Popular in Social Sciences

Books

City Time: On Being Sentenced to Rikers Island

Independent Scholar David A. Campbell

Culture and Society

The City as a Growth Machine: Toward a Political Economy of Place

New York University Harvey Moltoch

Economics

Gender and Economics

University of California Santa Barbara Shelly Lundberg

American Politics

Keeping the Republic

Boston College Marc Landy

Banks and Policy Organisations

Greening the financial sector

European Investment Bank Matteo Ferrazzi

British Politics

Entrepreneurial thinking: the politics and practice of policy impact

University of Sheffield Matthew Flinders

Culture and Society

Carceral Entrapments: Views from the Prison/Street Interface in India

Indian Institute of Technology Delhi Mahuya Bandyopadhya

Incentives in Healthcare Payment Systems

View more insights in Social Sciences - Economics - Health Economics

View more figures from Boston University

View the page of Albert Ma

Read the Study

Health services providers receive payments mostly from private or public insurers rather than patients. Provider incentive problems arise because an insurer misses information about the provider and patients, and has imperfect control over the provider’s treatment, quality, and cost decisions. Albert Ma discusses incentives in healthcare payment services.

Image courtesy of interviewee. July 24, 2024

Generating Workspace

Export as PDF

This video discusses the intricacies of payment incentives in the U.S. healthcare system, particularly focusing on the roles of three primary parties: consumers (patients), providers (medical personnel), and payers (insurance companies, government, or employers). The healthcare sector accounts for a significant portion of the U.S. GDP, making the study of payment mechanisms vital.

Initially, the text highlights the traditional fee-for-service model, where providers are reimbursed for actual costs incurred, creating a potential incentive for overutilization of services, which can contribute to rising healthcare costs. In contrast, the prospective payment system sets fixed prices for specific episodes of care, irrespective of the resources used. This video model encourages providers to be more efficient by internalizing costs.

The discussion delves into the internalization principle, which suggests that for a provider to align their actions with payer goals, they must be incentivized to internalize both costs and benefits of their service delivery. This video principle is illustrated through examples, such as incentivizing students through exams to ensure consistent studying over the semester.

This video also addresses potential issues with the prospective payment model, namely "dumping," where providers might avoid high-cost patients, and "cream skimming," where they attract low-cost patients, undermining the effectiveness of the payment system. Additionally, the conversation extends to nuances such as blending fee-for-service with prospective payment mechanisms to manage provider behavior effectively.

Lastly, the text notes that by developing more sophisticated payment models that account for organizational behavior—like paying for a group of providers instead of individual practitioners—the healthcare system can improve the quality of care while managing costs. It concludes by emphasizing the ongoing evolution and complexity of healthcare payment literature and its impact on provider behaviors and healthcare outcomes.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.015 --> 00:03.325
Healthcare is a very important part of any society,

00:04.255 --> 00:07.115
and it's particularly important in the US

00:07.115 --> 00:11.155
because it occupies up to 20% of GTP.

00:13.015 --> 00:15.925
There are various ways that we can think about healthcare

00:16.105 --> 00:18.765
and, today I'm just going to concentrate on

00:20.035 --> 00:22.005
this element called payment incentives.

00:23.145 --> 00:26.375
There are other issues to think about, like insurance

00:26.555 --> 00:29.415
and so on, but they will be left on another location.

00:30.465 --> 00:33.485
So why are payment incentive at all relevant?

00:34.415 --> 00:37.275
The basic issue is that there are sort

00:37.275 --> 00:40.875
of three parties in this, in this scenario.

00:41.585 --> 00:44.685
They're the consumers, they're the providers,

00:44.945 --> 00:46.245
and then they're the payers.

00:47.135 --> 00:50.805
It is like they interact. Consumers are patients.

00:52.135 --> 00:53.545
They want some services.

00:54.185 --> 00:56.345
Providers are medical personnel and facilities

00:56.345 --> 00:57.585
and healthcare professionals.

00:58.765 --> 01:02.825
And then the pay is a government, employers,

01:02.825 --> 01:05.025
insurance companies, public or private.

01:05.805 --> 01:07.945
It depends on various different countries.

01:09.335 --> 01:11.145
Basically, consumers demand

01:11.145 --> 01:13.265
and receive services from providers.

01:13.945 --> 01:18.565
The insurance will shield consumers from receiving, will

01:19.085 --> 01:21.525
shield consumers from part of the surface expenses.

01:22.225 --> 01:24.085
That's the whole point of health insurance

01:25.335 --> 01:29.225
because illness carries some financial risk.

01:29.725 --> 01:31.385
And the insurance part says

01:31.775 --> 01:35.125
that when you seek medical treatment, both

01:35.145 --> 01:37.685
or part of the medical expenses will be paid for

01:37.705 --> 01:38.805
by the insurance company.

01:40.045 --> 01:44.025
Now this creates an obvious issue, which I'll come

01:44.025 --> 01:45.145
to in just one moment.

01:46.625 --> 01:48.685
Uh, who pays the bill?

01:49.895 --> 01:52.915
If consumers don't pay, the payers would have,

01:53.895 --> 01:55.205
would have to pay.

01:55.865 --> 01:59.335
So how,

02:01.125 --> 02:04.025
providers paid would

02:04.025 --> 02:06.505
therefore determine their behavior?

02:07.905 --> 02:11.245
If the payer can make providers be

02:11.765 --> 02:13.765
efficient so much, the better.

02:13.905 --> 02:16.365
And what are the ways to make them efficient?

02:17.365 --> 02:21.135
Um, in the old days, like 50 years ago,

02:22.195 --> 02:25.255
60 years ago is Medicare in the US

02:25.355 --> 02:28.495
and many other countries use something called fee

02:28.495 --> 02:31.095
for service or cost reimbursement.

02:33.135 --> 02:37.145
Whatever cost a provider incurs to surface consumers,

02:37.885 --> 02:41.425
the payer will just reimburse the expenses.

02:43.115 --> 02:46.775
Uh, of course this may often include a profit margin.

02:48.185 --> 02:50.885
So the profit margin is, it can be an issue

02:50.885 --> 02:54.325
because the profit margin then create an incentive problems.

02:54.575 --> 02:58.685
There will be more services provided and maybe qualities

02:58.745 --> 03:00.565
Can be made inefficiently high

03:00.905 --> 03:04.365
to attract services demanded by, demanded by consumers.

03:05.685 --> 03:08.785
The cost reimbursement fee for service has been regarded

03:08.785 --> 03:12.065
as a major reason for the high cost in the us

03:13.815 --> 03:17.755
That's one of course base payment mechanism.

03:18.295 --> 03:21.755
And there's an obvious incentive issue, then

03:23.005 --> 03:25.205
a popular, and now very,

03:25.235 --> 03:28.285
very common payment mechanism is called prospective

03:28.625 --> 03:30.045
or fixed price mechanism.

03:31.235 --> 03:34.555
It works like this. It works in two steps.

03:35.415 --> 03:38.645
First, we sort of define an episode of care.

03:39.945 --> 03:42.455
Let's say a patient gets into a hospital,

03:43.455 --> 03:45.565
eventually the patient gets discharged,

03:46.145 --> 03:48.765
and let's say this is a pneumonia episode.

03:49.665 --> 03:52.845
At the end of the episode, appoint discharge at

03:52.845 --> 03:56.685
that primary diagnosis called pneumonia will be associated

03:56.755 --> 03:58.645
with the patient and the discharge.

03:59.515 --> 04:01.135
Now this category

04:01.475 --> 04:03.975
of pneumonia will be associated with a price.

04:04.935 --> 04:06.985
This price will be the payment

04:07.455 --> 04:09.705
that the payer makes to the provider.

04:10.365 --> 04:11.985
It is a fixed price mechanism.

04:12.765 --> 04:16.225
No matter what kind of resources the provider has incurred.

04:16.645 --> 04:20.665
No matter how many days the patient stays in the hospital,

04:20.925 --> 04:22.785
it will be the same price.

04:25.205 --> 04:27.745
And, and it is very different from cost reimbursement.

04:27.895 --> 04:29.985
Cost reimbursement says, well, if you stay longer,

04:30.315 --> 04:32.145
we'll pay you more, stay shorter,

04:32.395 --> 04:34.665
we'll pay you less fixed price.

04:34.665 --> 04:37.545
Prospective payment says no matter what kind

04:37.545 --> 04:38.825
of resources you use,

04:39.465 --> 04:41.985
provided that you know the the provider, the hospital has

04:41.985 --> 04:44.345
to treat you in a satisfactory way.

04:44.685 --> 04:46.465
Um, the payment will remains the same.

04:47.335 --> 04:49.155
So what does it aim to achieve?

04:51.065 --> 04:52.795
Well, think about the following scenario.

04:52.975 --> 04:54.795
The two things that the provider does.

04:56.055 --> 04:58.675
The provider is gonna make quality decision

04:59.455 --> 05:01.435
and it's gonna make cost decisions.

05:02.435 --> 05:06.095
The quality decision can be associated with,

05:06.125 --> 05:10.885
nursing care, the, facilities, tests

05:11.705 --> 05:14.925
and any other things that, consumer,

05:15.145 --> 05:20.125
the patient may enjoy raising qualities will

05:20.125 --> 05:23.645
be sort of good for consumers and will raise demand,

05:24.105 --> 05:27.285
and consumers can be happier with higher quality.

05:28.155 --> 05:30.695
But of course, qualities can be very costly

05:31.415 --> 05:33.905
when costs are high,

05:35.625 --> 05:37.485
the hospital doesn't get paid more.

05:38.145 --> 05:42.205
So there is in some sense a way an incentive for the

05:42.725 --> 05:44.445
provider to reduce costs.

05:44.985 --> 05:48.285
Reduce costs, reduction of costs can happen in many ways.

05:49.145 --> 05:51.885
Um, finding the right treatment,

05:52.435 --> 05:57.005
finding the appropriate treatment, and even the organization

05:57.345 --> 05:59.765
Of the hospital of the facility

06:00.425 --> 06:02.045
can sometimes affect costs.

06:02.765 --> 06:07.265
Okay, so these are the thing about this as the two decisions

06:07.265 --> 06:08.425
that a provider makes

06:09.345 --> 06:11.965
and, the cost reimbursement system that we,

06:12.165 --> 06:14.805
I talk about just, you know, few minutes ago, the fee

06:14.805 --> 06:18.445
for service says that cost will always be paid for,

06:18.915 --> 06:21.765
even if I reduce cost, my payment

06:22.775 --> 06:24.725
would not get me more profit

06:24.725 --> 06:29.225
because it would just be reducing the reimbursement for me.

06:29.655 --> 06:31.025
Okay? Now,

06:31.415 --> 06:35.025
prospective payment doesn't work in the same way.

06:35.295 --> 06:36.385
Okay? And then,

06:36.525 --> 06:39.145
and let me now tie tie it to this, the sort

06:39.145 --> 06:41.825
of the economic principles involved in all this,

06:42.735 --> 06:45.475
and I call this the internalization principle.

06:46.285 --> 06:48.465
It applies in a very general setup,

06:48.845 --> 06:50.105
is not limited to healthcare.

06:51.075 --> 06:52.335
It says the following.

06:53.155 --> 06:56.535
If A wants B to do something

06:56.965 --> 07:01.195
that A desires A wants B to do something,

07:01.675 --> 07:06.595
A cannot control B'S behavior A cannot co B's action, then

07:07.235 --> 07:10.115
B must be made to share a's preferences.

07:12.355 --> 07:14.655
In other words, B must be made

07:14.655 --> 07:16.895
to internalize ace benefits and costs.

07:17.825 --> 07:20.915
I'll give you a very simple, since we are at a university,

07:21.035 --> 07:22.035
a very simple example,

07:22.895 --> 07:24.915
why do we think we have midterms and final?

07:25.895 --> 07:28.275
Why not have just the final exam?

07:28.935 --> 07:32.555
You see, once we have a midterm, that will be an incentive

07:32.695 --> 07:35.635
for students to study during the midterm

07:35.635 --> 07:37.235
during the first half of the semester.

07:37.695 --> 07:40.555
And the final, of course, would be another way

07:40.575 --> 07:42.715
to motivate students to study during,

07:42.735 --> 07:44.595
during the second half of the semester.

07:45.055 --> 07:47.995
So if we want students to be studying

07:49.085 --> 07:53.715
throughout the semester, these two exams will be needed.

07:54.135 --> 07:58.675
So I as a professor warned my student to study gradually

07:58.775 --> 08:00.155
and consistently, and

08:00.155 --> 08:04.315
therefore I set up these two exams so that my preferences

08:04.535 --> 08:07.195
for their learning behavior will be more in, in line

08:07.195 --> 08:10.235
with their incentive to perform well in the exam.

08:11.185 --> 08:14.325
Now, of course, in quality choices to the satisfaction

08:14.325 --> 08:17.965
of the, of the payer, the provider makes two choices.

08:19.685 --> 08:24.495
Um, so think about the following, another sort

08:24.495 --> 08:25.575
of principle, one

08:25.575 --> 08:30.095
and a half, the internalization principle, right?

08:30.335 --> 08:33.635
I, I want the payer to think of a way

08:33.815 --> 08:35.755
to internalize this cost

08:35.895 --> 08:38.315
and quality incentives, motivation,

08:38.545 --> 08:40.595
perspective by the provider.

08:40.905 --> 08:42.595
Okay? Now you think about that.

08:43.135 --> 08:44.595
Two things that I want to do.

08:45.035 --> 08:46.915
I need them to study first half

08:47.415 --> 08:50.635
and I want them to study second half of the semester.

08:51.115 --> 08:55.775
I need two exams. One exam isn't quite good enough, right?

08:57.245 --> 08:59.545
So you think about cost and quality choices

08:59.725 --> 09:01.225
and cost in quality incentive.

09:01.565 --> 09:04.225
You naturally think, well, we need two instruments.

09:04.485 --> 09:08.825
One needs an instrument to motivate cost incentives.

09:09.325 --> 09:12.905
One needs, an instrument to motivate,

09:14.115 --> 09:15.165
quality incentives.

09:16.385 --> 09:18.605
Now, what is the beauty of the prospective payment?

09:19.715 --> 09:22.165
What does the pro prospective payment say?

09:22.345 --> 09:25.885
Is the provider is responsible for cost

09:27.695 --> 09:31.115
no matter what the cost is, the payment from the payer

09:31.375 --> 09:33.035
to the provider is constant.

09:33.735 --> 09:36.995
In other words, it internalizes cost, so to speak.

09:37.375 --> 09:41.555
The provider under this payment mechanism has already

09:42.295 --> 09:45.115
one instrument to align the cost incentives.

09:45.945 --> 09:47.805
So what are we left behind with?

09:48.305 --> 09:50.565
We are left behind with the second instrument.

09:50.745 --> 09:53.445
The price, higher price would

09:53.445 --> 09:57.565
therefore encourage higher quality, lower price can,

09:58.625 --> 10:00.675
encourage or discourage higher quality.

10:00.985 --> 10:03.195
Okay, let's say you have this patient

10:03.335 --> 10:07.075
who has pneumonia being admitted to the hospital, right?

10:07.075 --> 10:08.275
So how many days to stay?

10:08.735 --> 10:12.275
Uh, okay, well, the hospital has to bear the cost.

10:12.335 --> 10:14.195
You cannot just send away the patient when

10:14.215 --> 10:15.355
the patient is very sick.

10:16.385 --> 10:21.055
Uh, the quality of care now depends on what, depends on

10:21.055 --> 10:22.695
how much the hospital is paid.

10:23.605 --> 10:25.865
If this price upon this charge is high,

10:26.095 --> 10:29.305
then we have an incentive to provide, would've an incentive

10:29.845 --> 10:31.625
to make the patient happier.

10:32.695 --> 10:37.355
So the price can be adjusted to align some incentive.

10:37.775 --> 10:40.885
The price can be made, can be set.

10:40.905 --> 10:43.645
So the provider to internalizes the benefit.

10:44.305 --> 10:46.285
So that's the internalization principle.

10:47.035 --> 10:50.855
The provider wants, the, the,

10:50.875 --> 10:53.895
the payer wants a provider to have

10:54.935 --> 10:57.975
efficient cost and quality decisions.

10:58.975 --> 11:01.915
And the way to do it is to ask the provider

11:02.175 --> 11:06.805
to internalize the cost and benefit the desirable part

11:06.825 --> 11:09.325
and the, and the costly part of the cost

11:09.505 --> 11:10.565
and quality incentives.

11:11.835 --> 11:13.735
We have one instrument because the prospective

11:13.735 --> 11:15.175
payment is a single price.

11:16.185 --> 11:19.075
However, the beauty of prospective payment is

11:19.075 --> 11:21.155
that cost is already internalized.

11:21.295 --> 11:25.475
By definition, the provider has to bear the cost.

11:25.575 --> 11:27.595
So we have, we have full efficiency in

11:27.595 --> 11:28.875
this, in this scenario.

11:29.895 --> 11:33.075
Now, it also brings out the issue about fee for service.

11:33.895 --> 11:37.675
Fee for service would never induce efficiency simply

11:37.675 --> 11:40.275
because cost is always reimbursed.

11:41.165 --> 11:43.735
That part of the sort of the equation

11:43.955 --> 11:45.775
of the incentive is never internal

11:47.155 --> 11:48.755
provided, does not internalize.

11:50.145 --> 11:52.915
Alright? So that's the sort of the literature

11:53.045 --> 11:58.005
I I sort of started, about 30 years ago is, one

11:58.005 --> 12:00.445
of the first papers to recognize that,

12:01.165 --> 12:04.005
providers behavior can be multifaceted.

12:04.155 --> 12:05.805
They have many things that they can do,

12:06.515 --> 12:08.605
some concern causing some concern, quality

12:09.305 --> 12:12.645
and different payment mechanism is steering them towards

12:12.645 --> 12:16.335
different kinds of behavior, right?

12:16.885 --> 12:20.135
It's simple model, but we always start with simple model

12:20.135 --> 12:21.815
because we can derive insights

12:22.315 --> 12:26.255
and then we can continue to work on the same principle,

12:26.255 --> 12:30.025
but relaxing various, restrictions.

12:30.205 --> 12:31.225
So what is the catch?

12:32.285 --> 12:36.925
Uh, way back when people already spotted

12:37.035 --> 12:38.845
some issues with prospective payment.

12:40.475 --> 12:44.425
Um, the prospective payment suffers from some

12:44.675 --> 12:45.785
well-known problems.

12:45.815 --> 12:49.385
It's called dumping and cream skiing. So what is dumping?

12:49.895 --> 12:52.465
Dumping refers to the idea that

12:52.565 --> 12:55.945
to make money a provider can simply turn away very

12:55.975 --> 12:57.105
expensive patients.

12:57.895 --> 13:00.825
It's quite illegal in many countries, in most countries.

13:01.005 --> 13:04.785
Uh, but it doesn't stop people, providers from hinting,

13:04.845 --> 13:06.985
for example, that your case may,

13:06.985 --> 13:09.865
your maybe better off being treated elsewhere.

13:10.525 --> 13:12.585
Uh, this is for your own benefit and this

13:12.725 --> 13:16.825
and some persuasion can drive some expensive patients

13:16.965 --> 13:18.785
to other providers, right?

13:18.785 --> 13:19.945
This is called dumping

13:20.085 --> 13:24.505
and cream skimming, refers to higher

13:25.055 --> 13:30.045
quality to attract more attractive to, to,

13:30.145 --> 13:33.845
to, to get more, profitable patients.

13:34.345 --> 13:35.405
So this is the idea.

13:35.475 --> 13:39.045
Well, in, in the case of services,

13:39.215 --> 13:43.595
maybe we are going to have very good pre-op services

13:43.815 --> 13:47.955
before the operation, before, the actual procedure.

13:48.415 --> 13:52.075
We, we will advise you on what to do, what not to do, and

13:52.075 --> 13:54.795
therefore this will enhance your experience.

13:55.675 --> 13:57.955
Although, you know, surgery may never be a good experience,

13:58.295 --> 14:01.995
it is going to be sort of a, a more welcoming strategy.

14:02.345 --> 14:06.565
Okay? So what is the catch?

14:06.635 --> 14:08.765
What what is the problem with prospective payment?

14:09.265 --> 14:12.165
You ask consumers, you ask providers

14:12.165 --> 14:13.245
to bear the full cost,

14:13.545 --> 14:15.325
but they don't have to bear the full cost.

14:15.385 --> 14:18.285
If they understand that they can turn away some very

14:18.285 --> 14:19.845
expensive or costly patients.

14:20.555 --> 14:24.055
So this idea is called dumping and cream ischemic, okay?

14:24.365 --> 14:25.535
Cost variations

14:25.535 --> 14:27.135
therefore may limit the power

14:27.135 --> 14:31.755
of prospective payment selection incentive.

14:32.175 --> 14:33.715
The so-called dumping

14:33.715 --> 14:35.355
and cream skimming,

14:37.445 --> 14:39.345
may be avoided by fee for surface.

14:40.205 --> 14:43.065
So I'm not going to argue that fee

14:43.065 --> 14:47.825
for surface is completely useless in the case of,

14:48.375 --> 14:49.385
cost variation.

14:50.045 --> 14:53.525
Um, then fever surface can eliminate some sort

14:53.525 --> 14:55.725
of undesirable dumping or cream scheme.

14:57.705 --> 15:01.315
It's still a fairly simpl simplistic scenario.

15:01.695 --> 15:03.235
But notice what I've done.

15:04.565 --> 15:07.085
I have started with the internalization principle

15:07.465 --> 15:08.805
and then I've started counting.

15:09.995 --> 15:12.605
I've got a mechanism, I've got payment method

15:13.435 --> 15:15.335
to incentivize a provider,

15:16.285 --> 15:19.225
and I've listed two things that the provider is doing,

15:19.535 --> 15:20.985
like cost and quality.

15:22.225 --> 15:24.585
I have successfully introduced more

15:24.605 --> 15:26.985
and more actions available to the provider,

15:27.935 --> 15:30.815
selecting patients, getting the good risk,

15:30.965 --> 15:33.375
getting the less expensive patients.

15:33.635 --> 15:36.735
Uh, so now I have sort of three problems.

15:37.395 --> 15:41.855
And now prospective payment has only one single instrument,

15:42.535 --> 15:45.255
although it does force the provider

15:45.955 --> 15:48.325
to internalized costs, right?

15:48.625 --> 15:52.805
As you gradually make the situation more

15:52.805 --> 15:57.385
and more complicated, if you allow, the provider

15:57.725 --> 16:00.585
to have more power over its environment,

16:00.895 --> 16:04.115
over its profit maximization attempt, you're gonna see

16:04.115 --> 16:05.955
that none of the mechanism

16:06.065 --> 16:08.595
that we have ever considered is sort

16:08.595 --> 16:10.435
of completely foolproof.

16:11.085 --> 16:12.955
There will be some sort of trade off

16:13.845 --> 16:17.265
the extent literature has sort of documented this kind

16:17.265 --> 16:19.385
of trade off in the past 30 years

16:19.385 --> 16:22.705
or so, we have expanded this literature into

16:22.815 --> 16:24.385
various scenarios.

16:25.595 --> 16:29.135
Uh, the extent literature is about making the analysis more

16:29.135 --> 16:30.575
realistic and more practical.

16:32.185 --> 16:36.965
First, I was just talking about providers

16:37.065 --> 16:39.925
as if they were just trying to maximize their profit.

16:40.855 --> 16:43.235
In many circumstances, this may not really be

16:44.055 --> 16:45.435
the correct assumption.

16:45.625 --> 16:47.355
They may be a little bit altruistic.

16:47.705 --> 16:50.075
They may love their patients to some extent

16:50.785 --> 16:52.635
they care about consumer's wellbeing.

16:53.415 --> 16:58.315
Now that element alone is not that hard to to handle.

16:59.445 --> 17:02.665
Uh, if you love your patients 50%, well, okay, I'm, I'm,

17:02.665 --> 17:05.105
I'm going to ask you to bear 50% of the cost.

17:05.615 --> 17:08.905
I'll reimburse you 50% of the cost, but not everything.

17:10.085 --> 17:11.305
The price will be adjusted

17:11.305 --> 17:14.385
because I know that I don't need a lot of incentive for you

17:14.385 --> 17:16.235
to provide for it, right?

17:16.255 --> 17:18.595
So this mixing fee for service

17:19.215 --> 17:22.995
and prospective payment together is a very effective way

17:23.695 --> 17:27.915
to account for, so-called provider heterogeneity.

17:29.325 --> 17:31.465
And that has been done over the years,

17:31.525 --> 17:32.585
in, in various papers.

17:34.885 --> 17:38.455
Then in more recent years, we have expanded sort of

17:39.075 --> 17:42.685
discussions about, for example, pay for performance.

17:43.455 --> 17:46.115
Uh, we are not just paying for the discharge.

17:46.535 --> 17:50.285
We like to assess how well the patient is, is feeling

17:50.615 --> 17:53.765
after a certain amount of time period.

17:54.615 --> 17:57.665
Is there any way that we can assess,

17:57.895 --> 17:59.705
performance by the provider?

17:59.875 --> 18:03.145
There, there are some ways to do that, especially in,

18:03.285 --> 18:06.425
in many cases, in more, long-term illnesses

18:06.815 --> 18:09.185
that we are mission and so on and so forth.

18:09.645 --> 18:14.155
Um, complication, okay, so that's one another

18:15.025 --> 18:18.275
element that the literature has consisted as a study.

18:19.905 --> 18:24.265
Then the internalization principle also applies to

18:24.845 --> 18:26.465
organizational incentives.

18:26.975 --> 18:29.705
What do I mean by that? Patients get sick.

18:31.105 --> 18:35.325
It may involve a multi-stage, treatment protocol.

18:36.065 --> 18:40.285
You may be seeing, a a primary care doctor

18:41.185 --> 18:43.805
and given the complexity of your illness,

18:43.805 --> 18:47.085
primary care Dr. May refer you to a specialist

18:47.705 --> 18:52.255
and a specialist may consult and provide treatment,

18:52.395 --> 18:54.775
or maybe the specialist can send you back to the

18:55.355 --> 18:57.685
primary care, with suggestions.

18:57.685 --> 19:00.565
And so, right. So how do you make sure that this kind

19:00.565 --> 19:04.925
of referrals, this kind of reorganization re redirecting,

19:05.305 --> 19:09.155
is, is correct that simply

19:09.745 --> 19:14.145
will require you to pay the entire group of doctors.

19:15.525 --> 19:16.825
And they, and,

19:16.845 --> 19:20.145
and in the literature is sort of called the,

19:20.505 --> 19:22.385
organization incentives and payment.

19:23.045 --> 19:24.785
We don't pay a single doctor,

19:24.965 --> 19:26.425
we don't pay a single hospital.

19:27.045 --> 19:31.625
We pay a health plan to handle a sort of population

19:32.125 --> 19:33.845
of patients.

19:34.495 --> 19:37.275
And this will be maybe made on, on a yearly

19:37.295 --> 19:38.995
or long term, longer term basis.

19:40.615 --> 19:45.145
So this, sort of a summary

19:45.445 --> 19:50.145
of, of the ba the basic issue, the incentive in payment,

19:50.145 --> 19:52.185
what has been developed over the years,

19:52.325 --> 19:55.545
and it's a very short, of course, very concise summary.

19:56.165 --> 19:58.745
The, the literature is, is getting very big

19:59.325 --> 20:03.875
and we have many technical issues in terms of theoretical

20:04.395 --> 20:07.675
analysis and also in terms of the empirical work

20:07.785 --> 20:08.715
that has been done.

00:00 - Healthcare is a very important part of any society, and it's particularly important in the US because it occupies up to 20% of GTP. There are various ways that we can think about healthcare and, today I'm just going to concentrate on this element called payment incentives.

00:23 - There are other issues to think about, like insurance and so on, but they will be left on another location. So why are payment incentive at all relevant? The basic issue is that there are sort of three parties in this, in this scenario. They're the consumers, they're the providers, and then they're the payers. It is like they interact. Consumers are patients. They want some services. Providers are medical personnel and facilities

00:56 - and healthcare professionals. And then the pay is a government, employers, insurance companies, public or private. It depends on various different countries. Basically, consumers demand and receive services from providers. The insurance will shield consumers from receiving, will shield consumers from part of the surface expenses. That's the whole point of health insurance because illness carries some financial risk.

01:29 - And the insurance part says that when you seek medical treatment, both or part of the medical expenses will be paid for by the insurance company. Now this creates an obvious issue, which I'll come to in just one moment. who pays the bill? If consumers don't pay, the payers would have, would have to pay. So how,

02:01 - providers paid would therefore determine their behavior? If the payer can make providers be efficient so much, the better. And what are the ways to make them efficient? in the old days, like 50 years ago, 60 years ago is Medicare in the US and many other countries use something called fee for service or cost reimbursement. Whatever cost a provider incurs to surface consumers,

02:37 - the payer will just reimburse the expenses. of course this may often include a profit margin. So the profit margin is, it can be an issue because the profit margin then create an incentive problems. There will be more services provided and maybe qualities Can be made inefficiently high to attract services demanded by, demanded by consumers. The cost reimbursement fee for service has been regarded as a major reason for the high cost in the us That's one of course base payment mechanism.

03:18 - And there's an obvious incentive issue, then a popular, and now very, very common payment mechanism is called prospective or fixed price mechanism. It works like this. It works in two steps. First, we sort of define an episode of care. Let's say a patient gets into a hospital, eventually the patient gets discharged, and let's say this is a pneumonia episode. At the end of the episode, appoint discharge at

03:52 - that primary diagnosis called pneumonia will be associated with the patient and the discharge. Now this category of pneumonia will be associated with a price. This price will be the payment that the payer makes to the provider. It is a fixed price mechanism. No matter what kind of resources the provider has incurred. No matter how many days the patient stays in the hospital, it will be the same price.

04:25 - And, and it is very different from cost reimbursement. Cost reimbursement says, well, if you stay longer, we'll pay you more, stay shorter, we'll pay you less fixed price. Prospective payment says no matter what kind of resources you use, provided that you know the the provider, the hospital has to treat you in a satisfactory way. the payment will remains the same. So what does it aim to achieve?

04:51 - Well, think about the following scenario. The two things that the provider does. The provider is gonna make quality decision and it's gonna make cost decisions. The quality decision can be associated with, nursing care, the, facilities, tests and any other things that, consumer, the patient may enjoy raising qualities will be sort of good for consumers and will raise demand, and consumers can be happier with higher quality.

05:28 - But of course, qualities can be very costly when costs are high, the hospital doesn't get paid more. So there is in some sense a way an incentive for the provider to reduce costs. Reduce costs, reduction of costs can happen in many ways. finding the right treatment, finding the appropriate treatment, and even the organization Of the hospital of the facility can sometimes affect costs.

06:02 - Okay, so these are the thing about this as the two decisions that a provider makes and, the cost reimbursement system that we, I talk about just, you know, few minutes ago, the fee for service says that cost will always be paid for, even if I reduce cost, my payment would not get me more profit because it would just be reducing the reimbursement for me. Okay? Now, prospective payment doesn't work in the same way.

06:35 - Okay? And then, and let me now tie tie it to this, the sort of the economic principles involved in all this, and I call this the internalization principle. It applies in a very general setup, is not limited to healthcare. It says the following. If A wants B to do something that A desires A wants B to do something, A cannot control B'S behavior A cannot co B's action, then

07:07 - B must be made to share a's preferences. In other words, B must be made to internalize ace benefits and costs. I'll give you a very simple, since we are at a university, a very simple example, why do we think we have midterms and final? Why not have just the final exam? You see, once we have a midterm, that will be an incentive for students to study during the midterm during the first half of the semester.

07:37 - And the final, of course, would be another way to motivate students to study during, during the second half of the semester. So if we want students to be studying throughout the semester, these two exams will be needed. So I as a professor warned my student to study gradually and consistently, and therefore I set up these two exams so that my preferences for their learning behavior will be more in, in line with their incentive to perform well in the exam.

08:11 - Now, of course, in quality choices to the satisfaction of the, of the payer, the provider makes two choices. so think about the following, another sort of principle, one and a half, the internalization principle, right? I, I want the payer to think of a way to internalize this cost and quality incentives, motivation, perspective by the provider. Okay? Now you think about that.

08:43 - Two things that I want to do. I need them to study first half and I want them to study second half of the semester. I need two exams. One exam isn't quite good enough, right? So you think about cost and quality choices and cost in quality incentive. You naturally think, well, we need two instruments. One needs an instrument to motivate cost incentives. One needs, an instrument to motivate, quality incentives.

09:16 - Now, what is the beauty of the prospective payment? What does the pro prospective payment say? Is the provider is responsible for cost no matter what the cost is, the payment from the payer to the provider is constant. In other words, it internalizes cost, so to speak. The provider under this payment mechanism has already one instrument to align the cost incentives. So what are we left behind with? We are left behind with the second instrument.

09:50 - The price, higher price would therefore encourage higher quality, lower price can, encourage or discourage higher quality. Okay, let's say you have this patient who has pneumonia being admitted to the hospital, right? So how many days to stay? okay, well, the hospital has to bear the cost. You cannot just send away the patient when the patient is very sick. the quality of care now depends on what, depends on

10:21 - how much the hospital is paid. If this price upon this charge is high, then we have an incentive to provide, would've an incentive to make the patient happier. So the price can be adjusted to align some incentive. The price can be made, can be set. So the provider to internalizes the benefit. So that's the internalization principle. The provider wants, the, the, the payer wants a provider to have

10:54 - efficient cost and quality decisions. And the way to do it is to ask the provider to internalize the cost and benefit the desirable part and the, and the costly part of the cost and quality incentives. We have one instrument because the prospective payment is a single price. However, the beauty of prospective payment is that cost is already internalized. By definition, the provider has to bear the cost.

11:25 - So we have, we have full efficiency in this, in this scenario. Now, it also brings out the issue about fee for service. Fee for service would never induce efficiency simply because cost is always reimbursed. That part of the sort of the equation of the incentive is never internal provided, does not internalize. Alright? So that's the sort of the literature I I sort of started, about 30 years ago is, one

11:58 - of the first papers to recognize that, providers behavior can be multifaceted. They have many things that they can do, some concern causing some concern, quality and different payment mechanism is steering them towards different kinds of behavior, right? It's simple model, but we always start with simple model because we can derive insights and then we can continue to work on the same principle, but relaxing various, restrictions.

12:30 - So what is the catch? way back when people already spotted some issues with prospective payment. the prospective payment suffers from some well-known problems. It's called dumping and cream skiing. So what is dumping? Dumping refers to the idea that to make money a provider can simply turn away very expensive patients. It's quite illegal in many countries, in most countries.

13:01 - but it doesn't stop people, providers from hinting, for example, that your case may, your maybe better off being treated elsewhere. this is for your own benefit and this and some persuasion can drive some expensive patients to other providers, right? This is called dumping and cream skimming, refers to higher quality to attract more attractive to, to, to, to get more, profitable patients.

13:34 - So this is the idea. Well, in, in the case of services, maybe we are going to have very good pre-op services before the operation, before, the actual procedure. We, we will advise you on what to do, what not to do, and therefore this will enhance your experience. Although, you know, surgery may never be a good experience, it is going to be sort of a, a more welcoming strategy. Okay? So what is the catch? What what is the problem with prospective payment?

14:09 - You ask consumers, you ask providers to bear the full cost, but they don't have to bear the full cost. If they understand that they can turn away some very expensive or costly patients. So this idea is called dumping and cream ischemic, okay? Cost variations therefore may limit the power of prospective payment selection incentive. The so-called dumping

14:33 - and cream skimming, may be avoided by fee for surface. So I'm not going to argue that fee for surface is completely useless in the case of, cost variation. then fever surface can eliminate some sort of undesirable dumping or cream scheme. It's still a fairly simpl simplistic scenario. But notice what I've done. I have started with the internalization principle

15:07 - and then I've started counting. I've got a mechanism, I've got payment method to incentivize a provider, and I've listed two things that the provider is doing, like cost and quality. I have successfully introduced more and more actions available to the provider, selecting patients, getting the good risk, getting the less expensive patients. so now I have sort of three problems.

15:37 - And now prospective payment has only one single instrument, although it does force the provider to internalized costs, right? As you gradually make the situation more and more complicated, if you allow, the provider to have more power over its environment, over its profit maximization attempt, you're gonna see that none of the mechanism that we have ever considered is sort of completely foolproof.

16:11 - There will be some sort of trade off the extent literature has sort of documented this kind of trade off in the past 30 years or so, we have expanded this literature into various scenarios. the extent literature is about making the analysis more realistic and more practical. First, I was just talking about providers as if they were just trying to maximize their profit. In many circumstances, this may not really be

16:44 - the correct assumption. They may be a little bit altruistic. They may love their patients to some extent they care about consumer's wellbeing. Now that element alone is not that hard to to handle. if you love your patients 50%, well, okay, I'm, I'm, I'm going to ask you to bear 50% of the cost. I'll reimburse you 50% of the cost, but not everything. The price will be adjusted because I know that I don't need a lot of incentive for you

17:14 - to provide for it, right? So this mixing fee for service and prospective payment together is a very effective way to account for, so-called provider heterogeneity. And that has been done over the years, in, in various papers. Then in more recent years, we have expanded sort of discussions about, for example, pay for performance. we are not just paying for the discharge. We like to assess how well the patient is, is feeling

17:50 - after a certain amount of time period. Is there any way that we can assess, performance by the provider? There, there are some ways to do that, especially in, in many cases, in more, long-term illnesses that we are mission and so on and so forth. complication, okay, so that's one another element that the literature has consisted as a study. Then the internalization principle also applies to organizational incentives.

18:26 - What do I mean by that? Patients get sick. It may involve a multi-stage, treatment protocol. You may be seeing, a a primary care doctor and given the complexity of your illness, primary care Dr. May refer you to a specialist and a specialist may consult and provide treatment, or maybe the specialist can send you back to the primary care, with suggestions. And so, right. So how do you make sure that this kind of referrals, this kind of reorganization re redirecting,

19:05 - is, is correct that simply will require you to pay the entire group of doctors. And they, and, and in the literature is sort of called the, organization incentives and payment. We don't pay a single doctor, we don't pay a single hospital. We pay a health plan to handle a sort of population of patients. And this will be maybe made on, on a yearly

19:37 - or long term, longer term basis. So this, sort of a summary of, of the ba the basic issue, the incentive in payment, what has been developed over the years, and it's a very short, of course, very concise summary. The, the literature is, is getting very big and we have many technical issues in terms of theoretical analysis and also in terms of the empirical work that has been done.

Faculti AI can make mistakes