula Chatterjee (28:03):
Thank you so
much for the conversation.
Rob Lott (28:05):
To our listeners, if
you enjoyed this episode,
subscribe, recommend it to afriend, leave a review, and tune
in next week. Thank you. Thanksfor listening. If you enjoyed
today's episode, I hope you'lltell a friend about a health
plus.