The Politics of Alternative Payment Models

Let us know what you think about Health Affairs podcasts at [e-mail]. If you have 30 minutes to spare, let us know and we'll set up a 30-minute chat for the first 20 listeners that reach out. Coffee will be on us.

Health Affairs' Jeff Byers welcomes Senior Editor Michael Gerber back to the program to discuss the recent announcement from the Trump administration to cancel certain alternative payment models, their previous impact on health spending, and the difficult task of measuring success over a long period of time and across different administrations.

Order the April 2025 theme issue of Health Affairs focusing on food, nutrition, and health.

Join us for our theme issue briefing on April 8 featuring panels on Food is Medicine, Government Programs & Policies, Community Voices, and more. Save the date.

Also, join us April 23 for an exclusive Insider virtual event exploring site-neutral payments with health economist and health services researcher Brady Post of Northeastern University and Health Affairs' Meg Winchester.

Related Articles:

• Kennedy to slash 10,000 jobs in major overhaul of US health agencies (Reuters)
• Internal fallout at HHS delays 10,000 firings (Politico)
• Maryland Hospital All-Payer Model: Can It Be Emulated? (Health Affairs Forefront)
• CMMI cancels 4 payment models early (Healthcare Dive)
• Statement on CMS Innovation Center Aligning Portfolio with Statutory Obligation (CMS)

Episode Transcript

Jeff Byers (00:39):
Hello and welcome to Health Affairs This Week. I'm
your host, Jeff Byers. We'rerecording on 04/02/2025. As a
reminder, we have a theme issueon food, nutrition, and health
coming out next week. Be on thelookout for that, and there's a
virtual briefing on that themeissue on April 8.
Please join us if you can. Youcan find a registration link in

(01:00):
the show notes. Relatedly, ournext insider event is April 23
with Brady Post on-site NeutralPayments, and we're working on
some good late spring, earlysummer events as well. Today,
I'm joined by Michael Gerber todiscuss the announcement that
CMMI will cancel somealternative payment trials.
Before we get into that, thebiggest policy news this week

(01:23):
probably is the series oflayoffs inside HHS that began
this week.
Reuters reported last week thatHHS planned to cut 10,000
employees, which wouldultimately equate to a reduction
of about a quarter of the HHSworkforce. Reuters stated that
about 3,500 of these cuts wouldcome from the FDA, two thousand

(01:43):
four hundred at the CDC, twelvehundred at NIH, and quote, CMS
less affected, unquote. Axiosreported this morning in their
Axios Vitals newsletter that inthe immediate aftermath, health
care industry players and formalfederal workers say the
workforce reductions will almostcertainly affect drug approvals,

(02:04):
low income assistance, diseasetracking, and biomedical
research once held up as thegold standard, unquote. I'm sure
a lot of trade publications willhave more details on this, but
this is the newsletter that Iopened before we began prep
today. It's big news.
Perhaps we can dive into theimplications at a later date and
episode as more informationbecomes available. But with

(02:25):
that, Michael, welcome back tothe program.

Michael Gerber (02:28):
Hey, Jeff. Thanks for having me.

Jeff Byers (02:29):
Yeah. So the Trump administration announced this
month it would cancel sixalternative payment models, two
of which have had yet to start.Of the four that will be
canceled early, what can youtell

Michael Gerber (02:42):
us about them? Yeah. Well, as you said, they
they announced they canceledfour models about a month ago.
These include the Maryland TotalCost of Care model, the primary
care first model, the makingcare primary model, and the end
stage renal disease treatmentchoices model. Those models
varied in how long they'd beenin existence, some of them for

(03:03):
several years now, some reallywere just about to get started.
And they didn't give a lot ofreasons as to why, but it's not
totally uncommon. The Bidenadministration also came in four
years ago, paused some models,canceled at least at least one
that the Trump administrationhad just created. So it wasn't a

(03:24):
a shock at all.

Jeff Byers (03:25):
The administration estimated that the program
cancellations will save about750,000,000 and stated that this
will streamline the focus ofCMMI models and reduce spending.
How many models were there? Whenyou're looking to health
spending, how did CMMI relate tothe overall federal budget?

Michael Gerber (03:45):
Yeah. Great question because CMMI somewhat
controversial, but really in abipartisan way to some extent.
The goal of CMMI when it wascreated as part of the
Affordable Care Act was reallyto test models, specifically,
really, the value based paymentand that shift to value based
payment. And when they finally,a couple years ago, the CBO and

(04:09):
and CMS itself started, youknow, really trying to do big
evaluations of how did all thesemodels perform over years. And
that's, you know, upwards ofclose to a hundred models with
about two dozen active now thatthey increased spending by about
$5,000,000,000 as opposed to anexpected savings of several

(04:30):
billion dollars.
Now there are some caveats tothat. Some of those that even as
they increased spending, theydid find some portions of those
models that work that they wereable to adopt in sort of regular
CMS day to day activities. SoCMMI itself might have increased
spending, but, you know, I thinkas we know and as maybe some

(04:55):
people in the currentadministration would say, you
have to try things, you have tobreak things to see what works.
And so even if not everythingworked, a lot of people say a
lot was learned that is beingused to save money in other
areas today. That said, the theMaryland total cost of care
model is one that that was foundin in the evaluations at least
to save some money.

(05:15):
So

Jeff Byers (05:16):
Yeah. Yeah. We'll we'll get into that in a second.
But going back to your note ofhow, these kind of changes were
also made in the Bidenadministration and when you're
looking at the CBO report, isthis just kind of the ebb and
flow of how these trials and,like, general politics go?

Michael Gerber (05:34):
Yeah. It seems it might be. Again, some of
these, when they made theannouncement, you know, that
there's differences in how theyframe it for sure. The Biden
administration came in, pausedsome models, canceled some,
although shifted that really tojust slightly different
variations. You know, they saidthey were gonna shift models,

(05:54):
change models to align withCMS's new strategy, which in the
Biden administration focused onthings like health equity.
You know, no surprise that theTrump administration would not
make that argument. Theyheadlined their announcement
that they were gonna focus onsaving taxpayer money. So
they're focused more on youknow, they said canceling these
models will save 700, I think,$750,000,000 right, you know,

(06:16):
right away. You know, with eachmodel, it was a little
different. You know, with theMaryland model, yes, they
canceled that, but there werealready plans.
That model was, you know,already set to end in 2026 and
shift to, you know, what theycall the ahead model. If
anything, this this announcementseemed a little less, I mean,
it's still very political, butmore old fashioned political

(06:39):
than what we're seeing in a lotof other decisions, being made
right now.

Jeff Byers (06:44):
Yeah. Yeah. And so looking at that Marilyn model,
you mentioned that it doesn'thave much, life left in the
trial. It's gonna transition.What has the Maryland model
showed us when it comes to,like, savings and its actual
model?
What has the model shown? Was itaccomplished?

Michael Gerber (07:02):
The evaluations, you know, say that in the first
three years of the currentversion of the the Maryland
total cost of care model savedalmost $700,000,000 over three
years. So, you know, notbillions, but but a significant
chunk of money in in one onestate.

Jeff Byers (07:19):
Small state.

Michael Gerber (07:20):
Yeah. But the fact that it is one state and a
small state also makes somepeople question how much can be
copied from that to otherstates. Maryland has actually so
this model, really a big part ofit was these sort of global
payments to hospitals where CMSand they both with Medicare and
they were incorporating thiswith other payers, because CMS

(07:43):
is trying to get other payers toalign to make things simpler,
but would just have globalpayments to hospitals. So we're
gonna give you this much moneybased on, you know, our research
and past trends, and that'llcover your costs for treating
Medicare patients as opposed tofee for service for each
individual, you know, specificservice provided. So the thing
is in Maryland, that dates backactually decades to the state

(08:06):
trying similar models.
So this wasn't brand new. And Ithink that that also brings us,
if I can, transition real quickto to broader lessons from that,
is that one of the things ofsome of these models that had
just been, like, the newest onethat that the Trump
administration just paused or orcanceled, for example, the
making care primary model wasscheduled to go for ten years,

(08:29):
which most CMMI models had notgone for that long. And some
people say you really need togive it more time. So one
question is if eachadministration every, you know,
four or eight years or however,when they look at their strategy
as they have the right to do andchange priorities, if they're
canceling models or shiftingthese models is how hard will it
then be to evaluate whetherthey're successful, and how much

(08:52):
of a challenge is that for thehealth systems and providers,
clinicians out there who have tomean, these are not always easy.
One of the biggest problems withthese models is they can be
hard, especially for yourmidsize or smaller systems or
individual clinicians to have, Imean, just the paperwork sort of
in the data analysis and theresources, which that was one of

(09:12):
the things this making careprimary model was supposed to
address was making it easier foryour individual clinicians to
build up to being in the modeland giving it ten years to go.
So I think that's anotherquestion of whether the whole
idea of CMMI and testing theseinnovative models in something
as big as health care andMedicare can function if every
few years it changes over.

Jeff Byers (09:33):
Yeah. Yeah. You bring up a a great point. I
think with a lot of the I mean,I'm really just restating what
you just said, where a lot ofthese alternative payment models
have been questioned aboutwhether they can be emulated or
scaled up. While doing researchfor this episode, I did find
that we have a lot of writingson the subject on health affairs

(09:56):
forefront.
I didn't find as much in ourjournal, but some of the
interesting articles wereMaryland Hospital All Payer
Model, Can it be Emulated? Andthen a lot of interesting stuff
on Forefront when it comes toalternative payment models. We
talked a little bit about thisalready, but do these
cancellations or like what youjust mentioned, does it tell us

(10:16):
anything about the state of APMsor like primary care versus cost
of care or saving healthspending dollars?

Michael Gerber (10:23):
Yeah. I think from a big picture, it it
doesn't tell us a lot that wedidn't already know. I think one
thing one thing we're learningfrom it so so alternative
payment models value based care,I think, has become somewhat
popular bipartisan. Now you canreally get in there and debate
what those mean and thedifferent models, and and then

(10:44):
you start to, you know, findbreakdowns both partisan, but
also among differentstakeholders, certainly
physicians versus hospitalsversus payers. But in general,
you know, it is these models tosome extent have become popular
among the sort of health policywonks who are running the show
for Biden for the previous Trumpadministration.
So from what I read, forexample, the head of CMMI one of

(11:07):
the heads of CMMI in the firstTrump administration named Brad
Smith, who actually, isreportedly one of the high ups
at Doge, who's making some ofthese layoff decisions that are
impacting HHS this week, youknow, is still supportive of
value based payment in thebetween administrations, worked
for some companies that, youknow, are in this space. And

(11:30):
Politico even reported thatmaybe that's why CMS and CMMI
weren't hit as hard, it appearsso far, by some of these changes
because that's where he hadserved and and that he's still
supportive. And, again, as Isaid, if you look at this
announcement, you know, theythey canceled some of these
models, but they said they wereshifting some to other sort of
ACO or alternative paymentmodels. They didn't cancel a lot

(11:52):
of other models that are outthere, at least not yet. And and
they didn't they didn't come outand say, you know, we're against
value based payment.
And, again, if if you think ofvalue based payment as the big
picture, which includes thingslike Medicare Advantage, so it
might just mean shifting intoother kinds of value based
payment that might be lesspopular for some people and more

(12:14):
popular for others.

Jeff Byers (12:16):
Okay. Well, as we come to a close, is there
anything else we should touch onwith this topic?

Michael Gerber (12:22):
No. I think the one thing is to you know, this
announcement certainly, willaffect a lot of people who, in
the health care space who are inthese models and who may be
awaiting word on what this meansfor them exactly. The, again,
differently than I think a lotof other things we've read from
this administration, itspecifically said they will work

(12:44):
closely with the providers,clinicians, health systems
impacted by this to help withthe transition and that they are
working on a strategy.

Jeff Byers (12:53):
Yeah. And just as a quick note, I I have this in our
notes. We didn't we didn't touchon this, but there looks to be
about 90 models within the CMMIbase. 23 of them look to be
active. So kinda interestingwith that just to just to scale
that or put that in perspective.
Well, with that, Michael, thanksfor joining us today on Health
Affairs This Week. And to you,the listener, if you enjoyed

(13:17):
this episode, please send it tothe huckleberry in your life,
and we'll see you next week.Thanks.

Michael Gerber (13:23):
Thanks a lot, Jeff.