[Federal Register Volume 85, Number 229 (Friday, November 27, 2020)]
[Rules and Regulations]
[Pages 76180-76259]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2020-26037]



[[Page 76179]]

Vol. 85

Friday,

No. 229

November 27, 2020

Part II





Department of Health and Human Services





-----------------------------------------------------------------------





Centers for Medicare & Medicaid Services





42 CFR Part 513





-----------------------------------------------------------------------





Most Favored Nation (MFN) Model; Final Rule

  Federal Register / Vol. 85 , No. 229 / Friday, November 27, 2020 / 
Rules and Regulations  

[[Page 76180]]


-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Centers for Medicare & Medicaid Services

42 CFR Part 513

[CMS-5528-IFC]
RIN 0938-AT91


Most Favored Nation (MFN) Model

AGENCY: Centers for Medicare & Medicaid Services (CMS), HHS.

ACTION: Interim final rule with comment period.

-----------------------------------------------------------------------

SUMMARY: This interim final rule with comment period (IFC) implements 
the Most Favored Nation (MFN) Model, a new Medicare payment model under 
section 1115A of the Social Security Act (the Act). The MFN Model will 
test whether more closely aligning payment for Medicare Part B drugs 
and biologicals (hereafter, referred to as ``drugs'') with 
international prices and removing incentives to use higher-cost drugs 
can control unsustainable growth in Medicare Part B spending without 
adversely affecting quality of care for beneficiaries.

DATES: Effective date: These regulations are effective on November 27, 
2020.
    Comment date: To be assured consideration, comments must be 
received at one of the addresses provided below, no later than 5 p.m. 
on January 26, 2021.

ADDRESSES: In commenting, please refer to file code CMS-5528-IFC. 
Because of staff and resource limitations, we cannot accept comments by 
facsimile (FAX) transmission.
    Comments, including mass comment submissions, must be submitted in 
one of the following three ways (please choose only one of the ways 
listed):
    1. Electronically. You may submit electronic comments on this 
regulation to http://www.regulations.gov. Follow the ``Submit a 
comment'' instructions.
    2. By regular mail. You may mail written comments to the following 
address ONLY: Centers for Medicare & Medicaid Services, Department of 
Health and Human Services, Attention: CMS-5528-IFC, P.O. Box 8013, 
Baltimore, MD 21244-8013.
    Please allow sufficient time for mailed comments to be received 
before the close of the comment period.
    3. By express or overnight mail. You may send written comments to 
the following address ONLY: Centers for Medicare & Medicaid Services, 
Department of Health and Human Services, Attention: CMS-5528-IFC, Mail 
Stop C4-26-05, 7500 Security Boulevard, Baltimore, MD 21244-1850.
    For information on viewing public comments, see the beginning of 
the SUPPLEMENTARY INFORMATION section.

FOR FURTHER INFORMATION CONTACT: Andrew York, 410-786-7400.

SUPPLEMENTARY INFORMATION: 
    Inspection of Public Comments: All comments received before the 
close of the comment period are available for viewing by the public, 
including any personally identifiable or confidential business 
information that is included in a comment. We post all comments 
received before the close of the comment period on the following 
website as soon as possible after they have been received: http://www.regulations.gov. Follow the search instructions on that website to 
view public comments.

I. Executive Summary

A. Purpose

    High drug prices are impacting the wallets of Medicare 
beneficiaries, especially during the Coronavirus disease 2019 Public 
Health Emergency (PHE). Increases in drug prices are accelerating at a 
rate that significantly outpaces the growth in spending on other 
Medicare Part B services, and prices in the United States (U.S.) for 
most Medicare Part B drugs with the highest Medicare spending far 
exceed prices in other countries. Specifically, drugs have consistently 
been a major contributor to the overall Medicare Part B spending trend. 
Medicare Part B Fee-For-Service (FFS) spending for separately payable 
physician-administered drugs and drugs furnished in a hospital 
outpatient department represented about 11 percent of Medicare Part B 
FFS benefit spending in 2015, but accounted for about 37 percent of the 
change in Medicare Part B FFS benefit spending from 2015 to 2020, and 
spending on these Medicare Part B FFS drugs increased to represent 
roughly 14 percent of Medicare Part B FFS benefit spending in 2019.\1\ 
In addition to the continued growth in spending, the U.S. already pays 
almost twice as much on average as other developed countries pay. In 
one analysis of 27 drugs, acquisition costs in the U.S. were 1.8 times 
higher than in comparator countries.\2\ A more recent analysis using 
the prescription drugs and countries in the MFN Model suggests Medicare 
Part B paid at least 2.05 times as much as other higher-income 
countries in 2018.\3\ The Centers for Medicare & Medicaid Services' 
(CMS) Center for Medicare and Medicaid Innovation (Innovation Center) 
is taking action on President Trump's goal to lower drug costs and 
seeking to realign financial incentives by implementing the Most 
Favored Nation (MFN) Model as described in this IFC.
---------------------------------------------------------------------------

    \1\ 2020 Annual Report of the Boards of Trustees of the Federal 
Hospital Insurance and Federal Supplementary Medical Insurance Trust 
Funds. Accessed via: https://www.cms.gov/files/document/2020-medicare-trustees-report.pdf.
    \2\ ``Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' accessed via https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures.
    \3\ El-Kilani Z, Finegold K, Mulcahy A, and Bosworth A. Medicare 
FFS Part B and International Drug Prices: A Comparison of the Top 50 
Drugs. Washington, DC: Office of the Assistant Secretary for 
Planning and Evaluation, U.S. Department of Health and Human 
Services. November 20, 2020 (https://aspe.hhs.gov/pdf-report/medicare-ffs-part-b-and-international-drug-prices).
---------------------------------------------------------------------------

    Medicare pays substantially more than other countries for many of 
the highest-cost Medicare Part B drugs that beneficiaries receive in an 
outpatient setting for which Medicare Part B allows separate 
payment.\4\ In many instances, Medicare pays more than twice as much 
for certain drugs as other countries do.5, 6 This is because 
Medicare generally establishes the payment for separately payable 
Medicare Part B drugs using the methodology in section 1847A of the 
Act. In most cases, this means payment is based on the Average Sales 
Price (ASP) plus a statutorily mandated 6 percent add-on. Under this 
methodology, the Medicare program does not get the benefit of the 
substantial discounts provided in other

[[Page 76181]]

countries, because ASP is calculated using only the prices that 
manufacturers charge to certain U.S.-based purchasers. ASP-based 
payments may encourage the use of more expensive drugs because the 
dollar amount of the 6 percent add-on portion is larger for drugs with 
higher ASPs.\7\ As MedPAC noted in its June 2017 Report, ``Although, in 
some cases, drugs with patent protection may face competition from 
other brand drugs in the same therapeutic class, price competition 
between such products may be limited because the [Medicare] Part B drug 
payment system is not structured to facilitate competition among brand 
products with similar health effects.'' \8\ Thus, the ASP-based payment 
approach currently used in Medicare Part B may not promote price 
competition or provide sufficient incentive to minimize avoidable 
costs.
---------------------------------------------------------------------------

    \4\ ``Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' accessed via https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures; El-Kilani Z, Finegold K, 
Mulcahy A, and Bosworth A. Medicare FFS Part B and International 
Drug Prices: A Comparison of the Top 50 Drugs. Washington, DC: 
Office of the Assistant Secretary for Planning and Evaluation, U.S. 
Department of Health and Human Services. November 20, 2020 (https://aspe.hhs.gov/pdf-report/medicare-ffs-part-b-and-international-drug-prices).
    \5\ ``Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' accessed via https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures; El-Kilani Z, Finegold K, 
Mulcahy A, Bosworth A. Medicare FFS Part B and International Drug 
Prices: A Comparison of the Top 50 Drugs. Washington, DC: Office of 
the Assistant Secretary for Planning and Evaluation, U.S. Department 
of Health and Human Services. November 20, 2020 (https://aspe.hhs.gov/pdf-report/medicare-ffs-part-b-and-international-drug-prices).
    \6\ Individual countries differ in the regulatory processes and 
standards governing approval of drugs and biologicals. Use of 
international drug prices in the MFN Model should not be interpreted 
to connote FDA approval or to otherwise describe any scientific or 
regulatory relationship between U.S.-approved and non-U.S.-approved 
products.
    \7\ MedPAC, June 2017, ``Medicare Part B Drug Payment Policy 
Issues,'' accessed via http://medpac.gov/docs/default-source/reports/jun17_ch2.pdf.
    \8\ MedPAC, June 2017, ``Medicare Part B Drug Payment Policy 
Issues,'' accessed via: http://medpac.gov/docs/default-source/reports/jun17_ch2.pdf.
---------------------------------------------------------------------------

    The MFN Model aims to take a global approach to calculating 
Medicare Part B drug payment amounts, by testing a new payment 
methodology that takes into account the discounts that other countries 
enjoy, and pays providers and suppliers with a fixed add-on amount that 
does not reward the use of higher-cost drugs. We expect that this model 
will reduce Medicare program expenditures while preserving or enhancing 
quality of care furnished to Medicare beneficiaries, and will lower 
beneficiary cost-sharing through lower drug payment amounts. The MFN 
Model will be tested in all states and U.S. territories by the CMS 
Innovation Center for 7 performance years, from January 1, 2021 to 
December 30, 2027.

B. Summary of the Major Provisions

    The MFN Model will focus on a select cohort of separately payable 
Medicare Part B drugs. This cohort will initially include 50 single 
source drugs and biologicals (including biosimilar biological products) 
that encompass a high percentage of Medicare Part B drug spending. The 
MFN Model will require mandatory participation. Participants in the MFN 
Model will include all providers and suppliers that participate in the 
Medicare program and submit a separately payable claim for an MFN Model 
drug with limited exceptions, such as providers and suppliers that are 
paid for separately payable Medicare Part B drugs based on reasonable 
costs. The vast majority of providers and suppliers that furnish 
separately payable Medicare Part B drugs are physicians and non-
physician practitioners, supplier groups (such as a group of physicians 
or other practitioners), hospital outpatient departments (HOPDs), 
including on- or off-campus provider-based departments (PBDs), whether 
paid under the outpatient prospective payment system (OPPS) or the 
physician fee schedule (PFS), and ambulatory surgical centers (ASCs) 
paid under the ASC Payment System. Claims from these providers and 
suppliers will encompass approximately 88 percent of the annual 
Medicare Part B spending on the drugs we selected for inclusion in the 
MFN Model beginning in performance year 1. Other types of providers and 
suppliers that furnish separately payable selected drugs will also be 
required to participate in the MFN Model, but they may not often 
furnish the selected drugs or may not typically receive separate 
payment for Medicare Part B drugs.
    The MFN Model will--
     Calculate the payment amount for MFN Model drugs based on 
a price that reflects the lowest per capita Gross Domestic Product-
adjusted (GDP-adjusted) price of any non-U.S. member country of the 
Organisation for Economic Co-operation and Development (OECD) with a 
GDP per capita \9\ that is at least sixty percent of the U.S. GDP per 
capita, based on available data;
---------------------------------------------------------------------------

    \9\ For the purposes of this IFC, GDP means GDP based on 
purchasing power parity (PPP), rather than nominal GDP. A nation's 
GDP at purchasing power parity (PPP) exchange rate is the sum value 
of all goods and services produced in the country valued at prices 
prevailing in the U. S.
---------------------------------------------------------------------------

     Make an alternative add-on payment for MFN Model drugs 
that will remove or reduce the financial incentive to prescribe higher-
cost drugs more frequently; and
     Reduce beneficiary cost sharing on MFN Model drugs.

C. Summary of Costs and Benefits

    We believe the MFN Model will substantially lower drug payment 
amounts for the most costly Medicare Part B drugs, thereby lowering 
program expenditures and out-of-pocket costs for beneficiaries. As 
discussed in more detail in section VI. of this IFC, we estimate that 
the MFN Model will result in substantial overall Medicare savings 
during the 7-year model performance period (that is, 28 calendar 
quarters). In the CMS Office of the Actuary (OACT) estimate, OACT 
estimates savings of roughly $64.4 billion in Medicare FFS benefits, 
$49.6 billion in Medicare Advantage (MA) payments, and $9.9 billion in 
Medicaid \10\ spending ($5.7 billion in federal payments and $4.3 
billion in state payments). Overall, OACT estimates that the MFN Model 
will result in savings of $85.5 billion, net of the associated change 
in the Part B premium, in Medicare Part B spending. In addition, OACT 
estimates that all beneficiaries will save a total of $28.5 billion 
from a reduction in the Medicare Part B premium as a result of the MFN 
Model, and will also see their coinsurance reduced. In the HHS Office 
of the Assistant Secretary for Planning and Evaluation (ASPE) estimate, 
ASPE estimates roughly a net reduction of $87.8 billion in spending on 
MFN Model drugs by the federal government, state governments, and 
beneficiaries over the 7 years of the model. We note that there is much 
uncertainty around the assumptions for both the OACT and ASPE estimates 
and refer readers to section VI. of this IFC for a more complete 
discussion of potential impacts of the MFN Model.
---------------------------------------------------------------------------

    \10\ Medicaid savings estimates do not include impacts of 
changes in Average Manufacturer Price (AMP) and Best Price on 
manufacturer rebates under the Medicaid Drug Rebate Program.
---------------------------------------------------------------------------

II. Background on Need for Regulatory Action

    On May 11, 2018, President Trump released his Blueprint to Lower 
Drug Prices and Reduce Out-of-Pocket Costs,\11\ which outlined the 
steps his administration is taking to combat high drug prices, end 
foreign freeloading, and spur biomedical innovation.\12\
---------------------------------------------------------------------------

    \11\ American Patients First: The Trump Administration Blueprint 
to Lower Drug Prices and Reduce Out-of-Pocket Costs, Available at: 
https://www.hhs.gov/sites/default/files/AmericanPatientsFirst.pdf?language=es.
    \12\ ``President Donald J. Trump's Blueprint To Lower Drug 
Prices,'' accessed via: https://www.whitehouse.gov/briefings-statements/president-donald-j-trumps-blueprint-lower-drug-prices/.
---------------------------------------------------------------------------

    On October 25, 2018, CMS released an advance notice of proposed 
rulemaking (ANPRM) (83 FR 54546) \13\ (hereafter called the October 
2018 ANPRM) describing a potential model, referred to in the October 
2018 ANPRM as the International Pricing Index Model (IPI), that would 
test whether changing the payment amount for selected Medicare Part B 
drugs would reduce Medicare expenditures and preserve or enhance 
quality of care. In the October 2018 ANPRM, we sought comment on a 
model test that would--
---------------------------------------------------------------------------

    \13\ International Pricing Index Model for Medicare Part B 
Drugs; Medicare Program, 83 Fed. Reg (210) 54246 (Oct 30, 2018) 
available at: https://www.govinfo.gov/content/pkg/FR-2018-10-30/pdf/2018-23688.pdf.
---------------------------------------------------------------------------

     Calculate the Medicare payment amount for selected 
Medicare Part B

[[Page 76182]]

drugs to be phased down to more closely align with international 
prices;
     Allow private-sector vendors to negotiate prices for 
drugs, take title to drugs, and compete for physician and hospital 
business;
     Increase the drug add-on payment to reflect 6 percent of 
historical drug costs; and
     Pay physicians and hospitals the add-on based on a set 
payment amount structure.
    We considered the comments that we received in response to the 
October 2018 ANPRM in developing the MFN Model described in this IFC. 
In addition to considering these comments, we considered feedback and 
suggestions from a broad set of stakeholders gathered through comments 
on the President's Blueprint and through numerous meetings with 
stakeholders.
    President Trump discussed an Executive Order (E.O.) regarding an 
MFN payment model for Medicare Part B drugs on July 24, 2020, and 
subsequently published a superseding Executive Order on Lowering Drug 
Prices by Putting America First on September 13, 2020.\14\ In response 
to the September 13, 2020 Executive Order, we will implement the MFN 
Model described in this IFC.
---------------------------------------------------------------------------

    \14\ Executive Order 13948, https://www.govinfo.gov/content/pkg/FR-2020-09-23/pdf/2020-21129.pdf.
---------------------------------------------------------------------------

A. Medicare Part B Drug Benefit and ASP Payment Methodology

    Medicare Part B includes a limited drug benefit for drugs and 
biologicals described in section 1861(t) of the Act. The majority of 
drugs paid under Medicare Part B generally fall into three categories: 
Drugs furnished incident to a physician's service in the physician 
office, HOPD, or other outpatient setting; drugs administered via a 
covered item of durable medical equipment (DME); and other categories 
of drugs specified by statute (generally in section 1861(s)(2) of the 
Act).
    Many drugs covered under Medicare Part B are administered via 
injection or infusion in a physician's office, an HOPD, and certain 
other outpatient settings, such as ASCs, and, when Medicare allows 
separate payment for these drugs, the payment limit is typically based 
on the methodology described in section 1847A of the Act. The payment 
amount for these drugs does not include payment for administering the 
drug to a beneficiary; payment for drug administration services is made 
in accordance with the applicable payment policy for the setting in 
which the drug was furnished, such as the Physician Fee Schedule (PFS) 
(https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/PhysicianFeeSched/index.html), the Hospital Outpatient Prospective 
Payment System (OPPS) (https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/HospitalOutpatientPPS/index.html), or the Ambulatory 
Surgical Center Payment System (https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/ASCPayment/index.html). Medicare Part B also 
allows separate payment for drugs in less common situations such as 
osteoporosis drugs furnished by a home health agency, and when a 
beneficiary does not have benefits available under the Part A program.
    The payment methodology for drugs described in section 1847A of the 
Act is generally based on the volume-weighted ASP for all National Drug 
Codes (NDCs) that are assigned to a Healthcare Common Procedure Coding 
System (HCPCS) code for the drug plus a 6 percent add-on. The volume-
weighted ASP for a HCPCS code is calculated quarterly using 
manufacturer-submitted data \15\ on sales to all purchasers (with 
limited exceptions as articulated in section 1847A(c)(2) of the Act, 
such as sales at nominal charge and sales exempt from Medicaid best 
price \16\) with manufacturers' rebates, discounts, and price 
concessions included in the ASP calculation (that is, the sales price 
is net of these rebates, discounts, and price concessions). The ASP+6 
percent payment amount that Medicare pays for an individual Medicare 
Part B drug claim generally does not vary based on the exact price an 
individual provider or supplier pays to acquire the drug. In the case 
of multiple source drugs, the price of a brand name drug and its 
generic equivalent(s) included in the same billing code are averaged 
together to determine the payment allowance.\17\ As noted earlier, this 
payment methodology may create an incentive for the use of more 
expensive drugs, but, as noted in the MedPAC report (and by sources 
cited in the report; pages 68 and 79), an add-on may be needed to 
account for handling and overhead costs and additional mark-up in 
distribution channels that are not captured in the manufacturer-
reported ASP.
---------------------------------------------------------------------------

    \15\ OMB Control Number 0938-0921.
    \16\ Best price is defined in section 1927(c)(1)(C) of the Act.
    \17\ Under section 3139 of the Affordable Care Act (Pub. L. 111-
148) the add-on amount for a biosimilar is based on the ASP of the 
reference product. Biosimilars are not grouped together with one 
another or the reference product for payment purposes.
---------------------------------------------------------------------------

    Currently, under Medicare Part B, beneficiaries' cost-sharing \18\ 
is generally 20 percent of the Medicare-allowed amount. The term 
``Medicare-allowed amount'' means the maximum amount that a provider or 
supplier will be paid for a covered health care service or drug. 
However, for items and services paid under the OPPS, beneficiaries are 
only financially responsible for a copayment amount up to the amount of 
the inpatient hospital deductible.\19\ Medicare pays for the remaining 
portion of the Medicare-allowed amount.\20\
---------------------------------------------------------------------------

    \18\ Not including the annual deductible.
    \19\ Section 1833(t)(8)(C)(i) of the Act limits the amount of 
beneficiary copayment that may be collected for a procedure 
performed in a year to the amount of the inpatient hospital 
deductible for that year. This limit is $1,408 in 2020.
    \20\ 2020 Medicare Parts A & B Premiums and Deductibles: Fact 
Sheet, available at: https://www.cms.gov/newsroom/fact-sheets/2020-medicare-parts-b-premiums-and-deductibles.
---------------------------------------------------------------------------

B. Medicare and Beneficiary Spending

    Medicare Part B spending for separately payable physician-
administered drugs and drugs furnished in hospital outpatient 
departments represented about 11 percent of Medicare Part B FFS 
spending in 2015 but increased to represent roughly 14 percent of 
Medicare Part B FFS spending in 2019; spending on these Medicare Part B 
separately payable drugs accounted for about 37 percent of the change 
in Medicare Part B FFS spending from 2015 to 2019. Furthermore, 
Medicare Part B FFS spending per capita for separately payable drugs 
has increased at an average annual rate of 11.5 percent over this same 
period while Medicare Part B FFS spending per capita has increased by 
3.8 percent. From 2015 to 2019, Medicare Part B spending for separately 
payable drugs increased from $19.4 billion to $29.8 billion (a nearly 
55-percent increase) with per capita spending increasing from $583 to 
$900. This increase in Medicare Part B FFS spending for separately 
payable drugs during this period reflects increases in the prices of 
drugs, introduction of new drugs, changes in utilization of these 
drugs, changes in Medicare Part B FFS enrollment, and changes in the 
mix of drugs for those beneficiaries who received them.\21\ Since 
beneficiaries

[[Page 76183]]

without supplemental insurance typically pay 20 percent of the 
Medicare-allowed amount, as described in section II.A. of this IFC, 
they have faced similar increases in spending on Medicare Part B drugs 
as has Medicare.\22\
---------------------------------------------------------------------------

    \21\ The average annual growth in number of Medicare Part B FFS 
beneficiaries was less than 0 percent from 2015 to 2019, so the 
change in Medicare Part B beneficiaries does not fully account for 
the average annual growth (11.4 percent) in Medicare Part B spending 
for physician-administeredpayable drugs. Instead, the increase 
during this period is more fully explained by increases in the 
prices of drugs, introduction of new drugs, changes in drug 
utilization, and changes in the mix of drugs than by increases in 
Medicare enrollment.
    \22\ In 2016, 8 in 10 beneficiaries in traditional Medicare (81 
percent) had some type of supplemental insurance (which typically 
covers some or all of Medicare Part A and Medicare Part B cost-
sharing), including employer-sponsored insurance (30 percent), 
Medigap (29 percent), and Medicaid (22 percent). Nearly 1 in 5 
beneficiaries in traditional Medicare (19 percent)--6.1 million 
beneficiaries overall--had no source of supplemental coverage in 
2016. https://www.kff.org/medicare/issue-brief/sources-of-supplemental-coverage-among-medicare-beneficiaries-in-2016/.
---------------------------------------------------------------------------

    A new Issue Brief from the Office of the Assistant Secretary for 
Planning and Evaluation (ASPE) provides additional evidence of the need 
for the rule. Between 2006 and 2017, Medicare Part B FFS drug spending 
per enrollee grew at 8.1 percent, more than twice as high as per capita 
spending on Medicare Part D (3.4 percent) and nearly three times as 
high as overall retail prescription per capita drug spending (2.9 
percent). Spending and enrollment projections by OACT for the 2021 
President's Budget suggest that per capita spending on Medicare Part B 
physician-administered drugs and separately payable hospital outpatient 
drugs will grow at a very similar annual rate of 8 percent between 2020 
and 2027, before consideration of any COVID-19 pandemic impacts.\23\ 
Because biologics account for about 77 percent of Medicare Part B FFS 
prescription drug spending, there has been little opportunity to reduce 
Medicare Part B spending growth through generic substitution, as has 
occurred in Medicare Part D and in retail pharmacy overall.\24\
---------------------------------------------------------------------------

    \23\ ASPE analysis of OACT spending and enrollment projections.
    \24\ Nguyen X. Nguyen and Steve Sheingold. Medicare Part B 
Drugs: Trends in Spending and Utilization, 2006-2017. Washington, 
DC: Office of the Assistant Secretary for Planning and Evaluation, 
U.S. Department of Health and Human Services. November 20, 2020 
(https://aspe.hhs.gov/pdf-report/medicare-part-b-drugs-spending-and-utilization).
---------------------------------------------------------------------------

C. Relative High Price of Medicare Part B Drugs

    Drug acquisition costs in the U.S. exceed those in Europe, Canada, 
and Japan, according to an October 2018 ASPE analysis \25\ of Medicare 
Part B physician-administered drugs. This finding was generally 
consistent with the existing evidence base as described in the HHS 
analysis's background section, which found peer-reviewed literature on 
this topic to be relatively limited and dated, but with similar 
findings of higher drug prices in the U.S. compared to other 
countries.\26\ The HHS analysis compared U.S. drug acquisition costs 
for a set of Medicare Part B physician-administered drugs to 
acquisition costs in 16 other developed economies--Austria, Belgium, 
Canada, Czechia, Finland, France, Germany, Greece, Ireland, Italy, 
Japan, Portugal, Slovakia, Spain, Sweden, and the United Kingdom 
(UK).\27\ The main analysis in the HHS report focused on 27 drugs 
accounting for 64 percent of total Medicare Part B drug spending in 
2016.\28\ Among the 27 drugs included in the analysis, acquisition 
costs in the U.S. were 1.8 times higher than in comparator countries. 
Acquisition cost ratios ranged from U.S. prices being on par with 
international prices for one of the 27 drugs, to U.S. prices being up 
to 7 times higher than the international prices for others. There was 
variability across the 16 countries in the study as well, with no one 
country consistently acquiring drugs at the lowest prices. The U.S. had 
the highest drug prices for 19 of the 27 products.\29\
---------------------------------------------------------------------------

    \25\ Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' accessed via https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures.
    \26\ ``Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' accessed via https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures.
    \27\ Please refer to the HHS report (``Comparison of U.S. and 
International Prices for Top Medicare Part B Drugs by Total 
Expenditures'' accessed via https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures) for more information on the countries selected 
for analysis.
    \28\ ``Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' accessed via https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures.
    \29\ The ASPE report utilized ex-manufacturer prices (sometimes 
called the ex-factory price) stated in U.S. currency on the 
transaction date. The report defines ex-manufacturer prices as the 
price received by manufacturers of a product, including discounts 
applied at the point of sale.
---------------------------------------------------------------------------

    A new ASPE Issue Brief updates the earlier analysis for the set of 
Medicare Part B drugs and the set of countries in the MFN Model. In 
2018, based on available data, ASP rates were at least 2.05 times the 
value-weighted average price for these drugs in OECD countries with per 
capita GDP at least 60 percent of that in the U.S.\30\
---------------------------------------------------------------------------

    \30\ El-Kilani Z, Finegold K, Mulcahy A, and Bosworth A. 
Medicare FFS Part B and International Drug Prices: A Comparison of 
the Top 50 Drugs. Washington, DC: Office of the Assistant Secretary 
for Planning and Evaluation, U.S. Department of Health and Human 
Services. November 20, 2020 (https://aspe.hhs.gov/pdf-report/medicare-ffs-part-b-and-international-drug-prices).
---------------------------------------------------------------------------

    The results of these reports demonstrate that, save for a few 
outlier cases, the U.S. prices used to calculate ASP rates are 
significantly higher than the prices in international comparator 
countries.\31\ Based on this significant difference, which aligns with 
the analysis we present in this IFC, we will test the impact of more 
closely aligning payment for Medicare Part B drugs and biologicals with 
international prices in the MFN Model.
---------------------------------------------------------------------------

    \31\ ASP is defined in statute, and based on sales in the U.S.
---------------------------------------------------------------------------

III. Provisions of the Interim Final Rule With Comment Period

A. Model Performance Period

    In part 513, we codify the MFN Model that will be tested for 7 
performance years. We define ``model performance period'' to mean 
January 1, 2021, the date the model will begin, through December 31, 
2027. We are testing a 7-year performance period because it will allow 
a smooth transition to the MFN Price (described in section III.E.5. of 
this IFC) by performance year 4 and adequate duration to understand the 
impact of the MFN Model. As discussed in section III.N. of this IFC, we 
will assess for potential impacts of the MFN Model across quarterly 
time periods throughout the performance period. Further, we will assess 
initial impacts of the MFN Model on quality of care, including access 
to drugs, prior to beginning performance year 5.

B. Defined Population

    Our goal is to include all beneficiaries who are furnished an MFN 
Model drug by an MFN participant and who, on the date of service, are 
enrolled in Medicare Part B, have Medicare as the primary payer, and 
are not covered under Medicare Advantage or any other group health 
plan, including a United Mine Workers of America health plan, hereafter 
called MFN beneficiaries. Thus, the defined population for the MFN 
Model will be Medicare FFS beneficiaries who receive an MFN Model drug 
from an MFN participant where payment for such drug is allowed under 
the MFN Model. We define the term ``MFN beneficiary'' in Sec.  513.2.
    Testing the model in the population of beneficiaries who receive 
drugs with high annual Medicare Part B spending allows the MFN Model 
payment to apply to a broad set of conditions, drugs, medical 
specialties, clinical settings, and localities rather than having MFN 
Model payment focused on a particular clinical presentation, course of 
treatment or single type of care setting. Defining the population in 
this

[[Page 76184]]

manner allows CMS to observe the implications of a global approach to 
calculating Medicare Part B drug payment amounts and an alternative 
add-on approach across a broad set of providers and suppliers and 
beneficiaries, as well as a large set of manufacturers. Learnings from 
the MFN Model will inform CMS and other stakeholders about the effect 
of applying the innovative payment model to a broad set of drugs on a 
diverse set of beneficiaries and to the Medicare program.

C. MFN Participants

1. Eligible Providers and Suppliers
    A majority of Medicare spending on separately payable Medicare Part 
B drugs is for drugs that are furnished incident to a physician's 
service (see section 1861(s)(2)(A) of the Act), in a HOPD (see section 
1861(s)(2)(B) of the Act), including in an on- or off-campus PBD 
(regardless of whether those PBDs are excepted or nonexcepted),\32\ or 
in an ASC (see section 1832(a)(2)(F)(i) of the Act). Depending upon the 
circumstances, Medicare Part B allows separate payment for drugs to 
other providers and suppliers, such as pharmacies, home health 
agencies, hospices, radiation therapy centers, independent diagnostic 
testing facilities, ambulance suppliers, durable medical equipment 
(DME) suppliers, mass immunization suppliers, inpatient hospitals (when 
Part A payment is not permitted), and other types of providers and 
suppliers. Our goal is to broadly include providers and suppliers that 
receive separate payment for MFN Model drugs as MFN participants, with 
limited exceptions. MFN participants will consist of Medicare 
participating providers and suppliers that submit a claim for a 
separately payable drug that is an MFN Model drug furnished to an MFN 
beneficiary, unless otherwise excluded.\33\ Because separately payable 
Medicare Part B drugs (that is, potential MFN Model drugs) are most 
often furnished by physicians, non-physician practitioners, supplier 
groups (such as group practices), hospitals that are paid under the 
OPPS as defined in 42 CFR 419.20 (including off-campus PBDs paid under 
the PFS), and ASCs, these providers and suppliers will represent the 
vast majority of MFN participants. Other types of providers and 
suppliers (that are not excluded) also will be MFN participants to the 
extent that they submit a claim for an MFN Model drug furnished to an 
MFN beneficiary. For example, a home health agency that receives 
separate payment for an osteoporosis drug (defined in section 1861(kk) 
of the Act) will be an MFN participant if such drug is an MFN Model 
drug and the home health agency furnishes such drug to an included 
beneficiary and a claim is submitted.
---------------------------------------------------------------------------

    \32\ That is, regardless of whether those PBDs are excepted or 
nonexcepted under section 1833(t)(21)(B)(ii) of the Act, as added by 
section 603 of the Bipartisan Budget Act of 2015 (Pub. L. 114-74).
    \33\ These providers and suppliers will be included as 
participants in the MFN Model only if they participate in Medicare; 
this means that nonparticipating physicians and non-physician 
practitioners will not be MFN participants and will continue to be 
paid in accordance with current program policies.
---------------------------------------------------------------------------

    We will exclude certain types of providers and suppliers that are 
ultimately not paid for drugs based on ASP as well as those who are 
subject to the hold harmless provision in section 1833(t)(7)(D)(ii) of 
the Act. Thus, in Sec.  513.100(c), we exclude from the MFN Model the 
following providers and suppliers: Children's hospitals (defined under 
section 1886(d)(1)(B)(iii) of the Act); PPS-exempt cancer hospitals 
(defined under section 1886(d)(1)(B)(v) of the Act); critical access 
hospitals (CAHs) (defined under section 1820 of the Act); Indian Health 
Service (IHS) facilities (described in section 1880 of the Act), except 
when MFN Model drugs are furnished and such service is described in 
section 1880(e)(2)(B) of the Act; Rural Health Clinics (RHCs) (defined 
under section 1861(aa)(2) of the Act); Federally Qualified Health 
Centers (FQHCs) (defined under section 1861(aa)(4) of the Act); 
hospitals that are not subsection (d) hospitals (as defined in section 
1886(d)(1)(B) of the Act) and are paid on the basis of reasonable costs 
subject to a ceiling under section 1886(b) of the Act; and extended 
neoplastic disease care hospitals (defined in section 1886(d)(1)(B)(vi) 
of the Act). In addition, for the first quarter and second quarter of 
performance year 1, we will exclude acute care hospitals that 
participate in a CMS Innovation Center model under which they are paid 
for outpatient hospital services furnished to Medicare FFS 
beneficiaries, including MFN Model drugs, on a fully capitated or 
global budget basis in accordance with a waiver under such model of 
section 1833(t) of the Act. This exclusion, codified at Sec.  
513.100(c)(9), will apply during the first quarter and second quarter 
of performance year 1, and only if the hospital participates in a CMS 
Innovation Center model under which it is paid on a fully capitated or 
global budget basis. As codified at Sec.  513.100(c)(10), for the third 
quarter of performance year 1 (that is, beginning July 1, 2021) and 
beyond, acute care hospitals that participate in a CMS Innovation 
Center model under which they are paid for outpatient hospital services 
furnished to Medicare FFS beneficiaries, including MFN Model drugs, on 
a fully capitated or global budget basis in accordance with a waiver 
under such model of section 1833(t) of the Act will be excluded from 
the MFN Model if the parameters of the other CMS Innovation Center 
model adjust for the difference in payment for MFN Model drugs between 
the MFN Model and non-MFN Model drug payments such that savings under 
the MFN Model are incorporated into the other CMS Innovation Center 
model's parameters (for example, the annual global budget) for the 
duration of the MFN Model. Thus, acute care hospitals that are 
participating in the Maryland Total Cost of Care Model will not be MFN 
participants during the first two calendar quarters of 2021 while they 
are paid on a fully capitated or global budget basis. Further, if the 
parameters of the Maryland Total Cost of Care Model have been updated 
to adjust for the difference in payment for MFN Model drugs between the 
MFN Model and non-MFN Model drug payments such that savings under the 
MFN Model are incorporated into the parameters for the Maryland Total 
Cost of Care Model (for example, the annual global budget) for the 
duration of the MFN Model, then these acute care hospitals will remain 
excluded from the MFN Model beginning with the third quarter of 
performance year 1 and beyond. However, if the parameters of the 
Maryland Total Cost of Care Model change such that the participating 
acute care hospitals are no longer paid on a fully capitated or global 
budget basis or if a participating acute care hospital leaves the 
Maryland Total Cost of Care Model such that they are paid under section 
1833(t) of the Act, then such hospitals would no longer fall under this 
exclusion. This exclusion also applies on the same terms to acute care 
hospitals participating in the Pennsylvania Rural Health Model that 
otherwise meet the definition of MFN participant, unless the parameters 
of the Pennsylvania Rural Health Model change such that the 
participating acute care hospitals are no longer paid on a fully 
capitated or global budget basis or if a participating acute care 
hospital leaves the Pennsylvania Rural Health Model such that they are 
paid under section 1833(t) of the Act. We expect that the CMS 
Innovation Center will adjust the parameters of the Maryland Total Cost 
of Care Model and the Pennsylvania Rural Health Model such

[[Page 76185]]

that the participants in these CMS Innovation Center models will remain 
excluded from the MFN Model for the duration of the MFN Model. Further, 
as discussed in section III.J.1. of this IFC, the CMS Innovation Center 
intends to address model overlaps with other CMS Innovation Center 
models whether or not the participants in other models are MFN 
participants, for example we will account for changes in Medicare Part 
B drug payments that impact other models' financial calculations.
    We note that community mental health centers, comprehensive 
outpatient rehabilitation facilities (CORF), outpatient rehabilitation 
facilities (ORF), and certain other providers and suppliers do not 
submit claims for Medicare Part B drugs or are not paid separately for 
Medicare Part B drugs; thus, an express exclusion for these providers 
and suppliers is not necessary. We also note that including these 
providers and suppliers in the MFN Model would complicate the model 
design and make it challenging to test the impact of the MFN Model on 
these types of providers and suppliers because of the varied payment 
structures among these providers and suppliers.
    Table 1 shows the distribution of 2019 Medicare Part B allowed 
charges for separately payable Medicare Part B drugs by provider and 
supplier type using available final action claims where Medicare was 
the primary payer, with limited exclusions as noted. This table shows 
the distribution of Part B drug claims among provider and supplier 
types. To assign claims to a provider or supplier type, we considered 
the type of Medicare Administrative Contractor (MAC) that processed the 
claim, type of bill, provider number, revenue center, line place of 
service code, and specialty of the health care practitioner associated 
with the drug claim line.
BILLING CODE 4120-01-P

[[Page 76186]]

[GRAPHIC] [TIFF OMITTED] TR27NO20.000

BILLING CODE 4120-01-C
    To minimize the complexity of the MFN Model, we are not including 
in the MFN Model Medicare Part B drugs that are furnished in the 
inpatient setting,

[[Page 76187]]

administered through covered DME, orally administered, or paid under 
the End-Stage Renal Disease Prospective Payment System (ESRD PPS). 
Therefore, in Sec.  513.100(d), we provide an exception for claims 
submitted by acute care hospitals for separately payable Medicare Part 
B drugs that were administered during an inpatient stay or included on 
an inpatient claim, such as when a beneficiary has exhausted their Part 
A benefit days, claims administered by the Durable Medical Equipment 
Medicare Administrative Contractors (DME MACs) as described in 42 CFR 
421.404(c)(2), and claims paid under the ESRD PPS, including claims for 
drugs that are paid using the transitional drug add-on payment 
adjustment.
    Under the approach set forth in Sec.  513.100(b), all Medicare 
participating providers and suppliers that submit a claim for an MFN 
Model drug (excluding claims specified in Sec.  513.100(d)) furnished 
to an MFN beneficiary will be included as MFN participants unless 
otherwise excluded (as specified in Sec.  513.100(c)), regardless of 
the volume of MFN Model drugs for which they submit claims. As Table 1 
shows, a significant proportion of suppliers bill for a relatively 
lower volume of MFN Model drugs, such as less than $2,000 in total 
annual allowed charges, and will likely have limited claims paid under 
the MFN Model. We considered whether to make specific payment 
adjustments under the MFN Model for MFN participants that bill for a 
low volume of MFN Model drugs during a historical period or whether 
low-volume providers and suppliers could have the option to opt into or 
out of the MFN Model. However, we believe that requiring participation 
in the model only of providers and suppliers that bill for a higher 
volume of MFN Model drugs would not allow us to observe the impact of 
the MFN Model on a full range of providers and suppliers and would 
create opportunities for shifting sites of care and gaming. As such, we 
are including a broad set of providers and suppliers as MFN 
participants, regardless of their volume of billing for MFN Model 
drugs. As described in section III.I.2. of this IFC, the MFN Model 
includes a financial hardship exemption in the form of a potential 
reconciliation amount for MFN participants that are significantly 
affected by their participation in the MFN Model.
    We note that MFN Model drugs could be furnished to a beneficiary in 
an HOPD who is subsequently admitted to an inpatient hospital stay. 
When a beneficiary receives outpatient hospital services, including MFN 
Model drugs, during the 3 days immediately preceding admission to a 
hospital defined under section 1886(d) of the Act, the outpatient 
hospital services are treated as inpatient services if the beneficiary 
has Medicare Part A coverage and such services are not separately 
payable under Medicare Part B. We will apply this policy consistently 
under the MFN Model such that if a beneficiary receives an MFN Model 
drug in an HOPD that is an MFN participant and is admitted to this 
hospital within 3 days, then those services, including drugs, will be 
treated as inpatient services (in accordance with Medicare inpatient 
payment policies) and will not be separately payable under the MFN 
Model. We note that when a beneficiary receives outpatient hospital 
services during the day immediately preceding a hospital admission to a 
hospital not paid under the Inpatient Prospective Payment System 
(IPPS), such as psychiatric hospitals and units, inpatient 
rehabilitation hospitals and units, long-term care hospitals, 
children's hospitals, and cancer hospitals, the statutory payment 
window is one day preceding the date of the patient's admission; but 
because these categories of hospitals will be excluded from the MFN 
Model, as discussed previously, the payment window policy will not be 
applicable for this model.
    We are codifying these provisions in Sec. Sec.  513.100(a) through 
(d).
    We note that we include a limitation on the MFN Drug Payment Amount 
in Sec.  513.210(d)(5) that will apply to certain claims submitted by 
340B covered entities as described in section III.E.10. of this IFC to 
ensure that beneficiaries who are furnished MFN Model drugs by a 340B 
covered entity do not face increased cost-sharing under the MFN Model 
than would otherwise apply.
2. Mandatory Participation and Requirements
    Model participation will be mandatory for Medicare participating 
providers and suppliers that satisfy the MFN participant definition. 
There will be no specific enrollment activities for MFN participants; 
rather, their participation will be effectuated by the submission of a 
claim for an MFN Model drug furnished to an MFN beneficiary, and we 
will apply the MFN Model payment to such a claim.
    As we have described in previous rules implementing models with 
required provider or supplier participation, such as the Comprehensive 
Care for Joint Replacement (CJR) Model, mandatory participation can 
enhance the generalizability of model results, as mandatory model 
participants may be more broadly representative of all entity types 
that could be affected by a model. Requiring participation in the MFN 
Model will allow us to observe the experiences of providers and 
suppliers with diverse characteristics, such as geographies, patient 
populations, and specialty mixes. Mandatory participation (with 
specified exceptions) by providers and suppliers submitting claims for 
MFN Model drugs in a nationwide model, as further discussed in section 
III.C.3. of this IFC, will minimize administrative complexity and risk 
to the integrity of the MFN Model.
    In Sec.  513.100(e) and Sec.  513.100(f), we are codifying MFN 
participant requirements during and after the MFN Model. During the MFN 
Model performance period described in Sec.  513.1(c), MFN participants 
must--
     Adhere to the beneficiary protections requirements in 
Sec.  513.410 to ensure beneficiaries' access to care is not adversely 
impacted;
     Adhere to the MFN Model-specific billing instructions 
established by CMS and the MAC responsible for processing the MFN 
participant's claims, including without limitation those described in 
Sec.  513.200, to ensure appropriate and accurate Medicare payments; 
and
     Participate in MFN Model monitoring and evaluation 
activities in accordance with 42 CFR 403.1110(b), including collecting 
and reporting of information as the Secretary of Health and Human 
Services (the Secretary) determines is necessary to monitor and 
evaluate the MFN Model, including without limitation ``protected health 
information'' as that term is defined at 45 CFR 160.103.
    For 2 years after termination of the MFN Model, MFN participants 
must participate in MFN monitoring activities as described in Sec.  
513.420.
    MFN participants will continue to bill Medicare for separately 
payable MFN Model drugs furnished to MFN beneficiaries and be 
responsible for collecting beneficiary cost sharing amounts for MFN 
Drug Payment Amounts. As such, we anticipate MFN participants will have 
the same administrative requirements for collection of beneficiary 
cost-sharing amounts under the MFN Model as apply to collection of 
beneficiary cost-sharing outside the MFN Model.
    As discussed in section III.L. of this IFC, manufacturers will 
exclude from their calculation of ASP all units of MFN Model drugs that 
are furnished to MFN beneficiaries and for which payment under Sec.  
513.210 is allowed.

[[Page 76188]]

Manufacturers will need to determine the number of units to exclude and 
may adjust purchasing arrangements with MFN participants in order to 
obtain information about such units. While MFN participants are not 
required to provide data to manufacturers related to the number of 
units of MFN Model drugs that were furnished to MFN beneficiaries and 
for which payment under Sec.  513.210 was allowed, we anticipate that 
manufacturers may establish mechanisms to obtain such information, 
which also may create administrative burden for MFN participants 
related to the MFN Model. For example, manufacturers could require use 
of separate purchasing accounts, or reporting of information about 
units of MFN Model drugs that were furnished to MFN beneficiaries and 
for which payment under Sec.  513.210 was allowed in order to receive a 
more favorable purchase price.
3. Model Geographic Area
    In the October 2018 ANPRM, CMS anticipated the geographic area 
included in a potential IPI Model would encompass 50 percent of 
Medicare Part B drug spending. Several commenters expressed concern 
that having model participants subjected to multiple payment 
methodologies for included drugs based on having some but not all of 
their locations within the model's geographic area would be 
administratively burdensome. Additionally, some commenters expressed 
concern at the idea of requiring participation in some geographic areas 
but not others, noting that this approach would disproportionately 
affect some providers and suppliers and not others. Multiple commenters 
noted that reduced cost-sharing for patients in the model compared to 
those outside of the model would create potential differences in access 
for beneficiaries. One commenter noted that there would be a risk of 
patient steering if the model created a financial incentive for 
providers and suppliers to provide care at sites outside of the model 
geographic area rather than at sites in the model geographic area.
    Due to the administrative complexity and risk to model integrity 
associated with a limited scope, CMS believes that the MFN Model cannot 
realize its full potential in spending reductions for Medicare and its 
beneficiaries and improvement in quality of care without broad 
participation of Medicare participating providers and suppliers through 
a nationwide scope. Section 1115A(b) of the Act gives the Secretary 
discretion in the design of models, including the scope of models. 
Section 1115A(a)(5) of the Act states that the Secretary may elect to 
limit testing of a model to certain geographic areas. It follows that 
the Secretary could similarly elect not to limit testing to certain 
geographic areas, and instead test a nationwide model.
    The MFN Model requires mandatory, nationwide participation of 
Medicare participating providers and suppliers (with limited 
exclusions) to be able to successfully test the model for the reasons 
described later in this section. First, a nationwide scope avoids 
additional administrative burden on MFN participants with some service 
locations inside the MFN Model geographic area and others outside of 
the MFN Model geographic area, which could lead to such MFN 
participants needing to track and follow separate requirements for how 
drugs are acquired, furnished, and billed, depending on the service 
location. Second, a nationwide model geographic area eliminates the 
potential for MFN participants with service locations both inside and 
outside the MFN Model's geographic area to seek to influence 
beneficiaries' choice of treatment location in response to the 
differences between non-model payments and the MFN Model payments. This 
potential issue is of particular concern for the MFN Model given the 
broad use of MFN Model drugs and the ambulatory settings in which these 
drugs may be furnished, which can be geographically distributed over 
wide areas. Third, CMS also believes that a nationwide model geographic 
area maintains continuity with current treatment patterns by limiting 
disruption to beneficiary and health care provider treatment plans that 
may arise due to potential changes in the site of care. Fourth, a 
nationwide model geographic area allows all eligible beneficiaries who 
receive an MFN Model drug from an MFN participant where separate 
payment is allowed to benefit from the cost-sharing reductions under 
the MFN Model. Finally, CMS believes that a nationwide model geographic 
area along with mandatory participation creates the necessary market 
participation to increase the likelihood of MFN participants being able 
to acquire MFN Model drugs at lower prices as discussed in section VI. 
of this IFC. CMS notes that several of these points were commented on 
by several respondents to the October 2018 ANPRM. These points 
highlight the challenges that accompany a limited scope (non-
nationwide) model geographic area. CMS therefore believes a nationwide 
scope is the most appropriate for the MFN Model. Thus, we are codifying 
in Sec.  513.120 that the MFN Model geographic area includes all states 
and U.S. territories.
    As described in section VI. of this IFC, we anticipate that there 
could be potential challenges associated with a mandatory, nationwide 
model, namely greater impacts on manufacturers, a greater number of MFN 
participants that potentially receive lower payments for drugs under 
the model, and fewer non-participants who potentially increase their 
patient volume should beneficiaries need to locate alternative sites of 
care. We have designed the model to mitigate these potential challenges 
where possible.

D. MFN Model Drugs

    We will begin the MFN Model with 50 Medicare Part B drugs, 
identified by Healthcare Common Procedure Coding System (HCPCS) codes 
with high annual spending during 2019 (based on dates of service and 
after applying certain exclusions), that will be included on the MFN 
Model Drug HCPCS Codes List (described later in this section), and 
maintain approximately 50 Medicare Part B drugs on the MFN Model Drug 
HCPCS Codes List during the 7-year model performance period. We will 
focus the model on the separately payable, physician-administered 
Medicare Part B drugs with the highest annual spending which make up a 
portion of the roughly 550 HCPCS codes listed on the quarterly ASP 
pricing files, but encompass approximately three-quarters of annual 
Medicare Part B drug spending, \34\ and are furnished by the types of 
providers and suppliers that frequently bill under Medicare Part B. The 
MFN Model payments will apply only to MFN Model drugs when these drugs 
are administered by MFN participants to MFN beneficiaries and Medicare 
Part B allows separate payment as the primary payer.
---------------------------------------------------------------------------

    \34\ CMS publishes a Medicare Part B Drug Dashboard which can be 
used to view annual spending on drugs by HCPCS code. The 
downloadable file can be used to examine the proportion of annual 
spending for the included drugs. See: https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Information-on-Prescription-Drugs/MedicarePartB.
---------------------------------------------------------------------------

    In Sec.  513.130(b), we exclude some categories of Medicare Part B 
drugs from the model, such as certain vaccines, radiopharmaceuticals, 
oral drugs, compounded drugs, and intravenous immune globulin products. 
We also exclude drugs that are billed with HCPCS codes to which any 
generic drugs are assigned, including in applicable instances where 
single

[[Page 76189]]

source drugs or biologicals were within the same billing and payment 
code as of October 1, 2003. For purposes of the MFN Model, we consider 
a drug to be a generic drug if it is approved under an abbreviated new 
drug application (ANDA) under section 505(j) of the Federal Food, Drug, 
and Cosmetic Act. In accordance with President Trump's Blueprint to 
Lower Drug Prices, we are excluding such drugs because these drugs are 
already subject to competitive market forces and because the Medicare 
Part B payment allowances for these drugs already reflect price 
competition from generic products. In addition, we are excluding drugs 
for which there is an Emergency Use Authorization (EUA) or approval by 
the Food and Drug Administration (FDA) to treat patients with suspected 
or confirmed coronavirus disease 2019 (COVID-19). Since there may 
likely be urgent, high demand for such drugs and available supply may 
be targeted to certain populations, this exclusion allows maximum 
flexibility for potential changes in drug distribution for such drugs.
    To encourage introduction and use of biosimilars, the Trump 
Administration has taken several actions, including establishing 
separate HCPCS codes for Medicare Part B biosimilar biological 
products. We are not excluding biosimilar biological products from the 
MFN Model, however, given the relative lower annual Medicare Part B 
spending for HCPCS codes for separately payable biosimilar biological 
products through 2019, only one biosimilar biological product is 
included among the performance year 1 MFN Model Drug HCPCS Codes List 
in Table 2.
    We further discuss the drugs that will be included in or excluded 
from the MFN Model in the following four subsections.
1. MFN Model Drug HCPCS Codes List
    We will use an approach for including drugs in the MFN Model that 
is similar to what we described in the October 2018 ANPRM. However, 
rather than beginning with approximately 27 drugs, as discussed in the 
October 2018 ANPRM, and adding drugs annually, we will include 
approximately 50 Medicare Part B drugs in the MFN Model for each 
performance year. We will identify the top 50 Medicare Part B 
separately payable drugs with the highest aggregated Medicare Part B 
total allowed charges in the baseline period, after excluding certain 
claims, to result in an initial set of drugs that will be included in 
the model beginning in performance year 1. Thereafter, annual additions 
will follow a similar process using claims data for the subsequent 
year.
    Compared to beginning with a smaller number of drugs and phasing in 
additional drugs in each subsequent performance year, beginning with 50 
Medicare Part B drugs simplifies the model design and reduces 
complexity for MFN participants. Based on spending patterns over time 
for high spend Medicare Part B drugs,\35\ we expect the set of included 
Medicare Part B drugs to remain relatively stable over the model's 7-
year performance period, and we believe that a generally stable set of 
MFN Model drugs will help MFN participants plan their drug acquisition 
strategies. We believe the benefits of this stability outweigh the 
incremental challenge of beginning the MFN Model with a longer drug 
list than envisioned in the October 2018 ANPRM, and allows Medicare and 
its beneficiaries to benefit from the model payment methodology sooner 
for more of the highest spend Medicare Part B drugs, if anticipated 
savings are realized.
---------------------------------------------------------------------------

    \35\ CMS publishes a Medicare Part B Drug Dashboard, which can 
be used to view annual spending on drugs by HCPCS code. See: https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Information-on-Prescription-Drugs/MedicarePartB.html.
---------------------------------------------------------------------------

    By focusing the MFN Model on separately payable Medicare Part B 
drugs, payments for products that are bundled or otherwise included in 
payment for a procedure or other services will not be affected by the 
MFN Model and payments for such bundled services will not have to be 
separated or adjusted. This approach does not exclude drugs that are 
packaged under a Medicare payment system in certain settings and 
separately payable in other settings. However, the MFN Model payment 
only applies to such drugs in settings where separate payment is 
allowed.
    In Sec.  513.130, we describe the creation and periodic updates of 
an MFN Model Drug HCPCS Codes List, which designates the MFN Model 
drugs that are subject to the MFN Model payments specified in Sec.  513 
subpart C. Specifically, to select the list of drugs included in the 
MFN Model for the beginning of performance year 1 (that is, beginning 
January 1, 2021), the regulation text at Sec.  513.130(a)(1) codifies 
that, after making the exclusions specified in Sec.  513.130(b)(1) and 
(b)(2), CMS identifies the top 50 drugs by HCPCS code with the highest 
aggregate 2019 Medicare Part B total allowed charges, and adds those 
HCPCS codes to the MFN Model Drug HCPCS Codes List, after updating such 
HCPCS codes for any applicable changes. We will use HCPCS codes to 
identify drugs because they are an established way to identify, bill, 
and pay for separately payable Medicare Part B drugs in the Medicare 
claims processing system, and they are commonly used in other Medicare 
Part B drug payment resources like the ASP drug pricing files. For this 
process, we will use final action Medicare Part B claims for separately 
paid drugs with dates of service within calendar year 2019 and allowed 
charges greater than $0 where Medicare was the primary payer from all 
Medicare providers and suppliers as the baseline period. This period is 
the most recent full calendar year of claims data that was sufficiently 
available prior to the model performance period start on January 1, 
2021. Accordingly, we arrayed drugs, using HCPCS codes, in descending 
order based on the aggregate Medicare Part B total allowed charges in 
the 2019 baseline period, after making the exclusions specified in 
Sec.  513.130(b)(1) and (b)(2), and identified the 50 Medicare Part B 
drugs (identified by HCPCS codes) with the highest total Medicare Part 
B allowed charges. These HCPCS codes are included on the MFN Model Drug 
HCPCS Codes List for the beginning of performance year 1 as shown in 
Table 2 of this IFC.
    The MFN Model uses an annual calendar year baseline period for 
purposes of identifying the drugs that will be added to the MFN Model 
Drug HCPCS Codes List for performance year 1 (and annually thereafter, 
using the next subsequent calendar year as the baseline) because: The 
vast majority of HCPCS Code updates occur annually in the January HCPCS 
update; the model will use an annual baseline period to calculate the 
alternative add-on payment amount described in section III.F. of this 
IFC; and these baseline periods will be aligned for consistency in the 
model design.
    This approach for identifying the drugs that are included in the 
MFN Model at the beginning of performance year 1 captures most of the 
drugs listed in the October 2018 ASPE report,\36\ which used the 
Medicare Part B National Summary Drug file from 2016 to identify 
approximately 27 HCPCS codes associated with high amounts of spending, 
and nearly all the drugs listed in the November 20, 2020 ASPE report, 
which applied the criteria in the MFN Model to Medicare Part B claims 
data

[[Page 76190]]

for 2018.\37\ This approach also results in the inclusion of a variety 
of drugs and biologicals (including biosimilar biological products) 
that are used to treat common conditions in the Medicare Part B 
beneficiary population. These drugs and biologicals with high annual 
Medicare allowed charges are frequently prescribed and administered by 
various physician specialties to beneficiaries with various medical 
conditions. Examples of uses of the drugs included in the MFN Model 
are: Drugs and biologicals used to treat cancer and related conditions, 
biologicals used for the treatment of rheumatoid arthritis and other 
immune mediated conditions, and biologicals used to treat macular 
degeneration. Beneficiaries who receive such drugs, often on a 
recurring basis, face substantial cost-sharing liability directly or 
through their supplemental insurance, and such costs may be partly 
avoidable (that is, reduced) if Medicare payment for these drugs were 
not based on the current ASP methodology.
---------------------------------------------------------------------------

    \36\ ``Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures.
    \37\ El-Kilani Z, Finegold K, Mulcahy A, and Bosworth A. 
Medicare FFS Part B and International Drug Prices: A Comparison of 
the Top 50 Drugs. Washington, DC: Office of the Assistant Secretary 
for Planning and Evaluation, U.S. Department of Health and Human 
Services. November 20, 2020 (https://aspe.hhs.gov/pdf-report/medicare-ffs-part-b-and-international-drug-prices).
---------------------------------------------------------------------------

    Beginning with 50 of the highest spend HCPCS codes based on annual 
Medicare Part B allowed charges during 2019, after taking into account 
certain exclusions, focuses the MFN Model on a wide variety of 
frequently utilized Medicare Part B drugs and specialties that 
administer such drugs to Medicare FFS beneficiaries, and allows CMS to 
test the MFN Model payment on a broad set of drugs and biologicals that 
are furnished to many beneficiaries. We believe that including single 
source drugs and biologicals (including biosimilar biological products) 
that move into the top 50 HCPCS codes on an annual basis will capture 
potential shifts in utilization to drugs that had not yet been included 
in the MFN Model, if such shifting were to occur, and will mitigate the 
potential for medically unnecessary shifts in utilization.
    In developing this approach, we also considered comments we 
received in response to the October 2018 ANPRM on using drug classes to 
help inform which drugs to include in the MFN Model, as well as 
requests to consider how access to Medicare Part B drugs (as a whole 
and for specific subsets of drugs) might be affected by inclusion in 
the model. We considered these suggestions and believe that using 
annual Medicare Part B allowed charges as a primary factor is a more 
transparent, consistent, and clear approach because attempting to 
identify drugs for inclusion in the MFN Model based on groups or 
classes of drugs could become complicated and confusing for MFN 
participants. There are numerous drug classification approaches 
available; for example, drug classification can be based on a chemical 
class, site of action, mechanism of action, as well as other factors. 
These approaches can become difficult to apply consistently when drugs 
from different chemical classes are used to treat the same condition, 
when a drug has more than one mechanism of action, or when conditions 
are treated with drugs having more than one mechanism of action. For 
example, the Medicare Part B biological products commonly used to treat 
rheumatoid arthritis include a variety of monoclonal antibodies. Using 
broad terms such as monoclonal antibodies to identify a ``group'' of 
MFN Model drugs would include a variety of biologicals that are 
commonly also used in treating other conditions, such as Crohn's 
disease, ulcerative colitis, cancer, and multiple sclerosis. Attempting 
to select MFN Model drugs using more narrow terms, for example by 
specifying agents that exert effects on more specific inflammatory 
pathways, such as tumor necrosis factor and interleukins, would miss 
biologicals that affect other pathways, like T cell stimulation. These 
approaches may also miss products that are primarily used to treat 
other diseases, but may be used less frequently in rheumatoid 
arthritis, and these approaches may not be readily adaptable for novel 
products that may be introduced over the 7-year performance period of 
the model.
    In Sec.  513.130(a)(2), we are codifying the process for annual 
updates of the MFN Model Drug HCPCS Codes List to update the list of 
drugs that will be included in the MFN Model for the subsequent 
performance year, as further described in section III.D.3. of this IFC.
2. Exclusion of Certain HCPCS Codes and Claims
    In the October 2018 ANPRM, we discussed the potential exclusion of 
several groups of drugs from the potential IPI Model (83 FR 54555). 
Commenters generally agreed that these drugs should be excluded. As 
codified in Sec.  513.130(b)(1), the MFN Model excludes the following 
types of drugs, by excluding claims at the HCPCS code level, before 
identifying the top 50 drugs with the highest aggregate annual Medicare 
Part B total allowed charges:
     Medicare Part B vaccines specified in section 1861(s)(10) 
of the Act (that is, influenza, pneumococcal pneumonia, and Hepatitis B 
vaccines, and any future vaccine for COVID-19). These preventive 
products are paid under section 1842(o)(1)(A)(iv) based on average 
wholesale price (AWP), a price that does not include discounts or 
rebates. Including such drugs in the MFN Model also would not comport 
with our test of an alternative add-on payment amount (described in 
section III.F. of this IFC) because the statutory add-on percentage 
under section 1847A of the Act does not apply to these drugs.
     Radiopharmaceuticals. Many radiopharmaceuticals are 
typically acquired outside of the traditional drug supply chain. 
Nuclear pharmacies are frequently involved in the preparation of 
patient-ready doses of these drugs, and Medicare Part B payment is 
frequently based on contractor pricing. We are excluding 
radiopharmaceuticals from the MFN Model because it is unlikely that we 
will be able to obtain reliable international drug pricing information 
for radiopharmaceuticals.
     Oral Medicare Part B drugs, including oral anticancer 
drugs described in section 1861(s)(2)(Q) of the Act, oral antiemetic 
drugs described in section 1861(s)(2)(T) of the Act and 
immunosuppressive drugs described in section 1861(s)(2)(J) of the Act. 
Oral anticancer, antiemetic, and many immunosuppressive drugs are often 
used outside of the provider and supplier settings (for example, these 
drugs are often used at home); therefore, we are excluding these oral 
drugs from the MFN Model.
     Compounded drugs including products prepared by 
outsourcing facilities.\38\ Although subject to certain FDA 
requirements, these products are not approved by FDA per se, and with 
one exception under the OPPS \39\ are not billed under drug-specific 
HCPCS codes; they are typically billed using under ``not otherwise 
classified'' (NOC) codes. Also, compounded drugs are typically acquired 
outside of the traditional drug supply chain, and Medicare Part B 
payment for compounded drugs is generally based on contractor pricing, 
such as invoice pricing. We are excluding these drugs because it is 
unlikely that we will be able to obtain reliable international drug 
pricing information for compounded drugs.
---------------------------------------------------------------------------

    \38\ See section 503B of the Federal Food, Drug, and Cosmetic 
Act (21 U.S.C. 353b) with respect to the definition of outsourcing 
facilities and their regulation by FDA.
    \39\ C9257 Injection, bevacizumab, 0.25 mg.

---------------------------------------------------------------------------

[[Page 76191]]

     Intravenous immune globulin products. In response to the 
October 2018 ANPRM, a commenter suggested that CMS exclude plasma-
derived products and stated such products have potential unique 
sourcing and distribution, and past supply shortages. We note that FDA 
has identified a current shortage related to one of the HCPCS codes 
that is among the top drugs with high aggregate 2019 Medicare Part B 
total allowed charges (J1569, Gammagard liquid infusion). Three other 
immune globulin products are also among the top drugs in 2019, J1459 
(Inj ivig privigen 500 mg), J1561 (Gamunex-c/gammaked), and J1568 
(Octagam injection). After considering this concern, we are excluding 
intravenous immune globulin products from the MFN Model because these 
products are at higher risk of shortage based on their complex sourcing 
and production, and we are aware of the ongoing exploration of the 
potential benefit of plasma in the treatment of patients with COVID-19.
     Drugs that are subject to an EUA or receive FDA approval 
to treat patients with suspected or confirmed COVID-19. The exclusion 
of these drugs will minimize any potential for the MFN Model to impact 
rapid, widespread availability of such drugs in the U.S. to treat 
patients with suspected or confirmed COVID-19.
     Drugs without drug-specific HCPCS codes, that is, those 
billed under ``not otherwise classified'' (NOC) codes, such as J3490. 
NOC codes are used to bill for drugs not assigned to a particular HCPCS 
code. NOC codes typically include a variety of unrelated drugs that 
cannot be easily separated for the purpose of ranking allowed charges 
of the individual drugs. Also, significantly greater claims processing 
complexity for Medicare and MFN participants would result if we had to 
identify whether an MFN Model drug was billed under a NOC code during 
MFN Model operations. By excluding HCPCS codes for these types of 
drugs, these drugs will be fully excluded from the MFN Model.
    While we intend that the MFN Model drugs will encompass a wide 
variety of frequently utilized Medicare Part B drugs, we also intend 
that drugs will not be included on the basis of substantial use at 
home. Thus, in Sec.  513.130(b)(2), we codify the exclusion of claims 
that were processed and paid by the DME MACs as described in 42 CFR 
421.404(c)(2), and professional claims with a place of service code 
that indicates the drug was used in a home, including home-like 
settings, prior to identifying the top 50 drugs (by HCPCS code).\40\ 
The place of service exclusion applies only to professional claims 
because place of service codes are not used on institutional claims to 
identify home use. Specifically, professional claims with place of 
service codes 04--homeless shelter, 12--home, 13--assisted living 
facility, 14--group home, 16--temporary lodging, and 33--custodial care 
facility will be excluded prior to identifying the top 50 drugs (by 
HCPCS code).
---------------------------------------------------------------------------

    \40\ The DME MACs process Medicare Durable Medical Equipment, 
Orthotics, and Prosthetics (DMEPOS) claims for a defined geographic 
area or ``jurisdiction,'' servicing suppliers of DMEPOS. 
Professional claims must comply with the ASC X12 837 Professional 
guide (005010X222A1).
---------------------------------------------------------------------------

    For future years of model implementation, we seek comment on 
whether all blood related, plasma derived, and human tissue products 
should be included in or excluded from the MFN Model. We also seek 
comment on how CMS should define such products and what would be the 
supporting rationale for such an exclusion and how to address such 
considerations in the future. We note that we are also considering as a 
potential addition to the model design whether certain drugs, such as 
certain gene and cell therapies (for example, chimeric antigen receptor 
T-cell (CAR-T) products) and drugs approved by FDA after the start of 
the MFN Model that are indicated for and used to treat rare diseases or 
conditions, should be excluded from the MFN Model for all performance 
years, or for several years after the drug is first sold in the U.S. We 
note that under the MFN Model, annual Medicare Part B allowed charges 
would have to exceed tens of millions of dollars for such drugs to 
reach the top 50 and be added to the MFN Model. We also note that many 
of the top 50 drugs in 2019 are used to treat conditions with limited 
populations and were first approved within the last 5 years. In 
addition, we note that while drugs may initially be approved for one or 
a few very narrow indications, subsequently approved indications can 
quickly expand the use of the drug to a much larger patient population. 
We are considering whether we should exclude certain gene and cell 
therapies based on supply chain criteria, similar to our policy to 
exclude vaccines and compounded drugs. For future years, we seek 
comment on whether we should exclude certain gene and cell therapies or 
new drugs for the treatment of rare diseases and conditions from the 
MFN Model, and how CMS would identify such drugs for exclusion, 
particularly how we would define such drugs, identify rare diseases and 
conditions for purposes of the MFN Model, and determine the appropriate 
length of such exclusion (for example, all performance years or several 
years after the drug is first sold in the U.S.).
    Some commenters have suggested that drugs in short supply (based on 
inclusion on the FDA drug shortages list) should be excluded from drug 
payment models. As discussed previously, we are excluding intravenous 
immune globulin products from inclusion on the MFN Model Drug HCPCS 
Codes List, because these products are at higher risk of shortage based 
on their complex sourcing and production. Otherwise, based on our 
experience with ASP pricing, shortages of high cost single source drugs 
and biologicals are uncommon, of short duration, and generally apply to 
some but not all package sizes of a drug. As described in section 
III.E.12. of this IFC and codified in Sec.  513.210(d)(2), we include a 
quarterly payment exception for MFN drugs that are in short supply 
(based on inclusion on the FDA drug shortages list). We believe it will 
be less disruptive to the MFN Model to include a quarterly payment 
exception for MFN Model drugs during the time they are in short supply 
than to exclude such drugs from the MFN Model altogether because a 
quarterly payment exception approach will avoid changing the inclusion 
status of drugs should a shortage occur and again when the shortage is 
resolved, eliminate the need to consider developing a process to add 
and remove replacement drugs to maintain the number of MFN Model drugs, 
and avoid manufacturers having to change processes for capturing sales 
of such drugs in their ASP calculations as discussed in section III.L. 
of this IFC (under this policy, manufacturers will not include in their 
calculation of the manufacturer's ASP any units of MFN Model drugs 
billed by MFN participants where the MFN Drug Payment Amount is paid by 
Medicare as the primary payer).
    Finally, we considered whether an exception to inclusion on the MFN 
Model Drug HCPCS Codes List might be appropriate for MFN Model drugs in 
cases where pharmaceutical manufacturers that distribute the drug in 
the U.S. do not own the rights to the drug product for distribution 
outside the U.S. and therefore do not control ex-U.S. pricing for the 
drug product. To avoid a gaming opportunity whereby manufacturers' new 
or recent business arrangements create such cases, this type of 
exception could be defined such that only ownership rights that were 
transferred prior to the October 2018 ANPRM, when CMS announced a new

[[Page 76192]]

Medicare Part B drug payment model was being developed, would qualify. 
To avoid an exception being too broad, we are concerned that additional 
criteria should be required to qualify for it, such as whether the 
increase in the MFN Model drug's applicable ASP (a measure of U.S. 
prices) based on sales since October 2018 has been slower than 
inflation (that is, the change in the CPI-U from the end of October 
2018 through the ASP calendar quarter for the first calendar quarter of 
the model), and whether the U.S. manufacturer makes a legally 
enforceable commitment to future U.S. price increases being slower than 
inflation moving forward, if such an exception were to be granted. In 
addition, to maintain the exception for the remainder of the model, the 
increase in the MFN Model drug's applicable ASP since October 2018 
would need to be assessed quarterly to determine whether it continues 
to be slower than inflation. Given the complex and numerous 
relationships that manufacturers may have across U.S. and international 
markets, we are not including such an exception for the MFN Model.
    We seek comments for future years on our approach to identifying 
and maintaining the MFN Model Drug HCPCS Codes List and whether there 
is a need for an exception relating to manufacturers' ownership of drug 
products internationally, and if so, how such an exception might be 
defined and operated transparently.
3. Annual Updates to the MFN Model Drug HCPCS Codes List
    As discussed in section III.D.1. of this IFC, the MFN Model will 
begin with 50 drugs and biologicals by HCPCS code on the MFN Model Drug 
HCPCS Codes List for performance year 1. We will keep approximately 50 
drugs by HCPCS code in the MFN Model during the 7-year performance 
period so that drugs that continue to account for a large portion of 
Medicare Part B drug spending will continue to be included in the 
model. However, we believe that some adjustments to the MFN Model Drug 
HCPCS Codes List will likely be required from time to time as drugs 
enter and exit the market and as utilization of Medicare Part B drugs 
(measured by annual total allowed charges) changes. Thus, we will 
update the MFN Model Drug HCPCS Codes List annually. The annual update 
process will occur prior to the beginning of each performance year 
rather than more frequently, such as a quarterly process, because less 
frequent changes to the MFN Model Drug HCPCS Codes List will decrease 
the burden associated with participating in the model. We believe that 
making fewer changes to the MFN Model Drug HCPCS Codes List will result 
in MFN participants having to make fewer changes to acquisition 
arrangements, and this in turn will lessen any potential for disruption 
in workflow and care delivery compared to a quarterly update process. 
Additionally, as specified in Sec.  513.130(a)(4), some quarterly 
changes may be necessary to comport with HCPCS coding updates that are 
applicable to the HCPCS codes on the MFN Model Drug HCPCS Codes List, 
such as when a code is terminated and a successor code is established.
    For each annual update for performance years 2 through 7, as 
described in Sec.  513.130(a)(2), we will array in descending order all 
separately payable Medicare Part B drugs, using HCPCS codes, based on 
total allowed charges after applying the exclusions codified in Sec.  
513.130(b)(1) and (b)(2), using the most recent full calendar year's 
Medicare Part B claims from all providers and suppliers. Those drugs 
(as identified by HCPCS codes) that have total allowed charges that 
fall in the top 50 drugs by spending for that calendar year that are 
not already on the MFN Model Drug HCPCS Codes List will be added to the 
MFN Model Drug HCPCS Codes List to take effect on the first day of the 
next performance year and the MFN Drug Payment Amount that will apply 
will be based on the applicable MFN Price phase-in for that performance 
year and will follow the annual payment updates thereafter. This 
process will be used only to add HCPCS codes that are new to the top 
50--to maintain consistency, we will not remove any codes from the MFN 
Model Drug HCPCS Codes List on the grounds that the HCPCS code dropped 
out of the top 50. We will keep all HCPCS codes that were included on 
the MFN Model Drug HCPCS Codes List for the prior performance year on 
the MFN Model Drug HCPCS Codes List, except in certain circumstances as 
noted in section III.D.4. of this IFC, in order to have greater 
stability in the set of drugs that are included in the MFN Model across 
the performance years. As a result, in performance years 2 through 7, 
the number of HCPCS codes on the MFN Model Drug HCPCS Codes List may be 
greater than 50. We believe this approach has the potential to identify 
drugs that are alternative therapies to MFN Model drugs, such as 
competitor products, where MFN participants may shift utilization to 
avoid using drugs subject to the MFN Model payment, and will provide a 
mechanism for adding such drugs to the MFN Model. In addition, this 
approach will serve as a mechanism to identify newer drugs with high 
annual Medicare Part B spending for inclusion in the MFN Model.
    To maintain transparency, when we add HCPCS codes that are new to 
the top 50 or are replacement codes for HCPCS codes that are listed on 
the MFN Model Drug HCPCS Codes List, we will list the code's start date 
for inclusion in the MFN Model. In addition, we will revise HCPCS codes 
on the MFN Model Drug HCPCS Codes List as necessary to reflect 
quarterly HCPCS code updates that are applicable to the HCPCS codes on 
the MFN Model Drug HCPCS Codes List, for example when a permanent code 
replaces a temporary code, a HCPCS code is terminated and a replacement 
code is established, or a HCPCS code is established for Medicare use. 
In such case, we will include an end date on the MFN Model Drug HCPCS 
Codes List for the terminated code. We will notify MFN participants of 
updates to the MFN Model Drug HCPCS Codes List no less frequently than 
quarterly by adding the updated MFN Model Drug HCPCS Codes List to the 
MFN Model website (https://innovation.cms.gov/initiatives/most-favored-nation-model).
4. Approach for Removing Drugs From the MFN Model Drug HCPCS Codes List
    We do not anticipate that drugs will be removed from the MFN Model 
frequently. In accordance with Sec.  513.130(a)(3), we will remove 
drugs from the MFN Model Drug HCPCS Codes List only under the following 
limited circumstances, but no more frequent than quarterly, to align 
with quarterly MFN Model payment updates:
     If they are permanently withdrawn from the U.S. market;
     If a specific HCPCS code included on the MFN Model Drug 
HCPCS Codes List is terminated with no replacement code available or 
planned; or
     The drug is excluded from the MFN Model pursuant to the 
exclusions in Sec.  513.130(b)(1), for example a HCPCS code describes a 
generic drug approved under an ANDA or a drug with an EUA or FDA 
approval to treat patients with suspected or confirmed COVID-19.
    To maintain transparency, we will remove HCPCS codes by setting an 
end date on the MFN Model Drug HCPCS Codes List at the next quarterly 
update after CMS becomes aware, through environmental scanning 
activities, that all of the NDCs assigned to a HCPCS code have been 
withdrawn from the U.S. market and the drug is permanently withdrawn 
from the U.S. market, or the HCPCS code has been terminated with

[[Page 76193]]

no replacement code available or planned, or the exclusion in Sec.  
513.130(b) applies. HCPCS codes that are removed from the MFN Model 
Drug HCPCS Codes List will no longer be subject to the MFN Model 
payment, but rather will be subject to current Medicare payment 
policies. If the conditions for removal no longer exist, the HCPCS code 
could again qualify for inclusion on the MFN Model Drug HCPCS Codes 
List at the next annual update.
5. Performance Year 1 MFN Model Drug HCPCS Codes List
    To create the MFN Model Drug HCPCS Codes List for performance year 
1, we arrayed drug HCPCS codes by aggregate 2019 Medicare Part B total 
allowed charges \41\ after applying the exclusions in Sec.  
513.130(b)(1) and (b)(2). We then identified the top 50 drugs by HCPCS 
code with the highest aggregate 2019 Medicare Part B total allowed 
charges. This process excluded HCPCS codes for two influenza vaccines 
(90662 (Iiv no prsv increased ag im) and 90653 (Iiv adjuvant vaccine 
im)), two pneumococcal pneumonia vaccines (90732 (Ppsv23 vacc 2 yrs+ 
subq/im) and 90670 (Pcv13 vaccine im)), and a radiopharmaceutical 
(A9606 (Radium ra223 dichloride ther)) from the MFN Model Drug HCPCS 
Codes List. The exclusion of intravenous immune globulin products 
excluded four HCPCS codes: J1459, Inj ivig privigen 500 mg; J1561, 
Gamunex-c/gammake; J1568, Octagam injection; and J1569, Gammagard 
liquid injection. Additionally, one HCPCS code that describes a generic 
drug (J9395, Injection, fulvestrant) was excluded. Excluding claims 
that were processed and paid by the DME MACs resulted in the following 
HCPCS codes no longer falling within the top 50 drugs in 2019: J7605 
(Arformoterol non-comp unit); J7686 (Treprostinil, non-comp unit); and 
J3285 (Treprostinil injection). Excluding claims based on the place of 
service exclusion resulted in one HCPCS code, J7192 (Factor viii 
recombinant nos), no longer falling within the top 50 drugs in 2019.
---------------------------------------------------------------------------

    \41\ We used 2019 final action claims data that were available 
in the CMS Chronic Conditions Data Warehouse in September 2020 where 
Medicare was the primary payer.
---------------------------------------------------------------------------

    Using this approach for selecting MFN Model drugs, the resulting 
performance year 1 MFN Model Drug HCPCS Codes List includes single 
source drugs and biologicals that accounted for approximately 75 
percent of annual Medicare Part B drug allowed charges for separately 
payable drugs during 2019. Table 2 displays the list of MFN Model drugs 
(by HCPCS code) that are included on the MFN Model Drug HCPCS Codes 
List for the beginning of performance year 1, along with the top 
billing specialties.
    CMS will publish the MFN Model Drug HCPCS Codes List quarterly on 
the MFN Model website (https://innovation.cms.gov/initiatives/most-favored-nation-model), in advance of the calendar quarter, along with 
MFN Model Payment amounts and other MFN Model information and 
materials.
BILLING CODE 4120-01-P

[[Page 76194]]

[GRAPHIC] [TIFF OMITTED] TR27NO20.001


[[Page 76195]]


[GRAPHIC] [TIFF OMITTED] TR27NO20.002


[[Page 76196]]


BILLING CODE 4120-01-C

E. Model Payment Methodology for MFN Model Drugs

    The MFN Model will test an innovative approach to calculating drug 
payment through use of a more comprehensive set of drug pricing data to 
calculate an alternative payment amount for MFN Model drugs, along with 
an alternative add-on payment, which is described in section III.F. of 
this IFC. Payment for drug administration services, when applicable, 
will continue to be separately billed by model participants to 
Medicare; there will be no change in the payment for drug 
administration services under the MFN Model. Providers and suppliers 
will continue to purchase MFN Model drugs, furnish such drugs to 
beneficiaries, submit claims to Medicare, and collect applicable 
beneficiary cost-sharing. Under the MFN Model, payments for separately 
payable Medicare Part B drugs will include the alternative drug payment 
amount and the alternative add-on payment amount, both subject to 
sequestration, as applicable.
    Similar to the current approach under section 1847A of the Act, the 
MFN Model alternative payment limit for the ``drug portion'' of payment 
for MFN Model drugs (that is, not including the add-on amount) will be 
calculated by CMS quarterly. This amount is called the MFN Drug Payment 
Amount. The calculation of the MFN Drug Payment Amounts is codified in 
Sec.  513.210(b). Beneficiary cost-sharing will apply to the MFN Drug 
Payment Amount for included drugs.
    We will calculate an MFN Drug Payment Amount for each drug on the 
MFN Model Drug HCPCS Codes List based on an MFN Price, which will be 
derived from the lowest GDP-adjusted country-level price, based on non-
U.S. OECD member countries with a GDP per capita that is at least 60 
percent of the U.S. GDP per capita.\42\ We will use GDP per capita 
information that is based on purchasing power parity. We are also 
establishing limits such that the MFN Drug Payment Amount will not 
exceed non-model payment for the drug (excluding any non-model add-on 
payment amount), will not apply to drugs that are not separately 
payable, and certain other limitations discussed later in this section.
---------------------------------------------------------------------------

    \42\ Individual countries differ in the regulatory processes and 
standards governing approval of drugs and biologicals. Use of 
international drug pricing information in the MFN Model should not 
be interpreted to connote FDA approval or to otherwise describe any 
scientific or regulatory relationship between U.S.-approved and non-
U.S.-approved products.
---------------------------------------------------------------------------

    Section III.E.1. of this IFC identifies the data sources for the 
MFN Model drugs' international drug pricing information that we will 
use to calculate the MFN Price for each drug. Section III.E.2. of this 
IFC outlines the international drug pricing information we will include 
in these calculations and the included countries. Section III.E.3. of 
this IFC defines the MFN Drug Payment Amount. Section III.E.4. of this 
IFC outlines our approach to calculating each drug's MFN Drug Payment 
Amount. Section III.E.5. of this IFC describes the phase-in of the MFN 
Price. Section III.E.6. of this IFC describes the alternative 
calculation for the MFN Drug Payment Amount for situations where no 
international drug pricing information is available for an MFN Model 
drug. Section III.E.7. of this IFC provides illustrative MFN Drug 
Payment Amounts for each drug on the performance year 1 MFN Model Drug 
HCPCS Codes List in Table 2 using historical data. Section III.E.8. of 
this IFC describes the timing of data and MFN Drug Payment Amount 
updates. Section III.E.9. of this IFC describes adjustments to the 
phase-in formula and incentives for manufacturers to address rising 
U.S. drug prices. Section III.E.10. of this IFC describes the 
limitation on the MFN Drug Payment Amount. Section III.E.11. of this 
IFC describes the method for establishing MFN Drug Payment Amounts for 
MFN Model drugs added to the model for performance year 2 and 
subsequent performance years. Section III.E.12. of this IFC describes 
the quarterly payment exception for MFN Model drugs in short supply. 
Section III.E.13. of this IFC describes continued payment of the blood 
clotting factor furnishing fee under the MFN Model.
1. Data Sources on International Drug Pricing Information
    We will rely on existing data sources to obtain data that we will 
use to calculate and update the MFN Drug Payment Amounts. We will use 
existing data sources that contain international drug pricing 
information, including list prices, sales and/or volume data (for 
example, package size and number of packages sold), as available, in 
order to optimize operational efficiency. Sales may be based on ex-
manufacturer prices (sometimes called the ex-factory price), that 
represent actual or calculated prices paid to the manufacturer by 
wholesalers and other distributors, retail prices, prices for other 
distribution channels, or a combination thereof. Confidential 
manufacturer rebates will not likely be accounted for within these 
data; therefore, existing sources for international drug sales data may 
overstate actual prices realized by manufacturers.
    In the October 2018 ANPRM, we considered establishing a data 
collection system for manufacturers to report to CMS their 
international drug sales data for prices and units sold to support the 
calculation of the model payment for each drug. In response to the 
October 2018 ANPRM, we received comments stating that CMS should use 
existing data sources for international drug pricing information in 
order not to place burden on manufacturers. Some commenters expressed 
concerns that new data reporting would greatly increase burdens and 
costs for manufacturers, further limiting their ability to invest in 
research and development for innovative therapies, and would be 
impractical because defining price reporting for foreign markets would 
be too complex and could not adequately capture fluid pricing policy 
changes. We appreciate these concerns, and as such, we will rely on 
existing data sources for purposes of calculating MFN Drug Payment 
Amounts. We believe that existing data sources are adequate for 
purposes of calculating country-level prices, GDP-adjusted country-
level prices, and the MFN Prices, as described in this IFC, that will 
be used to calculate the MFN Drug Payment Amount.
    Commenters also noted that one potential adverse reaction to the 
model described in the October 2018 ANPRM may be a shift 
internationally to a high price and high rebate pricing strategy. 
Specifically, commenters expressed concern that if the international 
drug pricing information used to establish payment under a model relied 
on the list prices in the included countries, then manufacturers would 
restructure their pricing arrangements to increase the list prices of 
the model's drugs in those countries, and offer higher rebates to 
offset the increased list price. CMS appreciates this concern, and we 
will prioritize use of available international drug pricing information 
that incorporate discounts and rebates to the extent possible, rather 
than just the list prices.
    We have assessed several existing data sources to determine the 
availability and sufficiency of international drug pricing information. 
In Sec.  513.140(c), we are codifying the use of one or more 
international drug pricing data sources. Specifically, we will use one 
or more data sources, available to CMS at least 20 business days prior 
to the start of a calendar

[[Page 76197]]

quarter, that utilize a standardized method for identifying drugs 
across countries within that data source, such as using an 
internationally recognized method for identifying scientific and 
nonproprietary names (for example, active ingredient name) and a 
standard method for identifying drug forms that at a minimum 
distinguishes among injectable, oral, and other forms of a drug. For 
example, the data source might use the International Nonproprietary 
Names (INN), as applicable.\43\ This process requires mapping between 
the data source's standardized method for identifying scientific and 
nonproprietary names and HCPCS codes, as discussed and illustrated in 
section III.E.7. of this IFC. Further, we will use one or more data 
sources that contain international drug pricing information stated in 
U.S. currency, such as list prices, ex-manufacturer prices (sometimes 
called the ex-factory price) that represents actual or calculated 
prices paid to the manufacturer by wholesalers and other distributors, 
actual or calculated sales for retail and other distribution channels, 
or volume data (for example, number of units sold).
---------------------------------------------------------------------------

    \43\ World Health Organization, International Nonproprietary 
Names accessed via https://www.who.int/medicines/services/inn/en/.
---------------------------------------------------------------------------

    If more than one data source is available for an MFN Model drug, as 
noted previously, we will prioritize the data sources using a hierarchy 
that we describe later in this section. Thus, for each MFN Model drug, 
we will identify and use the most comprehensive data source available, 
using the hierarchy codified in Sec.  513.140(c)(3). We will use only 
one data source for an MFN Model drug for a quarter, meaning we will 
not combine data from different data sources or time periods to 
calculate the MFN Drug Payment Amount for an MFN Model drug for a 
quarter.
    Whenever possible, we will use international drug pricing 
information from two calendar quarters prior to the calendar quarter to 
which the MFN Drug Payment Amount will apply since the ASP payment 
limits that apply to that calendar quarter are based on manufacturers' 
U.S. sales from two calendar quarters prior such that the U.S. and 
international drug pricing data will be based on information from the 
same calendar quarter. We use the term applicable ASP calendar quarter 
to mean the period that is two calendar quarters prior to the calendar 
quarter to which the MFN Drug Payment Amount will apply.
    The hierarchy of data sources we will use is as follows:
     A data source with sales and volume data for the 
applicable ASP calendar quarter from at least one included country, 
that is, a non-U.S. OECD member country at the end of the applicable 
ASP calendar quarter with a GDP per capita that is at least 60 percent 
of the U.S. GDP per capita.
     A data source that does not have sales and volume data for 
the applicable ASP calendar quarter, but contains sales and volume data 
for any prior calendar quarter beginning on or after October 1, 2019 
from at least one included country.
     The extracted data used by CMS to determine the most 
recent MFN Price used to calculate an MFN Drug Payment Amount posted on 
the MFN Model website.
     A data source with ex-manufacturer price data for the 
applicable ASP calendar quarter from at least one included country.
     A data source with list price data for the applicable ASP 
calendar from at least one included country.
    In each of these cases, if there is more than one data source 
meeting the requirements in Sec.  510.140(c), we will use the data 
source at the highest level of the hierarchy that contains information 
from the highest number of included countries, and, if available, 
incorporates discounts and rebates into its drug pricing information. 
It is possible that we will use different data sources for different 
drugs over different quarters. We will use the data as available from 
the data source, and we will not make adjustments to account for 
differences between the data sources or for confidential rebates. We 
note that, based on the performance year 1 MFN Model Drug HCPCS Codes 
List shown in Table 2, levels 4 and 5 of the hierarchy will only apply 
to MFN Model drugs that are added to the MFN Model Drug HCPCS Codes 
List after performance year 1 and perhaps for Q2043 (Sipuleucel-t auto 
cd54+) \44\ and J2507 (Pegloticase injection), because for other MFN 
Model drugs in performance year 1, the first three levels of the 
hierarchy will always result in an available data source as we consider 
the data used by CMS to create the illustrative MFN Prices and MFN Drug 
Payment Amounts in Table 6 of this IFC to satisfy level 3 of our 
hierarchy. To illustrate: Suppose we identified four data sources 
meeting the requirements of Sec.  510.140(c), where Data Source 1 
contains sales and volume data for MFN Model drug X for the applicable 
ASP calendar quarter from 10 included countries, Data Source 2 contains 
sales and volume data for MFN Model drug X for the applicable ASP 
calendar quarter from 15 included countries, Data Source 3 contains 
sales and volume data from the third calendar quarter of 2020 for MFN 
Model drug X from 16 included countries, and Data Source 4 contains 
list price information for the applicable ASP calendar quarter from all 
included countries. In this scenario, we would use information solely 
from Data Source 2 to determine the MFN Price for MFN Model drug X by 
calculating unadjusted country-level prices for each of the 15 
countries for which Data Source 2 contains information, and we would 
not use Data Sources 1, 3, or 4 to calculate the MFN Price for MFN 
Model drug X for that quarter. For further illustration of how we will 
apply the hierarchy in calculating MFN Drug Payment Amounts, see 
section III.E.4.a. of this IFC.
---------------------------------------------------------------------------

    \44\ No data on international pricing or sales of Sipuleucel-t 
auto cd54+ were available in the data source used for Table 6, but 
international drug pricing information for this drug could be 
available in other sources.
---------------------------------------------------------------------------

    We will use international sales and volume information from as 
early as the third calendar quarter in 2020 to minimize the possibility 
of having no international sales and volume information with which to 
calculate the MFN Price and to mitigate the potential effect of 
manufacturers' limiting the reporting of international drug pricing 
information during the model performance period.
    In addition, the one or more data sources we will use will have 
mechanisms in place to maintain, update, and correct, if necessary, the 
data source on at least a quarterly basis. Further, the data sources we 
will use will be maintained by organizations that seek to limit the lag 
inherent in data to no more than 180 days from the end of the calendar 
quarter for which drug pricing information is compiled to the time that 
the organization makes such updates available to users of the data 
source.
    We plan to monitor the implementation of a World Health Assembly 
(WHA) resolution to ``improve the transparency of markets for 
medicines, vaccines, and other health products.'' This resolution aims 
to help Member States make more informed decisions when purchasing 
health products, negotiate more affordable prices, and ultimately 
expand access to health products for their populations. In particular, 
the WHA resolution \45\--
---------------------------------------------------------------------------

    \45\ World Health Assembly Update, 28 May 2019, accessed via: 
https://apps.who.int/gb/ebwha/pdf_files/WHA72/A72_R8-en.pdf.
---------------------------------------------------------------------------

     Urges Member States to publicly share information on net 
prices paid for

[[Page 76198]]

health products, to promote greater transparency on pharmaceutical 
patents and clinical trial results and to improve suppliers' reporting 
of information such as sales revenues and units sold; and,
     Requests the WHO secretariat to support the development 
and implementation of national policies relevant to transparency and to 
monitor the impact of transparency on affordability and availability of 
health products, including the effect of differential pricing.
    We will monitor developments related to this WHA resolution and 
assess its impact on the availability of data we will use to calculate 
and update MFN Drug Payment Amounts.
    As discussed previously, we will use a hierarchy when selecting 
from available data sources and start by using data sources that 
incorporate discounts and rebates to the extent possible in order to 
address commenters' concerns about a shift internationally to a high 
price and high rebate pricing strategy. We believe that using one or 
more data sources will help to ensure that we will capture sufficient 
information to monitor the international drug pricing landscape and to 
calculate and update MFN Drug Payment Amounts. Data sources that 
include the information described previously, as determined by CMS, 
will be considered sufficient, and as such, we will calculate MFN Drug 
Payment Amounts for MFN Model drugs using information extracted from 
such data sources. Specifically, as necessary, for each MFN Model drug, 
we will extract and use data that align with the data sources' 
standardized method for identifying scientific and nonproprietary names 
and dosage forms (for example, injectable forms), and with the HCPCS 
code's long descriptor, including dosage form, for the HCPCS codes on 
the MFN Model Drug HCPCS Codes List, as applicable. Further, we will 
only use the extracted data for dosage formulations that could be 
described by the MFN Model drug's HCPCS code descriptor as determined 
by CMS when such limitation is not feasible prior to extracting the 
data. For example, for a drug, one HCPCS code may include drug products 
that are a certain type of formulation, such as short-acting, 
intravenously administered drug products, and another HCPCS code may 
include drug products with the same scientific and nonproprietary name 
but a different formulation (such as a long-acting suspension for 
intramuscular injection), and the extracted data contains international 
drug pricing information for both formulations. In such case, we will 
align the extracted data in accordance with the HCPCS code descriptor 
for the MFN Model drug. In order to align with our existing policies 
for how we utilize manufacturer-reported ASP data to calculate payment 
limits, we may find it necessary to make adjustments to the data that 
we extract from international drug pricing information data sources. 
For example, in calculating payment amounts based on ASP we do not 
adjust the volume or units of a drug (that is, the amount of a drug in 
a package) for intentional overfill (see 75 FR 73466). If we find that 
a data source from which we obtain international drug pricing 
information makes adjustments for overfill, we will make adjustments to 
the data that we extract from such source so that the extracted data is 
comparable to ASP data. There could be other cases where we will have 
to examine the extracted data and make adjustments to align the data 
with a HCPCS code descriptor for an MFN Model drug. Specifically, we 
will adjust the extracted international drug pricing information for 
MFN Model drugs when the data source shows the package size of a drug 
product that is inconsistent with the manufacturer's information about 
that product as determined by CMS. In such cases where we confirm a 
difference, we will make adjustments to the pricing, sales and volume 
data as necessary before calculating the unadjusted country-level price 
for the drug at the HCPCS code level. We believe that such cases will 
be rare. However, we identified the need to make such adjustment to the 
international drug pricing information we used to illustrate the MFN 
Drug Payment Amounts for J9311 (Inj rituximab, hyaluronidase) shown in 
Table 6 to align the package size volume with manufacturer labeling and 
the HCPCS code dosage descriptor. We note that there could be 
additional cases if international drug pricing data sources that we 
will select show prices, sales or volume data that are adjusted for 
intentional overfill, include multiple ingredients for a single drug 
product, or are in error (for example, the package size represents the 
maximum volume of a vial instead of the volume of drug in a package).
    We will only use the extracted data that have complete package size 
information. As discussed previously, we will use a hierarchy to 
determine which data source to use for each MFN Model drug for a 
quarter, in which we will select a data source that includes sales and 
volume data first. Data without both sales and volume data will not be 
able to be combined with other data, therefore we will exclude such 
observations. For data sources with international sales and volume data 
for a given MFN Model drug, we will exclude from the calculation of the 
unadjusted country-level price data that fall below a minimum threshold 
or are incomplete, that is, international pricing data with less than 
$1,000 in quarterly sales, with less than 1,000 units in quarterly 
volume, or where both sales and volume data are not present. We believe 
that $1,000 in quarterly sales and 1,000 units in quarterly volume for 
a package size is an appropriate minimum necessary to establish 
sufficient sales and volume for data to be included in the calculation 
of a meaningful and reliable unadjusted country-level price for an MFN 
Model drug and will minimize inclusion of potential outlier data. We 
will exclude presentations with low volume or low sales to prevent 
outlier presentations from exerting undue influence.
    In developing the illustrative MFN Prices shown in Table 6, we 
applied these exclusions. Minimal sales and volume across all countries 
were excluded because of the low volume or sales exclusion criteria. We 
explored the impact of different volume and expenditure thresholds, and 
determined that $1,000 in quarterly sales and 1,000 units are a 
reasonable threshold to reduce risk associated with extremely low 
values. We found that data with potential outlier sales remained 
relatively common with lower thresholds (that is, below $1,000 in 
quarterly sales). While using higher thresholds may further reduce 
potential inclusion of outlier sales data, doing so would result in 
having less data to calculate unadjusted country-level prices.
    The exclusion of international pricing data with less than $1,000 
in quarterly sales or with less than 1,000 units in quarterly volume 
from the calculation of the unadjusted country-level price will greatly 
minimize the potential risk for including possible outlier or errant 
data. To better understand this potential issue, we considered the 
impact of including or excluding data with less than $1,000 in 
quarterly sales or less than 1,000 units in quarterly volume in the 
calculation of the unadjusted country-level price. There was little 
impact from including these data but, as a potential safeguard to 
prevent inclusion of inappropriately low or high international drug 
pricing information in our calculations for the MFN Model, we will 
exclude such data from the calculation of the unadjusted country-level 
price. Overall, where this approach had more than a 1 percent

[[Page 76199]]

impact, there tended to be an increase in the MFN Prices.
    We also considered whether pricing information that is greater than 
or less than 95 percent of the mean across all data for the drug at the 
equivalent of the HCPCS code billing unit level should be considered a 
possible outlier or error and whether trimming such data or removing 
such data would be warranted. In our experience with international drug 
pricing information data sources, outlier or potentially erroneous data 
appear only in isolated instances and are often suggestive of 
unintended differences in the unit at which data is shown. For example, 
the pricing data for a product with a standard unit of one gram in one 
country could appear to be 1,000 times lower than the pricing data for 
that same product from other countries in the data source; in such a 
case, it seems likely that the data for the one country with a very low 
relative price represents the price per milligram not per gram and such 
data would likely be corrected over time by the data source. We believe 
international drug pricing data sources have mechanisms to correct such 
discrepancies based on market research of currently available 
international drug pricing information data sources. Further, as 
codified in Sec.  513.140(c), the international drug pricing 
information data sources that we will obtain will have mechanisms in 
place to maintain, update, and correct, if necessary, the information 
on international drug pricing in the database on at least a quarterly 
basis. As such, because we will revise the MFN Drug Payment Amounts 
quarterly, we will recalculate the MFN Drug Payment Amounts for up to 
four prior quarters when revised international drug pricing information 
is available in the data source that we used to calculate the MFN Model 
drug's MFN Price for the relevant quarter or ASP updates for the 
relevant quarter are available.\46\ In cases where an MFN Drug Payment 
Amount for a prior quarter is recalculated by CMS, CMS will 
prospectively apply the recalculations in the quarterly update 
following the availability of revised international drug pricing 
information and ASP updates, and will not automatically reprocess 
claims to apply the recalculation, but reserves the right to do so. To 
the extent that MFN Model claims are reprocessed due to revisions to 
the international drug pricing information, the Medicare payment amount 
and beneficiary cost sharing will be recalculated to reflect the 
revised prices. If prior to calculating the unadjusted-country level 
prices for a quarter, the data source confirms that there is an error 
that they plan to correct in a future version of the dataset and we 
have the corrected information, we will make the correction to avoid 
the need to reprocess claims later. Therefore, we do not believe it is 
necessary to take further steps to trim or remove potential outlier or 
erroneous international drug pricing information before calculating the 
unadjusted country-level prices. We note that CMS does not make outlier 
adjustments to ASP data.
---------------------------------------------------------------------------

    \46\ We will apply the recalculations in the quarterly update 
following the availability of revised international drug pricing 
information and ASP updates.
---------------------------------------------------------------------------

    In addition, for future years, we seek comment on whether a 
threshold should be applied to determine whether the MFN Drug Payment 
Amount should be recalculated for a prior quarter. Specifically, we are 
interested in comments on whether recalculations should only occur when 
the international drug pricing information data source used corrects 
its data and the impact on the MFN Price is more than a nominal amount. 
We seek comment on the appropriate amount of such threshold and how a 
nominal amount should be defined. Finally, in the event that the 
international drug pricing information data source that we used to 
calculate the MFN Drug Payment Amount for an MFN Model drug for a 
quarter identifies an error in their data and does not correct such 
error within 180 days after the applicable ASP calendar quarter, we 
seek comment on whether CMS should recalculate the MFN Drug Payment 
Amount for such MFN Model drug and quarter using international drug 
pricing information in accordance with the hierarchy in Sec.  
513.140(c)(3) after excluding the data source we initially used. We 
also seek comment on whether CMS should adopt an alternative approach 
to remediating such data errors.
2. International Data Included in the MFN Model
    In the October 2018 ANPRM, for purposes of a potential IPI Model, 
we stated that we were considering using pricing data from the 
following countries: Austria, Belgium, Canada, Czechia, Denmark, 
Finland, France, Germany, Greece, Ireland, Italy, Japan, Netherlands, 
and the United Kingdom. We considered including these countries' data 
as they are either economies comparable to the U.S. or they are 
included in Germany's market basket for reference pricing for their 
drug prices, and existing data sources contain pricing information for 
these countries. We received wide-ranging and helpful feedback in 
response to the October 2018 ANPRM regarding which countries' data to 
include in a model. In addition to comments received to the October 
2018 ANPRM, we also conducted significant outreach to stakeholders, 
such as stakeholder meetings and conference calls, to gather targeted 
feedback. There was also a substantial number of media and press 
reports surrounding which countries' data to include in the MFN Model.
    Generally, we received a significant number of comments that 
expressed opposition to including data from countries that have health 
care systems that are substantially dissimilar to the U.S.'s health 
care system. Specifically, many commenters stated that data from 
countries utilizing government-run health care systems or imposing 
strict drug price controls should be excluded. Alternatively, other 
commenters noted that CMS should consider broadening the scope to 
include more countries, because the more countries that are included in 
the index, the harder it would be for pharmaceutical companies to 
manipulate or game the pricing changes. Commenters also recommended 
utilizing various criteria for selecting the countries that would be 
included, such as the launching speed of new drugs, the presence of 
rigorous health technology assessment, the proportions of public and 
private markets, the economies of those countries, and Human 
Development Index (HDI).\47\
---------------------------------------------------------------------------

    \47\ The Human Development Index is utilized by the United 
Nations and is ``is a summary measure of average achievement in key 
dimensions of human development: a long and healthy life, being 
knowledgeable and have a decent standard of living. The HDI is the 
geometric mean of normalized indices for each of the three 
dimensions.'' Please see the United Nations Development Programme's 
Human Development Reports for more information: http://hdr.undp.org/en/content/human-development-index-hdi.
---------------------------------------------------------------------------

    Based on the comments received, we believe the most appropriate 
criteria for considering a country for MFN pricing is membership in the 
OECD and GDP per capita relative to the U.S. The current list of OECD 
countries includes all countries included in the October 2018 ANPRM as 
well as Australia, Chile, Colombia, Estonia, Hungary, Iceland, Israel, 
Latvia, Lithuania, Luxembourg, New Zealand, Norway, Poland, Portugal, 
Republic of Korea, Slovakia, Slovenia, Spain, Switzerland, and Turkey. 
OECD countries comprise a set of countries that share with the U.S. 
democratic principles and commitment to market-based economies, and 
these countries' GDP per capita (based on purchasing power parity) 
range from

[[Page 76200]]

approximately 25 percent of the U.S. GDP per capita to over 175 percent 
of the U.S. GDP per capita. Based on this wide range of GDP per capita 
data, we believe it is most appropriate to include available 
international drug pricing information for countries with a GDP per 
capita of at least 60 percent of the U.S. GDP per capita, as codified 
in Sec.  513.140(b). We believe that applying a minimum of 60 percent 
of the U.S. GDP per capita strikes a balance between--(1) having too 
low a GDP per capita threshold and including data from countries with 
economies that are substantially different from the U.S., while; (2) 
also not having such a high GDP per capita threshold that the list of 
countries would be very short, which commenters suggested we should 
avoid. To avoid creating a potential incentive for countries to 
discontinue their membership in the OECD, we will include available 
international drug pricing information for countries that were OECD 
members as of October 1, 2020, regardless of whether they remain OECD 
members after October 1, 2020, unless the country's GDP per capita, as 
determined by CMS quarterly, falls below the threshold of 60 percent of 
the U.S. GDP per capita. Based on available data, this means that we 
will calculate the MFN Price for the first quarter of performance year 
1 based on available international drug pricing information from 
Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, 
Iceland, Ireland, Israel, Italy, Japan, Republic of Korea, Luxembourg, 
Netherlands, New Zealand, Norway, Spain, Sweden, Switzerland, and the 
United Kingdom.\48\ These 22 OECD countries are among the countries 
with the highest GDP per capita worldwide.\49\
---------------------------------------------------------------------------

    \48\ The non-U.S. OECD countries that will not be an included 
country for purposes of calculating the MFN Price for MFN Model 
drugs for the first quarter of performance year 1 will be Chile, 
Colombia, Czechia, Estonia, Greece, Hungary, Latvia, Lithuania, 
Mexico, Poland, Portugal, Slovakia, Slovenia, and Turkey.
    \49\ https://www.cia.gov/library/publications/the-world-factbook/fields/211rank.html.
---------------------------------------------------------------------------

    We considered alternative approaches to including data from 
countries for the MFN Model. Specifically, we considered including all 
non-U.S. OECD countries or selecting countries based on factors such as 
World Health Organization (WHO) recognition as a Stringent Regulatory 
Authority (SRA) and intellectual property protections. We also 
considered including data from only countries that may represent large 
markets for drug manufacturers such as the European Union, Canada, 
Japan, and United Kingdom. Additionally, the Foundation for Research on 
Equal Opportunity (FREOPP) recommended an alternative approach called 
the Market-Based International Index (MBII) as a benchmark for 
evaluating other countries' prescription drug pricing systems; \50\ 
this approach would include data from the following countries that 
FREOPP identified as having market-based health care systems: Austria, 
Belgium, Czechia, Denmark, France, Germany, Ireland, Japan, 
Netherlands, Portugal, Singapore, Slovakia, and Switzerland.
---------------------------------------------------------------------------

    \50\ Roy, A. (2018) The Foundation for Research on Equal 
Opportunity, ``What Medicare Can Learn From Other Countries on Drug 
Pricing,'' accessed via https://freopp.org/what-medicare-can-learn-from-other-countries-on-drug-pricing-bf298d390bc5.
---------------------------------------------------------------------------

    Based on analyses examining potential alternatives, we believe that 
none of these alternative approaches would be as objective and 
predictable for purposes of calculating MFN Prices as our approach. Our 
approach will result in a large set of countries that are economically 
similar, have reasonably comparable purchasing power to the U.S., and 
generally have existing international drug pricing information that is 
available. We considered an alternative that would phase-in countries 
over time based on a defined set of characteristics, such GDP per 
capita or average drug prices. We believe that phasing in countries 
over time would create instability in the MFN Price. Thus we are 
adopting a set of included countries that meet the requirements in 
Sec.  513.140(b), which allows for the inclusion of data from countries 
that were non-U.S. OECD member countries as of October 1, 2020, when 
CMS calculates the MFN Drug Payment Amounts for a calendar quarter. 
That means that at the end of each applicable ASP calendar quarter, CMS 
will assess the non-U.S. OECD member countries as of October 1, 2020, 
that have a GDP per capita that is at least 60 percent of the U.S. GDP 
per capita. Because available GDP data are updated infrequently, we 
believe this approach will result in a highly stable process for 
developing the MFN Prices.
    We will include available international drug pricing information 
from the included countries when such data are contained in the data 
sources that we have described in Sec.  513.140(c), as described in 
section III.E.1. of this IFC.
    There are several existing sources for GDP data, including the 
Central Intelligence Agency (CIA) World Factbook,\51\ the World 
Bank,\52\ and the International Monetary Fund.\53\ Upon examining these 
sources, we noted that the GDP data across these sources are highly 
associated with one another. We will use the CIA World Factbook as our 
source for GDP data as it is issued by a U.S. government agency and 
includes estimates for all OECD member countries. We will use the 
following process to determine the countries that were non-U.S. OECD 
member countries as of October 1, 2020, with a GDP per capita that is 
at least 60 percent of the U.S. GDP per capita. For each country, we 
will assess the GDP per capita based on purchasing power parity that is 
available in the CIA World Factbook at the end of the applicable ASP 
calendar quarter. The CIA World Factbook contains the most recent 
estimate of GDP per capita based on purchasing power parity for a 
country as well as historical data. We will identify whether a country 
has a GDP per capita that is at least 60 percent of the U.S. GDP per 
capita by dividing the most recent estimate of GDP per capita based on 
purchasing power parity for a country by the U.S. GDP per capita, using 
data for the same year, and assessing the results. We will use the GDP 
per capita from the same year as the international drug pricing 
information that is used to calculate the unadjusted country-level 
price, if available, or the most recent prior year.
---------------------------------------------------------------------------

    \51\ https://www.cia.gov/library/publications/the-world-factbook/fields/211rank.html.
    \52\ https://data.worldbank.org/indicator/ny.gdp.mktp.cd.
    \53\ https://www.imf.org/external/pubs/ft/weo/2019/01/weodata/weoselgr.aspx.
---------------------------------------------------------------------------

3. Definition of the MFN Drug Payment Amount
    As described later in this section, we will calculate the MFN Drug 
Payment Amount for a calendar quarter for the MFN Model drug based on a 
phased-in blend of the applicable ASP and the MFN Price, which we will 
determine by selecting the lowest GDP-adjusted country-level price from 
the included countries for the applicable ASP calendar quarter.
4. Calculation of the MFN Drug Payment Amounts
    We will calculate an MFN Drug Payment Amount for each MFN Model 
drug for which there is international drug pricing information from at 
least one data source that meets our criteria for at least one included 
country. Section III.E.6. of this IFC describes an alternative approach 
for calculating the MFN Drug Payment Amount for situations where no 
international drug

[[Page 76201]]

pricing information is available for an MFN Model drug, for example, 
because the MFN Model drug is not approved for marketing by any 
included country.
    When using international drug pricing information to calculate the 
MFN Drug Payment Amounts, we want to account for the relative economic 
resources of non-U.S. countries to be able to fairly compare country-
level prices. We will address relative economic resources in two ways: 
(1) We will only use available international drug pricing information 
from non-U.S. OECD member countries with a GDP per capita that is at 
least 60 percent of the U.S. GDP per capita; and (2) we will adjust the 
extracted country-level prices using a GDP adjuster that adjusts for a 
country's GDP per capita if it is lower than that of the U.S.
    Specifically, to calculate the MFN Drug Payment Amounts for a 
calendar quarter in a performance year, we will follow a multi-step 
process using the corresponding quarterly ASP pricing file, as well as 
the available international drug pricing information for included 
countries for the applicable ASP calendar quarter, where available. The 
key steps to calculate the MFN Drug Payment Amount for each MFN Model 
drug will be--
     Identify the available international drug pricing 
information for the MFN Model drug (by applying the hierarchy of data 
sources obtained by CMS and extracting the relevant data); \54\
---------------------------------------------------------------------------

    \54\ Applicable subsequent steps depend upon the level of the 
hierarchy for the selected data source. For example, when there are 
no international sales and volume data available for the drug for an 
applicable ASP calendar quarter or from any quarter beginning on or 
after October 1, 2019, in accordance with level 3 of the hierarchy, 
we will use the extracted data used by CMS to determine the most 
recent MFN Price used to calculate an MFN Drug Payment Amount posted 
on the MFN Model website, including the data used by CMS to create 
the illustrative MFN Prices and MFN Drug Payment Amounts in Table 6 
of this IFC. In such case, it will not be necessary to redo steps to 
extract data from the data source; however, CMS will follow the 
remaining steps in the MFN Drug Payment Amount calculation.
---------------------------------------------------------------------------

     Remove incomplete and low sales and volume data, as 
applicable;
     Convert extracted volume data to the HCPCS code unit level 
and adjust for volume issues such as intentional overfill, as 
applicable;
     Calculate \55\ the unadjusted country-level price 
(representing the average price per unit of drug where the unit of drug 
is the same as the HCPCS code billing unit) for the MFN Model drug for 
each included country with available data in the selected data source 
for that drug;
---------------------------------------------------------------------------

    \55\ The calculation used depends upon whether volume data is 
available.
---------------------------------------------------------------------------

     Calculate the GDP adjuster for each included country;
     Apply the GDP adjuster to the unadjusted country-level 
price;
     Select the lowest GDP-adjusted country-level price for 
each MFN Model drug, which, if available, will be the MFN Price;
     Identify the applicable ASP (which we define as the 
payment amount determined in accordance with 1847A of the Act, less the 
applicable add-on percentage, for the MFN Model drug's HCPCS code); 
\56\
---------------------------------------------------------------------------

    \56\ In general, the ASP Pricing File contains payment limits 
based on 106 percent of the volume weighted average of 
manufacturers' ASP for a given HCPCS code. To identify the 
applicable ASP, we will divide the payment limit by 1.06 after 
removing the blood clotting factor furnishing fee, if applicable. 
For a biosimilar, we will remove the amount that represents 6 
percent of the reference biological product's ASP.
---------------------------------------------------------------------------

     Compare the MFN Price to the applicable ASP (to apply 
limit, if applicable);
     Identify the applicable phase-in formula and adjustments; 
and
     Apply the applicable phase-in formula and adjustments, if 
applicable, to calculate the MFN Drug Payment Amount.
    The following paragraphs further describe how we will calculate the 
MFN Model Drug Payment Amounts for each MFN Model drug for each 
calendar quarter during the model:
a. Identify the Available International Drug Pricing Information for 
the MFN Model Drug
    Using the data sources that we obtain and applying the hierarchy 
described previously in this IFC, we will extract the available 
international drug pricing information for an MFN Model drug for the 
applicable time period (that is, the applicable ASP calendar quarter) 
by aligning the MFN Model drug's HCPCS code long description (in terms 
of name and dosage form) with the data sources' standard method for 
identifying scientific names or nonproprietary names (such as the 
International Nonproprietary Names). That is, for an MFN Model drug, we 
will identify the data sources' standardized scientific name or 
nonproprietary name for that drug, and then use that naming to identify 
data for all products within that data source with an applicable 
formulation. We will extract the applicable data (for example, data for 
all package sizes for injectable forms of the drug aligned with the 
identified scientific or nonproprietary name and formulations, for the 
included countries) from the data source for the applicable ASP 
calendar quarter, and in accordance with our hierarchy, select the data 
source for the MFN Model drug for that quarter.
    As previously discussed in this IFC, we will only use extracted 
data from the selected data source that appears complete and represent 
dosage formulations that could be described by the MFN Model drug's 
HCPCS code descriptor, as determined by CMS. For example, J0178, 
Aflibercept injection, represents injectable ophthalmic formulations 
whereas a data source may contain data for aflibercept for both 
ophthalmic and systemic formulations; only data for ophthalmic 
formulations will be used to calculate the MFN Price for such drug. The 
international drug pricing data used to calculate the MFN Price will 
not be limited to distinguish between products with different inactive 
ingredients (for example, different excipients) or whether or not the 
product is protein bound. However, we will limit the international drug 
pricing data for combination drugs that contain multiple active 
ingredients or biological products to the extent feasible, as 
determined by CMS. This approach is particularly relevant for four of 
the MFN Model drugs for performance year 1, aflibercept injection 
(J0178), which represents ophthalmic formulations compared to systemic 
formulations; paclitaxel protein bound (J9264), which represents 
protein bound formulations compared to formulations of paclitaxel that 
are not protein bound; ferric carboxymaltos (J1439), which represents 
injected formulations compared to oral formulations; and rituximab, 
hyaluronidase (J9311), which represents formulations for subcutaneous 
administration compared to formulations of rituximab for intravenous 
administration.
    In accordance with the hierarchy for selecting international drug 
pricing information data sources, we will prioritize use of 
international drug pricing information that includes sales and volume 
data for the applicable ASP calendar quarter if such information is 
available for a drug for one or more included countries. If more than 
one such data source is available, we will select the data source with 
international drug pricing information for the applicable ASP calendar 
quarter, even if another data source includes a higher number of 
included countries. For example, if the applicable ASP calendar quarter 
is the third quarter of 2021 and an available data source has sales and 
volume data for a drug for 20 of the included countries for the second 
quarter of 2021 and for 15 included countries for the third quarter of 
2021, we would extract and then calculate unadjusted country-level 
prices for that drug based on sales and volume data

[[Page 76202]]

from the third quarter of 2021 only for the 15 included countries for 
which data from that quarter are available.
    If there are available data from a data source at the second level 
of our hierarchy (that is, no international sales and volume data for 
the applicable ASP calendar quarter, but sales and volume data from any 
quarter beginning on or after October 1, 2019), for a drug, we will use 
available international sales and volume data from that data source for 
the most recent prior quarter that begins on or after October 1, 2019 
for that drug for included countries.
    If there are no international sales and volume data available for 
the drug, we will use the extracted data used by CMS to determine the 
most recent MFN Price used to calculate an MFN Drug Payment Amount 
posted on the MFN Model website, in accordance with the third level of 
the hierarchy.
    If no MFN Drug Payment Amount has been publicly posted for the 
drug, we will use a data source at the fourth level of our hierarchy if 
available (the data source contains ex-manufacturer price data but does 
not include volume data for the applicable ASP calendar quarter).
    If ex-manufacturer price data for the applicable ASP calendar 
quarter are not available, we will use a data source at the fifth level 
of our hierarchy (the data source contains list price data for the 
applicable ASP calendar quarter).
b. Remove Incomplete Low Sales and Volume Data, as Applicable
    If the data source we select has sales and volume data at the 
package level for an included country, we will apply the exclusions for 
data with incomplete data and low sales and volume. That is, we will 
exclude data without both sales and volume data, with less than $1,000 
in quarterly sales (expressed as U.S. currency), or with less than 
1,000 units in quarterly volume.
c. Convert the Extracted Volume Data to the HCPCS Code Unit Level and 
Adjust for Volume Issues, Such as Intentional Overfill, as Applicable
    We will adjust the remaining volume data to the same level as the 
HCPCS billing unit, as applicable. For example, if the data for a 
package size shows the volume is 1,000 units and each unit represents a 
1 MG vial package and for another package size the volume is 500 units 
and each unit represents a 10 MG vial package, and both of these data 
are for a drug assigned to the same HCPCS code with a HCPCS billing 
unit of 1 MG, the adjusted volume data for these packages would be 
1,000 units and 5,000 units, respectively, for a total adjusted volume 
of 6,000 units. The volume for the 1 MG vial package is unchanged 
because the amount of drug in one package (that is, 1 MG) equals the 
amount of drug in one HCPCS billing unit. The volume for the 10 MG vial 
package is changed to be 10 times higher because the amount of drug in 
one vial (that is, 10 MG) equals 10 times the amount of drug in one 
HCPCS billing unit.
    Before this step is performed, as applicable, we will adjust the 
extracted volume information before converting it to the HCPCS billing 
unit level when the data source shows the package size of a drug 
product that is inconsistent with the manufacturer's information about 
that product based on the available product information, such as 
package labeling, compared to the data extracted from the data source. 
In addition, we will limit the number of billing units in a package 
when the available package labeling specifies use of a limited amount 
of drug is to be used from the package. For example, we will limit the 
number of billing units in a package for an aflibercept vial to one 2 
mg dose in accordance with available package labeling, which specifies 
that each vial, regardless of the labeled volume, has one 2 mg dose. 
For injectable formulations for HCPCS codes with dosage specified as 
per dose, we will limit the number of billing units in a package to no 
more than one per vial. This approach was applied to illustrate the MFN 
Prices for J7324 (Orthovisc inj per dose) in Table 6.
d. Calculate the Unadjusted Country-Level Price for the MFN Model 
Drug's HCPCS Code for Each Included Country With Available Data in the 
Selected Data Source for That Drug
    Using the data available after completing the prior steps, we will 
calculate the unadjusted country-level price for each included country 
with available data. The unadjusted country-level price represents the 
average price per unit of drug where the unit of drug is the same as 
the HCPCS code billing unit.
    We will use a calculation that is applicable to the data available 
at this step. If volume data are available, we will use a calculation 
that includes volume-weighting across the different data (which often 
represent different package sizes) of the drug included in the data 
source for the country to calculate the unadjusted country-level price. 
If volume data are not available, we will use a calculation that treats 
all packages of the drug included in the data source for the country 
equally, after converting the pricing data to the HCPCS code unit 
level, in calculating the unadjusted country-level price.
    If sales and volume data are available, we will first sum the 
adjusted volume data for all package sizes for the drug. We will then 
sum the total sales for all package sizes for the drug, and divide that 
sum by the sum the adjusted volume data for all package sizes for the 
drug, resulting in an average price per unit of drug where the unit of 
drug is the same as the HCPCS code billing unit. If the data source we 
select has ex-manufacturer or list prices and does not have volume 
data, we will calculate the number of HCPCS billing units in a package 
and divide the ex-manufacturer price or list price for a package by the 
number of HCPCS billing units in the package, resulting in a price per 
unit of drug for each package listed in the data source. We will then 
sum the price per unit of drug for each package listed in the data 
source for the drug and divide the sum by the number of packages listed 
in the data source for the drug, resulting in an average price per unit 
of drug where the unit of drug is the same as the HCPCS code billing 
unit.
    We will repeat this process for each country specified in Sec.  
513.140(b), to the extent international drug pricing information for 
the drug for the country is available from the selected data source. As 
explained previously and specified in Sec.  513.140(c)(3)(i), we will 
use the highest tier data source, in accordance with the hierarchy, 
which includes data for the drug in at least one included country. If 
the selected data source for a drug for a calendar quarter does not 
include data from a particular included country, we will still 
calculate the MFN Price for that drug using the data from the selected 
data source based on the included countries from which there are data 
for the drug. We will not include any information from countries that 
did not have data in the selected data source for that drug. In cases 
where there is no data source that meets our criteria for using 
international drug pricing information (that is, there are no 
international sales, volume, or other pricing data available from any 
of the included countries in our international drug pricing information 
data sources, including data used by CMS to determine the most recent 
MFN Price used to calculate an MFN Drug Payment Amount posted on the 
MFN Model website, for an MFN Model drug for any quarter beginning on 
or after October 1, 2019 up to and including the model performance 
period, we will not calculate an unadjusted country-level price (or 
GDP-adjusted country-level price) and will instead use the applicable 
ASP (which we will define as

[[Page 76203]]

the payment amount determined in accordance with section 1847A of the 
Act minus the applicable add-on percentage, for the MFN Model drug's 
HCPCS code) as the MFN Model Drug Payment Amount, as described in 
section III.E.6. of this IFC.
e. Calculate the GDP Adjuster for Each Included Country
    As discussed previously, we want the MFN Price to account for the 
relative economic resources and purchasing power for each included 
country to be able to fairly compare country-level prices. As such, we 
will calculate a GDP adjuster, using a country's GDP per capita based 
on purchasing power parity, that will be used to adjust the unadjusted 
country-level price for each drug (whether based on international sales 
and volume data or international ex-manufacturer or list prices) to 
reflect the country's economic resources relative to the U.S. We 
believe that GDP per capita based on purchasing power parity represents 
a broadly used and reliable measure of a country's economic resources 
to ensure a meaningful comparison of country-level prices.
    As previously mentioned, there are several existing sources for GDP 
data, including the CIA World Factbook,\57\ the World Bank,\58\ and the 
International Monetary Fund.\59\ Our analyses suggest that the GDP data 
across these sources are highly associated with one another. We will 
use the CIA World Factbook as our source for GDP data as it is issued 
by a U.S. government agency and includes estimates for all current OECD 
member countries. The GDP adjuster will be based on the GDP per capita 
available from the CIA World Factbook at the end of the applicable ASP 
calendar quarter. We will use the most recent GDP per capita data 
available for each included country and the U.S. GDP per capita from 
the same year as the GDP per capita data that is available from the 
included country. For example, if the most recent GDP per capita from 
the comparison OECD country is from 2016 and the most recent U.S. GDP 
per capita is 2017, then we will use the GDP per capita from 2016 for 
both countries when comparing. In cases where we use international drug 
pricing information from a quarter other than the applicable ASP 
calendar quarter (that is, an earlier time period) to determine the 
unadjusted country-level price, we will use the GDP per capita data for 
that time period, if available, or the most recent earlier data 
available. That is, CMS will use the GDP per capita for the same year 
as the data used to calculate the unadjusted country-level price, if 
available, or the most recent earlier year available.
---------------------------------------------------------------------------

    \57\ https://www.cia.gov/library/publications/the-world-factbook/fields/211rank.html.
    \58\ https://data.worldbank.org/indicator/ny.gdp.mktp.cd.
    \59\ https://www.imf.org/external/pubs/ft/weo/2019/01/weodata/weoselgr.aspx.
---------------------------------------------------------------------------

    To create a simple, easily understandable GDP adjuster, each 
country's GDP adjuster will be a straight ratio of its GDP per capita 
based on purchasing power parity divided by U.S. GDP per capita, 
subject to the limitation described later in this section. The U.S. GDP 
per capita for 2017, the most current data available, was $59,800. 
Table 4 presents GDP per capita for 2017 and the GDP adjusters for each 
non-U.S. OECD member country, based on the U.S. GDP per capita of 
$59,800 for 2017, that we will use to calculate the MFN Drug Payment 
Amounts for performance year 1, quarter 1. In cases when an included 
country's GDP per capita and the U.S. GDP per capita are not updated in 
the CIA World Factbook at the same time, we will use the most recent 
GDP per capita for the included country and the U.S. GDP per capita 
from the same year to ensure that the GDP adjuster for an included 
country is calculated using GDP data from both countries from the same 
time period. For example, if at the end of an applicable calendar 
quarter a 2018 estimate of a country's GDP per capita based on 
purchasing power parity becomes available in the CIA World Factbook but 
the most recent U.S. GDP per capita available in the CIA World Factbook 
continues to be for 2017, we will continue to use data from 2017 for 
both countries to calculate the GDP adjuster for that country.
    The GDP adjuster will be capped at 1 such that the adjuster will 
only increase the unadjusted country-level price for a drug; it will 
not decrease it. We will cap the GDP adjuster at 1 because its purpose 
is to adjust for countries' economic resources when lower than those of 
the U.S. Capping the GDP adjuster at 1 will ensure that we do not make 
an adjustment that would result in an amount that would be lower than 
the unadjusted country-level price. For example, if Country X with a 
higher GDP per capita based on purchasing power parity than the U.S., 
such as a GDP per capita ratio of 2, has an unadjusted country-level 
price of $100 for an MFN Model drug, we would use a GDP adjuster of 1.0 
and calculate a GDP-adjusted country-level price of $100 rather than 
using a GDP adjuster of 2.0 and calculating a GDP-adjusted country-
level price of $50.

     Table 4--Non-U.S. OECD Member Country GDP per Capita (Based on
   Purchasing Power Parity) and GDP Adjusters for Performance Year 1,
                                Quarter 1
------------------------------------------------------------------------
                                        CIA GDP  per
                                       capita,  based  GDP adjuster  for
           OECD countries              on  purchasing  performance  year
                                        power parity     1,  quarter 1
                                           (2017)
------------------------------------------------------------------------
The following countries have a GDP
 per capita of at least 60 percent
 of U.S. GDP per capita:[dagger]
------------------------------------------------------------------------
    Australia.......................          $50,400              0.843
    Austria *.......................           50,000              0.836
    Belgium *.......................           46,600              0.779
    Canada *........................           48,400              0.809
    Denmark *.......................           50,100              0.838
    Finland *.......................           44,500              0.744
    France *........................           44,100              0.737
    Germany *.......................           50,800              0.849
    Iceland.........................           52,200              0.873
    Ireland *.......................           73,200           ** 1.000
    Israel..........................           36,400              0.609

[[Page 76204]]

 
    Italy *.........................           38,200              0.639
    Japan *.........................           42,900              0.717
    Republic of Korea...............           39,500              0.661
    Luxembourg......................          105,100           ** 1.000
    Netherlands *...................           53,900              0.901
    New Zealand.....................           39,000              0.652
    Norway..........................           72,100           ** 1.000
    Spain...........................           38,400              0.642
    Sweden..........................           51,200              0.856
    Switzerland.....................           62,100           ** 1.000
    United Kingdom *................           44,300              0.741
------------------------------------------------------------------------
The following countries have a GDP per capita below 60 percent of U.S.
 GDP per capita:
------------------------------------------------------------------------
    Chile...........................           24,600              0.411
    Colombia........................           14,400              0.241
    Czechia *.......................           35,500              0.594
    Estonia.........................           31,700              0.530
    Greece *........................           27,800              0.465
    Hungary.........................           29,600              0.495
    Latvia..........................           27,700              0.463
    Lithuania.......................           32,400              0.542
    Mexico..........................           19,900              0.333
    Poland..........................           29,600              0.495
    Portugal........................           30,500              0.510
    Slovakia........................           33,100              0.554
    Slovenia........................           34,500              0.577
    Turkey..........................           27,000              0.452
------------------------------------------------------------------------
* Indicates countries that were listed as potential included countries
  in the October 2018 ANPRM (83 FR 54557).
** Indicates that the GDP adjuster is capped at 1.000.
[dagger] The 2017 U.S. GDP per capita is $59,800.

f. Apply the Applicable GDP Adjuster To Calculate the GDP-Adjusted 
Country-Level Price for the MFN Model Drug
    Next, we will apply the country-specific GDP adjuster to the 
unadjusted country-level price for that country by dividing the 
unadjusted country-level price by the country's GDP adjuster. The 
result will be the GDP-adjusted country-level price for the MFN Model 
drug for that country. We will repeat this calculation to produce a 
GDP-Adjusted Price for every country for which we have calculated an 
unadjusted country-level price for the MFN Model drug.
g. Identify the Lowest GDP-Adjusted Country-Level Price for the MFN 
Model Drug
    We will examine the GDP-adjusted country-level prices for the MFN 
Model drug, and identify the lowest GDP-adjusted country-level price 
for the MFN Model drug. The lowest GDP-adjusted country-level price 
will be the MFN Price for the MFN Model drug.
h. Compare the MFN Price to the Applicable ASP
    As a safeguard for beneficiaries, we will compare the MFN Price to 
the applicable ASP in order to ensure that beneficiaries are always 
paying the lowest amount of coinsurance available. If the applicable 
ASP is less than the MFN Price, we will establish the MFN Price as 
equal to the applicable ASP.
i. Identify the Applicable Phase-In Formula and Adjustments
    As described in section III.E.5. of this IFC, we will phase-in the 
use of the MFN Price over the course of the MFN Model. As discussed in 
section III.E.9. of this IFC, we will also accelerate the applicable 
phase-in formula when the applicable ASP for an MFN Model drug rises 
faster than both a designated inflation factor and the change in MFN 
Price, and lower the MFN Drug Payment Amount below the MFN Price by a 
certain percentage if the applicable ASP for an MFN Model drug 
continues to increase faster than the inflation factor and the MFN 
Price after the full phase-in of the MFN Price. In this step of the 
process to calculate the MFN Drug Payment Amount, we will determine the 
applicable phase-in formula and whether any of these adjustments will 
apply.
j. Calculate the MFN Drug Payment Amount
    As the last step, we will calculate the MFN Drug Payment Amount for 
the MFN Model drug using the applicable phase-in formula, which blends 
the applicable ASP and the MFN Price as described in section III.E.5. 
of this IFC. This calculation, including any adjustments that apply, 
will result in the MFN Drug Payment Amount for the MFN Model drug 
(except as otherwise specified).
5. Phase-In of the MFN Price
    We will use a phase-in approach that will blend the MFN Price with 
the applicable ASP to allow MFN participants time to adjust to the 
model payment amounts and processes. The phase-in formula will be 
stable for a given performance year, whereas the MFN Price and 
applicable ASP will vary quarterly based on fluctuations in drug

[[Page 76205]]

prices in the U.S. and in included countries. We will phase-in the MFN 
Price by 25 percent per year for performance years 1 to 3 of the model, 
reaching 100 percent of the MFN Price for performance years 4 through 7 
of the model. The phase-in formula uses a blend of the applicable ASP 
and MFN Price for an MFN Model drug as shown in Table 5. The MFN Drug 
Payment Amount will be based on 100 percent of the MFN Price starting 
in performance year 4, unless an adjustment that accelerates the phase-
in applies as described in section III.E.9. of this IFC. Thus, the 
phase-in represents the outer bound in terms of the amount of time it 
will take for the MFN Drug Payment Amount to transition to 100 percent 
of the MFN Price.
    We believe that a phase-in approach during the initial years of the 
model will enable MFN participants and the markets to adjust to the 
model's payment methodology, while enabling CMS to test the full phase-
in of the MFN Price over a 7-year model performance period. As noted in 
section III.E.11. of this IFC, when MFN Model drugs get added to the 
MFN Model Drug HCPCS Codes List during the model performance period, 
their MFN Drug Payment Amount gets determined as set forth for the 
corresponding performance year, meaning that if an MFN Model drug were 
to be added during performance year 4, the MFN Drug Payment Amount will 
equal 100 percent of the MFN Price.

           Table 5--Phase-In of MFN Prices by Performance Year
------------------------------------------------------------------------
                               Blend of the ASP and MFN price for an MFN
       Performance year            model drug at the HCPCS code level
------------------------------------------------------------------------
Year 1.......................  75 percent applicable ASP and 25 percent
                                MFN Price.
Year 2.......................  50 percent applicable ASP and 50 percent
                                MFN Price.
Year 3.......................  25 percent applicable ASP and 75 percent
                                MFN Price.
Year 4.......................  100 percent MFN Price.
Year 5.......................  100 percent MFN Price.
Year 6.......................  100 percent MFN Price.
Year 7.......................  100 percent MFN Price.
------------------------------------------------------------------------

    We are codifying the phase-in formula in Sec.  513.210(b)(8).
6. Alternative Calculation for the MFN Drug Payment Amount
    Over the course of the MFN Model, we may determine that the 
international drug pricing information data sources that we obtain do 
not contain any international drug pricing information (meaning no 
sales, volume, ex-manufacturer price, or list price data from any 
included country from any quarter beginning in the fourth calendar 
quarter of 2018 through the applicable quarter in the model performance 
period) for an MFN Model drug, for example, because the MFN Model drug 
is not approved for marketing in the included countries. For such 
cases, we will establish the MFN Drug Payment Amount at the applicable 
ASP for the applicable calendar quarter, subject to any adjustment in 
Sec.  513.210(d) that applies, until international drug pricing 
information is available.
    Because international drug pricing information may become available 
for a subsequent calendar quarter, we will use this method to establish 
the MFN Drug Payment Amount instead of excluding or removing drugs 
without any international drug pricing information from the model until 
international drug pricing information becomes available. We believe 
having a stable list of MFN Model drugs will be more predictable for 
MFN participants, lessening MFN participants' need to monitor changes 
to the MFN Model Drug HCPCS Codes List, and will avoid creating an 
opportunity for manufacturers to get their products out of the model by 
stopping the reporting of international drug pricing information. Based 
on our experience with international drug pricing information data 
sources, we expect the potential of no international drug pricing 
information for an MFN Model drug across all included countries will be 
limited. We note that our approach may increase model payments compared 
to non-model payments for MFN Model drugs with no international drug 
pricing information because the alternative add-on payment, a single 
flat add-on amount per dose (see section III.F. of this IFC), could be 
greater than the add-on payment outside of the model.
7. Illustrative MFN Drug Payment Amounts
    To illustrate how CMS will calculate the MFN Drug Payment Amounts 
under the MFN Model in accordance with Sec. Sec.  513.130 and 513.140, 
we applied the methodology for determining the applicable ASPs, MFN 
Prices, and MFN Drug Payment Amounts using historical ASP-based payment 
limits,\60\ available international drug pricing information from 2019 
for the included countries, and the MFN Model performance year 1 phase-
in formula. Table 6 shows illustrative data for applicable ASPs, MFN 
Prices, and MFN Drug Payment Amounts for one billing unit for the HCPCS 
codes that are included on the performance year 1 MFN Model Drug HCPCS 
Codes List in Table 2. Actual MFN Drug Payment Amounts per billing unit 
for performance year 1, quarter 1, and thereafter will be calculated as 
specified in Sec.  513.210. We will publish the quarterly MFN Drug 
Payment Amounts on a CMS website (such as the MFN Model website), 
similar to how the ASP Drug Pricing Files are posted online prior to 
the start of the calendar quarter. The performance year 1, quarter 1 
MFN Drug Payment Amounts will be published on a CMS website before the 
start of the MFN Model.
---------------------------------------------------------------------------

    \60\ We used the 2019 Quarter 3 and Quarter 4 and 2020 Quarter 1 
and Quarter 2 ASP data that align with manufacturer-reported data 
based on sales during 2019 to identify the applicable ASPs. The ASP 
pricing files are posted at links available here: https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Part-B-Drugs/McrPartBDrugAvgSalesPrice/index.html.
---------------------------------------------------------------------------

    Illustrative MFN Drug Payment Amounts per billing unit are listed 
in Table 6 by HCPCS code. For this illustration, we partnered with 
ASPE, which purchases licenses to data products maintained by IQVIA\TM\ 
(formerly known as Quintiles-IMS). IQVIA's proprietary MIDAS data set 
\61\ is a widely used source of drug sales and volume data.
---------------------------------------------------------------------------

    \61\ https://www.iqvia.com/solutions/commercialization/brand-strategy-and-management/market-measurement/Midas.

---------------------------------------------------------------------------

[[Page 76206]]

MIDAS data contain estimates of drug sales (called ``Monetary Value '' 
within the MIDAS data set) and volume (called ``Quantity'' within the 
MIDAS data set) that are based on audits of drug transactions in 
different countries and distribution channels (for example, retail 
pharmacies and hospitals). The audits underlying the MIDAS data collect 
sales and volume information at the ex-manufacturer (that is, prices as 
drugs are sold by manufacturers), ex-wholesaler, and/or retail levels. 
IQVIA applies a set of country- and channel-specific assumptions on 
markups between manufacturer, wholesale, and retail prices to estimate 
ex-manufacturer and retail sales. Sales information within the database 
is stated in local and U.S. currency, as of the transaction date or 
current date, and are expressed as ex-manufacturer, trade, and public 
(retail) sales.\62\ MIDAS uses a variable called ``Molecule List'' 
(also called ``Moleculelist'') which identifies scientific and 
nonproprietary names for drug and biological products. Users extract 
data from the MIDAS database by selecting report filters, which are 
values for various data fields included in the database, such as 
``Molecule List'' and ``NFC123'' (or ``New Form Code,'' a 3-digit code 
which identifies the dosage form \63\). The database has a standard 
method for identifying drugs within the U.S. and across countries, and 
a standard method for identifying drug forms. MIDAS data is updated 
monthly and retains up to 12 years of history.
---------------------------------------------------------------------------

    \62\ Ex-manufacturer sales are: Manufacturer Selling Price or 
Wholesaler Purchasing Price or Price to Wholesaler (PTW). Trade 
sales are: Wholesaler Selling Price or Pharmacy Purchase Price or 
Price to Chemist (PTC). Public (retail) sales are: Pharmacy Selling 
Price or Consumer Purchase Price or Price to the Public (PTP).
    \63\ For more information on the New Form Codes see: https://www.ephmra.org/classification/new-form-codes/.
---------------------------------------------------------------------------

    CMS obtained a MIDAS data extract of available 2019 international 
drug pricing information for the included countries for the MFN Model 
drugs for performance year 1 from ASPE. After identifying the MFN Price 
for each drug, we applied the phase-in formula for performance year 1 
(75 percent of the applicable ASP and 25 percent of the MFN Price) and 
applied the exceptions in Sec.  513.210(d) when no international drug 
pricing information was available in the MIDAS data. In Table 6, the 
illustrative MFN Prices, calculated using available international drug 
pricing sales and volume information at the ex-manufacturer level, 
represent the lowest of the GDP-adjusted country-level prices available 
in the single data source we used. For a complete discussion of how CMS 
used international drug pricing information available through IQVIA and 
CMS data to calculate the illustrative applicable ASPs, MFN Prices, and 
MFN Drug Payment Amounts displayed in Table 6, we refer readers to the 
supplemental documentation available on the MFN Model website.\64\ We 
also refer readers to the Medicare Part B Drug Spending Dash board \65\ 
that can be used to search for brand name or generic name; search 
results present certain manufacturer information when available.
---------------------------------------------------------------------------

    \64\ See: https://innovation.cms.gov/initiatives/most-favored-nation-model/.
    \65\ See: https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Information-on-Prescription-Drugs/MedicarePartB.
---------------------------------------------------------------------------

BILLING CODE 4120-01-P

[[Page 76207]]

[GRAPHIC] [TIFF OMITTED] TR27NO20.003


[[Page 76208]]


[GRAPHIC] [TIFF OMITTED] TR27NO20.004


[[Page 76209]]


[GRAPHIC] [TIFF OMITTED] TR27NO20.005


[[Page 76210]]


[GRAPHIC] [TIFF OMITTED] TR27NO20.006


[[Page 76211]]


[GRAPHIC] [TIFF OMITTED] TR27NO20.007


[[Page 76212]]


[GRAPHIC] [TIFF OMITTED] TR27NO20.008

BILLING CODE 4120-01-C

[[Page 76213]]

    We note that, as codified in Sec.  513.210(d)(5), and described in 
section III.E.10. of this IFC, the MFN Drug Payment Amount will not 
exceed the non-model drug payment amount for line items submitted with 
the JG modifier (or any successor modifier used to identify drugs 
purchased under the 340B program) after removing any add-on amount, if 
applicable.
8. Timing of Data and MFN Drug Payment Amount Calculations
    As discussed in section III.E.4. of this IFC, we will calculate the 
MFN Drug Payment Amounts on a calendar quarter basis using the most 
recent ASP and correlated international drug pricing information (that 
is, data from the highest level of hierarchy available). Under the 
reporting requirements outlined in section 1927(b)(3)(A)(iii), 
manufacturers that report ASPs are required to submit them to CMS no 
later than 30 days after the last day of the previous quarter. CMS uses 
these data to calculate the ASP-based Medicare payment amounts for the 
next calendar quarter. As a result, there is a two-quarter lag between 
the time when sales reflected in the ASP occur and the time when these 
sales become the basis for Medicare payment amounts.
    We will use international drug pricing information from the same 
time period (that is, the same calendar quarter), if available in 
accordance with the hierarchy specified in Sec.  513.140(c)(3), in 
order to align information across the ASP Drug Pricing files and the 
data sources for international drug pricing information that we will 
use. This approach will consistently correspond to the two-quarter lag 
used for the ASP pricing files when an international drug pricing 
information data source at the highest level of the hierarchy specified 
in Sec.  513.140(c)(3) is available. Table 7 illustrates how the 
information we will use to calculate the MFN Drug Payment Amounts for 
each calendar quarter during performance year 1 using data from the 
applicable ASP calendar quarter will align when an international drug 
pricing information data source at the highest level of the hierarchy 
specified in Sec.  513.140(c)(3) is available. We will use the same 
approach for each performance year.

 Table 7--Alignment of Performance Year Calendar Quarters for ASP and MFN Price Data Based on January 2021 Model
                                                      Start
----------------------------------------------------------------------------------------------------------------
                                                       ASP pricing file
       Performance year           Performance year       for calendar       Applicable ASP       MFN price for
                                  calendar quarter         quarter         calendar  quarter   calendar quarter*
----------------------------------------------------------------------------------------------------------------
1.............................  2021, Quarter 1....  2021, Quarter 1....  2020, Quarter 3...  2020, Quarter 3
1.............................  2021, Quarter 2....  2021, Quarter 2....  2020, Quarter 4...  2020, Quarter 4
1.............................  2021, Quarter 3....  2021, Quarter 3....  2021, Quarter 1...  2021, Quarter 1
1.............................  2021, Quarter 4....  2021, Quarter 4....  2021, Quarter 2...  2021, Quarter 2
----------------------------------------------------------------------------------------------------------------
*When an international drug pricing information data source at the highest level of the hierarchy specified in
  Sec.   513.140(c)(3) is available.

    For example, for the initial calculations to calculate payment 
amounts for the start of the MFN Model on January 1, 2021, the 
beginning of the first calendar quarter in 2021, we will use the 
January 2021 ASP Pricing File (which will be based on manufacturers' 
ASP for the third quarter of 2020, from July 1, 2020, to September 30, 
2020) and international drug pricing information for the third quarter 
of 2020, from July 1, 2020, to September 30, 2020. For each subsequent 
calendar quarter for a performance year, the MFN Drug Payment Amount 
will be established by calculating the MFN Price based on more recent 
international drug pricing information, using data for the applicable 
ASP calendar quarter, if available, as illustrated in Table 7, and 
calculating the MFN Drug Payment Amount.
9. Adjustments to Phase-In Formula and Incentives for Manufacturers To 
Address Rising U.S. Drug Prices
    In response to the October 2018 ANPRM, we received several comments 
asking that we consider including model design features to address 
potential spillover effects and cost-shifting to the commercial market 
and Medicare payment outside of the model geographic area. The 
commenters requested that CMS carefully consider the potential impacts 
of a potential model on other markets--including the potential for 
cost-shifting to other segments of the Medicare program, the Medicaid 
program, and the commercial market. The commenters recommended that in 
order to avoid unintended consequences and cost-shifting, CMS should 
closely monitor prices for included drugs and consider additional 
policies or actions if drug prices in other markets rise above certain 
pricing thresholds (for example, above the Consumer Price Index (CPI) 
or inflation).
    We appreciate these concerns, as it is possible that, in response 
to the MFN Model, manufacturers may take steps to increase U.S. prices 
outside of the MFN Model, such as in the commercial and Medicare 
Advantage markets, which may be seen in increases in manufacturers' 
ASPs. In response to the concerns expressed in the October 2018 ANPRM 
comments and to minimize the possibility of a spillover impact on 
beneficiaries outside of the MFN Model, we will make adjustments to the 
phase-in formula in order to mitigate cost-shifting in the market and 
incentivize manufacturers of MFN Model drugs to maintain stable ASPs of 
MFN Model drugs to minimize the potential for spillover impacts. In 
addition to creating spillover impacts, rapid increases in ASP that 
outstrip not only U.S. inflation but also changes in international 
prices over time would reduce our ability to test the phase-in of the 
MFN Price over time, as the MFN Price's contribution to the MFN Drug 
Payment Amount could be obscured by a significant increase in the MFN 
Model drug's ASP.
    As discussed in section III.E.5. of this IFC, we will phase-in the 
MFN Prices to allow MFN participants time to adjust to the MFN Model 
payment amounts and processes. Calculating the MFN Prices and MFN Drug 
Payment Amounts each calendar quarter will allow manufacturers to 
address the large difference between prices in the U.S. and in other 
countries for MFN Model drugs during the course of the MFN Model and 
serves as an incentive for manufacturers to refrain from raising U.S. 
prices faster than a reasonable inflation allowance. Furthermore, as 
discussed in section III.M. of this IFC, we are waiving requirements of 
section 1847A in order to exclude units of MFN Model drugs from the 
calculation of the manufacturer's ASP. However, if these incentives 
prove to be insufficient to deter manufacturers from raising U.S. 
prices for MFN Model drugs faster than a reasonable inflation 
allowance, we will adjust the calculation of the MFN Drug Payment 
Amount by adjusting the

[[Page 76214]]

phase-in formula for MFN Model drugs where such concerns are observed.
    Specifically, to preserve the integrity of the model test as 
described previously, we will make an adjustment to the phase-in 
formula for an MFN Model drug if the applicable ASP or monthly U.S. 
list price (defined as Wholesale Acquisition Cost (WAC) available in a 
U.S. drug pricing compendium or if WAC is not available, other 
available list prices, such as Average Wholesale Price (AWP) available 
in a U.S. drug pricing compendium) increases faster than both inflation 
and the MFN Price. CMS will accelerate the phase-in of the MFN Price by 
5 percentage points at the next quarterly update for each MFN Model 
drug with: (1) A greater cumulative percentage increase in either the 
applicable ASP \66\ or any monthly U.S. list price for any of the NDCs 
assigned to the MFN Model drug's HCPCS code compared to the cumulative 
percentage increase in the Consumer Price Index for All Urban Consumers 
(CPI-U) \67\ based on all items in U.S. city average and not seasonally 
adjusted; and (2) a greater cumulative percentage increase in either 
the applicable ASP or any monthly U.S. list price for any of the NDCs 
assigned to the MFN Drug's HCPCS code compared to the cumulative 
percentage increase in the MFN Price. To apply these conditions for an 
MFN Model drug, we will identify the cumulative percentage increase 
from a baseline to the applicable ASP calendar quarter. For all MFN 
Model drugs with an applicable ASP for the first quarter of performance 
year 1, we will set the baseline as the ASP calendar quarter for the 
applicable ASP for the first quarter of performance year 1 (that is, 
the third calendar quarter of 2020 (July 2020 through September 2020)). 
For all MFN Model drugs that do not have an applicable ASP for the 
first quarter of performance year 1 (for example, a drug that is first 
marketed in the U.S. after the start of the model), the baseline will 
be the ASP calendar quarter for the first applicable ASP based on the 
manufacturer's average sales price for that MFN Model drug that occurs 
after the third quarter of 2020. For example, the baseline for an MFN 
Model drug with its first applicable ASP based on the manufacturer's 
average sales price occurring in the second quarter of performance year 
1 (that is, April 2021 through June 2021) will have a baseline of the 
fourth calendar quarter of 2020 (October 2020 through December 2020).
---------------------------------------------------------------------------

    \66\ We note that the manufacturers' ASPs will be based on non-
model sales only as codified in Sec.  513.600(b) and as discussed in 
section III.M. of this IFC.
    \67\ All references to CPI-U are based on all items in U.S. city 
average and not seasonally adjusted.
---------------------------------------------------------------------------

    The cumulative percentage change will be calculated from the end of 
the baseline to the end of the applicable ASP calendar quarter. We will 
apply the adjustment to the phase-in formula similarly for all MFN 
Model drugs regardless of when the MFN Model drug is added to the MFN 
Model Drug HCPCS Codes List.
    Further, if both conditions are not met, such as the cumulative 
percentage increase in any monthly U.S. list prices for the NDCs 
assigned to the MFN Drug's HCPCS code outpaces the cumulative 
percentage increase in CPI-U but is less than the cumulative percentage 
increase in the MFN Price, then the trigger conditions will not be met 
and the phase-in formula will not be accelerated. If the cumulative 
percentage change in the CPI-U or MFN Price is negative, we will use 
zero as the cumulative percentage increase in the CPI-U or MFN Price, 
as applicable, for the relevant quarter.
    We will accelerate the phase-in formula by 5 percentage points as 
we believe this amount strikes a balance between moving the MFN Drug 
Payment Amount more quickly toward the MFN Price while still retaining 
the stepwise nature of the phase-in. As an example, in the case that 
both trigger conditions are met for an MFN Model drug during the 
applicable ASP calendar quarter for the second quarter of performance 
year 1, the phase-in formula would be 70 percent applicable ASP and 30 
percent MFN Price for that quarter and remaining quarters in 
performance year 1, assuming both trigger conditions are not met in the 
ASP calendar quarters for the third and fourth quarter of performance 
year 1.
    We will apply the acceleration of the phase-in formula for each 
calendar quarter of the MFN Model where both trigger conditions are 
met. That is, for an MFN Model drug that is subject to the accelerated 
phase-in of the MFN Price, we will further accelerate the phase-in of 
the MFN Price by an additional 5 percentage points at the next 
quarterly update if the cumulative percentage increase in the 
applicable ASP or any of the monthly U.S. list prices for the NDCs 
assigned to the MFN Model drug's HCPCS code continues to be greater 
than the cumulative percentage increase in the CPI-U and MFN Price. In 
the previous example, if both of the trigger conditions were met for 
the same MFN Model drug during the applicable ASP calendar quarter for 
quarters 3 and 4 of performance year 1, the phase-in formula would be 
65 percent applicable ASP and 35 percent MFN Price for quarter 3 of 
performance year 1, and 60 percent applicable ASP and 40 percent MFN 
Price for quarter 4 of performance year 1. The accelerated phase-in of 
the MFN Price will not be reversed, but will remain in place for the 
duration of the model performance period for that drug, even if the 
manufacturer lowers its ASP and U.S. list prices after the accelerated 
phase-in is in effect.
    Further, after the full phase-in of the MFN Price is reached, if 
both of the trigger conditions are met, there will be a decrease in MFN 
Model Drug Payment Amount equal to the largest difference in the 
cumulative percentage increase in the applicable ASP or any of the 
monthly U.S. list prices for the NDCs assigned to the MFN Model drug's 
HCPCS code compared to the cumulative percentage increase in the CPI-U 
and in the MFN Price. This additional adjustment will lead to the 
affected drug's MFN Drug Payment Amount falling below the MFN Price for 
that drug. For example, for an MFN Model drug, if 100 percent of the 
MFN Price was already applied in the calculation of the MFN Model Drug 
Payment Amount for a quarter and its applicable ASP cumulatively 
increased by 14 percent, the largest cumulative percentage increase of 
any of the monthly U.S. list prices for the NDCs assigned to the HCPCS 
code was 13 percent, the CPI-U cumulatively increased by 12 percent, 
and the MFN Price cumulatively increased by 11 percent, we would reduce 
the MFN Drug Payment Amount for the quarter (in this case, previously 
established as to equal the MFN Price) by 3 percent (that is, the 
difference between 14 and 11) of the MFN Price.
    Any such additional adjustment will apply for the duration of the 
model performance period, unless a larger additional adjustment is 
triggered. As with the adjustment before the full phase-in is reached, 
we will update the calculation for the additional adjustment for each 
additional calendar quarter of the model. That is, for an MFN Model 
drug that is subject to the additional adjustment of the MFN Price, 
each calendar quarter thereafter, we will calculate the largest 
difference between the cumulative percentage increase in the applicable 
ASP or any of the monthly U.S. list prices for the NDCs assigned to the 
MFN Model drug's HCPCS code and the cumulative percentage increase in 
CPI-U and in MFN Price and increase the additional adjustment if the 
result of the updated calculation results in a larger additional 
adjustment. CMS will not reduce the

[[Page 76215]]

additional adjustment based on the results of the updated calculation. 
We believe this policy will serve as a strong incentive for 
manufacturers to avoid taking steps that could cause spillover impacts 
and will help to address commenters' concerns.
10. Limitation on MFN Drug Payment Amount To Protect Beneficiaries
    To avoid potentially increasing beneficiary cost-sharing or 
coinsurance, we are codifying in Sec.  513.210(b)(6) to compare the MFN 
Price to the applicable ASP in order to ensure that beneficiaries are 
always paying the lowest amount of coinsurance available. If the 
applicable ASP is less than the MFN Price, we will establish the MFN 
Price as equal to the applicable ASP. In addition, in Sec.  513.210(a), 
we are codifying that the allowed MFN Drug Payment Amount will not 
exceed the billed amount on the claim for the MFN Model drug. In 
addition, to maintain beneficiary protections for all claims paid under 
the OPPS, we are codifying in Sec.  513.210(d)(4) that the MFN Drug 
Payment Amount cannot exceed the non-model drug payment amount for line 
items submitted with the JG modifier (or any successor modifier used to 
identify drugs purchased under the 340B program) after removing any 
add-on amount, if applicable. We will apply this limitation to line 
items submitted with the JG modifier. We refer readers to the Calendar 
Year (CY) 2021 OPPS/ASC Notice of Proposed Rulemaking (CMS-1736-P) \68\ 
(85 FR 48880) for a discussion of CMS's proposal for CY 2021 and 
subsequent years to pay for drugs acquired under the 340B program at 
ASP minus 34.7 percent, plus an add-on of 6 percent of the product's 
ASP, for a net payment rate of ASP minus 28.7 percent based on the 
results of the Hospital Acquisition Cost Survey for 340B--Acquired 
Specified Covered Drugs. If CMS finalizes the proposed OPPS payment 
policy to pay for drugs acquired under the 340B program at ASP minus 
34.7 percent, plus an add-on of 6 percent of the product's ASP, the MFN 
Drug Payment Amount for an MFN Model drug furnished by an MFN 
participant and billed with the JG modifier will be capped at ASP minus 
34.7 percent. In such cases, the MFN participant will also receive the 
per-dose add-on payment amount described in section III.F. of this IFC.
---------------------------------------------------------------------------

    \68\ https://www.govinfo.gov/content/pkg/FR-2020-08-12/pdf/2020-17086.pdf.
---------------------------------------------------------------------------

    In the CY 2021 OPPS/ASC Notice of Proposed Rulemaking, CMS proposed 
in the alternative to continue its current policy of paying ASP minus 
22.5 percent for 340B-acquired drugs. If CMS finalizes this alternative 
proposal, the MFN Drug Payment Amount for an MFN Model drug furnished 
by an MFN participant and billed with the JG modifier will be capped at 
ASP minus 22.5 percent (85 FR 48890). In such cases, the MFN 
participant will also receive the per-dose add-on payment amount 
described in section III.F. of this IFC.
11. Method for Establishing MFN Drug Payment Amounts for Drugs Added to 
the MFN Model
    We will add annually any top 50 drugs that are not already included 
on the MFN Model Drug HCPCS Codes List, after taking the exclusions in 
Sec.  513.130(b) into account. In accordance withSec.  513.210, we will 
calculate the MFN Price that will apply to drugs that are added to the 
list of MFN Model drugs and the applicable phase-in formula for a given 
performance year and adjustments will apply. We will apply the 
applicable phase-in formula for drugs that are added to the MFN Model 
Drug HCPCS Codes List, in order to simplify and maintain consistent 
payment policies for all MFN participants and MFN Model drugs. For 
example, for a drug added as an MFN Model drug for performance year 2, 
the phase-in formula will be a blend of 50 percent of the ASP and 50 
percent of the MFN Price for the drug. Thus, Medicare Part B drugs that 
will be added to the MFN Model Drug HCPCS Codes List for performance 
year 2 and beyond will have an MFN Drug Payment Amount that will start 
more heavily based on the MFN Price than drugs that were included in 
earlier performance years. We believe this approach is appropriate 
because the MFN Model seeks to test a new payment methodology that 
takes into account the discounts that other countries enjoy and 
delaying the phase-in of the MFN Price for drugs that will be added to 
the MFN Model Drug HCPCS Codes List for performance year 2 and beyond 
will not allow CMS to fully evaluate the model payment test for such 
drugs during the model performance period.
    For drugs added to the MFN Model Drug HCPCS Codes List in a later 
performance year, this approach could result in a more significant 
change in payment for the drug upon entry to the model compared to 
drugs that are included from the beginning of the model. Although there 
is the potential for a larger change in payment for drugs that are 
added later in the model, we believe that it is necessary to maintain 
the same phase-in for all included drugs to enable us to test the full 
phase-in of the MFN Price by performance year 4. We also believe that 
MFN participants are aware of which separately payable Medicare Part B 
drugs have high annual spending and therefore will have a basis for 
assessing which drugs that are not on the MFN Model Drug HCPCS Codes 
List in performance year 1 are more likely be added to the MFN Model 
Drug HCPCS Codes List in a later performance year. For future years, we 
seek comment on whether additional information that CMS could provide 
would be helpful to MFN participants for their planning purposes, for 
example drug utilization reports developed through the model monitoring 
activities that CMS could make available on the model website.
12. Payment Exceptions for MFN Model Drugs in Short Supply
    Rather than broadly excluding drugs that are in short supply from 
the model, we will keep MFN Model drugs in the model while they are in 
short supply, but revert the MFN Drug Payment Amount to the applicable 
ASP, which could be the amount determined under section 1847A(e) of the 
Act if the conditions set forth in that provision are met, beginning 
with the first day of the next calendar quarter after the date on which 
the MFN Model drug is reported as ``Currently in Shortage'' by FDA, as 
available on these websites: https://www.accessdata.fda.gov/scripts/drugshortages/ and https://www.fda.gov/vaccines-blood-biologics/safety-availability-biologics/cber-regulated-products-current-shortages, and 
continuing for subsequent calendar quarters as warranted. Once the MFN 
Model drug is no longer reported as ``Currently in Shortage'' by FDA, 
the MFN Model payment will resume the first day of the next quarter 
after the date on which it is no longer reported in shortage. For 
example, as noted in section III.D.2. of this IFC, one of the HCPCS 
codes with high aggregate 2019 Medicare Part B total allowed charges 
(J1569, Gammagard liquid infusion) represents a drug that is currently 
on the FDA shortages list. If this HCPCS code were to be included on 
the MFN Model Drug HCPCS Codes List and remain on the FDA shortages 
list, the MFN Drug Payment Amount will be the applicable ASP until the 
first day of the next quarter of the model performance period after it 
is no longer reported as ``Currently in Shortage'' by FDA. However, we 
note that we are excluding HCPCS codes that describe intravenous immune 
globulin from the MFN Model Drug HCPCS Codes List as discussed in 
section III.D.2 of this IFC.

[[Page 76216]]

13. Payment of Blood Clotting Factor Furnishing Fee Under the MFN Model
    Currently, payment for the blood clotting factor furnishing fee 
under 42 CFR 410.63(c) is made along with payment for the blood 
clotting factor. Under the MFN Model, a HCPCS code that is used to bill 
for a blood clotting factor may be an MFN Model drug if such HCPCS code 
is included on the MFN Model Drug HCPCS Codes List. To maintain the 
current payment approach for the blood clotting factor furnishing fee 
during the MFN Model, we are codifying in Sec.  513.210(e), that when 
applicable, the blood clotting furnishing fee under Sec.  410.63(c) 
will be payable along with the MFN Drug Payment Amount. We believe this 
approach will eliminate the need to establish different billing 
instructions for MFN Model drugs that are blood clotting factors.

F. MFN Model Alternative Add-On Payment

1. Overview of the Alternative Add-On Payment
    In the October 2018 ANPRM, we sought public comment on testing an 
alternative add-on payment to the current system, required by section 
1847A of the Act, under which Medicare Part B pays a fee based on 6 
percent of the ASP of the drug so that the dollar amount of the add-on 
increases with the price of the drug rather than reflecting the service 
being performed. In general, the amount of add-on realized by providers 
and suppliers has been described by commenters as 4.3 percent as a 
result of sequestration.\69\ In the October 2018 ANPRM, we described 
our belief regarding how a potential model could pay a drug add-on 
amount that would be different from the current drug add-on amount. We 
sought public comment on potential ways to structure the alternative 
add-on, including but not limited to: An amount based on drug class, 
the physician's specialty, or the practice's historical billing 
patterns, with a possible bonus pool tied to clinically appropriate 
utilization. We requested feedback on several design topics, such as 
how we could best define and determine the alternative add-on payment 
amount, whether CMS should develop an encounter-based or monthly add-on 
payment approach, and potential inclusion of a quality bonus pool to 
incentivize evidence-based care. We stated that our goal was to 
maintain relative stability in provider and supplier revenue through an 
alternative drug add-on payment for furnishing drugs that removes the 
current percentage-based drug add-on payments.
---------------------------------------------------------------------------

    \69\ Stakeholders have reported that the add-on percentage is 
slightly further reduced when the OPPS beneficiary cost-sharing 
limitation applies. Further, we note that current payments under the 
OPPS for certain drugs when the drug is acquired under the 340B 
program are made based on ASP-22.5 percent and are not considered to 
include a drug add-on payment amount. We refer readers to to the 
Calendar Year (CY) 2021 OPPS/ASC Notice of Proposed Rulemaking (CMS-
1736-P) (85 FR 48880) for a discussion of CMS's proposal for CY 2021 
and subsequent years to pay for drugs acquired under the 340B 
program at ASP minus 34.7 percent, plus an add-on of 6 percent of 
the product's ASP, for a net payment rate of ASP minus 28.7 percent 
based on the results of the Hospital Acquisition Cost Survey for 
340B-Acquired Specified Covered Drugs. We also refer readers to the 
alternative proposal in the CY 2021 OPPS/ASC Notice of Proposed 
Rulemaking (85 FR 48890) to continue the current policy of paying 
ASP-22.5 percent for 340B drugs and biologicals under the OPPS.
---------------------------------------------------------------------------

    In response to the October 2018 ANPRM, we received feedback from a 
number of stakeholder groups on the structuring of an alternative add-
on payment. Overall, there was no consensus on the best approach to 
designing an alternative add-on payment, though several commenters 
supported calculating the alternative add-on payment in such a way that 
model participants would be held harmless. Some commenters supported 
the idea of testing an alternative add-on payment that is not tied to 
increases in drug prices over time, with one commenter noting that this 
could promote revenue stability. One commenter noted an approach that 
varies the alternative add-on payment between different drugs would 
risk creating perverse incentives in prescribing decisions between 
alternative treatment options. Several commenters supported a flat fee 
with more than one tier. Several commenters expressed concern about 
linking a bonus pool to prescribing lower cost drugs. One commenter 
opposed reducing the add-on amount to allow for a bonus pool.
    After considering the comments we received, we were persuaded that 
potential model requirements to qualify for a modest quality bonus 
would be challenging and may be burdensome for MFN participants to 
implement and adhere to consistently for all MFN beneficiaries, and 
would add potential financial risk for MFN participants, which is not 
necessary for purposes of testing an alternative add-on payment 
approach under the MFN Model. Thus, we are not including a quality 
bonus in the MFN Model. We were also persuaded that the alternative 
add-on should be designed in as straightforward a manner as possible to 
minimize administrative burden for MFN participants and potential 
confusion for beneficiaries.
    We will pay MFN participants a single add-on payment amount per 
dose of an MFN Model drug; this payment will not vary based on the 
amount of drug furnished in a dose, billing units billed on the claim 
line, or by MFN participant or specialty. The goals for the model's 
approach to the alternative add-on payment are to test an innovative 
way to pay the add-on portion of the drug payment, boost add-on revenue 
for MFN participants on average based on historical overall add-on 
revenue, create an incentive to encourage appropriate drug utilization 
by breaking the link between the manufacturer's drug price and the 
calculation of the Medicare Part B payment for the add-on amount, and 
remove or reduce the incentive to furnish higher-cost drugs inherent in 
the current methodology.
    With the MFN alternative add-on payment, we will test a single add-
on payment amount that will paid per dose, where ``dose'' for the 
purposes of the MFN alternative add-on payment is defined as the number 
of HCPCS billing units reported on a claim line \70\ (also called 
service line or line item). We are codifying this alternative add-on 
payment at Sec.  513.220. We will waive beneficiary cost-sharing for 
the add-on payment. As such, the add-on approach will test a separate 
standardized add-on payment amount per dose that is not tied to the 
Medicare Part B payment amount for a drug. We will start with an amount 
that is calculated based on 6.1224 percent of historical applicable 
ASPs for 2019 final action claim lines for the selected MFN Model drugs 
for the beginning of performance year 1 as further described in Sec.  
513.220, trended forward using an inflationary adjustment for the start 
of performance year 1. With this approach, the per-dose add-on payment 
amount will be calculated once at the beginning of the model and will 
not be recalculated as the MFN Model Drug HCPCS Codes List changes. For 
each calendar quarter thereafter, beginning with performance year 1, 
quarter 2, we will update the per-dose add-on payment amount using an 
inflation factor.
---------------------------------------------------------------------------

    \70\ An exception is when a claim line is billed with the 
modifier JW, indicating discarded drug.
---------------------------------------------------------------------------

    For the MFN Model drugs for the beginning of performance year 1 
that are biosimilar biological products, we will use 6.1224 percent of 
the historical applicable ASPs for the reference biological product in 
the calculation of the per-dose add-on amount rather than 6.1224 
percent of the historical

[[Page 76217]]

applicable ASPs for the biosimilar biological product to align with the 
determination of the add-on amount to such products under section 
1847A. Based on the performance year 1 MFN Model Drug HCPCS Codes List 
in Table 2, this applies to Q5111 (Injection, udenyca 0.5 mg).
    We selected 6.1224 percent because that amount results in an add-on 
pool that will allow MFN participants to realize, on average, a 6 
percent add-on per dose after sequestration, which generally 
applies.\71\ In the absence of actual drug acquisition costs for 
eligible providers and suppliers, we believe it is appropriate to use 
an amount for the add-on pool that represents, on average, a 40 percent 
increase compared to 4.3 percent of ASP in use in the baseline period 
to achieve a goal of the model to provide increased add-on revenue for 
MFN participants on average.
---------------------------------------------------------------------------

    \71\ Note that Section 3709 of the Coronavirus Aid, Relief, and 
Economic Security (CARES) Act temporarily suspends Medicare 
sequestration from May 1, 2020 to December 31, 2020. Available at: 
https://www.govinfo.gov/content/pkg/BILLS-116hr748enr/pdf/BILLS-116hr748enr.pdf.
---------------------------------------------------------------------------

2. Per-Dose Add-On Payment Amount Methodology
a. Calculation of the Single Per-Dose Add-On Payment Amount
    In Sec.  513.220(b), we specify how we calculated a single per-dose 
add-on payment amount for the start of the MFN Model. Using 2019 
historical claims data, we calculated a per-dose add-on payment amount 
by applying the applicable ASP (that is, the payment amount determined 
in accordance with section 1847A of the Act for a quarter minus the 
applicable add-on percentage) to the identified 2019 claims lines, 
based on the calendar quarter in which the claim's date of service 
falls, which corresponds to the manufacturer-reported ASPs from two 
calendar quarters prior, with an exception for biosimilar biological 
products as described previously. We used all 2019 Medicare Part B FFS 
claims lines for separately paid drugs (by HCPCS code) included on the 
MFN Model HCPCS Codes List for the beginning of performance year 1 that 
were furnished by eligible Medicare-participating providers and 
suppliers (that is, entities that are eligible to be an MFN 
participant). We excluded claims submitted by excluded providers and 
suppliers described in Sec.  513.100(c) (such as CAHs, and cancer 
hospitals) as well as certain claims described in Sec.  513.100(d) 
(such as claims processed by the DME MAC), as applicable in 2019, as 
well as claims where Medicare was not the primary payer. We included 
all relevant claim lines for an MFN Model drug with an allowed charge 
greater than zero dollars in the calculation. As we used nearly all 
2019 claims for drugs included on the MFN Model HCPCS Codes List for 
the beginning of performance year 1 furnished from any eligible 
Medicare-participating provider or supplier, we believe that one 
calendar year provided sufficient data for purposes of calculating a 
single per-dose add-on payment amount. Calendar year 2019 represents 
the same baseline year that we used to select the MFN Model drugs for 
the beginning of performance year 1, as identified in Table 2.
    Once all relevant 2019 claim lines were identified for each drug 
(by HCPCS code) on the MFN Model HCPCS Codes List for the beginning of 
performance year 1, we multiplied the number of HCPCS units billed on 
each claim line by 6.1224 percent of the 2019 applicable ASP (which we 
define as the payment amount determined in accordance with 1847A of the 
Act less the applicable add-on percentage for the MFN Model drug's 
HCPCS code) for the calendar quarter that matches the claim line's date 
of service and then summed across all claim lines for that drug to 
yield a total add-on spending amount for that drug. For biosimilar 
biological products, we used the applicable ASP for the reference 
biological product.
    Then we pooled together the total add-on spending amounts for all 
drugs on the MFN Model HCPCS Codes List for performance year 1 and the 
total number of claim lines for those drugs (excluding claim lines 
billed with the JW modifier). Lastly, we calculated the per-dose add-on 
payment amount as the total pooled add-on spending amount divided by 
the total pooled number of claim lines.
    Using the drugs (by HCPCS code) included on the performance year 1 
MFN Model Drug HCPCS Codes List in Table 2, available 2019 claims data 
subject to the exclusions and exception previously noted, and 
applicable ASPs from 2019, we calculated a single per-dose add-on 
payment amount in the amount of $146.55. This amount represents the 
single per-dose add-on payment amount for a dose of any MFN Model drug 
prior to application of the inflationary factor as described in section 
III.F.2.b. of this IFC.
b. Trending the Single Per-Dose Add-On Payment Amount Forward Each 
Calendar Quarter During the MFN Model
    We will trend forward the single per-dose add-on payment amount 
each calendar quarter during the MFN Model to account for inflation 
over time by using a cumulative inflationary factor as described in 
this section of this IFC. We will not use changes in ASP or MFN Drug 
Payment Amount to trend forward the single per-dose add-on payment 
amount to align with our intention to test the removal of the link 
between a drug's add-on payment and its price.
    As specified in Sec.  513.220(b)(7), after calculating the single 
per-dose add-on payment amount, we multiplied the single per-dose add-
on payment amount ($146.55) by an inflationary factor, which equals the 
percentage increase in the CPI-U from the midpoint of the baseline year 
(2019) through the first month of the calendar quarter prior to the 
start of the model (that is, the percentage increase in CPI-U from July 
2019 through October 2020). The resulting per-dose alternative add-on 
payment amount for the first calendar quarter of performance year 1 
(January 1, 2021 through March 31, 2021) is $148.73.
    To calculate the per-dose alternative add-on payment amount for 
each subsequent calendar quarter during the model performance period, 
as specified in Sec.  513.220(c), we will multiply the performance year 
1, quarter 1 alternative add-on payment amount by a cumulative 
inflation factor that will ensure the amount will remain equal to or 
greater than the alternative add-on payment amount calculated for 
performance year 1, quarter 1. We will calculate a cumulative inflation 
factor as equal to the percentage increase in the CPI-U from October 
2020 through the first month after the end of the applicable ASP 
calendar quarter. If the cumulative percentage change in the CPI-U is 
negative, we will use an inflation factor of 1. For example, the 
cumulative inflation factor for performance year 1, quarter 2 (that is, 
April 1, 2021 through June 30, 2021) will be the percentage increase in 
the CPI-U from October 2020 through January 2021. Similarly, the 
cumulative inflation factor for performance year 1, quarter 3 will be 
the percentage increase in the CPI-U from October 2020 through April 
2021.
    As discussed in section III.G. of this IFC, MFN participants will 
use a new HCPCS code (M1145, MFN drug add-on, per dose) to bill for and 
receive the alternative add-on payment amount for each dose of an MFN 
Model drug that is billed on the claim.
3. Discussion of the Per-Dose Add-On Payment Approach
    The per-dose add-on payment amount approach will test an 
alternative way to

[[Page 76218]]

calculate the add-on payment that is not tied to the sales price of the 
drug that is furnished. This approach also aims to boost add-on 
revenue, on average, for MFN participants by setting the per-dose add-
on payment amount based on 6.1224 percent of historical ASP payment 
allowances trended forward for inflation. However, the impact on MFN 
participants will vary based on the MFN participant's prescribing 
patterns, including the amount and types of MFN Model drugs they 
furnish to Medicare FFS beneficiaries.
    Compared with the current add-on payment policy, on an average per 
dose basis based on 2019 historical claims, the single per-dose add-on 
approach will initially decrease add-on payments for MFN Model drugs 
with relatively higher historical applicable ASP-based payment amounts 
per dose and increase add-on payments for MFN Model drugs with 
relatively lower historical applicable ASP-based payment amounts per 
dose. Average 2019 historical add-on payment amounts per dose for the 
MFN Model drugs for performance year 1 ranged from $10.44 to $2,575.47 
per average dose for a drug. Based on 2019 claims, on average, a single 
per-dose add-on payment amount, calculated as described in this IFC and 
after sequestration is applied, will represent an increase in the add-
on payment amount for 70 percent of doses on average compared to the 
effective historical add-on amount of 4.3 percent of the applicable ASP 
after sequestration.
    To examine the potential impact of the single per-dose add-on 
approach on MFN participants using 2019 claims data, we considered the 
overall potential change in the add-on payment amount at the eligible 
entity level, specialty level, and type of provider and supplier. That 
is, for this entity level analysis, we grouped 2019 claim lines for the 
drugs (by HCPCS code) identified in Table 2 based on the provider's or 
supplier's CMS Certification Number (``CCN'') or Taxpayer 
Identification Number (``TIN''). To examine the potential impact of the 
single per-dose add-on payment amount at the specialty level, we 
assigned claims to a specialty category based on the primary specialty 
of the National Provider Identifier (NPI) associated with the 
furnishing of the drug as listed in the Medicare Provider Enrollment, 
Chain, and Ownership System (PECOS). Eligible providers were assigned 
to the specialty that was most frequently associated with their 2019 
claims for the drugs (by HCPCS code) identified in Table 2. We also 
used the type of bill to examine the potential impacts on various types 
of providers and suppliers.
    These analyses highlight that different subsets of providers and 
suppliers will potentially gain (or lose) under the single per-dose 
add-on approach. For 340B covered entities that were paid under the 
OPPS during calendar year 2019, the entirety of the alternative add-on 
payment amount represent an increase in payment when drugs are acquired 
under the 340B program. Thus, we removed these entities from the 
following analyses.
    To explore the potential entity level change in the add-on amount 
for the single per-dose add-on payment approach, we assigned each CCN 
or TIN to only one specialty based on the specialty code with the 
highest total allowed spending for the entity's claim lines, regardless 
of setting (for example, hospitals, ASCs, and physician office). We 
also assigned each specialty a value of ``low,'' ``medium,'' or 
``high,'' based on the percentage of its Medicare revenue that is 
related to Part B drugs, such that ``high'' means the specialty's drug 
revenue is more than 50 percent of its total Medicare revenue, 
``medium'' means the specialty's drug revenue is 25 to 50 percent of 
its total Medicare revenue, and ``low'' means the specialty's drug 
revenue is less than 25 percent of its total Medicare revenue.
    Based on the single per-dose add-on payment amount of $146.55 
(prior to the application of the inflationary factor that applies 
during the model) and using 2019 drug utilization, MFN participants 
will fare, on average, 40 percent better overall across all specialties 
with the per-dose add-on payment amount than they did historically 
based on 4.3 percent of ASP after sequestration. Some MFN participants 
will see more than a 40 percent increase in revenue related to the MFN 
add-on payment amount compared to their 2019 historical Part B drug 
claims, and others will see less than a 40 percent increase, including 
some who will see a reduction in add-on revenue. Based on our analysis, 
in general, physician practices will be better off under the per-dose 
add-on payment approach than hospital outpatient departments, and 
single specialty practices will be better off than multi-specialty 
practices. Table 8 shows the estimated variation in impacts for the top 
specialties by comparing 2019 baseline add-on payments based on 4.3 
percent of the applicable ASP with a post-sequestration single per-dose 
add-on payment amount (that is, for this comparison, we used the per-
dose add-on payment amount prior to the application of the inflationary 
factor ($146.55) and applied the effects of sequestration for this 
comparison). The Entity-Level Percentage Change By Percentile portion 
of Table 8 shows the distribution of entities based on size of the 
difference between their 2019 baseline add-on payments (based on 4.3 
percent of the applicable ASP) and the single per-dose add-on amount 
(post-sequestration). Each row shows the size of the impact for the 
given specialty. The 5th percentile will experience the largest 
negative impact whereas the 95th percentile will experience the largest 
positive impact.
BILLING CODE 4120-01-P

[[Page 76219]]

[GRAPHIC] [TIFF OMITTED] TR27NO20.009

BILLING CODE 4120-01-C
    Based on these data, as shown in Table 8, all but 9 of the top 35 
specialties (in terms of overall 2019 allowed dollars) impacted by the 
MFN

[[Page 76220]]

Model will on average see increases in add-on revenue compared to 4.3 
percent of the applicable ASP with a single payment amount (the 
exceptions are hematology/oncology, medical oncology, neurology, 
hematology, gastroenterology, gynecological/oncology, infectious 
disease, hematopoietic cell transplantation & cellular therapy, and 
dermatology). At the 25th percentile, 57 percent of the entities will 
see increased add-on revenue for the top 35 specialties with the single 
per-dose add-on payment amount; whereas at the 50th percentile, 83 
percent of the entities will see increased add-on revenue for the top 
35 specialties with the single per-dose add-on payment amount. Please 
note that some of the large percentage increases seen shown in the 95th 
percentile column are likely driven by the small volume of drugs 
furnished by entities in this percentile.
    We observed that volume is not consistently associated with whether 
an entity will be better or worse off under the per-dose add-on payment 
approach when we look at the single per-dose add-on amount approach for 
the top five specialties in terms of total aggregate Medicare spending 
on MFN Model drugs in 2019: internal medicine, hematology/oncology, 
ophthalmology, rheumatology, and medical oncology. When we specifically 
looked at the top, middle, and bottom of a distribution of all entities 
based on how much better or worse off each entity will be under the 
per-dose add-on payment amount compared to their add-on revenue (based 
on their 2019 claims), we found that entities in the top 5 percent 
(that is, those that will do the best) had very low volume (that is, 
few claims for these drugs in 2019 claims). Entities in the bottom 5 
percent (that is, those that will do the worst) tended to have lower 
volume than the middle 10 percent, though volume was highest in the 
bottom 5 percent of entities in the internal medicine and ophthalmology 
specialties. Overall, entities that will be worse off compared to their 
add-on revenue (based on their 2019 claims) under the per-dose add-on 
payment approach tended to furnish more drugs with higher drug add-on 
payment amounts per dose more frequently than the entities that will be 
better off. We estimate that similar impacts will be experienced across 
the performance years unless ASPs for MFN Model drugs rise faster than 
inflation, in which case the overall increase in add-on revenue 
compared to non-model add-on revenue will diminish over time.
4. Beneficiary Cost-Sharing Responsibilities
    In response to the October 2018 ANPRM, which suggested continuing 
beneficiary cost-sharing for the alternative add-on payment, some 
commenters suggested that CMS should ensure any alternative add-on 
payment does not increase out-of-pocket costs for beneficiaries. Other 
commenters noted that an alternative add-on payment could be confusing 
to beneficiaries since currently they pay cost-sharing based on a 
single amount, versus separate amounts, such as the MFN Model Drug 
Payment Amount and alternative add-on that we are including in the MFN 
Model. We appreciate these commenters' feedback.
    To support reducing out-of-pocket drug costs and minimizing 
potential confusion for MFN beneficiaries related to the alternative 
add-on payment amount, and decreasing administrative burden for MFN 
participants, we will waive beneficiary cost-sharing (coinsurance and 
deductible amounts) on the portion of the allowed MFN Model Payment 
amount that is based on the alternative add-on payment. Under the MFN 
Model, the MFN Drug Payment Amount will be subject to beneficiary 
coinsurance and the annual deductible amount. MFN participants will 
continue to collect beneficiary cost-sharing applicable to the portion 
of the allowed payment amount that is based on the MFN Drug Payment 
Amount. For the alternative add-on, Medicare will pay the entire 
allowed payment amount that is based on the alternative add-on payment 
to ensure that beneficiaries do not experience an increase in cost-
sharing under the MFN Model as a result of testing an alternative add-
on amount. That is, beneficiaries will not owe any coinsurance or 
amount for the annual deductible for the per-dose add-on payment 
amount.

G. Billing and Claims Processing Approach

    We intend to issue model-specific claims submission instructions 
that MFN participants will be required to follow. Currently, for 
separately payable Part B drugs, providers and suppliers submit 
separate claim lines for each drug. Among the information included in 
each claim line for the applicable bill type, providers and suppliers 
specify the appropriate HCPCS code to indicate the drug that was 
furnished, the number of billing units to indicate the total amount of 
the drug that was furnished, billing code modifiers as necessary, and a 
billing amount (or charge). In general, providers and suppliers 
routinely use one claim line to bill for a furnished drug dose, and 
using billing modifiers when doing so may be necessary to comply with 
billing instructions. In certain situations, a second claim line may be 
necessary to report the amount of drug that was furnished, for example, 
when the number of billing units necessary to indicate the dosage given 
exceeds the character size of the units field or when appropriately 
discarded drug is billed. When applicable, a separate line item is 
billed with the modifier JW to identify the amount of unused drugs (or 
biologicals) from single use vials or single use packages that was 
appropriately discarded. The Medicare claims processing system 
calculates payment for the amount of discarded drug when the modifier 
JW is present. MFN participants will be required to submit a separate 
claim line using a new model-specific HCPCS code (M1145, MFN drug add-
on, per dose) to bill for and receive the alternative add-on payment 
amount for each dose of an MFN Model drug that is billed on the claim. 
The MFN participant will indicate in the units field of the claim line 
with HCPCS code M1145 the number of doses of a separately payable MFN 
Model drug that are billed on the claim. To do so, the MFN participant 
will count the number of claim lines with a HCPCS code that is included 
on the applicable MFN Model Drug HCPCS Codes List (based on the date of 
service), including all claim lines when the number of billing units 
necessary to indicate the dosage given exceeds the character size of 
the units field and the claim has more than one claim line for such MFN 
Model drug (we note that this is expected to be a rare situation), and 
excluding the number of claim lines billed with the JW modifier. This 
approach will allow the Medicare claims processing system to apply the 
alternative add-on payment amount for each dose, and not apply 
beneficiary cost-sharing to the alternative add-on payment amount. MFN 
participants will still bill for wastage as they otherwise would, using 
a separate claim line and the JW modifier, and the payment for such 
claim lines will be based on the MFN Drug Payment Amount (the 
alternative add-on payment amount is not applicable to such claim 
lines).
    This billing and claims processing approach will initiate from the 
MFN participant's billing system and will establish a clear mechanism 
for MFN participants to track when the alternative add-on amount was 
billed and paid. This approach will simplify Medicare claims processing 
changes for the MFN Model. However, this

[[Page 76221]]

approach may increase administrative burden for MFN participants and 
requires MFN participants to count the number of claim lines for MFN 
Model drugs included on a claim, indicate this number in the units 
field of the claim line for the alternative add-on (using HCPCS code 
M1145), and submit a billing amount (or charge) on the claim line for 
the alternative add-on. In addition, the alternative add-on payment 
amount will be updated quarterly. Because Medicare allows the lesser of 
the applicable payment amount or the billed amount, MFN participants 
will have to ensure that they submit an appropriate billing amount (or 
charge) for the alternative add-on for the applicable quarter. Because 
the same HCPCS code will be used to bill for the alternative add-on for 
all MFN Model drugs, we believe this approach minimizes, but does not 
eliminate, the additional administrative burden for MFN participants.
    We are waiving program requirements in section 1833(a)(1)(S), 
section 1833(a)(1)(G) and section 1833(t) of the Act, respectively to 
allow flexibility in the way in which claims subject to the MFN Model 
payment will be processed. Section 1833(a)(1)(S) of the Act specifies 
that the Medicare payment for drugs and biologicals not paid on a cost 
or prospective payment basis is 80 percent of the lesser of actual 
charge or the amount established in section 1842(o) of the Act. 
Similarly, section 1833(a)(1)(G) of the Act specifies that the amounts 
paid with respect to facility services furnished in connection with 
certain surgical procedures and with respect to services furnished to 
an individual in an ASC shall be 80 percent of the lesser of the actual 
charge for the services or the amount determined by the Secretary under 
such revised payment system. Section 1833(t) of the Act specifies how 
payment under the OPPS is calculated including beneficiary copayment. 
Specifically, we are waiving these program requirements to the extent 
necessary to allow the total allowable model payment for the service as 
specified in Sec.  513.210 and Sec.  513.220 (that is, the sum of the 
allowed MFN Drug Payment Amount and the allowed alternative add-on 
payment amount) and to not apply beneficiary cost-sharing to the 
alternative add-on payment amount.

H. Quality Measures

    The October 2018 ANPRM stated our intention to include quality 
measures as part of the potential IPI Model, and our interest in 
several categories of potential measures, specifically: patient 
experience measures, medication management measures, medication 
adherence measures, and measures related to patient access and 
utilization. We sought public input on ways to assess quality of care 
for purposes of real-time monitoring of utilization, hospitalization, 
mortality, shifts in site-of-service and other important indicators of 
patient access and outcomes, without requiring providers or suppliers 
to report additional data. We received numerous comments in response to 
the October 2018 ANPRM on this topic. Several commenters expressed 
concern that testing alternative payments for Part B drugs in general 
may impact beneficiaries' access to care and may impact the overall 
patient experience of care. Some commenters requested that any quality 
measurement not add burden to model participants. Some commenters also 
discussed the importance of adherence to nationally recognized clinical 
guidelines in treatment decisions, stating that adherence to nationally 
recognized clinical guidelines would reduce drug spending while also 
maintaining and possibly increasing quality of care.
    We appreciate the public feedback on ways we could structure a 
model to enhance and monitor quality of care. In the MFN Model, we will 
implement robust monitoring activities, such as analyzing claims data, 
using patient survey data, and site visits, to identify any unintended 
consequences and ensure that MFN beneficiaries' access to medications 
is not impeded and that quality of care is preserved or enhanced. 
Further, we believe the following principles are appropriate for a 
quality measurement approach for the MFN Model: (1) Use quality 
measures for the purpose of monitoring quality of care and beneficiary 
access to treatment and experience with care; (2) avoid unnecessary 
participant reporting burden as many providers and suppliers are 
currently reporting quality measures to other programs and payers, for 
example, the MFN Model should use claims-based measures where 
appropriate; and (3) establish standards for adding quality measures, 
if necessary, during the model. We believe that this approach will 
allow CMS to test the MFN Model's alternative drug payment methodology, 
while creating a safeguard for beneficiary access and quality of care, 
as well as a means to monitor patient access and quality of care. We 
are also sensitive to concerns regarding adding administrative burden 
to MFN participants and beneficiaries and, thus, seek to minimize 
burden on them. As such, in Sec.  513.400(b)(1) we will collect only 
one quality measure, focused on patient experience, to help better 
understand the impact of the MFN Model on beneficiary access and 
quality of care. This survey will be fielded by CMS to avoid any 
quality measure reporting burden for MFN participants, although there 
will be reporting burden on beneficiaries. CMS will also monitor for 
quality as outlined in section III.I.4. of this IFC, including 
monitoring access to medications through rapid analysis of claims data, 
using monthly claims extracts that will provide frequent assessments of 
beneficiary access to MFN Model drugs and that complement existing 
methods to receive, assess, and respond to beneficiary and health care 
provider feedback on the MFN Model.
    For the patient experience focused quality measure, we will use a 
patient experience survey, which we will field periodically to a sample 
of Medicare beneficiaries, beginning in performance year 1. The patient 
experience survey will be administered to these beneficiaries by a 
third party contractor throughout the model performance period. A 
sample of beneficiaries will be surveyed regarding their experience of 
care, access, or other issues they experienced under the MFN Model, and 
we may also sample beneficiaries who are not in the MFN Model. 
Beneficiaries will not be required to complete the survey.
    Survey results will be used to monitor the impact of the MFN Model 
on MFN beneficiaries' care experience and potentially to inform 
educational materials for MFN participants. As is outlined in section 
III.I.4. of this IFC, claims data will also be monitored to assess 
patient access and outcomes.
    If during the model the patient experience of care quality measure 
and claims-based monitoring strategies are found to be insufficient to 
adequately measure the quality of care that MFN beneficiaries are 
receiving or MFN participants are providing, CMS may specify additional 
measures to monitor quality. If additional quality measures are added, 
they will meet the following criteria: (1) Additional measures would be 
among one or more of the following categories: Patient experience of 
care, patient activation, shared decision making, adherence, 
utilization, and process measures; (2) Additional measures would not 
add significant burden to MFN participants or beneficiaries; and (3) 
Additional measures would utilize an instrument that CMS has used 
previously in a model to adjust payment or for monitoring or 
evaluation. We are codifying the inclusion of the patient

[[Page 76222]]

experience quality measure and its use as well as the criteria for 
adding measures during the MFN Model in Sec.  513.400.

I. Beneficiary Protections and Monitoring Actions

    We are interested in enhancing protections for beneficiaries 
included in the MFN Model. In addition to existing beneficiary 
protections, we will actively monitor the MFN Model to ensure it is 
operating effectively and meeting the needs of beneficiaries, providers 
and suppliers, and the Medicare program. We will coordinate with the 
Medicare Beneficiary Ombudsman and other customer facing components to 
ensure that any MFN Model-related beneficiary complaints, grievances, 
or requests for information submitted are responded to in an 
appropriate and timely manner, per CMS protocol.
    We believe it will also be necessary to have additional protections 
in place in the MFN Model to ensure that beneficiaries retain their 
existing rights and are not harmed by the model test. Further, we 
believe it is important for beneficiaries to know and understand their 
rights as beneficiaries who are receiving care from MFN participants. 
We therefore believe it is necessary to include certain policies 
regarding beneficiary choice, appeals, and the availability of 
services.
1. Beneficiary Freedom of Choice
    A beneficiary's ability to choose his or her provider or supplier 
is an important principle of Medicare fee-for-service and is reflected 
in section 1802 of the Act. We are codifying in Sec.  513.410(a) that 
any MFN participant must not commit any act or omission, nor adopt any 
policy that inhibits a beneficiary from exercising his or her freedom 
to choose to receive care from any Medicare participating provider or 
supplier or any provider or supplier who has opted out of Medicare. We 
believe these provisions are necessary to ensure the MFN Model does not 
prevent beneficiaries from the general rights and guarantees provided 
under Medicare.
2. Appeals Processes and Financial Hardship Exemption
a. Appeals Processes
    In Sec.  513.410(b), we are codifying that MFN beneficiaries and 
their assignees will have access to the existing formal claims appeals 
process under 42 CFR part 405, subpart I. In other words, once an MFN 
Model drug is furnished by an MFN participant to a beneficiary and a 
claim is submitted and processed for payment, that claim will be 
eligible for the current Medicare claims appeals processes. If a 
beneficiary receives an MFN Model drug from an MFN participant it does 
not mean that he or she should lose this right, but instead this right 
should necessarily be applicable to included beneficiaries as it would 
be if they were not a part of the MFN Model.
b. Financial Hardship Exemption
    To include financial protection for physicians and other MFN 
participants, specifically those who furnish substantial amounts of MFN 
Model drugs as part of the services they furnish to Medicare FFS 
beneficiaries, especially MFN Model drugs with the greatest difference 
between the MFN Price and the applicable ASP, we are including a 
financial hardship exemption codified in Sec.  513.230. The financial 
hardship exemption process for MFN participants will be available in 
the event unintended consequences arise to ensure access to MFN Model 
drugs for MFN beneficiaries and financial protections for MFN 
participants who are unable to obtain MFN Model drugs at or below the 
MFN Model Payment for such drugs and are significantly affected by 
their participation in the MFN Model.
    The financial hardship exemption process will occur independently 
of existing Medicare claims processing and appeals processes. In Sec.  
513.230(a), we codify that a financial hardship exemption for a 
performance year may be granted to an MFN participant by CMS, in its 
sole discretion and will not be subject to appeal, when the provisions 
in Sec.  513.230 are met. This means that a financial hardship 
exemption, if granted, will be applied at the MFN participant level (as 
defined in Sec.  513.2). As further described in this section of this 
IFC, a financial hardship exemption will be limited to cases where the 
MFN participant experienced a financial loss.
    Specifically, to be eligible for a financial hardship exemption, 
the MFN participant must submit its request for a financial hardship 
exemption to CMS in accordance with the submission process that CMS 
will post on the MFN Model website prior to October 1, 2021, and in the 
form and manner and with the content that will be specified by CMS, 
including without limitation the requirements specified in Sec.  
513.230(b). Such requests must be submitted to CMS within 60 calendar 
days following the end of the performance year for which the MFN 
participant seeks a financial hardship exemption. The MFN participant 
must include the following in its request for a financial hardship 
exemption:
     Evidence of methods used to obtain each MFN Model drug 
that was furnished by the MFN participant during the performance year 
to any patient;
     Average net acquisition cost for each MFN Model drug 
(inclusive of all on-invoice prices and price reductions, off-invoice 
discounts, any adjustments thereto, and any other price concessions 
related to the purchase of the MFN Model drug) that was furnished by 
the MFN participant during the performance year to MFN beneficiaries;
     Average net acquisition cost for each MFN Model drug 
(inclusive of all on-invoice prices and price reductions, off-invoice 
discounts, any adjustments thereto, and any other price concessions 
related to the purchase of the MFN Model drug) that was furnished by 
the MFN participant during the performance year to patients who were 
not MFN beneficiaries;
     Statement of any remuneration received by the MFN 
participant from manufacturers of MFN Model drugs, wholesalers, and 
distributors that is not reflected in the MFN participant's average net 
acquisition costs with a justification of why such remuneration should 
not be treated as a price concession related to the purchase of an MFN 
Model drug;
     Administrative information, including: MFN participant's 
name, TIN or CCN (as applicable), contact name, phone number, and email 
address; and
     The MFN participant's attestation that--
    ++ It experienced a reduction in Medicare Part B FFS payments for 
separately payable drugs on a per beneficiary basis during the 
performance year as compared to the prior year (that is, the four 
calendar quarters immediately preceding the performance year) due to 
its inability to obtain one or more of the MFN Model drugs at or below 
the MFN Model Payments for such drugs during the performance year;
    ++ It has not received and will not receive any remuneration from 
manufacturers of MFN Model drugs, wholesalers, and distributors related 
to the purchase of an MFN Model drug that was furnished by the MFN 
participant during the performance year that is not reflected in the 
MFN participant's submission; and
    ++ Its submission is true, accurate, and complete.
    In addition, MFN participants must use a template that CMS will 
post on the MFN Model website for submission of their net acquisition 
costs for MFN Model drugs and administrative information. This template 
will be

[[Page 76223]]

similar to the template CMS provided for the 2020 Hospital Survey for 
Specified Covered Outpatient Drugs (SCODs) Average Acquisition 
Cost.\72\ The MFN participant will submit the other required materials 
to CMS along with the template.
---------------------------------------------------------------------------

    \72\ The template for the 2020 Hospital Survey for Specified 
Covered Outpatient Drugs (SCODs) (CMS-10709; OMB 0938-1374) 
available at: https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/HospitalOutpatientPPS/index.
---------------------------------------------------------------------------

    In Sec.  513.230(c), we codify the standards that CMS will use to 
determine if an MFN participant is granted a financial hardship 
exemption. Specifically, to be eligible for the financial hardship 
exemption, we codify in Sec.  513.230(c)(2)(i) that the MFN participant 
must submit a timely, complete request for a financial hardship 
exemption in accordance with the requirements specified in Sec.  
513.230(b) that in the sole discretion of CMS demonstrates all of the 
following:
     The MFN Participant exhausted all reasonable methods to 
obtain the MFN Model drugs at or below the MFN Model Payments for such 
drugs during the performance year.
     The MFN participant's average net acquisition cost for 
each MFN Model drug (including on- and off-invoice discounts or 
adjustments) that was furnished by the MFN participant during the 
performance year to patients who were not MFN beneficiaries was not 
less than the MFN participant's average net acquisition costs for such 
MFN Model drug (including on- and off-invoice discounts or adjustments) 
that was furnished by the MFN participant during the performance year 
to MFN beneficiaries.
     Any remuneration the MFN participant received from 
manufacturers of MFN Model drugs, wholesalers, and distributors that 
was not reflected in the MFN participant's average net acquisition 
costs was not a price concession related to the purchase of an MFN 
Model drug.
    In addition, in Sec.  513.230(c)(2)(ii), we are codifying that the 
agency in its sole discretion must also determine that the MFN 
participant's excess reduction amount per beneficiary (as determined by 
CMS in accordance with Sec.  513.230(d)(6)) is greater than zero. That 
is, the MFN participant must have experienced a reduction in Medicare 
FFS allowed charges for separately payable Medicare Part B drugs on a 
per beneficiary basis during the performance year as compared to the 
prior year (that is, the four calendar quarters immediately preceding 
the performance year) that is greater than 25 percent of the MFN 
participant's total Medicare Part A and Medicare Part B FFS allowed 
charges on a per beneficiary basis during the prior year. We are 
establishing a threshold of 25 percent of the MFN participant's total 
Medicare Part A and Medicare Part B FFS allowed charges on a per 
beneficiary basis as a criterion to qualify for the financial hardship 
exemption because the exemption is designed to be limited to MFN 
participants that experience a significant year-to-year reduction in 
total allowed charges as a result of the MFN Model. We believe this 
threshold will protect MFN participants from significant financial 
hardship under the MFN Model while also preserving the model test of 
aligning payment for Medicare Part B drugs with the lowest 
international prices using a phase-in approach.
    Incomplete financial hardship exemption requests will not be 
considered by CMS.
    In Sec.  513.230(d), we are codifying how CMS will calculate the 
MFN participant's excess reduction amount per beneficiary. CMS will 
calculate the MFN participant's excess reduction amount per beneficiary 
using available final action claims data that are estimated to be more 
than 90 percent complete (claims are generally complete within 2 months 
after the service month) where Medicare was the primary payer, as 
determined by CMS. This approach will not include non-claims based 
payments or other transactions, for example, performance-based payment 
or repayments. CMS will calculate, for dates of service within the 
performance year, the MFN participant's total allowed charges for 
separately payable Medicare Part B drugs, and the total number of 
beneficiaries that had at least one claim for a service furnished by 
the MFN participant with a Medicare Part A or Medicare Part B allowed 
charge greater than $0. Then, CMS will divide the MFN participant's 
total allowed charges for separately payable Medicare Part B drugs for 
dates of service within the performance year by the total number of 
beneficiaries that had at least one claim for a service furnished by 
the MFN participant with a Medicare Part A or Medicare Part B allowed 
charge greater than $0 with a service date within the performance year. 
CMS will repeat this calculation using the available claims data for 
the prior year, to calculate the MFN participant's average per 
beneficiary total allowed charges for separately payable Medicare Part 
B drugs for the prior year. Then, CMS will subtract the MFN 
participant's average per beneficiary total allowed charges for 
separately payable Medicare Part B drugs for the performance year from 
the MFN participant's average per beneficiary total allowed charges for 
separately payable Medicare Part B drugs for the prior year. This 
difference will then be compared to 25 percent of the MFN participant's 
average per beneficiary total allowed charges for all Medicare Part A 
and Part B claims with dates of service within the prior year, using 
subtraction as described in Sec.  513.230(d)(6). The latter quantity 
will be calculated by identifying 25 percent of the MFN participant's 
total allowed charges for all Medicare Part A and Part B claims with 
dates of service within the prior year, then dividing this amount by 
the total number of beneficiaries that had at least one claim for a 
service furnished by the MFN participant with a Medicare Part A or 
Medicare Part B allowed charge greater than $0 with a date of service 
within the prior year. If the resulting amount, called the excess 
reduction amount per beneficiary, is greater than zero, then the MFN 
participant will meet this eligibility criterion for the financial 
hardship exemption.
    In Sec.  513.230(e)(1), we are codifying that if CMS in its sole 
discretion grants a financial hardship exemption to an MFN participant 
for a performance year, CMS shall provide to such MFN participant, a 
reconciliation payment for the performance year. To calculate the 
reconciliation amount for the MFN participant, CMS will multiply the 
excess reduction amount per beneficiary by the total number of 
beneficiaries that had at least one claim for a service furnished by 
the MFN participant with a Medicare Part A or Medicare Part B allowed 
charge greater than $0 with a service date within the performance year.
    The reconciliation payment amount will be paid by a CMS contractor 
using Medicare Part B funds as soon as practical after CMS notifies the 
MFN participant of CMS's decision regarding the MFN participant's 
financial hardship exemption request and the amount of the 
reconciliation payment, if any, to be made to the MFN participant. In 
Sec.  513.230(e)(2), we are codifying that there will be no appeal of 
the amount of the reconciliation payment, if any, to be made to the MFN 
participant. In addition, the reconciliation payment amount will not be 
subject to beneficiary cost sharing (including any deductible or 
coinsurance) because the reconciliation payment will not be tied to 
specific beneficiary claims, beneficiaries will have been responsible 
for 20 percent cost-sharing on the allowed payment amounts for the

[[Page 76224]]

Medicare Part B drugs they received during the performance year, and 
steps to seek additional cost-sharing from beneficiaries would likely 
cause significant confusion and burden for beneficiaries and MFN 
participants.
    We do not foresee that many MFN participants will qualify for a 
reconciliation payment for performance year 1, because the estimated 
overall reduction in Medicare Part B drug payment during performance 
year 1 is 7 percent on average. This reflects the MFN Price phase-in 
formula in section III.E.5. of this IFC which will begin with the MFN 
Price making up 25 percent of the MFN Drug Payment Amount and the 
alternative add-on payments in section III.F. of this IFC will 
represent a 40 percent increase on average for MFN participants 
relative to historical Medicare add-on payments. Given the financial 
hardship execption threshold of 25 percent of the MFN participant's 
total Medicare Part A and Medicare Part B FFS allowed charges on a per 
beneficiary basis in the prior year will be determined at the entity 
level, MFN participants with a high proportion of their overall 
Medicare payments related to MFN Model drugs will be more likely to 
qualify for the hardship exemption if their Medicare Part B drug 
allowed charges on a per beneficiary basis during a performance year 
were to decrease significantly compared to the prior year. MFN 
participants that are hospitals will likely have significant Medicare 
Part A revenues and purchasing abilities that will lessen the 
likelihood that they will qualify for a financial hardship exemption 
based on their experience in the MFN Model during performance year 1. 
Non-hospital MFN participants will be more likely to potentially 
qualify in later performance years.
    For future years, we seek comment on whether an alternative 
threshold might better protect beneficiary access to MFN Model drugs or 
mitigate impacts on physicians and other MFN participants under the MFN 
Model. For example, we are interested in whether a uniform threshold 
should be applied for all MFN participants, and whether certain 
physician specialties or types of MFN participants would find the 
threshold insufficient in protecting beneficiary access to MFN Model 
drugs. For future rulemaking, we also seek comment on how CMS could 
refine the design of the financial hardship exception to advance the 
model goals to reduce program expenditures and maintain or improve 
quality of care.
    CMS pledges to maintain confidentiality of individual financial 
hardship exemption requests to the extent provided by law. However, CMS 
may make public descriptive information about MFN participants that are 
granted a financial hardship exemption and the extent to which they 
were unable to obtain MFN Model drugs at or below the MFN Model Payment 
for such drugs. We do not intend to make such information available in 
an individually identifiable manner.
3. Availability of Services
    The MFN Model is designed to test potential improvements to the 
delivery of and payment for healthcare to reduce Medicare expenditures 
while preserving or enhancing the quality of care for beneficiaries. As 
such, an important aspect of testing models is that beneficiaries must 
continue to have access to and receive needed care.
    In Sec.  513.410(c), we are codifying that MFN participants must 
not take any action to select or avoid treating beneficiaries based on 
their diagnoses, care needs, income levels, or other factors that would 
render them ``at-risk beneficiaries'' as that term is defined at 42 CFR 
425.20 (``lemon dropping''). We will use monitoring to ensure that MFN 
participants are complying with this requirement. We believe that this 
is a necessary precaution to protect beneficiaries against potential 
beneficiary selection bias from MFN participants and ensure that MFN 
beneficiaries retain access to medically necessary treatment.
4. Monitoring and Compliance Activities
    Consistent with other CMS Innovation Center models, CMS will 
implement a monitoring program for the MFN Model to ensure that the MFN 
Model is implemented safely and appropriately. Given that MFN 
participants will receive model-specific payments and access to payment 
rule waivers while participating in the MFN Model, we believe that 
enhanced compliance review and monitoring of MFN participants is 
necessary and appropriate to ensure the integrity of the MFN Model. In 
addition, as part of the CMS Innovation Center's assessment of the 
impact of new models such as the MFN Model, we have a special interest 
in ensuring that model tests do not interfere with ensuring the 
integrity of the Medicare program. Our interests include ensuring the 
integrity and sustainability of the MFN Model and the underlying 
Medicare program from both a financial and policy perspective, as well 
as protecting the rights and interests of Medicare beneficiaries. For 
these reasons, as a part of the models currently being tested by the 
CMS Innovation Center, CMS or its designee(s) monitors model 
participants to assess compliance with model terms and with other 
applicable program laws and policies. We believe our monitoring efforts 
help ensure that model participants are furnishing medically necessary 
covered services and are not falsifying data, increasing program costs, 
or taking other actions that compromise the integrity of the model or 
are not in the best interests of the model, the Medicare program, or 
Medicare beneficiaries.
    In Sec.  513.420, we are codifying a framework for conducting 
compliance monitoring activities for the MFN Model that is consistent 
with the standard practices in other CMS Innovation Center models. 
Under the monitoring policy at Sec.  513.420(b), MFN participants will 
be monitored to assess compliance with the MFN Model requirements, to 
determine the effects of the MFN Model on MFN beneficiaries, providers, 
suppliers, and on the Medicare program and to facilitate real time 
identification and response to potential issues. Further, under Sec.  
513.420(a)(2), an MFN participant will be required to notify CMS within 
15 calendar days after becoming aware that the MFN participant is under 
investigation or has been sanctioned by the federal, state, or local 
government, or any licensing authority (including, without limitation, 
the imposition of program exclusion, debarment, civil monetary 
penalties, corrective action plans, and revocation of Medicare billing 
rights).
    In Sec.  513.420(b)(2), we are codifying that when we are 
conducting compliance monitoring and oversight activities, CMS or our 
designees will be authorized to use any relevant data or information, 
including without limitation Medicare claims submitted for items or 
services furnished to MFN beneficiaries. In Sec.  513.420(b)(3), we are 
codifying that MFN participants will be required to cooperate with the 
model monitoring and evaluation activities, comply with the 
government's right to audit, inspect, investigate, and evaluate any 
documents or other evidence regarding implementation of the MFN Model, 
and to retain and provide the government with access to records.
    In Sec.  513.420(b)(1), we are codifying that monitoring activities 
will include, but will not be limited to: (1) Documentation requests 
sent to the MFN participant, including surveys and questionnaires; (2) 
audits of claims data, medical records, and other data from the MFN 
participant; (3) interviews with any individual or entity participating 
in the MFN Model, including members of the MFN participant's 
leadership,

[[Page 76225]]

management, and staff; (4) interviews with beneficiaries and their 
caregivers; (5) site visits to the MFN participant; and (6) tracking 
complaints and appeals. We believe these specific monitoring 
activities, which align with those currently used in other models being 
tested by the CMS Innovation Center, are necessary in order to ensure 
compliance with the terms and conditions of the MFN Model and to 
protect beneficiaries from potential harms that may result from 
activities of an MFN participant, such as attempts to reduce access to 
medically necessary covered services or appropriate drugs.
    We anticipate that monitoring of the MFN Model activities will 
include gathering and analyzing data captured through the Ombudsman's 
service, the evaluation of the MFN Model, the patient experience 
survey, and audits of charts, claims data, medical records, among other 
data as available. As previously noted in this IFC, one purpose of 
monitoring and analyzing these data sources will be to provide timely 
information about the effects of the MFN Model on MFN beneficiaries, 
providers, suppliers, and on the Medicare program, and to facilitate 
real time identification and response to potential issues. We 
anticipate that these findings will inform model oversight and the 
potential need for action to address identified issues.
    In Sec.  513.420(c), we outline parameters for site visits. We will 
require that MFN participants cooperate in periodic site visits 
conducted by CMS or its designee. Such site visits will be conducted to 
facilitate the model implementation.
    In order to operationalize this model, CMS or its designee will 
provide the MFN participant with no less than 15 calendar days advance 
notice of a site visit, to the extent practicable. Furthermore, to the 
extent practicable, CMS will attempt to accommodate a request that a 
site visit be conducted on a particular date, but that the MFN 
participant will be prohibited from requesting a date that was more 
than 60 calendar days after the date of the initial site visit notice 
from CMS. We believe the 60-calendar day period will reasonably 
accommodate MFN Model participants' schedules while not interfering 
with the operation of the MFN Model. Further, we will require MFN 
participants to ensure that personnel with the appropriate 
responsibilities and knowledge pertaining to the purpose of the site 
visit be available during any and all site visits. We believe this is 
necessary to ensure an effective site visit and prevent the need for 
unnecessary follow-up site visits.
    Finally, CMS or its designee can perform unannounced site visits to 
all physical locations of MFN participants at any time to investigate 
concerns related to the health or safety of beneficiaries or other 
patients or other program integrity issues, notwithstanding these 
provisions. Further, nothing in part 513 will limit CMS from performing 
other site visits as allowed or required by applicable law. We believe 
that, regardless of the model being tested, CMS must always have the 
ability to timely investigate concerns related to the health or safety 
of beneficiaries or other patients, or program integrity issues, and to 
perform functions required or authorized by law. In particular, we 
believe that it will be necessary for us to monitor, and for MFN 
participants to be compliant with our monitoring efforts, to ensure 
that they are not denying or limiting the coverage or provision of 
medically necessary covered services to beneficiaries in an attempt to 
change the MFN Model results or their MFN Model payments, including 
discrimination in the provision of services to at-risk beneficiaries 
(for example, due to eligibility for Medicaid based on disability).
    We intend to monitor MFN participants through any of the previously 
described monitoring activities (such as documentation requests, audits 
of claims data, audits of medical records, etc.) to ensure that MFN 
Model drugs are not being inappropriately billed (for example, 
excessive doses or units). We anticipate that this monitoring activity 
will discourage MFN participants from furnishing smaller and more 
frequent doses of MFN Model drugs to beneficiaries in order to maximize 
the alternative add-on payments. If it is found that an MFN participant 
has been engaged in inappropriate billing, then we will use applicable 
remedial actions set forth in Sec.  513.440(a)(2).
    We may employ longer-term analytic strategies to confirm our 
ongoing analyses and detect more subtle or hard-to-determine changes in 
care delivery and beneficiary outcomes. Some determinations of 
beneficiary outcomes or changes in treatment delivery patterns may not 
be able to be built into ongoing claims analytic efforts and may 
require longer-term study.
a. Reduced Access
    We will monitor claims data from MFN participants--for example, to 
compare MFN participants' case mix relative to a pre-model historical 
baseline to determine whether complex patients are being systematically 
excluded. To the extent that the use of a patient experience survey 
includes items focused on access, we will analyze these data as well to 
determine whether MFN beneficiaries continue to be able to access the 
right drug at the right time. We will use these data to promote 
transparency and develop an understanding of the MFN Model's effects. 
We intend to review and audit MFN participants if we have reason to 
believe that they are compromising beneficiary access to care.
    We intend to conduct analyses of claims data, such as monthly 
updates and historic comparisons of trends including drug utilization, 
program spending, and prescribing patterns (including observing for any 
shift to compounded or other categories of drugs that are not included 
in the MFN Model) as well as changes in site of service delivery, 
mortality, hospital admissions, and other indicators present in claims 
data. We will monitor physician visits, days in a hospital, and other 
services as part of the thorough look at how MFN beneficiaries are 
receiving care to determine whether any treatment patterns are changing 
systematically. We will use the monitoring results to detect potential 
issues with beneficiary access to care or potential provider and 
supplier payment issues.
b. Quality of Care Monitoring
    We anticipate that quality monitoring activities may include claims 
and survey data analytics, site visits, medical record review, and 
tracking patient complaints and appeals. We will also use the most 
recent claims data available to track utilization and beneficiary 
outcomes under the MFN Model. We believe this type of monitoring is 
important as we want to ensure to the greatest extent possible that 
patients continue to receive high-quality care.
    We believe that this set of monitoring activities will allow us to 
promptly identify any unintended consequences of the MFN Model. We 
anticipate that by identifying unintended potential consequences of the 
MFN Model, that we will then be able to determine methods to address or 
alleviate those potential consequences.
c. Remedying Improper Payment
    We anticipate that our monitoring activities may identify instances 
of incorrect MFN Model payments. As such, we are codifying that CMS is 
authorized to correct model-specific payments under Sec.  513.420(d). 
Specifically, under this section if CMS discovers that it has made or 
received

[[Page 76226]]

an incorrect model-specific payment under the terms of the MFN Model, 
then CMS may make payment to, or demand payment from, the MFN 
participant. Should these monitoring activities identify a need for 
additional protections, we will consider appropriate action.
d. Compliance With Laws
    MFN participants will remain subject to all existing requirements 
and conditions for Medicare participation as set out in Federal 
statutes and regulations and provider and supplier agreements, unless 
waived under the authority of section 1115A(d)(1) of the Act solely for 
purposes of testing the MFN Model. In Sec.  513.420(a)(1), we therefore 
require that MFN participants must comply with all applicable laws and 
regulations. We note that a law or regulation is not ``applicable'' to 
the extent that its requirements have been waived under section 
1115A(d)(1) of the Act solely for purposes of testing the MFN Model.
5. Enforcement Authority and Remedial Action
    We are codifying at Sec.  513.440(b) that nothing contained in the 
terms of the MFN Model or part 513 will limit or restrict the authority 
of the HHS Office of Inspector General (OIG) or any other Federal 
Government authority, including its authority to audit, evaluate, 
investigate, or inspect the MFN participant.
    It is necessary for CMS to have the ability to impose remedial 
actions to address non-compliance with the requirements of the MFN 
Model and to ensure that the MFN Model does not interfere with the 
program integrity interests of the Medicare Program. Thus, in Sec.  
513.440(a)(1), CMS may take remedial action against an MFN participant 
if CMS determines, in CMS' sole discretion, that the MFN participant--
     Has failed to comply with any applicable Medicare program 
requirement, rule, or regulation;
     Has failed to comply with any of the terms of the MFN 
Model, including applicable requirements of part 513;
     Systematically engaged in the under delivery or over 
delivery of an MFN Model drug;
     Has taken any action that threatens the health or safety 
of an MFN beneficiary or other patient;
     Has undergone a change of control that presents a program 
integrity risk;
     Has submitted false data or made false representations, 
warranties, certifications or attestations in connection with any 
aspect of the MFN Model;
     Has avoided at-risk beneficiaries, as this term is defined 
in Sec.  425.20;
     Has avoided patients on the basis of payer status;
     Is subject to any sanctions or final actions of an 
accrediting organization or a Federal, State, or local government 
agency;
     Takes any action that CMS determines for program integrity 
reasons is not in the best interests of the MFN Model, or the Medicare 
program, or fails to take any action that CMS determines for program 
integrity reasons should have been taken to further the best interests 
of the MFN Model or Medicare program;
     Is subject to investigation or action by HHS (including 
the HHS Office of the Inspector General (OIG)) or the Department of 
Justice due to an allegation of fraud or significant misconduct, 
including being subject to the filing of a complaint, filing of a 
criminal charge, being subject to an indictment, being named as a 
defendant in a False Claims Act qui tam matter in which the Federal 
Government has intervened, or similar action;
     Is the subject of administrative enforcement action 
imposed by CMS; or
     Has failed to demonstrate improved performance following 
any remedial action imposed by CMS.
    In Sec.  513.440(a)(2), we are codifying that if CMS determines 
that one or more grounds for remedial action exists, CMS may take one 
or more of the following remedial actions:
     Notify the MFN participant of the violation.
     Require the MFN participant to provide additional 
information to CMS or its designees.
     Require the MFN participant to develop and implement a 
corrective action plan in a form and manner and by a deadline specified 
by CMS.
     Subject the MFN participant to additional monitoring, 
auditing, or both.
     Remove the MFN participant from the MFN Model;
     Recoup model-specific payments.
     Such other action as may be permitted under the terms of 
Sec.  513.420.
6. Audits and Record Retention
    By virtue of participation in the MFN Model, MFN participants will 
receive model-specific payments and access to payment rule waivers. We 
therefore believe that CMS' ability to audit, inspect, investigate, and 
evaluate records and other materials related to participation in the 
MFN Model is necessary and appropriate. In order to expand a phase 1 
model tested by the CMS Innovation Center, among other things, the 
Secretary must first determine that such expansion would not deny or 
limit the coverage or provision of benefits under the applicable title 
for applicable individuals. Thus, there is a particular need for CMS to 
be able to audit, inspect, investigate, and evaluate records and 
materials related to participation in CMS Innovation Center models to 
allow us to ensure that the model is not denying or limiting the 
coverage or provision of benefits for beneficiaries.
    We note that there are audit and record retention requirements 
under the Medicare Shared Savings Program (42 CFR 425.314) and in 
current models being tested under section 1115A (such as under 42 CFR 
510.110 for the CMS Innovation Center's Comprehensive Care for Joint 
Replacement Model). Building off those existing requirements, in Sec.  
513.430(a), the Federal Government, including, but not limited to, CMS, 
HHS, and the Comptroller General, or their designees, have a right to 
audit, inspect, investigate, and evaluate any documents and other 
evidence regarding implementation of the MFN Model. Additionally, in 
order to align with the policy of current models being tested by the 
CMS Innovation Center, we are codifying in Sec. Sec.  513.430(b) and 
(c) that MFN participants must--
     Maintain and give the Federal Government, including, but 
not limited to, CMS, HHS, and the Comptroller General, or their 
designees, access to all documents (including books, contracts, and 
records) and other evidence sufficient to enable the audit, evaluation, 
inspection, or investigation of the MFN Model, including without 
limitation, documents and other evidence regarding all of the 
following:
    ++ The MFN participant's compliance with the terms of the MFN 
Model, including new subpart E of part 513.
    ++ Quality measure information and the quality of services 
performed under the terms of the MFN Model, including new subpart E of 
part 513.
    ++ Patient safety.
    ++ The accuracy of model-specific payments under the MFN Model.
    ++ Utilization of items and services furnished under the MFN Model.
    ++ Any other program integrity issues.
     Maintain the documents and other evidence for a period of 
6 years from the last payment received by the MFN participant under the 
MFN Model or from the date of completion of any audit, evaluation, 
inspection, or

[[Page 76227]]

investigation, whichever is later, unless--
    ++ CMS determines there is a special need to retain a particular 
record or group of records for a longer period and notifies the MFN 
participant at least 30 calendar days before the normal disposition 
date; or
    ++ There has been a termination, dispute, or allegation of fraud or 
similar fault against the MFN participant in which case the records 
must be maintained for an addition 6 years from the date of any 
resulting final resolution of the termination, dispute, or allegation 
of fraud or similar fault.
    If CMS notifies the MFN participant of the special need to retain 
records or group of records at least 30 calendar days before the normal 
disposition date, the records must be maintained for such period of 
time determined by CMS.

J. Interaction With Other Models and Programs

1. Approach for Overlap With Other Models
    In designing each CMS Innovation Center model, CMS considers 
potential overlap between a new model and other ongoing and potential 
models and programs. Based on the type of overlap, such as health care 
provider or beneficiary, operating rules may be established for whether 
or not health care providers and beneficiaries can be part of both 
models as well as how to handle overlap when it occurs. These policies 
help to ensure that the evaluation of model impact is not compromised 
by issues of model overlap and that double counting of beneficiaries 
and dollars across different models does not occur.
    In response to the October 2018 ANPRM, several commenters expressed 
concern regarding model overlap, specifically with the Oncology Care 
Model (OCM) and initiatives involving accountable care organizations 
(ACOs). Some commenters noted that OCM participants should be excluded 
from the potential IPI Model or excluded from mandatory participation. 
Some commenters also requested that ACO initiatives take precedence in 
terms of calculating shared savings as well as for clarity on how 
overlap between ACO initiatives and the potential IPI Model would work.
    We appreciate commenters' request for detailed information about 
model overlap policies. In developing the MFN Model, CMS conducted an 
internal review of which models will have potential overlap with the 
MFN Model. As a result of our review, we expect there will be 
situations where a Medicare beneficiary who receives an MFN Model drug 
will also be assigned, aligned, or attributed to another CMS Innovation 
Center model or CMS program. Overlap could also occur among providers 
and suppliers at the individual or organization level, for example, a 
health care practitioner or a physician group practice could 
participate in multiple CMS Innovation Center models and CMS programs 
concurrently. Of note, some existing models and programs will not have 
overlap at the health care practitioner or participant level due to the 
way in which the model or program operates and makes payments.
    We believe that the MFN Model is operationally compatible with 
existing models and programs that provide opportunities to improve care 
and reduce spending, especially total cost of care-focused CMS programs 
and Innovation Center models. The MFN Model will test an innovative way 
to pay for Medicare Part B drugs that seeks to address any existing 
incentives for prescribing higher cost drugs and ways to lower costs 
for beneficiaries and the Medicare program; total cost of care-focused 
CMS programs and Innovation Center models incentivize more appropriate 
provision of care across multiple clinical areas, including use of 
Medicare Part B drugs; the MFN Model addresses only use of certain 
Medicare Part B drugs. To some degree, incentives for inappropriate use 
of higher cost drugs are reduced, and intended effects of the MFN Model 
are already built into total cost of care-focused models, so the 
addition of the MFN Model should not have further effects in those 
programs. We do not plan to make adjustments to the MFN Drug Payment 
Amount or MFN alternative add-on payment due to overlap between the MFN 
Model and another model or program, unless such model tests an 
alternative approach to the add-on portion of payment for Medicare Part 
B drugs as specified in Sec.  513.220(d)(2). However, for certain 
models and programs, adjustments to those models and programs may be 
necessary to account for payment changes under the MFN Model.
    Because the MFN Model will focus on approximately 50 separately 
payable Medicare Part B drugs, when claims are considered from all 
beneficiaries aligned with or assigned to some other Innovation Center 
models or CMS programs that focus on total cost of care, such as the 
Medicare Shared Savings Program, we do not expect that the MFN Model 
will have a significant impact on shared savings, total cost of care, 
or other benchmarks and measures. Therefore, changes to benchmarks, 
targets, and reconciliation methodologies may not be necessary, and 
will be determined by each other model, program, or initiative as 
appropriate.
    However, we recognize that the design of some other models, 
programs, and initiatives could create unique challenges at the 
organization, clinician, or beneficiary level. As a result, we will 
work with such models, programs, or initiatives to resolve any 
potential overlaps that could result in overpayment of savings due to 
double counting of the impact of a result that could be attributed to 
the interventions from two different models. For example, OCM focuses 
on improved care management and coordination for Medicare beneficiaries 
with cancer who receive chemotherapy during 6-month episodes of care. 
An OCM practice has the opportunity to receive a performance-based 
payment if it reduces the total cost of care in its OCM episodes 
compared to a target. Based on the performance year 1 MFN Model Drug 
HCPCS Codes List, we anticipate substantial overlap between MFN 
participants and MFN beneficiaries with OCM practices and OCM 
beneficiaries. To avoid paying performance-based payments in OCM that 
are due simply to the drug payment change that will occur under the MFN 
Model and not to changes in care delivery, for OCM, we will adjust 
reconciliation calculations such that the drug payments included in OCM 
episode expenditures will be calculated as if the MFN Model were not 
occurring. OCM participants will be notified and provided with further 
information through OCM's typical channels of communication.
    As discussed in the section III.C.1. of this IFC, CMMI has already 
waived section 1833(t) of the Act for certain acute care hospitals due 
to their participation in models under section 1115A of the Act for 
which payment for outpatient hospital services furnished to Medicare 
FFS beneficiaries, including MFN Model drugs, is made under such model 
on a fully capitated or global budget basis. For the first and second 
quarters of performance year 1, we will exclude these entities from the 
MFN Model with limitation. That is, the acute care hospitals that 
participate in another CMS Innovation Center model under which they are 
paid for outpatient hospital services furnished to Medicare FFS 
beneficiaries, including MFN Model drugs, on a fully capitated or 
global budget basis under a waiver under such model of section 1833(t) 
of the Act, such as the Maryland Total Cost of Care Model and the 
Pennsylvania Rural Health Model, will be excluded

[[Page 76228]]

from the MFN Model. For the third quarter of performance year 1 and 
beyond, acute care hospitals that participate in a CMS Innovation 
Center model under which they are paid for outpatient hospital services 
furnished to Medicare FFS beneficiaries, including MFN Model drugs, on 
a fully capitated or global budget basis under a waiver under such 
model of section 1833(t) of the Act will be excluded from the MFN Model 
if the parameters of the other CMS Innovation Center model adjust for 
the difference in payment for MFN Model drugs between the MFN Model and 
non-MFN Model drug payments such that savings under the MFN Model are 
incorporated into the other CMS Innovation Center model's parameters 
(for example, the annual global budget) for the duration of the MFN 
Model. These exclusions will apply only during the period of the 
hospital's participation in such model under which it is paid on a 
fully capitated or global budget basis. Upon termination of such 
participation for any reason or if the model is revised such that the 
waiver of section 1833(t) of the Act no longer applies under such 
model, the hospital--if it otherwise meets the definition of MFN 
participant--will be required to participate in the MFN Model.
    We anticipate model overlap may occur between the MFN Model and 
future CMS models or programs not yet implemented. As discussed in 
section III.F.5. of this IFC, if there are MFN participants that 
concurrently participate in a future CMS model that also tests an 
alternative approach to the add-on portion of payment for Part B drugs, 
we will not make the MFN alternative add-on payment to those MFN 
participants for those MFN Model drugs that overlap with the other 
model. Instead, we will follow the other model's approach to making an 
alternative add-on payment. We expect this overlap policy will maintain 
the intended financial effects of the MFN Model, while allowing 
operational compatibility with other models that test alternative 
approaches to Medicare Part B drug payment.
2. Quality Payment Program
    The MFN Model will not qualify as an Advanced APM under the Quality 
Payment Program. Specifically, the MFN Model does not require 
participant health care providers to use CEHRT, does not base payment 
to participant health care providers on quality measures, and does not 
satisfy the financial risk criteria because it does not involve 
requiring participating APM Entities to bear risk for monetary losses 
of more than nominal amounts under the APM and is not a Medical Home 
Model expanded under section 1115A(c) of the Act. The MFN Model also 
will not qualify as a MIPS APM, because it does not hold participant 
health care providers financially accountable for both the cost and 
quality of care provided to Medicare beneficiaries.

K. Interaction With Other Federal Programs

    The MFN Model may have impacts on other federal programs, such as 
Medicaid, the 340B Program, the Veterans Health Administration, the 
Department of Defense, the Public Health Service, the Coast Guard, and 
Medicare.
1. Impact on Medicaid
a. Impact on Medicaid ``Best Price''
    With respect to single source or innovator multiple source drugs 
(which Medicaid recognizes to include biologicals), the term ``Medicaid 
Best Price'' is the lowest price available from the manufacturer during 
the rebate period to any wholesaler, retailer, provider, health 
maintenance organization, non-profit entity or governmental entity 
within the U.S. with certain exclusions. That is, a manufacturer's best 
price determination represents the lowest price available from the 
manufacturer during a rebate period (a quarter) to best price eligible 
entities or purchasers in the U.S. only.
    Since the MFN Drug Payment Amount will be paid to MFN participants 
for each MFN Model drug as a Medicare payment, and it will not be a 
``price available from the manufacturer,'' the MFN Drug Payment Amounts 
themselves will not be included in the manufacturer's determination of 
best price. However, in order for MFN participants to purchase MFN 
Model drugs at prices that does not lead to financial loss, the 
manufacturer will need to make available prices that are competitive 
with the MFN Drug Payment Amounts. We expect that the MFN Drug Payment 
Amounts will likely drive manufacturer drug prices available to MFN 
participants down over the course of the model, and the model may 
indirectly impact a manufacturer's best price to the extent that a 
manufacturers' U.S. best price will be lower than what it would be 
otherwise. In other words, if during the course of the MFN Model, 
market forces result in manufacturers reducing prices available to MFN 
participants, such available prices to MFN participants will be 
considered in a manufacturer's determination of best price and could 
potentially lower best price and possibly increase Medicaid rebates.
    Specifically, if the manufacturer lowers prices available to an MFN 
participant at or below the MFN Drug Payment Amount, such prices will 
be considered in the manufacturer's determination of best price and may 
reset the manufacturer's best price if the reduced price is lower than 
the manufacturer's best price that would otherwise apply. This is 
particularly possible because the MFN Drug Payment Amount, which is 
expected to be lower than the payment amounts for the same drugs 
outside of the model, will include the impact of pricing outside of the 
U.S., which is typically lower than prices in the U.S., and will likely 
impact the prices made available by the manufacturer in the U.S.
b. Impact on Average Manufacturer Price (AMP)
    AMP is defined at section 1927(k)(1) of the Act. Generally, AMP is 
determined based on the average price paid to the manufacturer for a 
covered outpatient drug in the U.S. by wholesalers for drugs 
distributed to retail community pharmacies and retail community 
pharmacies that purchase drugs directly from the manufacturer with 
certain exclusions. Because the MFN Model will focus on certain Part B 
drugs that are furnished in the outpatient setting and these drugs are 
most likely injected or infused, the AMP for an MFN Model drug is 
likely determined using the AMP computation for 5i drugs,\73\ which 
includes sales that are not generally dispensed through retail 
community pharmacies (see section 1927(k)(1)(B)(i)(IV) of the Act, 42 
CFR 447.504(d)), such as sales to physicians, pharmacy benefit managers 
(PBMs) and hospitals. Thus, a manufacturer's sales of MFN Model drugs 
to MFN participants (or price paid by MFN participants) will be 
included in the AMP or 5i AMP. If, as described in section III.K.1.a. 
of this IFC, the manufacturer lowers prices available to an MFN 
participant at or below the MFN Drug Payment Amount, the manufacturer's 
AMP for an MFN Model drug may be lower. If a drug's AMP decreases, it 
may result in potentially lowering the applicable Medicaid drug rebate 
paid (the rebate, in part, is based on a percentage of AMP). However, 
the MFN Model may also lower a manufacturer's best price for an MFN 
Model drug as previously discussed. The resulting effect on the 
Medicaid

[[Page 76229]]

drug rebate will depend upon the relationship of any AMP change and any 
best price change.
---------------------------------------------------------------------------

    \73\ Inhalation, infusion, instilled, implanted or injectable 
drugs.
---------------------------------------------------------------------------

    We also note that if the AMP for an MFN Model drug is lowered it 
may be more likely that, in accordance with section 1847A of the Act, 
the Inspector General may find that the ASP for an MFN Model drug 
exceeds the AMP for such drug, and that the circumstances in which 103 
percent of AMP is substituted for ASP in CMS's determination of the 
non-model payment allowance for such drug would occur. We refer readers 
to section III.L. of this IFC for a discussion of excluding units of 
MFN Model drugs from manufacturers' ASP, which may also increase the 
likelihood that the ASP for an MFN Model drug will be greater than the 
AMP for such drug.
2. Interaction With 340B Program
    The Health Resources and Services Administration (HRSA) administers 
the 340B Drug Pricing Program that allows certain hospitals and other 
health care providers (``covered entities'') to obtain discounted 
prices on ``covered outpatient drugs'' (as defined at 1927(k)(2) of the 
Act) from drug manufacturers. HRSA calculates a 340B ceiling price for 
each covered outpatient drug, which represents the maximum price a 
manufacturer can charge a covered entity for the drug that is provided 
to an eligible patient. Several types of hospitals as well as clinics 
that receive certain federal grants from the HHS may enroll in the 340B 
program as covered entities. Such entities will be included in the MFN 
Model and will be subject to the MFN Model payment test. That is, these 
340B covered entities will be MFN participants and receive the MFN Drug 
Payment Amount and alternative add-on payment. To the extent these 
entities receive payment under the model that is lower than their 
current Medicare payment, there may be fewer resources available for 
their 340B program activities.
    Under the MFN Model, MFN participants will be paid for MFN Model 
drugs according to the payment approach discussed in section III.E. of 
this IFC. If the MFN participant is a 340B covered entity, the drug 
portion of the model payment will be the lower of the MFN Drug Payment 
Amount or the non-model payment amount paid to 340B covered entities 
for 340B drugs under the OPPS for the MFN Model drug for that 
corresponding calendar quarter. The MFN alternative add-on payment will 
be paid to MFN participants that are 340B covered entities in the same 
way as MFN participants that are non-340B covered entities.
    We are including certain 340B covered entities in the MFN Model in 
order to test the innovative payment approach, including the 
alternative (per-dose) add-on payment amount, broadly. MFN participants 
that are 340B covered entities may need to enhance their direct 
contracting with manufacturers in order to obtain MFN Model drugs 
within the MFN Drug Payment Amount. Our analyses estimate that 340B 
covered entities will realize a total add-on percentage amount of 4.5 
percent in the first year of the model due to the mix of MFN Model 
drugs they historically furnish. The amount of the alternative add-on 
that 340B entities realize will be an increase in revenue compared to 
their historical baseline. However, these entities will face the same 
or increased burden from model participation. Thus, we believe the 
modest increase in add-on revenue that will be paid to these entities 
through the alternative add-on payment approach will potentially be 
offset through higher facility costs for acquiring included drugs (for 
example, higher costs for direct contracting). Programs that support 
vulnerable Americans are a vital safety net. We refer readers to 
section III.C. of this IFC where we discuss providers and suppliers 
that will be MFN participants. We discuss potential impacts on 340B 
covered entities in more detail in section VI. of this IFC.
a. Impact on 340B Ceiling Price
    Covered entities that enroll in the 340B Program can purchase 
covered outpatient drugs at no more than a ``ceiling price,'' which is 
calculated as AMP minus Medicaid unit rebate amount. We note that a 
ceiling price is just a ceiling; some 340B hospitals can obtain covered 
outpatient drugs at less than the ceiling price. Since the Medicaid 
unit rebate amount is based partly on AMP minus best price, to the 
extent the MFN Model affects a drug's AMP and best price, the 340B 
prices will be affected. We discuss the potential impacts on a drug's 
AMP and best price in section III.K.1. of this IFC.
3. Interaction With Medicare
a. Medicare Part B
    As discussed in section VI. of this IFC, we believe the MFN Model 
will result in lower Medicare spending for MFN Model drugs, including 
lower program spending and lower beneficiary cost-sharing, and in 
overall reduced Medicare Part B Trust Fund expenditures, which in turn 
will lower Medicare FFS expenditures and beneficiaries' Part B 
premiums.
    As discussed in section III.K. of this IFC, manufacturers' ASPs for 
MFN Model drugs may be higher or lower than they otherwise would be 
absent the MFN Model. In turn, non-model Medicare Part B FFS payment 
for MFN Model drugs could be higher or lower. We are excluding from the 
calculation of the manufacturer's ASP any units of an MFN Model drugs 
furnished to MFN beneficiaries and billed by MFN participants. Thus, 
during the MFN Model, manufacturers' ASPs for MFN Model drugs could be 
higher or lower than they might be absent the model, resulting in 
Medicare payments to providers and suppliers that are not MFN 
participants that would be higher or lower than what the payments would 
have been absent the model.
    We note that if the AMP for an MFN Model drug is lowered it may be 
more likely that, in accordance with section 1847A of the Act, the 
Inspector General may find that the ASP for an MFN Model drug exceeds 
the AMP for such drug, and that the circumstances in which 103 percent 
of AMP is substituted for ASP in CMS's determination of the non-model 
payment allowance for such drug would occur.
b. Medicare Advantage
    Medicare Advantage (MA) plans will not be MFN participants. We note 
that when MA plans pay non-contracted, out of network providers who 
have administered an MFN model drug to an enrollee, the amount paid 
will be based on the non-model Medicare FFS payment amount (that is, 
the amount that MA plans pay to these providers will not be the MFN 
Model payment amounts).
    As discussed in section VI. of this IFC, we expect the MFN Model 
will lower overall Medicare FFS expenditures; that is, Medicare Part B 
MFN Drug Payment Amounts will be lower than such payment would be 
absent the model, the Medicare Part B alternative add-on payments will 
be greater than such payment would be absent the model, there could be 
increases in Medicare Part A spending, and taken together the model 
will result in an overall reduction in Medicare expenditures. The 
overall decrease in Medicare FFS expenditures will be considered in 
determining the historical FFS claims experience for calculating the 
rates for plan service areas. Payments to Medicare Advantage 
Organization plans are anticipated to be lower than they would be 
absent the model. At a high level, the FFS

[[Page 76230]]

component of the non-ESRD MA rates is based on the product of the 
projected national per-capita spending and a county-level relative cost 
index. Thus, the MA ratebook calculations will reflect changes in 
actual FFS spending due to the impact of the MFN Model. We note that 
this approach is consistent with treatment of payments made under other 
CMS Innovation Center models and the Medicare Shared Savings Program.
    As discussed in more detail in section VI. of this IFC, we estimate 
that total payments to MA plans over the 7-year course of the model 
will be substantially lower as a result of reduced FFS spending under 
the MFN Model, that is, total payments to MA plans may be approximately 
$49.6 billion lower in the OACT estimate and $28.5 billion lower in the 
ASPE estimate. We note that there is much uncertainty around the 
assumptions for these estimates.

L. Exclusion of Certain MFN Model Sales From Manufacturers' Calculation 
of ASP for MFN Model Drugs

    In accordance with sections 1847A and 1927(b)(3)(A)(iii) of the 
Act, manufacturers 74 75 submit ASP data for their products 
to CMS on a quarterly basis. The manufacturer's ASP is based on sales 
to all purchasers in the U.S. with limited exceptions (that is, 
exclusions are limited to sales exempt from best price (as defined in 
section 1927(c)(1)(C)(i) of the Act), sales at a nominal charge, and 
units sold to a CAP vendor), and is net of discounts such as volume 
discounts, prompt pay discounts, cash discounts, free goods that are 
contingent on any purchase requirement, chargebacks, and rebates (other 
than certain rebates specified in section 1927 of the Act). Specific 
ASP reporting requirements are set forth in section 1927(b)(3) of the 
Act. In accordance with sections 1847A and 1927(b)(3) of the Act, 
manufacturers report most ASP data by National Drug Code (NDC), which 
identifies products in terms of the labeler, product, and package size 
and type. The reported ASP data are used to establish the Medicare 
payment amounts. In general, Medicare's payment limit for most 
separately payable Part B drugs is based on the methodology in section 
1847A of the Act, that is, 106 percent of the volume-weighted average 
of manufacturers' ASP for a drug (at the billing and payment code 
level), and is updated quarterly. The payment requirements in section 
1847A of the Act will be waived for purposes of testing the MFN Model 
as discussed in section III.M.1. of this IFC, but will continue to 
apply outside of the model as discussed in this section.
---------------------------------------------------------------------------

    \74\ For the purposes of reporting under section 1847A of the 
Act, the term ``manufacturer'' is defined in section 1927(k)(5) of 
the Act and means any entity engaged in the production, preparation, 
propagation, compounding, conversion or processing of prescription 
drug products; either directly or indirectly by extraction from 
substances of natural origin, or independently by means of chemical 
synthesis, or by a combination of extraction and chemical synthesis; 
or in the packaging, repackaging, labeling, relabeling, or 
distribution of prescription drug products. The term manufacturer 
does not include a wholesale distributor of drugs or a retail 
pharmacy licensed under State law. However, manufacturers that also 
engage in certain wholesaler activities are required to report ASP 
data for those drugs that they manufacture. Note that the definition 
of manufacturers for the purposes of ASP data reporting includes 
repackagers.
    \75\ Manufacturer is also defined in 42 CFR 447.502.
---------------------------------------------------------------------------

    Section 1115A of the Act authorizes the CMS Innovation Center to 
test innovative payment and service delivery models to reduce program 
expenditures, while preserving or enhancing the quality of care 
furnished to beneficiaries. The MFN Model will test an alternative 
approach for determining Medicare's payment limit for MFN Model drugs, 
which will phase down the Medicare payment amount for selected Part B 
drugs to more closely align with available international prices, and 
test an alternative add-on payment. Under the MFN Model, the model's 
payment test will apply when Medicare makes separate payment for an MFN 
Model drug that was furnished on an outpatient basis by an MFN 
participant to an MFN beneficiary within the model's nationwide 
geographic area.
    In designing the MFN Model, we considered ways to mitigate 
potential impacts on manufacturers' ASPs stemming from price 
concessions given to MFN participants for purchases related to the MFN 
Model and on Medicare payment for units of MFN Model drugs that are not 
subject to the MFN Model payment test. For example, sales to MFN 
participants may include larger price concessions than are typical 
today, resulting in lower net sales prices as compared to what net 
sales prices would be absent the MFN Model. As such, the manufacturer's 
ASP for an MFN Model drug, which will reflect the average price for all 
non-excluded sales--including sales to MFN participants to the extent 
applicable--may be lower than the manufacturer's ASP would be absent 
the MFN Model. Because CMS will base the non-model Medicare payment 
limit for an MFN Model drug on 106 percent of the manufacturer's ASP, 
payment to providers and suppliers for such drug outside of the model 
may be lower than it otherwise would be absent the MFN Model. To 
conduct the MFN Model test it is necessary to minimize this potential 
spillover effect for providers and supplier that are not MFN 
participants to best observe the impacts of the payment change. Thus, 
we will exclude from the calculation of the manufacturer's ASP any 
units of MFN Model drugs billed by MFN participants where the MFN Drug 
Payment Amount is based on available international drug pricing 
information and Medicare Part B is the primary payer. policy will only 
apply when the MFN Price is based on available international drug 
pricing information. That is, the policy will not apply when there is 
no available international drug pricing information and the MFN Price 
is equal to the applicable ASP because there will be no concern for 
spillover impacts in such cases. We are waiving requirements of section 
1847A of the Act as necessary to exclude such units of MFN Model drugs 
from the calculation of the manufacturer's ASP. We will also indicate 
the MFN Drug Payment Amounts that are (and are not, when applicable) 
based on available international drug pricing information within the 
quarterly MFN Model drug pricing files posted on a CMS website.
    This approach is responsive to comments we received in response to 
the October 2018 ANPRM. Several commenters requested clarification 
about how sales for purposes of the model would be taken into account 
in computing the ASP under section 1847A of the Act. Some commenters 
who expressed concern about potential spillover effects of the 
potential model payment test recommended that purchases made for use 
under the potential model be excluded from the ASP calculation. Based 
on our interactions with stakeholders, particularly those with 
experience operating chargebacks related to the 340B program, we 
believe our exclusion of units of MFN Model drugs that are billed by 
MFN participants and have the MFN Drug Payment Amount paid by Medicare 
from manufacturers' ASPs will be feasible. Manufacturers have existing 
processes and tools to exclude various prices from the calculation of 
their ASPs, and excluding certain MFN Model related units of MFN Model 
drugs could be similar.
    Distribution management systems are employed throughout the drug 
distribution system to order drugs, track sales and shipments, trace 
custody, manage price and customer lists, record financial 
transactions, and support other industry processes. Separate purchasing 
accounts are often used to align with purchasing arrangement terms, and

[[Page 76231]]

through a process called the ``chargeback process,'' manufacturers 
reduce the final drug prices to wholesalers and other distributors to 
reflect the purchasing terms and contract prices that apply to the end 
purchaser. End purchasers of drugs who purchase under more than one 
contract use virtual inventory or replenishment purchasing tools or 
business processes to manage their purchases under their various 
contract arrangements. For example, a provider or supplier that belongs 
to more than one group purchasing organization could use such tools or 
business processes to track drug purchasing, maintain records toward 
volume targets and, should the need to return a product occur, conduct 
returns. However, based on stakeholder feedback, we understand that all 
MFN participants are unlikely to have such tools in place. Hospitals, 
particularly those that participate in the 340B program, are more 
likely to currently have these tools compared to other hospitals, 
physician offices and ASCs. Thus, manufacturers may establish 
mechanisms to obtain information from MFN participants about the number 
of units of MFN Model drugs that were furnished to MFN beneficiaries 
and for which payment under Sec.  513.210 was allowed, which would 
increase MFN participants' activities related to the model.
    CMS also seeks to minimize the potential for excessive increases in 
non-model Medicare drug payment amounts during the MFN Model. For 
example, during the model, manufacturers' ASPs may increase causing a 
concomitant increase in non-model Medicare drug payment amounts outside 
of the model if: (1) The policy that manufacturers not include units of 
an MFN Model drug billed by MFN participants where the MFN Drug Payment 
Amount is paid by Medicare and Medicare Part B is the primary payer in 
the manufacturer's ASP for the MFN Model drug results in higher ASPs; 
or (2) manufacturers raise drug prices or lower existing discounts for 
U.S. sales that are not subject to the model's payment test. Because 
manufacturers will continue to have the ability to set their own drug 
prices, as a behavioral response to the MFN Model, manufacturers could 
raise prices for MFN Model drugs in the United States in part to make 
up for price concessions that may be given to model participants.
    We believe the policy for manufacturers not to include in the 
manufacturer's ASP units of an MFN Model drug administered to an MFN 
beneficiary and billed by MFN participants where the MFN Drug Payment 
Amount applied by Medicare is based on available international drug 
pricing information and Medicare is the primary payer will minimize the 
potential for manufacturers to choose to increase purchase prices for 
non-model participants and for MFN participants' purchases of MFN Model 
drugs for use outside of the MFN Model. Additionally, we believe that 
the adjustments to the MFN Price phase-in, as described in section 
III.E. of this IFC, will also minimize the potential for manufacturers 
to increase prices for non-model participants and non-model purchases. 
We also believe this policy is necessary for a rigorous test of the 
model payment for MFN drugs because price concessions tied to the model 
will not lower Medicare payment when MFN Model drugs are purchased for 
use outside the model, which would limit our ability to observe the 
impacts of the payment change.
    We will not collect the number of units that manufacturers exclude 
from ASP as part of their ASP submission to CMS to avoid establishing a 
new data collection effort and to minimize administrative burden for 
manufacturers.
    As an alternative approach, we considered whether manufacturers 
should exclude from the manufacturer's ASP for the MFN Model drug price 
concessions on units of an MFN Model drug billed by MFN participants 
where the MFN Drug Payment Amount applied by Medicare is based on 
available international drug pricing information and Medicare is the 
primary payer. We believe that excluding from the manufacturer's ASP 
price concessions on units of an MFN Model drug billed by MFN 
participants where the MFN Drug Payment Amount applied by Medicare is 
based on available international drug pricing information and Medicare 
is the primary payer, and not excluding the manufacturer's ASP the 
units of an MFN Model drug billed by MFN participants where the MFN 
Drug Payment Amount is applied by Medicare is based on available 
international drug pricing information and Medicare is the primary 
payer would inappropriately raise the ASP. We believe this is the case 
because those units would likely be factored into the manufacturer's 
ASP calculation as undiscounted sales. Thus, this approach, while it 
may be less complex, would likely lead to inappropriately higher 
Medicare payment outside of the model.
    We are waiving requirements in section 1847A(c) to the extent 
necessary to exclude from the calculation of the manufacturer's ASP any 
units of an MFN Model drug administered to an MFN beneficiary and 
billed by MFN participants where the MFN Drug Payment Amount applied by 
Medicare is based on available international drug pricing information 
and Medicare is the primary payer. Consistent with section 1847A(c)(5) 
of the Act, we will issue program instructions to further describe how 
the waiver will impact manufacturers' calculation of the manufacturer's 
ASP. For example, we envision that manufacturers will take reasonable 
steps and make reasonable assumptions to exclude applicable units. We 
note that all other existing statutory requirements and regulations 
will continue to apply. For example, manufacturers who misrepresent or 
fail to report manufacturer ASP data will remain subject to civil 
monetary penalties, as applicable and described in sections 1847A and 
1927(b) of the Act and codified in regulations at Sec.  414.806.

M. Program Waivers and Model Termination

1. Waivers of Medicare Program Requirements for Purposes of Testing the 
Model
    We will test the MFN Model under the authority of section 1115A of 
the Act and waive certain Medicare program requirements as necessary 
solely for purposes of testing the model. Under section 1115A(d)(1) of 
the Act, the Secretary may waive the requirements of Titles XI and 
XVIII and of sections 1902(a)(1), 1902(a)(13), 1903(m)(2)(A)(iii), and 
1934 of the Act (other than subsections (b)(1)(A) and (c)(5) of such 
section) as may be necessary solely for purposes of carrying out 
section 1115A of the Act with respect to testing models described in 
section 1115A(b) of the Act. The purpose of these waivers will be to 
allow Medicare to test the MFN Model described in this IFC, with the 
goal of reducing Medicare expenditures while improving or maintaining 
the quality of beneficiaries' care.
    In Sec.  513.500, we waive program requirements that are necessary 
solely for purposes of testing the MFN Model--
     Sections 1833(t)(6) and 1833(t)(14) of the Act and 42 CFR 
419.62 and 419.64 related to Medicare payment amounts for drugs and 
biologicals under the OPPS as necessary to permit testing of an 
adjusted payment amount for MFN Model drugs using the pricing 
approaches described in this IFC;
     Section 1833(i)(2)(D) of the Act related to Medicare 
payment to ASCs for drugs and biologicals as necessary to permit 
testing of an adjusted payment

[[Page 76232]]

amount for MFN Model drugs using the pricing approaches described in 
this IFC;
     Sections 1847A(b) and 1847A(c) of the Act and 42 CFR 
414.904 and 414.802 related to use of the ASP-based, WAC-based, or 
other applicable payment methodology and calculation of manufacturers' 
ASP as necessary to permit testing of an adjusted payment for MFN Model 
drugs and to exclude certain units of MFN Model drugs from 
manufacturers' ASPs;
     Section 1833(a)(1) of the Act related to Medicare payment 
portion of the allowed payment amount for an included MFN Model drug 
that is determined under Sec.  513.220 as necessary to permit testing 
of an innovative payment approach for the alternative add-on payment 
amount;
     Section 1833(a)(1)(S) related to Medicare payment for 
drugs and biologicals at 80 percent of the lesser of actual charge or 
the amount established in section 1842(o) of the Act as necessary to 
allow CMS to not apply beneficiary cost-sharing to the alternative add-
on payment amount;
     Section 1833(a)(1)(G) of the Act related to the amounts 
paid with respect to facility services furnished in connection with 
certain surgical procedures and with respect to services furnished to 
an individual in an ASC shall be 80 percent of the lesser of the actual 
charge for the services or the amount determined by the Secretary under 
such revised payment system as necessary to allow CMS to not apply 
beneficiary cost-sharing to the alternative add-on payment amount;
     Section 1833(t) of the Act related to how Medicare payment 
under the OPPS is calculated including beneficiary copayment to allow 
CMS to not apply beneficiary cost-sharing to the alternative add-on 
payment amount; and
     Section 1833(t)(9)(B) of the Act related to the 
requirement that Medicare account for adjustments to ensure that the 
amount of expenditures under the OPPS for the year does not increase or 
decrease from the estimated amount of expenditures under the OPPS that 
would have been made if the adjustments had not been made (that is, 
OPPS budget neutrality). CMS intends to continue to maintain budget 
neutrality under the OPPS as it currently does, including as described 
in 42 CFR 419.32(d)(1). This includes continuing to use the applicable 
payment amount for each separately payable drug under that payment 
system, rather than the MFN Drug Payment Amount and alternative add-on 
payment amount. CMS may consider using volume for drugs included in the 
MFN Model for purposes of the budget neutrality calculations under the 
OPPS beginning in 2022, but would utilize the applicable OPPS payment 
amount for the drug or biological, rather than the MFN Drug Payment 
Amount. We believe a waiver of the OPPS budget neutrality requirements 
for Part B drugs furnished under the MFN Model is necessary solely for 
purposes of testing the MFN Model because if reductions in Medicare 
Part B drug expenditures were redistributed through the OPPS budget 
neutrality process to non-drug Part B services under the OPPS, the 
model would change pricing for numerous other services that are not 
related to Part B drugs. This would make it difficult to determine the 
independent impact of a change in Part B drug payment levels to MFN 
Model pricing if there is also a corresponding change in the payment 
amount for all non-drug hospital outpatient items and services as a 
result of the OPPS budget neutrality requirements.
    Our intent is to include a waiver for all program requirements in 
title XVIII of the Act as may be necessary solely to test separate 
payment for MFN Model drugs furnished to MFN beneficiaries by MFN 
participants. To the extent that MFN participants receive separate 
payment for MFN Model drugs under program requirements that we have not 
listed in Sec.  513.500, we waive such requirements as necessary to 
effectuate part 513.
2. Model Termination
    CMS may terminate the MFN Model for reasons including, but not 
limited to, the following: CMS determines that it no longer has the 
funds to support the model; or CMS terminates the model in accordance 
with section 1115A(b)(3)(B) of the Act. As provided by section 
1115A(d)(2) of the Act, termination of the model under section 
1115A(b)(3)(B) of the Act is not subject to administrative or judicial 
review. We are codifying these policies in Sec.  513.1000.

N. Evaluation

    We will conduct an evaluation of the MFN Model, as required under 
section 1115A(b)(4) of the Act. The evaluation of the MFN Model will 
include an analysis of the quality of care furnished under the model 
and the changes in spending under Medicare by reason of the model.
    There will be several populations of interest for the MFN Model 
evaluation. A population of interest for the evaluation will be 
Medicare beneficiaries who are likely to receive one of the MFN Model 
drugs based on recent diagnoses and/or prior treatment. One possible 
prescriber behavior change due to the MFN Model could be shifts from 
prescribing MFN Model drugs to other alternative Part B or Part D drugs 
or vice versa. A population defined by recent diagnoses and/or prior 
treatment will capture the model's impact on beneficiaries affected by 
these prescribing behavioral changes due to the model. Other 
populations such as, but not limited to, MFN Model drug users and 
subgroups of particular patient populations (for example, cancer, 
rheumatoid arthritis, ophthalmologic conditions) will be considered in 
the evaluation.
    For each of the populations of interest, we will create separate 
impact estimates for two types of outcomes: Medicare spending and drug/
other health care utilization. Medicare spending will be examined in 
terms of total Part B drug spending for MFN Model drugs, total Part B 
drug spending for any Part B drugs, total Parts A and B spending, and 
potentially other spending measures for specific types of health care 
services (for example, inpatient hospital spending). The evaluation of 
the model's impact on quality of care will examine drug access, 
measured by utilization (for example, rates of any use and duration of 
use) of both Part B (both MFN Model drug and non-MFN Model drugs) and 
Part D drugs. We will also examine non-drug health care utilization 
that may change as a result of the MFN Model to estimate any impacts on 
access to care. Examples of other non-drug health care utilization 
include hospitalizations, emergency department visits, and condition-
specific utilization related to a given subgroup of beneficiaries. The 
impact estimates will reflect the collective effect of the MFN Model's 
changes to Medicare payments and beneficiary cost-sharing for MFN Model 
drugs.
    Because the MFN Model will be a nationwide, mandatory model, we 
must employ an evaluation design that does not require an independent 
comparison group to establish the counterfactual (what would have 
happened in the absence of the model). The term ``interrupted time 
series'' (ITS) refers to the situation in which multiple observations 
for the treatment group are available both before and after the 
intervention is implemented.\76\ ITS models can be employed both with 
and

[[Page 76233]]

without comparison groups, and be used to imply causality without 
comparison groups.\77\ The design is used when data are available both 
for the pre-intervention period and the post-intervention period, and 
the intervention takes place at a specific, identifiable point in 
time.\78\ The time-relationship between the data points can then be 
used to estimate treatment effects. The trends from the pre-
intervention period establish a baseline that is used to project what 
would be expected in the absence of the intervention. The typical ITS 
approach assumes linear trends before and after the intervention, but 
ITS models can be made more general to address potential non-linear 
trends.79 80 Intervention effects are demonstrated when 
observations gathered after the intervention start period deviate from 
the baseline projections.
---------------------------------------------------------------------------

    \76\ Wagner, A.K., Soumerai, S.B., Zhang, F. and Ross-Degnan, D. 
(2002), Segmented regression analysis of interrupted time series 
studies in medication use research. Journal of Clinical Pharmacy and 
Therapeutics, 27: 299-309. doi: 10.1046/j.1365-2710.2002.00430.x.
    \77\ Sheingold, S., Bir, A, (2019), Evaluation for Health Policy 
and Health Care: A Contemporary Data Driven Approach, Sage 
Publications.
    \78\ Bernal, J.L., Cummins, S., Gasparrin, A., (2017) 
Interrupted time series regression for the evaluation of public 
health interventions: A tutorial, International Journal of 
Epidemiology, 2017, Vol. 46, No. 1: 348-355, doi: 10.1093/ije/
dyw098.
    \79\ Kontopantelis, E., Doran, T., Springate, D.A., Buchan, I 
Reeves, D, (2015), Regression based quasi-experimental approach when 
randomisation is not an option: Interrupted time series analysis, 
BMJ 2015;350:h2750, doi: 10.1136/bmj.h2750.
    \80\ Valsamis, E.M., Ricketts, D., Husband, H., and Rogers, B.A. 
(2019), Segmented linear regression models for assessing change in 
retrospective studies in healthcare. Computational and Mathematical 
Methods in Medicine. doi: 10.1155/2019/9810675.
---------------------------------------------------------------------------

    Using this design for evaluating the effects of an intervention--
that is, implying a causal relationship between the intervention and 
its target outcomes--relies on a strict set of conditions. As 
previously described, when there is no comparison group, the 
counterfactual is established as the continuation of the pre-
intervention trend for the treated group. The intervention impact is 
estimated as the difference between the actual post-intervention trend 
and the pre-intervention trend extended.
    The most common statistical method for analyzing ITS data is called 
segmented regression.81 82 Segmented regression focuses on 
two parameters, the level (intercept) and the trend (slope). For 
observations before the model, we will have a level (intercept) and 
trend (slope). After the model begins, the data may exhibit changes in 
any one of these features. The fundamental idea behind segmented 
regression is to estimate a regression specification with a linear 
trend for the data points before the model and estimate a regression 
specification with a linear trend for the data points after the model 
start. The level and trend before and after the model start will then 
be compared. We will use quarterly observations for the pre- and post- 
model start time periods ending with the most recent data that will be 
currently available. Given the MFN Model design, we provide our 
specification in this section of this IFC for the longitudinal 
regression using a more general specification of the trends to capture 
the non-linear nature of the data.
---------------------------------------------------------------------------

    \81\ Wagner, A.K., Soumerai, S.B., Zhang, F. and Ross-Degnan, D. 
(2002), Segmented regression analysis of interrupted time series 
studies in medication use research. Journal of Clinical Pharmacy and 
Therapeutics, 27: 299-309. doi: 10.1046/j.1365-2710.2002.00430.x.
    \82\ Valsamis, E.M., Ricketts, D., Husband, H., and Rogers, B.A. 
(2019), Segmented linear regression models for assessing change in 
retrospective studies in healthcare. Computational and Mathematical 
Methods in Medicine. doi: 10.1155/2019/9810675.
---------------------------------------------------------------------------

    In the longitudinal regression equation provided in this section of 
this IFC, the vector Xit consists of factors that will 
change from the pre-model time period to the model performance period 
and may include, but is not necessarily limited to, the medical care 
component of the Consumer Price Index (CPI-U), national unrelated 
policy changes, economic factors (for example, unemployment rate). The 
unit of analysis (for example, a hospital referral region (HRR) as 
defined by the Dartmouth Atlas \83\ or beneficiary) on which the 
quarterly observations are measured will be allowed to vary in order to 
estimate the model's impact at these different levels of aggregation. 
The anticipated statistical model specification includes a polynomial 
time trend variable f(t) to account for trends in spending and 
utilization over time. In addition, the statistical model includes 
separate indicator variables (It=k) for each of the model 
performance period quarters, which will allow for estimates of the 
model's impact in each performance period quarter relative to the 
entire pre-period after adjusting for the time trend and other factors.
---------------------------------------------------------------------------

    \83\ https://www.dartmouthatlas.org/faq/#research-methods-faq.

Yit = b0 + b1 [middot] Xit 
+ b2 [middot] f(t) + a1 [middot] It=1 
+ a2 [middot] It=2 + a3 [middot] 
It=3 + a4 [middot] It=4 + . . . + 
---------------------------------------------------------------------------
uit

Where:

Yit = outcome (see the previous section for cost and 
utilization measures), for a particular unit of analysis in a 
specific quarter
Xit = vector of adjustment factors
f(t) = polynomial function to account for time trend
It=k = denotes an indicator for time period k (all after 
model implementation)
uit = unaccounted variation
i = unit of analysis (for example, beneficiary, HRR)
t = time quarter (-12, -11 . . . 0, 1, 2, 3 . . .) using a 3 year 
pre-model time period, with 0 indicating the start of the model
b0, b1, b2, a1 . . . . 
an are the statistical model coefficients
b0 = the statistical model intercept
b1 = vector of estimates for the adjustment factors
b2 = estimate of the time trend f(t) across the pre-
period and model performance period
a1 thru an = estimate of change per model 
performance period quarter (t) relative to entire pre-period

    With the statistical model specification as previously described, 
in an initial, exploratory data assessment, the null hypothesis 
(Ho: a1 = a2 = a3 = 
a4 = . . . = an = 0) will be that there is no 
change in each of the model performance period quarters when compared 
to the pre-period after adjusting for the time trend and the other 
factors. The corresponding alternate hypothesis (Ha: 
a1 or a2 or a3 or a4 or . . 
. an [ne] 0) will be that any of the model performance 
period quarters is statistically significantly different than the pre-
model time period, suggesting that the model either positively or 
negatively impacted Medicare spending and quality of care in at least 
one model performance period quarter. These null and alternate 
hypotheses will apply to each outcome and population of interest.
    The assessment just described will not directly indicate success or 
failure of the model. CMS will need to observe a consistent 
statistically significant directional pattern over multiple consecutive 
time periods for the outcome and population of interest in order to 
draw sound conclusions about the model's impact. Based on a combination 
of results from exploratory data assessment and policy goals, CMS will 
set a hypothesis that encompasses the chosen outcome and population of 
interest. This hypothesis will be tested using data that is different 
from what was used in the exploratory assessment--for instance, due to 
being gathered later in time or consisting of a different randomly 
assigned subset of contemporaneous data.
    Statistical inference will be conducted using cluster-robust 
standard errors.\84\ Cluster-robust standard errors account for serial 
correlation as well as spatial correlation within geographies (such as 
an HRR). We will conduct hypothesis testing using an alpha-level of 5 
percent

[[Page 76234]]

and CMS will report the p-value and standard error to allow for 
inferences at other alpha-levels.
---------------------------------------------------------------------------

    \84\ Cameron, A.C., & Miller, D.L. (2015). A practitioner's 
guide to cluster-robust inference. Journal of Human Resources, 
50(2), 317-372.
---------------------------------------------------------------------------

    As an illustration of a potential subgroup analysis and the 
expected changes that could be detected in the MFN Model evaluation, 
CMS identified two groups of Medicare cancer patients using 2018 data. 
CMS defined the first narrower group as Medicare cancer patients who 
received an MFN Model drug. CMS defined the second broader group as 
Medicare cancer patients who either received an MFN Model drug or would 
have been considered eligible to receive an MFN Model drug. 
Specifically, CMS estimated that in 2018 approximately 400,000 Medicare 
beneficiaries were being treated for the most prevalent cancer types 
(that is, colorectal, endometrial, breast, lung, prostate, and certain 
forms of leukemia and lymphoma) and received an MFN Model drug. These 
400,000 Medicare beneficiaries were identified using the inclusion and 
exclusion criteria for the model, including the use of an MFN Model 
drug. Cancer treatment was determined by the utilization of Part B and/
or Part D cancer drugs and the presence of cancer diagnosis codes on 
Parts A and B claims. A subgroup analysis that requires MFN Model drug 
use, as in the narrower definition that identified 400,000 Medicare 
beneficiaries being treated for cancer and who received an MFN Model 
drug, would exclude cancer patients using an alternative non-MFN Model 
drug cancer therapy. A broader cancer population definition based on 
any Part B and/or Part D cancer drug use or just an incident cancer 
diagnosis based on new evidence of diagnosis codes on Parts A and B 
claims in the current year would capture the model's impact on 
beneficiaries affected by prescribing behavioral changes due to the 
model. This second broader cancer subgroup population definition 
applied to approximately 1.1 million Medicare beneficiaries in 2018.
    CMS believes that looking for unintended consequences will be 
critical for the monitoring and evaluation of the MFN Model. In the 
narrower definition of the cancer subgroup, CMS expects that 
approximately 100,000 Medicare cancer patients who receive a MFN Model 
drug will be eligible for inclusion in the quarterly evaluation 
analysis. In the broader cancer subgroup population, CMS expects that 
approximately 280,000 Medicare cancer patients will be included in the 
quarterly evaluation analysis. With a nationwide MFN Model (and the 
assumptions of an alpha-level of 5 percent and power of 80 percent), 
CMS will have the sample sizes needed in these two populations to 
detect small changes in Medicare total cost of care (approximately a 1 
percent change), drug access, and other important measures of quality 
of care. With multiple quarterly assessments of the impact of the model 
on subgroup populations, CMS will be able to intervene early in the 
model's performance period should any potential unintended consequences 
be detected in the potential subgroups of interest. Although CMS uses 
the cancer subgroup patient population in the previously discussed 
example, we recognize that other patient populations (for example, 
patients diagnosed with rheumatoid arthritis and wet macular 
degeneration) and certain types of providers could be differentially 
impacted by the MFN Model. These other patient and provider subgroups 
will be of interest in the evaluation. The model's impact on the 
Medicaid program and commercial insurance (including Medicare 
Advantage) population is also of interest.
    The evaluation will explore the experiences of MFN participants 
(beneficiaries and providers) and other stakeholders affected by the 
changes in payment and conditions included in the model. In particular, 
CMS will interview MFN participants and beneficiaries, either by focus 
groups, surveys, or one-on-one stakeholder interviews, to assess the 
model's influence on access to and quality of care, and administrative 
burden from their perspectives. Further, CMS intends to ask 
beneficiaries about their total out of pocket costs under the MFN Model 
to determine if those costs were reduced. MFN participants will be 
asked for their opinions about the MFN Model's payment changes to the 
drug and add-on payment amounts separately. The evaluation will also 
include qualitative analyses of primary data collected from MFN 
participants and beneficiaries. The results of the qualitative analyses 
will be used to provide additional context for the results of the 
quantitative analyses on health care spending and to help further 
explain the observed changes.
    Evaluation reports detailing the results and findings will be 
developed and publicly posted on the CMS website. The evaluation 
reports will include the results of the quantitative and qualitative 
analyses of the MFN Model's impact on spending and quality of care and 
the model's implementation as described in this section. The evaluation 
reports covering the earlier performance years of the MFN Model will be 
used in the decision making process on whether or not to continue the 
MFN Model into performance years 5 to 7.
    The evaluation may require that MFN participants collect and submit 
additional data specifically for the evaluation (please see Sec.  
513.100(e) and Sec.  513.100(f)). Such requirements for additional data 
to carry out model evaluation will be in compliance with 42 CFR 
403.1110(b), which requires entities participating in the testing of a 
model under section 1115A to collect and report such information, 
including protected health information (as defined at 45 CFR 160.103), 
as the Secretary determines is necessary to monitor and evaluate the 
model.

O. Limitations on Review

    In Sec.  513.450, we are codifying the preclusion of administrative 
and judicial review under section 1115A(d)(2) of the Act. Section 
1115A(d)(2) of the Act states that there is no administrative or 
judicial review under section 1869 or 1878 of the Act or otherwise for 
the all of the following:
     The selection of models for testing or expansion under 
section 1115A of the Act.
     The selection of organizations, sites, or participants to 
test models selected.
     The elements, parameters, scope, and duration of such 
models for testing or dissemination.
     Determinations regarding budget neutrality under section 
1115A(b)(3) of the Act.
     The termination or modification of the design and 
implementation of a model under section 1115A(b)(3)(B) of the Act.
     Determinations about expansion of the duration and scope 
of a model under section 1115A(c) of the Act, including the 
determination that a model is not expected to meet criteria described 
in paragraph (1) or (2) of such section.
    We interpret the preclusion from administrative and judicial review 
regarding the CMS Innovation Center's selection of organizations, 
sites, or participants to test models selected to preclude from 
administrative and judicial review CMS' selection of an MFN 
participant, as well as CMS' decision to terminate an MFN participant, 
as these determinations are part of CMS' selection of participants for 
CMS Innovation Center model tests.
    We interpret the preclusion from administration and judicial review 
regarding the elements, parameters, scope, and duration of models for 
testing or dissemination to preclude from administrative and judicial 
review the following CMS determinations made in connection with the MFN 
Model:

[[Page 76235]]

     The selection of the model geographic area for the MFN 
Model by CMS;
     The selection of MFN Model drugs by CMS; and
     The selection of included international data, including 
selection of countries, international drug pricing databases, and 
international drug pricing information.
    In addition, we interpret the preclusion from administrative and 
judicial review regarding the elements of the MFN Model to preclude 
from administrative and judicial review the methodology for determining 
MFN Prices, MFN Drug Payment Amounts, Alternative Add-on Amounts, and 
reconciliation payments related to financial hardship exemptions.

V. Collection of Information Requirements

    As stated in section 1115A(d)(3) of the Act, Chapter 35 of title 
44, United States Code, shall not apply to the testing and evaluation 
of CMS Innovation Center Models. As a result, the information 
collection requirements contained in this IFC need not be reviewed by 
the Office of Management and Budget. However, costs incurred through 
information collections are included in section VI.C.5. of this IFC.

V. Response to Comments

    Because of the large number of public comments we normally receive 
on documents, we are not able to acknowledge or respond to them 
individually. We will consider all comments we receive by the date and 
time specified in the DATES section of this preamble, and, when we 
proceed with a subsequent document, we will respond to the comments in 
the preamble to that document.

VI. Regulatory Impact Analysis

A. Statement of Need

    This IFC is necessary to address the current Medicare Part B 
payment system for separately payable Medicare Part B drugs, which has 
several features that may be incentivizing avoidable costs and causing 
greater utilization of higher priced drugs. By testing ways to address 
these payment issues, the MFN Model seeks to improve quality of care, 
address features of the current payment system that may be 
incentivizing unnecessary Medicare Part B drug spending and utilization 
of high cost drugs, and ensure that the Medicare program and its 
beneficiaries pay generally comparable prices for Medicare Part B drugs 
relative to certain other countries.
    As detailed in section III of this IFC, this IFC will establish a 
7-year nationwide MFN Model alternative payment test for approximately 
50 separately payable Medicare Part B drugs furnished by certain 
providers and suppliers. As discussed in section III.C. of this IFC, 
MFN participants will include Medicare-participating providers and 
suppliers that furnish MFN Model drugs, with certain exclusions. Most 
of the MFN participants will be: Physicians; non-physician 
practitioners; supplier groups; HOPDs (including on- and off-campus 
outpatient provider-based departments, but excluding cancer hospitals, 
children's hospitals, CAHs, and other hospitals exempt from the OPPS); 
and ASCs. When other providers and suppliers that are not excluded bill 
for separately payable MFN Model drugs (for example, pharmacies and 
independent diagnostic testing facilities), they will be included in 
the MFN Model as MFN participants; based on 2018 Medicare Part B claims 
data, their aggregate annual volume of separately payable Part B drugs 
was less than $3.6 million. MFN participants will be subject to the 
participation requirements described in section III. of this IFC.

B. Overall Impact

    We have examined the impacts of this IFC, as required by Executive 
Order 12866 on Regulatory Planning and Review (September 30, 1993), 
Executive Order 13563 on Improving Regulation and Regulatory Review 
(February 2, 2013), the Regulatory Flexibility Act (RFA) (September 19, 
1980, Pub. L. 96-354), section 1102(b) of the Act, section 202 of the 
Unfunded Mandates Reform Act of 1995 (UMRA) (March 22, 1995, Pub. L. 
104-4), Executive Order 13132 on Federalism (August 4, 1999), and the 
Congressional Review Act (5 U.S.C. 804(2)), and Executive Order 13771 
on Reducing Regulation and Controlling Regulatory Costs (January 30, 
2017).
    Executive Orders 12866 and 13563 direct agencies to assess all 
costs and benefits of available regulatory alternatives and, if 
regulation is necessary, to select regulatory approaches that maximize 
net benefits (including potential economic, environmental, public 
health and safety effects, distributive impacts, and equity). Section 
3(f) of Executive Order 12866 defines a ``significant regulatory 
action'' as an action that is likely to result in a rule: (1) Having an 
annual effect on the economy of $100 million or more in any 1 year, or 
adversely and materially affecting a sector of the economy, 
productivity, competition, jobs, the environment, public health or 
safety, or state, local or tribal governments or communities (also 
referred to as ``economically significant''); (2) creating a serious 
inconsistency or otherwise interfering with an action taken or planned 
by another agency; (3) materially altering the budgetary impacts of 
entitlement grants, user fees, or loan programs or the rights and 
obligations of recipients thereof; or (4) raising novel legal or policy 
issues arising out of legal mandates, the President's priorities, or 
the principles set forth in the Executive Order. This IFC triggers 
these criteria.
    A regulatory impact analysis (RIA) must be prepared for major rules 
with economically significant effects ($100 million or more in any 1 
year). We estimate that this rulemaking is ``economically significant'' 
as measured by the $100 million threshold and hence also a major rule 
under the Congressional Review Act. Accordingly, we have prepared a RIA 
that, to the best of our ability, reflects the economic impact of the 
policies contained in this IFC.

C. Detailed Economic Analysis

    The MFN Model will test different payment rates for certain 
separately payable Medicare Part B drugs and their associated drug add-
on payment. The payment rates for these Medicare Part B drugs will be 
phased in over 4 years, ultimately arriving at the lowest price for a 
particular drug from a selected group of countries. Eligible providers 
and suppliers participating in the 340B program will be paid the lesser 
of this amount or the payment outside the model for MFN Model drugs 
they purchase under the 340B program. This IFC includes a single 
alternative add-on payment, with MFN participants receiving an amount 
that represents 6 percent (after sequestration) of the average sales 
price (ASP) baseline for the initial set of included drugs trended 
forward. The phased-in MFN Price discount relative to applicable ASP is 
shown in Table 9, assuming the relationship remains constant.

[[Page 76236]]



                         Table 9--Most Favored Nation Discount From ASP by Calendar Year
----------------------------------------------------------------------------------------------------------------
        Calendar year            2021        2022        2023        2024        2025        2026        2027
----------------------------------------------------------------------------------------------------------------
MFN Price impact............       -16%        -33%        -49%        -65%        -65%        -65%        -65%
----------------------------------------------------------------------------------------------------------------

    The model will require participation by eligible providers and 
suppliers for the selected separately payable Medicare Part B drugs 
included in the model. Certain provider types, defined previously in 
this IFC, will be excluded from the model. We assume that acute care 
hospitals that are paid for outpatient hospital services on a fully 
capitated or global budget basis under a waiver under such model of 
section 1833(t) of the Act will be excluded from the MFN Model.
    Because current payment rates for 340B covered entities that are 
paid under the OPPS (hereafter called 340B providers) are different 
from those for other providers and suppliers (hereafter called non-340B 
providers), the impact of the MFN Model varies between the two provider 
types, and therefore OACT and ASPE estimated the financial impacts 
separately. Similarly, both analyses calculated the impact of the drug 
add-on payment separately from the MFN Price impact. Since the drug 
add-on payment inside the model will not be subject to beneficiary cost 
sharing, and will be an additional payment to 340B covered entities, 
the associated Medicare expenditures are higher.
    The baseline for these analyses is shown in Table 10, separately 
for OPPS 340B providers, OPPS non-340B providers, and physician 
settings. These values include all drugs, exclude providers and 
suppliers that are exempt from the model, and assume that 53% of the 
hospital outpatient claims will be from 340B providers. These payments 
were then adjusted for beneficiary responsibility, add-on payments, and 
federal payments relative to ASP. These values are on a pre-COVID-19 
basis, and the baseline is not are adjusted for the effects of the 
pandemic. Similarly, the impact analysis does not include the effects 
of the COVID-19 pandemic. Many assumptions such as utilization, 
mortality, and morbidity are more uncertain than usual due to the 
pandemic. The direction and magnitude of the financial impact of the 
pandemic on Part B drug spending is uncertain. For example, higher 
mortality due to COVID-19 could lead to lower drug utilization. A 
COVID-19-related drug discovery could lead to higher drug utilization. 
Beneficiaries seeking treatment for quality of life improvement may 
defer care during the pandemic.

                      Table 10--Baseline Expenditures for Claims Included in the MFN Model
----------------------------------------------------------------------------------------------------------------
                                                                 (In billions)
                              ----------------------------------------------------------------------------------
                                 2020     2021     2022     2023     2024     2025     2026     2027    2020-27
----------------------------------------------------------------------------------------------------------------
OPPS Non-340B Providers......     $6.1     $6.7     $7.5     $8.3     $9.2    $10.1    $11.2    $12.3      $71.4
OPPS 340B Providers..........      6.9      7.6      8.4      9.4     10.4     11.4     12.6     13.9       80.5
Other Providers and Suppliers     19.4     21.2     23.3     25.7     28.1     30.8     33.8     37.0      219.3
                              ----------------------------------------------------------------------------------
    Total....................     32.4     35.5     39.2     43.4     47.6     52.4     57.5     63.2      371.3
----------------------------------------------------------------------------------------------------------------

    As the model does not dictate the price that a drug manufacturer 
must charge an MFN participant, there are many possible behavioral 
responses by manufacturers, providers, suppliers, and beneficiaries. 
Because the estimates are highly sensitive to these behavioral 
assumptions, OACT provided three scenarios: (i) An OACT estimate; (ii) 
an illustrative estimate based on pricing-effects only; and (iii) an 
additional illustration under the assumption that manufacturers will 
refuse to change prices and MFN participants will be unwilling to 
administer drugs for which model payment will be below their 
acquisition cost. ASPE also developed a bottom-up estimate built from 
analysis of the IFC's likely potential effects on different types of 
separately payable Part B drugs.
    To better understand the values shown in the three OACT scenarios, 
the ASPE estimate, and the policy of the model, consider the following 
example. Suppose the current ASP for a given drug is $100. The total 
payment to the provider for this drug under the current system is 
$104.30, inclusive of the federal payment for the drug and the add-on, 
beneficiary cost-sharing, and net of sequestration. Now suppose the MFN 
Price of this drug is also $100. The total payment to the provider 
under the model would be $104.40. Under the model, the drug payment 
after sequestration is unchanged ($98.40) but the add-on increases from 
$5.90 to $6.00.
1. OACT Estimate
    Manufacturers could adopt several strategies in response to the 
model, such as (i) charging a lower price to providers and suppliers 
inside the model; (ii) refusing to adjust their price from the non-
model amounts; or (iii) altering the availability and terms of their 
international prices. Given that the international price data represent 
a challenge to their U.S. market revenues, manufacturers are expected 
to devote considerable resources to the third option. This assumption 
is included in the OACT estimate as a different discount relative to 
ASP compared with the values in Table 9. For drugs with significant use 
outside of Medicare, manufacturers may be willing to sacrifice 
utilization and revenue within the model. For drugs that are used 
primarily in the Medicare program, manufacturers may believe that 
offering some pricing relief is necessary to preserve a significant 
portion of their revenue.
    Eligible providers and suppliers will need to decide if the 
difference between the amount that Medicare will pay and the price that 
they must pay to purchase the drugs would allow them to continue 
offering the drugs. For 340B providers, the payment rates in the first 
year will match their payments outside the model. Accordingly, no 
change to utilization or costs is expected under the model in the first 
year for 340B providers. In later years, the impact varies depending on 
the assumed change to international price data. For non-340B providers, 
some may be

[[Page 76237]]

willing to provide the drugs under a lower payment rate to retain 
utilization on other associated services.
    Should an eligible provider or supplier be unable to offer access 
to the included drugs, beneficiaries will be left with several options. 
They could seek access to the drugs by traveling to an excluded 
provider or supplier, access the drugs through a 340B provider in the 
model, or forgo access.
    It should be noted that this model does not have a reliable 
precedent in the U.S. market; consequently, there is an unusually high 
degree of uncertainty in these assumptions, particularly with respect 
to the behavioral responses. To illustrate this uncertainty, three 
potential financial effects are included in this analysis; a full range 
of potential behavioral effects are presented under an Extreme 
Disruption scenario where non-340B utilization of affected drugs drops 
to zero percent and under a Pricing-Effects Only scenario where all 
currently projected utilization is assumed to be retained. The OACT 
estimate reflects one reasonable set of assumptions for potential 
changes in manufacturer, provider, and supplier behavior. Other 
estimates outside the range of the three scenarios could be reasonable 
as well, due to the wide range of potential responses.
    The OACT assumptions consider that the separately payable Medicare 
Part B drugs make up approximately 5 percent of the overall U.S. 
prescription drug market. Drug manufacturers could see this model as an 
obstacle to their pricing throughout the market, which could cause 
strong resistance to the model. The OACT assumptions reflect that some 
manufacturers will adhere to their current pricing instead of lowering 
sales prices in response to the model. This behavior may persist in 
spite of pricing in other sectors of the market or other countries that 
demonstrates an ability to offer the drug at the model payment rates, 
and would result in unmet demand for these Medicare Part B drugs. After 
considering the relative size of the Medicare Part B market, the 
current price control of drug manufacturers, the size of the model 
price reductions, the nature of the Medicare Part B drug providers and 
suppliers, the flexibility that manufacturers may have in adjusting 
pricing and arrangements in other countries, and many other factors, 
actuarial judgment was applied to determine the assumptions that are 
reflected in the OACT estimate, as shown in Table 11.
    Beneficiaries lacking continued availability of their drugs through 
their current provider or supplier are assumed to seek access outside 
the model, to obtain their drugs through 340B providers, or to forgo 
access. The schedule of the phase-in to the MFN price gives 
manufacturers incentive to adjust or reduce access to international 
price data quickly. Accordingly, manufacturers are assumed to raise the 
published international prices beginning in 2022 and to retain a 25-
percent MFN Price discount relative to applicable ASP.
    As a result of this expected behavior from manufacturers, 340B 
provider payments will see a 3-percent reduction compared to the 
current Medicare payment in 2022 and subsequent years. This 3-percent 
reduction represents the impact of the 25-percent MFN Price discount 
relative to the OPPS payment to 340B providers of ASP less 22.5 
percent, as that is the current payment formula for 340B providers. 
This represents a relatively small price change and is assumed to occur 
later in the model, so will be more predictable than the payment 
changes for non-340B providers. As a result, manufacturers and 340B 
providers are assumed to come to an agreement to continue to provide 
for all of their utilization.
    Because all regions are covered under the model, beneficiaries 
seeking a provider outside of the model will be limited to an excluded 
provider or supplier, such as a critical access hospital. Based on the 
historical trend of drug spending by excluded providers and suppliers 
as a percentage of total Medicare Part B drugs, the OACT estimate 
reflects only 1 percent of use shifting to non-model providers. 
Furthermore, because the OPPS payment to 340B providers will be reduced 
year two through year seven of the model, and because their capacity is 
limited, 10 percent of use is assumed to shift to 340B providers. Other 
utilization not covered by providers and suppliers continuing to 
provide access in the model or by excluded providers and suppliers is 
assumed to be utilization not covered by the Medicare benefit.

                                Table 11--Assumptions Reflected in OACT Estimate
----------------------------------------------------------------------------------------------------------------
                                                     2021     2022     2023     2024     2025     2026     2027
                                                     (%)      (%)      (%)      (%)      (%)      (%)      (%)
----------------------------------------------------------------------------------------------------------------
Non-340B providers:
    Behavior:
        Continued Availability...................       80       75       70       70       70       70       70
        Altered Availability:
            Move to non-MFN......................        1        1        1        1        1        1        1
            Move to 340B.........................       10       10       10       10       10       10       10
            No Access............................        9       14       19       19       19       19       19
                Total............................      100      100      100      100      100      100      100
MFN Price impact.................................      -16      -25      -25      -25      -25      -25      -25
340B providers:
    Behavior:
        Continued Availability...................      100      100      100      100      100      100      100
MFN Price impact.................................        0       -3       -3       -3       -3       -3       -3
----------------------------------------------------------------------------------------------------------------

    Table 12 shows the estimated financial impacts under the model 
based on the assumptions in Table 11. Medicare savings are estimated to 
be $85.5 billion, net of the premium offset. While there are 
significant savings as a result of this model, a portion of the savings 
is attributable to beneficiaries not accessing their drugs through the 
Medicare benefit, along with the associated lost utilization. This 
estimate does not capture any impacts to other program costs as a 
result of lower utilization. This estimate is on a pre-COVID-19 basis, 
and is not adjusted for the effects of the pandemic.
    To the extent that manufacturers discount their products for 
Medicare sales, there may be a reduction in Medicaid Best Price or AMP. 
Reductions in Best Price could result in increased Medicaid rebates and 
thus lower Medicaid costs. However, reductions in AMP generally result 
in

[[Page 76238]]

lower statutory and inflationary rebates under the Medicaid program. 
Therefore, if the manufacturer discounts a drug so that it is closer to 
the Medicaid best price, there is a possibility of increased Medicaid 
costs as a result of the model. Furthermore, the effects on AMP may be 
reduced or eliminated, if manufacturers respond by increasing prices in 
the private health insurance market. These estimates do not include 
secondary impacts to other sectors of the market as a result of the 
changes in Medicare payments under the model in part due to the 
significant uncertainty around manufacturer pricing behavior in 
response to this model.

                                Table 12--Estimated Financial Impact of MFN Model
----------------------------------------------------------------------------------------------------------------
                                                                  (In billion dollars)
                                       -------------------------------------------------------------------------
                                          2021     2022     2023     2024     2025     2026     2027    2021-27
----------------------------------------------------------------------------------------------------------------
Drug price reduction:
    FFS impact *......................     -4.7     -7.5     -9.3    -10.2    -11.2    -12.3    -13.5      -68.7
    Gross impact (FFS+MA) **..........     -4.7     -7.5    -17.6    -19.5    -21.6    -24.0    -26.5     -121.4
    Net of premium offset ***.........     -3.5     -5.6    -13.2    -14.6    -16.2    -18.0    -19.9      -91.1
    Medicaid impact...................     -0.4     -0.6     -1.3     -1.5     -1.6     -1.8     -2.0       -9.1
        Federal.......................     -0.2     -0.3     -0.8     -0.8     -0.9     -1.0     -1.1       -5.2
        State.........................     -0.2     -0.2     -0.6     -0.6     -0.7     -0.8     -0.9       -3.9
Drug add-on payment:
    FFS impact........................      0.6      0.6      0.5      0.6      0.6      0.7      0.8        4.4
    Gross impact (FFS+MA).............      0.6      0.6      1.0      1.1      1.2      1.4      1.5        7.4
    Net of premium offset.............      0.4      0.4      0.7      0.8      0.9      1.0      1.2        5.6
    Medicaid impact...................     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1     -0.2       -0.8
        Federal.......................     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1       -0.5
        State.........................      0.0      0.0      0.0     -0.1     -0.1     -0.1     -0.1       -0.4
Total impact:
    FFS impact........................     -4.1     -7.0     -8.8     -9.6    -10.6    -11.6    -12.7      -64.4
    Gross impact (FFS+MA).............     -4.1     -7.0    -16.6    -18.4    -20.4    -22.6    -25.0     -114.0
    Net of premium offset.............     -3.1     -5.2    -12.4    -13.8    -15.3    -16.9    -18.7      -85.5
    Medicaid impact...................     -0.4     -0.7     -1.4     -1.6     -1.8     -1.9     -2.1       -9.9
        Federal.......................     -0.3     -0.4     -0.8     -0.9     -1.0     -1.1     -1.2       -5.7
        State.........................     -0.2     -0.3     -0.6     -0.7     -0.8     -0.8     -0.9       -4.3
----------------------------------------------------------------------------------------------------------------
* Projected spending impact in the traditional Medicare FFS program under the model.
** Projected spending impact in both Medicare FFS and Medicare Advantage (MA).
*** Premium offset represents the change in the Part B premium income that would result from the change in Part
  B drug expenditures.

    These impacts are based on the President's Fiscal Year 2021 Budget 
baseline for Medicare Part B drugs, including those dispensed by 340B 
providers. Due to rounding, the sum of values in the table may differ 
slightly from the total results in the table. In addition to the 
behavioral assumptions in Table 11, these estimates reflect a number of 
other technical assumptions, including the following:
     Amounts illustrate the potential impact on Medicare Part B 
drug spending, assuming the reductions are achievable and realized.
     Amounts are presented by calendar year and are based on 
the date the service is incurred and have therefore not been adjusted 
to reflect when payment is made.
     The model runs from January 1, 2021 through December 31, 
2027. If any of the provisions of this rule are not effective on 
January 1, 2021, the impacts will differ.
     The model will include the top 50 Medicare Part B drugs 
with the highest spending each year and will account for roughly 73 
percent of Medicare Part B drug spending in each affected year.
     All included providers and suppliers receive an add-on 
payment of 6 percent (after sequestration) of the average sales price 
(ASP) and this add-on payment is not subject to beneficiary cost 
sharing.
     The impacts reflect changes to payments to Medicare 
Advantage plans starting in 2023.
     The premium offset is 25 percent of the gross impact.
     The Medicaid impact represents the portion of beneficiary 
cost sharing paid on behalf of dual-eligible beneficiaries (split 57 
percent/43 percent between Federal and State).
     The Medicaid impact does not account for the potential 
impacts to AMP or Best Price in the Medicaid program.
a. Pricing Effects Only Illustration
    As mentioned previously, there is much uncertainty around the 
behavioral assumptions underlying the estimated financial impacts. To 
show the effects of the model absent any provider or beneficiary 
behavioral responses, OACT calculated the impacts of the payment 
changes alone. These values reflect the pricing changes inside the 
model, as shown in Table 9, and the assumption that manufacturers and 
MFN participants are able to continue to provide access to all drugs. 
Again, because 340B providers will receive the lesser of the model 
payment amount or the amount outside the model for the drug, no impact 
to their costs is expected for the first year. Results for this 
illustration are shown in Table 13, and they reflect the same technical 
assumptions as the OACT estimate. The net impact on Medicare after the 
premium offset is a savings of $155.6 billion over the 7-year period, 
and none of the impact would be due to lost utilization.

[[Page 76239]]



                         Table 13--Estimated Impact of Pricing Effects Only Illustration
----------------------------------------------------------------------------------------------------------------
                                                                  (In billion dollars)
                                       -------------------------------------------------------------------------
                                          2021     2022     2023     2024     2025     2026     2027    2021-27
----------------------------------------------------------------------------------------------------------------
Drug price reduction:
    FFS impact *......................     -3.1     -7.3    -13.1    -20.1    -23.0    -25.3    -27.7     -119.7
    Gross impact (FFS+MA) **..........     -3.1     -7.3    -24.7    -38.5    -44.4    -49.2    -54.5     -221.8
    Net of premium offset ***.........     -2.4     -5.5    -18.5    -28.9    -33.3    -36.9    -40.9     -166.4
    Medicaid impact...................     -0.2     -0.5     -1.9     -2.9     -3.3     -3.7     -4.1      -16.6
        Federal.......................     -0.1     -0.3     -1.1     -1.6     -1.9     -2.1     -2.3       -9.5
        State.........................     -0.1     -0.2     -0.8     -1.2     -1.4     -1.6     -1.8       -7.2
Drug add-on payment:
    FFS impact........................      0.9      1.0      1.1      1.2      1.3      1.4      1.6        8.3
    Gross impact (FFS+MA).............      0.9      1.0      2.0      2.2      2.5      2.8      3.1       14.4
    Net of premium offset.............      0.7      0.7      1.5      1.7      1.9      2.1      2.3       10.8
    Medicaid impact...................     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1       -0.8
        Federal.......................      0.0     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1       -0.5
        State.........................      0.0      0.0      0.0      0.0     -0.1     -0.1     -0.1       -0.3
Total impact:
    FFS impact........................     -2.3     -6.4    -12.0    -19.0    -21.7    -23.9    -26.2     -111.4
    Gross impact (FFS+MA).............     -2.3     -6.4    -22.7    -36.3    -41.9    -46.5    -51.4     -207.4
    Net of premium offset.............     -1.7     -4.8    -17.0    -27.2    -31.5    -34.9    -38.6     -155.6
    Medicaid impact...................     -0.3     -0.6     -2.0     -3.0     -3.5     -3.8     -4.2      -17.4
        Federal.......................     -0.2     -0.4     -1.1     -1.7     -2.0     -2.2     -2.4       -9.9
        State.........................     -0.1     -0.3     -0.8     -1.3     -1.5     -1.6     -1.8       -7.5
----------------------------------------------------------------------------------------------------------------
* Projected spending impact in the traditional Medicare FFS program under the model.
** Projected spending impact in both Medicare FFS and Medicare Advantage (MA).
*** Premium offset represents the change in the Medicare Part B premium income that would result from the change
  in Medicare Part B expenditures.

b. Extreme Disruption Illustration
    To cover the spectrum of possible outcomes, the impact of a greater 
behavioral response from manufacturers and MFN participants was also 
considered. Under this scenario, it is assumed that non-340B providers 
and suppliers will not be able to obtain any of the current drugs 
inside the model. All non-340B utilization will then be divided among 
the three beneficiary choices of traveling to an excluded provider or 
supplier, using a 340B provider, or forgoing access. Because there are 
a small number of excluded providers and suppliers, OACT assumed they 
only have capacity for a 25 percent increase in utilization. 
Additionally, manufacturers are assumed to not change the international 
prices; as a result, 340B providers will have reduced reimbursement 
beginning in 2022, when the MFN Price dips below the baseline payment 
of ASP less 22.5 percent--leading to reduced beneficiary access through 
340B providers as well. The financial hardship exemption could possibly 
apply under this scenario, but as this payment is retrospective and the 
losses prior to the payment would be severe, it is unclear whether 
providers will be in a position to request the exemption.
    The illustrative results under these assumptions are shown in Table 
14. They weredeveloped with the same technical assumptions listed under 
the OACT estimate. The overall impact of the model would be a 
substantial savings to Medicare of $286.3 billion, but nearly half of 
that impact would be due to lost utilization.

                          Table 14--Estimated Impact of Extreme Disruption Illustration
----------------------------------------------------------------------------------------------------------------
                                                                  (In billion dollars)
                                       -------------------------------------------------------------------------
                                          2021     2022     2023     2024     2025     2026     2027    2021-27
----------------------------------------------------------------------------------------------------------------
Drug price reduction:
    FFS impact *......................    -17.6    -21.2    -26.9    -30.5    -33.7    -37.0    -40.6     -207.5
    Gross impact (FFS+MA) **..........    -17.6    -21.2    -50.9    -58.4    -65.0    -72.0    -79.7     -364.8
    Net of premium offset ***.........    -13.2    -15.9    -38.2    -43.8    -48.7    -54.0    -59.8     -273.6
    Medicaid impact...................     -1.3     -1.6     -3.8     -4.4     -4.9     -5.4     -6.0      -27.4
        Federal.......................     -0.8     -0.9     -2.2     -2.5     -2.8     -3.1     -3.4      -15.6
    State.............................     -0.6     -0.7     -1.6     -1.9     -2.1     -2.3     -2.6      -11.8
Drug add-on payment:
    FFS impact........................     -0.6     -0.8     -1.2     -1.5     -1.6     -1.8     -1.9       -9.4
    Gross impact (FFS+MA).............     -0.6     -0.8     -2.3     -2.8     -3.1     -3.4     -3.8      -16.9
    Net of premium offset.............     -0.5     -0.6     -1.8     -2.1     -2.3     -2.6     -2.9      -12.7
    Medicaid impact...................     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1     -0.2       -0.8
        Federal.......................     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1       -0.5
        State.........................      0.0      0.0      0.0     -0.1     -0.1     -0.1     -0.1       -0.4
Total impact:
    FFS impact........................    -18.2    -22.0    -28.2    -32.0    -35.3    -38.7    -42.5     -217.0
    Gross impact (FFS+MA).............    -18.2    -22.0    -53.2    -61.2    -68.1    -75.5    -83.5     -381.7
    Net of premium offset.............    -13.7    -16.5    -39.9    -45.9    -51.1    -56.6    -62.6     -286.3
    Medicaid impact...................     -1.4     -1.7     -3.9     -4.5     -5.0     -5.5     -6.1      -28.2
        Federal.......................     -0.8     -1.0     -2.2     -2.6     -2.9     -3.2     -3.5      -16.1

[[Page 76240]]

 
        State.........................     -0.6     -0.7     -1.7     -1.9     -2.2     -2.4     -2.6      -12.1
----------------------------------------------------------------------------------------------------------------
* Projected spending impact in the traditional Medicare FFS program under the model.
** Projected spending impact in both Medicare FFS and Medicare Advantage (MA).
*** Premium offset represents the change in the Medicare Part B premium income that would result from the change
  in Medicare Part B expenditures.

c. Additional Considerations
    Because the model will make substantial changes to payment for 
Medicare Part B drugs, there are many other potential responses not 
considered in this analysis. It is possible that manufacturers could 
increase prices for non-Part B drugs, which would affect both private 
market and Part D expenditures, although that potential impact has not 
been quantified for this estimate. It is also possible that moving to a 
flat add-on payment from a percentage of drug cost will have additional 
effects, which are not considered in the OACT analysis. The analysis is 
on a pre-COVID-19 basis, and neither the baseline nor the impact 
analysis are adjusted for the effects of the pandemic.
2. ASPE Estimate
    The behavioral responses of manufacturers, providers, suppliers, 
and beneficiaries to the MFN Model are critical to estimating its 
impact on key outcomes. Lack of direct experience with policies such as 
the MFN Model, however, results in great uncertainty for making these 
behavioral assumptions. For a robust approach, ASPE made a number of 
assumptions based on published literature and expert consensus, and 
applied such assumptions on a drug-by-drug basis. Please note that ASPE 
has not adjusted the assumptions and estimates based on the effects of 
the COVID-19 pandemic.
    The behavioral assumptions in this approach first address 
manufacturers' responses in the international market that might 
increase MFN Prices; and then the potential responses to the MFN Drug 
Payment Amounts by the manufacturers and providers and suppliers that 
purchase MFN Model drugs and submit a claim to Medicare after 
administering such drugs to beneficiaries. In general, these 
assumptions represent the proposition that manufacturers prefer to sell 
their products, even at lower prices, as long as net revenues (net 
sales prices minus production and distribution costs) remain positive; 
and that providers and suppliers are committed to maintaining effective 
treatments for beneficiaries either by negotiating lower prices, 
accepting reduced revenue, or finding effective Medicare Part B or Part 
D alternative treatments.
    To assess the likelihood of each of the alternative manufacturer 
responses to the MFN Model, ASPE reviewed published literature on the 
impacts and interviewed a small cohort of experts regarding the 
potential impacts. Published literature suggests that when a large 
country establishes an international reference price, smaller reference 
countries experience price increases and longer launch delays for new 
products.\85\ ASPE's conversations with experts suggested that as a 
result of the MFN Model, prices in other countries could increase at 
the ex-manufacturer level, potentially up to current ASP levels, and 
manufacturers could change formulations of MFN Model drugs to lessen 
the impact of the model. The experts generally believe that 
manufacturers will be able to price discriminate between the Medicare 
Part B market and other markets within the U.S. Potential utilization 
impacts will thus be limited to Medicare Part B beneficiaries, as 
payments to providers and suppliers for drugs provided to other 
patients will not be affected by the model.
---------------------------------------------------------------------------

    \85\ Patricia M. Danzon, ``The Economics of the 
Biopharmaceutical Industry'', in Sherry Glied and Peter C. Smith 
(eds.), The Oxford Handbook of Health Economics, Oxford University 
Press 2011, pp. 520-554.
---------------------------------------------------------------------------

    Considering this information, ASPE made a series of assumptions for 
a base analysis. First, ASPE considered a static group of 50 drugs for 
this analysis. Based on the literature and interviews with experts, 
ASPE assumed manufacturers of newly launched brand products that become 
MFN Model drugs would adjust their international pricing strategies so 
that the MFN Payment Amount will be equal to ASP absent of the MFN 
Model. This assumption does not necessarily mean that net international 
prices (ex-manufacturer sales prices minus the value of rebates or 
other financial concessions) will be equal to the ASP. In addition, 
ASPE assumed that manufacturers of currently marketed drugs outside but 
near the top 50 Medicare Part B drugs based on annual allowed charges 
(with certain exclusions and exemptions) will lower their U.S. prices 
in an attempt to prevent them from becoming MFN Model drugs. To 
compensate for this response, ASPE assumed that manufacturers will 
increase prices for non-MFN Model drugs. Since companies often sell 
many different drugs, ASPE assumed they will have some flexibility to 
allocate discounts between different drugs to ensure no currently 
marketed non-MFN Model drugs enter the top 50 while maintaining near 
constant revenues. In some cases, there are relatively new drug 
products that may not have launched or may be recently launched in the 
included countries that may enter the top 50. In those cases, ASPE 
assumed the manufacturers will re-evaluate their international pricing 
strategies to ensure the MFN Price is comparable to ASP absent of the 
MFN Model. ASPE assumed that these changes to U.S. prices of non-MFN 
Model drugs will ultimately fully offset one another in terms of 
Medicare Part B drug spending as well.
    For the 50 MFN Model drugs, the MFN Price ultimately depends on the 
prices for the drugs in the included countries. The exact mechanisms in 
which prices are determined in included countries differ by country and 
sometimes by product. These mechanisms include national (or sub-
national tendering \86\), therapeutic-level reference pricing, 
international reference pricing, cost-effectiveness analysis, and 
negotiation. These mechanisms generally result in lower observed prices 
in other countries compared to the U.S., and these differences tend to 
be larger for products that have more competition than in the U.S. 
(such as more biosimilar competition) or have only a marginally better 
clinical profile than a cheaper

[[Page 76241]]

therapy. Since the U.S. price under this model depends on the prices in 
other countries, the model will likely result in increased observed 
prices in other countries. This does not mean that net prices will 
necessarily increase as countries will try to find ways to prevent 
spending increases while limiting disruption in their drug markets. In 
this analysis, ASPE considered the potential impact at the drug-level 
because the context of each drug may determine the MFN Price.
---------------------------------------------------------------------------

    \86\ Tendering is a formal procedure to purchase medications 
using competitive bidding for a particular contract.https://
www.ncbi.nlm.nih.gov/pmc/articles/PMC5628685/.
---------------------------------------------------------------------------

    ASPE modeled the pricing response to the change in direct drug 
payment for each of the 50 MFN Model drugs shown in Table 6 of this 
IFC. ASPE assumed that any changes in international sales prices for 
included countries would not occur until the beginning of the second 
performance year of the MFN Model. ASPE modeled the manufacturer 
pricing response based on available 2019 international drug pricing 
information, using the sales and volume data that CMS used to calculate 
the MFN Prices shown in Table 6 of this IFC. ASPE did not model how 
manufacturers and providers might take into account the changes to the 
add-on.
    If there was only one related brand for the included countries,\87\ 
then ASPE assumed the MFN Price for a drug will increase to the average 
price of the drug for the included countries plus 10 percent (with the 
cap of ASP). ASPE made this assumption because at this point the market 
size of the included countries is roughly the size of the Medicare Part 
B market for many of the MFN Model drugs. ASPE applied this approach to 
34 of the 50 MFN Model drugs. ASPE assumed that the MFN Price will not 
likely increase by more than this because, even if the net price is 
constant for purchasers in the included countries, these countries may 
seek to avoid larger increases in transaction prices. In the case of 
drugs with no international spending in 2019, ASPE assumed that the 
model would have no impact. ASPE applied this approach to 2 of the 50 
MFN Model drugs. When the MFN Price was calculated based on 
international drug pricing information for a country with access to 
biosimilar products or a competitor brand product that is not one of 
the MFN Model drugs, ASPE assumed smaller international price increases 
because the MFN Model would reduce the incentive for the manufacturer 
of an MFN Model drug to compete in those international markets. This 
approach applied to 8 of the 50 MFN Model drugs. When the MFN Price was 
calculated using international drug pricing information for a non-
innovator unbranded product, ASPE assumed that the MFN Price would not 
increase. This assumption applied to 6 of the 50 MFN Model drugs.
---------------------------------------------------------------------------

    \87\ For this analysis, we included available sales and volume 
data for the brand drug manufacturer and any parallel importers of 
the brand drug.
---------------------------------------------------------------------------

    After analyzing price changes internationally, ASPE analyzed the 
potential for beneficiaries to switch to other products with, for 
example, the same active ingredient within the U.S. and billed with 
HCPCS codes that are not among the MFN Model drugs. First, ASPE assumed 
that when a manufacturer has multiple branded products with different 
indications represented by the same HCPCS code, the manufacturer will 
work to obtain a new HCPCS code for the product in which Medicare Part 
B makes up a smaller portion of its overall market. In addition, the 
manufacturer will restrict the amount of product sold that could be 
billed under this new HCPCS code so that such products will not become 
included in the MFN Model. This assumption applied to one of the MFN 
Model drugs. ASPE also assumed that if an MFN Model drug is available 
within the U.S. in a formulation that will be covered under Medicare 
Part D, the manufacturer will work to shift 90 percent of the 
utilization from Medicare Part B to Medicare Part D. This assumption 
impacted 2 of the 50 MFN Model drugs.
    In addition to these assumptions, ASPE made assumptions about 
potential generic entry for some of the MFN Model drugs. ASPE assumed 
that MFN Model drugs with generic drugs approved within the included 
countries or currently subject to on-going Paragraph 4 patent 
challenges would have generic competition by performance year 3. This 
assumption impacted 6 of the 50 MFN Model drugs.
    After examining the potential price impacts and other utilization 
changes described previously, ASPE examined the potential for 
utilization impacts. In general, economic theory and the experts ASPE 
interviewed suggested that manufacturers will adjust U.S. prices to 
maintain sales as long as price is greater than marginal costs of 
producing and distributing the drug. ASPE also assumed that 
manufacturers will have substantial ability to price discriminate--that 
is, adjust pricing for Medicare-participating providers and suppliers 
to reflect discounts for their Medicare Part B patient share as opposed 
to all patients. Nonetheless, ASPE still considered the potential that 
price discrimination will be less than perfect for some drugs. In these 
cases, a manufacturer might refuse to negotiate lower prices for MFN 
beneficiaries if doing so threatens its ability to sell in other 
segments of the U.S. at a positive margin. That is, would the loss in 
revenues from selling for all purchasers at a reduced price exceed the 
loss in revenues from losing the MFN beneficiary share of business for 
that drug? To examine this issue, ASPE estimated the Medicare Part B 
share of each MFN Model drug compared with the estimated U.S. market. 
If it seemed likely that a manufacturer will have higher revenues 
selling to all purchasers at prices slightly above the MFN Drug Payment 
Amount than not selling to MFN participants for MFN beneficiary use, 
ASPE assumed the manufacturer will not restrict MFN beneficiaries' 
access to an MFN Model drug under Medicare Part B. This included 
examining if the MFN Model drugs had U.S. competitors. Since MFN 
participants likely treat both Part B beneficiaries and non-Part B 
beneficiaries (including individuals with employer, individual market, 
or Medicaid coverage), an MFN participant may select an alternative 
therapy marketed by a competitor that can be provided to both types of 
patients. As a result, manufacturers will have an incentive to work to 
maintain utilization so long as the MFN Payment Amount is not too low.
    In cases where manufacturers might refuse to lower U.S. prices 
sufficiently to make it financially feasible for MFN participants to 
furnish the drug and receive the MFN Payment Amount, ASPE examined 
whether there were products that had similar therapeutic effects to a 
MFN Model drug. ASPE assumed that Medicare Part B beneficiaries will be 
switched to the potential alternative products. ASPE made these 
assessments for each performance year. ASPE assumed that half of 
Medicare Part B beneficiaries will continue accessing their current 
drugs through 340B providers. Such changes in drug utilization or 
service providers will likely result in additional burdens for 
patients. ASPE did not quantify these impacts.
    Additionally, for biological drugs for which there are licensed 
biosimilar products, ASPE assumed that there will be at least one 
biosimilar manufacturer that is willing to provide its product at MFN 
payment levels if the reference manufacturer would not supply this 
drug. We note however that if reference manufacturers are willing to 
sell at MFN payment levels, providers may not have any incentive to use 
biosimilar products. The extent to which providers may use biosimilar 
products will depend on whether they are easier to

[[Page 76242]]

access instead of a product subject to the model. The biosimilar 
manufacturers will need to balance those considerations with the 
possibility that sufficiently large sales may also result in that 
product becoming an MFN Model drug. ASPE assumed any utilization 
changes that occur will result in zero net changes in spending. ASPE 
made no assumptions about the potential entry of biosimilar products 
for reference products that currently do not have biosimilar 
competition in the U.S. or referenced countries.
    The overall utilization impact is the sum of the impacts for each 
of the 50 MFN Model drugs. These impacts reflect, on a drug by drug 
basis, the assumptions outlined previously. Specifically, where 
estimates reduced utilization, it reflects assumptions that either 
manufacturers will be unwilling to reduce prices to MFN participants, 
viable substitute drugs are not available for all affected patients, or 
both. In such cases, ASPE assumed that half of the impacted 
beneficiaries will be able to still access the MFN Model drug through a 
340B provider.
    ASPE calculated the potential impacts of the MFN Model by calendar 
year. ASPE assumed that at the end of the MFN Model, there will be no 
continued impacts because Medicare Part B payments for MFN Model drugs 
will immediately be based on non-model payment policies at the end of 
the MFN Model. Given the predictable 7-year model performance period, 
ASPE assumed manufacturers and MFN participants will have sufficient 
time to structure their agreements to ensure a seamless transition 
after the end of the MFN Model.
    Table 15 summarizes the results of the ASPE analysis.

                                Table 15--Assumptions Reflected in ASPE Estimate
----------------------------------------------------------------------------------------------------------------
                                                     2021     2022     2023     2024     2025     2026     2027
                                                     (%)      (%)      (%)      (%)      (%)      (%)      (%)
----------------------------------------------------------------------------------------------------------------
Non-340B providers:
    Behavior:
        Continued Availability...................    100.0    100.0     97.7     95.9     96.2     96.5     96.7
        Altered Availability:
            Shift to other drugs.................      0.0      0.0      1.1      2.1      1.9      1.8      1.6
            Move to 340B.........................      0.0      0.0      1.1      2.1      1.9      1.8      1.6
            No Access............................      0.0      0.0      0.0      0.0      0.0      0.0      0.0
                Total............................      100      100      100      100      100      100      100
MFN Price impact.................................    -11.4    -14.3    -18.1    -20.5    -19.4    -17.9    -16.5
340B providers:
    Behavior:
        Continued Availability...................      100      100      100      100      100      100      100
MFN Price impact.................................        0        0        0        0        0        0        0
----------------------------------------------------------------------------------------------------------------

    ASPE estimated the Medicare FFS program impacts of the change from 
ASP-based payment to MFN-based payment.\88\ The Medicare FFS impact 
includes changes in spending for Medicare Parts B and D.
---------------------------------------------------------------------------

    \88\ An indirect benefit of this IFC may be reduced distortions 
in the labor markets taxed to support the Medicare Trust Fund. Such 
distortions are sometimes referred to as marginal excess tax burden 
(METB), and Circular A-94--OMB's guidance on cost-benefit analysis 
of federal programs, available at https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/circulars/A94/a094.pdf-suggests that METB 
may be valued at roughly 25 percent of the estimated transfer 
attributed to a policy change; the Circular goes on to direct the 
inclusion of estimated METB change in supplementary analyses. If 
secondary benefits--such as reduced marginal excess tax burden is, 
in the case of this IFC--are included in regulatory impact analyses, 
then secondary costs must be as well, in order to avoid 
inappropriately skewing the net benefits results, and including METB 
only in supplementary analyses provides some acknowledgement of this 
potential imbalance.
---------------------------------------------------------------------------

    For patients that switch to 340B providers, ASPE estimated the 
spending change based on the difference in the MFN Model payment for 
drugs acquired under the 340B program and the current Medicare Part B 
OPPS payment policy.
    These impacts are generally considered transfer impacts of the 
model. To estimate these impacts, ASPE took an approach similar to 
OACT. ASPE used the direct reduction in Medicare Part B payments due to 
lower MFN payment amounts and translated that into transfers from the 
healthcare system to the government, beneficiaries, and Medicaid. In 
addition to the direct effects of lower payments and associated cost-
sharing, the model results in downstream transfers associated with 
changes in Part B premiums and government payments to Medicare 
Advantage Plans. Like OACT, ASPE estimated Medicaid impacts based on 
changes to federal and state shares of prescription drug costs for dual 
eligibles but did not estimate impacts on Medicaid that may result from 
changes in net payments under the Medicaid Drug Rebate Program.
    Overall, the model results in changes to federal spending in 
Medicare (including Part B, and Part D) from the model price and 
utilization impacts, changes in federal and state spending on Medicaid 
resulting from changes to the governmental obligation of Medicare cost-
sharing for dual eligible beneficiaries, and changes in federal 
spending associated with add-on payment changes in the model. The model 
also results in changes to beneficiary spending resulting from changes 
in cost-sharing for drugs, changes in beneficiary premiums, and changes 
to cost-sharing associated with the add-on payment. These transfers on 
net balance out with reduced revenues for healthcare providers (which 
may be completely or mostly offset by the reduced cost of acquiring 
drugs), reduced revenues for pharmaceutical manufacturers, and reduced 
revenues for MA plans.
    Based on our estimates of annual impacts on prescription drug 
pricing and annual add-on payments, ASPE did not model any impacts from 
the provider hardship payments. Eligibility for the hardship exemption 
will be based on year-over-year losses above 25 percent of total 
Medicare Part A and Part B payments, including payments for Medicare 
Part B drugs outside the model and payments for Medicare Part A and 
Medicare Part B services other than prescription drugs. We expect that 
few, if any, providers will have annual losses above this level, and 
that those who do may be insolvent and therefore unable to obtain 
retrospective hardship payments. We note in this regard that a 
hypothetical provider could experience revenue losses of 24.9 percent 
per year in each of the model's seven years, resulting in an 86.5 
percent loss of revenue in Performance Year 7

[[Page 76243]]

compared with the pre-model base year and a 62.7 percent loss of 
revenue over the seven-year demonstration period, without qualifying 
for the hardship payments in any year.
    Table 16 shows the net transfer impacts resulting from changes in 
Medicare B, and D. According to the ASPE estimate, this model would 
result in a net reduction of $87.8 billion in beneficiary, federal 
government, and state government spending over the 7 years of the 
model.

                         Table 16--Estimated Transfer Impact of MFN Model--ASPE Estimate
----------------------------------------------------------------------------------------------------------------
                                          2021     2022     2023     2024     2025     2026     2027    2021-27
----------------------------------------------------------------------------------------------------------------
Part B Drug Price Reduction:
    Federal Government Spending.......     -2.4     -3.4     -8.4    -10.0    -10.3    -10.7    -10.8      -56.0
    State Government Spending.........     -0.1     -0.1     -0.4     -0.4     -0.4     -0.5     -0.5       -2.4
    Beneficiary Spending *............     -1.4     -2.0     -5.0     -5.9     -6.2     -6.4     -6.4      -33.4
    MA Plan Revenue...................      0.0      0.0     -4.8     -5.9     -6.1     -6.5     -6.5      -29.8
    Health Care System Revenue **.....     -4.0     -5.5     -8.9    -10.5    -10.9    -11.1    -11.1      -61.9
Part D Drug Switching:
    Federal Government Spending.......      0.0      0.0      0.3      0.3      0.3      0.3      0.3        1.7
    State Government Spending.........      0.0      0.0      0.0      0.0      0.0      0.0      0.0        0.1
    Beneficiary Spending *............      0.0      0.0      0.1      0.1      0.1      0.1      0.1        0.7
    Health Care System Revenue **.....      0.0      0.0      0.5      0.5      0.5      0.5      0.5        2.5
Add-on Payment Impact:
    Federal Government Spending.......      0.2      0.2      0.4      0.3      0.4      0.4      0.4        2.2
    State Government Spending.........      0.0      0.0      0.0      0.0      0.0      0.0      0.0       -0.3
    Beneficiary Spending *............     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1     -0.1       -0.5
    MA Plan Revenue...................      0.0      0.0      0.2      0.3      0.3      0.3      0.3        1.3
    Health Care System Revenue **.....      0.0      0.0      0.0      0.0      0.0      0.0      0.0        0.1
Total Impact:
    Federal Government Spending.......     -2.2     -3.2     -7.7     -9.3     -9.7    -10.0    -10.0      -52.1
    State Government Spending.........     -0.1     -0.2     -0.4     -0.4     -0.5     -0.5     -0.5       -2.5
    Beneficiary Spending *............     -1.6     -2.1     -4.9     -5.9     -6.1     -6.3     -6.3      -33.2
    MA Plan Revenue...................      0.0      0.0     -4.6     -5.6     -5.8     -6.2     -6.2      -28.5
    Health Care System Revenue **.....     -3.9     -5.5     -8.4    -10.0    -10.3    -10.6    -10.6      -59.3
----------------------------------------------------------------------------------------------------------------
* Beneficiary spending includes spending by beneficiary Medigap plans.
** Health care system revenue includes revenue accrued by health care providers, hospitals, pharmacies, and
  pharmaceutical manufacturers.

    Based on this analysis, the model has the potential to generate 
impacts internationally. In particular, this model may result in higher 
prices or longer launch delays for new products in other OECD 
countries. ASPE did not attempt to quantify the impact of higher prices 
on utilization or the impact of these delays. The health effects of 
such delays depend on which products experience these delays and the 
potential alternative treatments. In addition, foreign governments may 
seek to mitigate these impacts by accepting higher prices for the 
products or pursuing alternative price arrangements that are less 
transparent.
3. Aggregate Effects on the Market
    There may be spillover effects in the non-Medicare market, or even 
in the Medicare market outside Part B as a result of the MFN Model. 
Testing changes in Medicare Part B drug payment policy may have 
implications for non-Medicare payers. During the MFN Model, 
manufacturers' ASPs may increase or decrease, which may cause the 
payment limits in the quarterly Medicare ASP payment files to increase 
or decrease. Other payers that align their payments for drugs included 
in the MFN Model with the quarterly Medicare ASP payment files could 
therefore be impacted. Because the extent to which other payers align 
with Medicare Part B drug payments is unknown, we are not able to 
quantify the potential impacts of the MFN Model in this regard.
    Private secondary payers that pay for beneficiary cost-sharing, 
such as Medigap plans and employer retiree coverage, will likely be 
impacted by the MFN Model. For MFN beneficiaries, cost-sharing on MFN 
Model drugs would be less than the amount that will apply outside of 
the model. If manufacturers generally raise drug prices in response to 
the MFN Model, the amount of cost-sharing paid by beneficiaries and 
secondary payers may increase; the opposite will occur if manufacturers 
decrease drug prices. Similarly, private primary insurers may be 
impacted if manufacturers change drug pricing as a result of the MFN 
Model. Market-wide changes in drug prices, including drugs not covered 
by Medicare Part B, will impact any individual who receives such drugs. 
In addition, to the extent manufacturers lower their overall prices for 
drugs, manufacturers may realize lower revenue as a result of the MFN 
Model. It is possible that manufacturers will increase international or 
domestic drug prices, reduce marketing and other expenses, or implement 
other efficiency measures to reduce their operating costs. Given the 
uncertainty of manufacturers' potential behavioral responses to the MFN 
Model, we are unable to quantify these potential spillover effects of 
the MFN Model. We welcome comments on these potential impacts and 
evidence on how this rule could affect other payers, patients, and drug 
manufacturers.
    Some of this final rule's important tradeoffs occur over the long 
run. We request comment on whether the drug products affected by this 
IFC are likely to be currently over- or under-incentivized, including 
evidence from the research literature on optimal patent length, and on 
the effects of the IFC on drug manufacturers' incentives.
4. Estimated Effect and Burden of MFN Model Changes on Medicare 
Beneficiaries
    We estimate that aggregate beneficiary Medicare Part B cost-sharing 
within the context of the MFN Model will decrease as the MFN Drug 
Payment Amount will not exceed 100 percent of the amount that applies 
outside the MFN Model (that is, the applicable ASP or WAC or payment 
limit that applies to drug acquired under the 340B program) and that 
beneficiaries will not have a cost-sharing liability for the 
alternative drug

[[Page 76244]]

add-on payment amount. Coinsurance for most separately payable drugs is 
set at 20 percent of the payment rates, subject to limitation in the 
hospital outpatient and ASC settings. To the extent that prescribing 
patterns shift toward lower cost drugs under the MFN Model, in 
aggregate, beneficiaries could benefit along with the Medicare program. 
If prescribing patterns shift toward Part D drugs, beneficiary cost-
sharing may increase or decrease depending upon the drugs they take, 
which phase of the Part D benefit such use occurs in, the beneficiary's 
eligibility for help with drug costs, and their plan choice. In 
addition, as a result of the MFN Model, we expect Medicare Part B 
premiums to decrease. Beneficiaries will benefit from 25 percent of any 
premium reduction that may result as this is the portion of annual 
premiums that beneficiaries pay.
    If MFN participants choose not to provide MFN Model drugs or 
prescribe alternative therapies instead, beneficiaries may experience 
access to care impacts by having to find alternative care providers 
locally, having to travel to seek care from an excluded provider, 
receiving an alternative therapy that may have lower efficacy or 
greater risks, or postponing or forgoing treatment. There is 
significant uncertainty with these potential effects of the MFN Model. 
CMS will carefully monitor for evidence of these potential effects and 
conduct beneficiary surveys to assess impacts of the MFN Model on 
beneficiaries.
    Given the uncertainty of these impacts, we are unable to quantify 
these potential effects of the MFN Model.
    In section III.H. of this IFC, we describe our intention to include 
quality measures as part of the MFN Model, and our plan to collect one 
quality measure, focused on patient experience, to help better 
understand the impact of the MFN Model on beneficiary access and 
quality of care. This information collection will be one part of robust 
monitoring activities to ensure that MFN beneficiaries' access to 
medications and quality of care is preserved or enhanced. We will use a 
patient experience survey, which we will field to a sample of MFN 
beneficiaries, beginning in performance year 1. We will include 
additional items in the patient experience survey that focus on patient 
access, to the extent that valid and reliable items are available. The 
patient experience survey will be administered to these beneficiaries 
by a third party contractor throughout the model performance period. 
Beneficiaries will not be required to complete the survey.
    The patient experience survey will be based on a standardized 
instrument, designed to assess patients' experiences with health care 
providers and staff in an ambulatory setting. We will use the most 
current version of the instrument plus additional survey questions as 
applicable to meet CMS's monitoring needs.
    Based on drug claims analyses and the scope of the MFN Model, we 
assume the patient experience survey will be administered to 75,000 
beneficiaries and be completed by 30,000 beneficiaries per year. The 
survey will take approximately 30 minutes to complete. Therefore, the 
annual total number of hours for this information collection will be 
15,000 hours (30,000 beneficiaries times 0.5 hours per beneficiary 
responding).
    To derive average costs for individuals we used data from the U.S. 
Bureau of Labor Statistics' May 2019 National Occupational Employment 
and Wage Estimates for our salary estimate (www.bls.gov/oes/current/oes_nat.htm). We believe that the burden will be addressed under All 
Occupations (occupation code 00-0000) at $25.72 per hour since the 
group of individual respondents varies widely from working and 
nonworking individuals and by respondent age, location, years of 
employment, and educational attainment, etc. We are not adjusting this 
figure for fringe benefits and overhead since the individuals' 
activities will occur outside the scope of their employment. Therefore, 
the estimated cost for this information collection will be $385,800 
(15,000 hours x $25.72). Beneficiaries will have annual costs 
associated with responding to the patient experience survey, which we 
estimate will be $385,800 annually during the model.
5. Estimated Effect and Burden on MFN Participants and Manufacturers
    MFN participants and drug manufacturers will have administrative 
costs related to adjusting to and complying with the regulations. These 
costs may include adjusting purchasing arrangements, which for some 
affected businesses may mean substantially changing their pricing 
models and engaging in negotiations with other businesses; tracking 
units of MFN Model drugs that are paid under the MFN Model and excluded 
from manufacturers' ASPs; recordkeeping requirements, which may require 
acquisition of new tools and information sharing; and adjusting to any 
spillover effects. Additionally, MFN participants may be subject to 
site visits for the purposes of monitoring the MFN Model.
    During the model performance period, MFN participants must 
participate in MFN Model monitoring and evaluation activities in 
accordance with 42 CFR 403.1110(b), as the Secretary determines is 
necessary to monitor and evaluate the MFN Model, including without 
limitation collecting and reporting of information, including 
``protected health information'' as that term is defined at 45 CFR 
160.103. These monitoring activities may include a sample of site 
visits to verify any monitoring concerns. We anticipate that these 
monitoring and compliance requirements will not diverge from general 
monitoring requirements for Medicare Part B providers. We believe that 
these requirements do not add additional burden or impose regulatory 
impact on participants. The MFN Model monitoring will likely include 
beneficiaries and eligible providers and suppliers completing surveys. 
Burden for the patient survey is described previously, and burden for 
any provider and supplier survey will depend on the length, complexity, 
and frequency of surveys administered as needed to ensure confidence in 
the survey findings. We will make an effort to minimize the length, 
complexity, and frequency of any provider and supplier surveys. A 
typical survey on average requires about 20 minutes of the respondent's 
time. In other evaluations of models where a survey is required, the 
frequency of surveys varies from a minimum of one round of surveys to 
annual surveys. We estimate the burden for annual surveys from 
clinicians, assuming one per eligible provider and supplier, will be 7 
surveys [annual] times \1/3\ hour [20 min.] times $200 [median 
physician/surgeon hourly rate plus fringe benefits] times 22,888 
[eligible providers and suppliers] = $10,702,429.
    Finally, MFN participants may choose to apply for a financial 
hardship exemption that requires the submission of a timely, complete 
request for a financial hardship exemption. We think that approximately 
900 MFN participants will submit a request for a financial hardship 
exemption each performance year of the model. We expect that a medical 
health service manager will need approximately 15 hours to compile the 
necessary supporting documentation and submit a complete financial 
hardship exemption request. We estimate the burden for applying for the 
financial hardship exemption per year for all performance year of the 
model will be 900 [number of MFN participants that submit

[[Page 76245]]

hardship exemption requests in each performance year] times 15 hours 
times $111 [medical health service manager hourly rate plus fringe 
benefits] = $1,498,500. Note, the financial hardship exemption requests 
for performance year 1 (2021) will be submitted in 2022, and the 
requests for performance year 7 will occur in 2028.
    We expect that manufacturers will need to update their ASP 
reporting. However, we expect the burden to be de minimis compared to 
existing ASP reporting requirements and can likely be automated based 
on existing processes.
6. Regulatory Review Cost Estimation
    In order to comply with the regulatory changes in this IFC, 
affected businesses will need to review the rule and MFN participants 
will need to review MFN-specific billing guidance on how to bill for 
the alternative add-on payment. We expect that a medical health service 
manager reading 250 words per minutes could review the rule in 
approximately 6 hours [(approximately 300 pages * 300 words/per page)/
250 words per minute)/60 minutes)]. We estimate 1 hour to review the 
relevant MLN matters publication and 2 hours to read MFN Model billing 
guidance for a total of 3 hours of billing specific training. Since all 
MFN participants have experience billing HCPCS codes to Medicare, we do 
not expect any additional specific burden related to the alternative 
add-on payment M code during model implementation after the MFN-
specific billing guidance is reviewed. We estimate the salary of a 
medical and health service manager is $111 per hour, using the wage 
information from the 2019 BLS including overhead and fringe benefits 
(BLS occupation code 11-9110). For each provider or supplier that 
reviews the rule and MFN-specific billing guidance, the estimated cost 
based on the expected time and salary of the person reviewing the rule 
is $999 ($111 * 9 [6 hours for reviewing the rule and 3 hours for 
billing training). We estimate that the cost for providers and 
suppliers to review this IFC and MFN-specific billing guidance will be 
approximately $117.9 million (118,101 entities times $999).
7. Regulatory Flexibility Act (RFA)
    The RFA requires agencies to analyze options for regulatory relief 
for small entities. For purposes of the RFA, small entities include 
small businesses, nonprofit organizations, and small governmental 
jurisdictions. Most hospitals, practitioners and most other providers 
and suppliers are small entities, either by nonprofit status or by 
having annual revenues that qualify for small business status under the 
Small Business Administration standards. (For details, see the SBA's 
website at http://www.sba.gov/content/table-smallbusiness-size-standards (refer to the 620000 series)). Individuals and states are not 
included in the definition of a small entity. The RFA requires that CMS 
analyze regulatory options for small businesses and other entities 
unless CMS certifies that a rule will not have a significant economic 
impact on a substantial number of small entities. The analysis must 
include a justification concerning the reason action is being taken, 
the kinds and number of small entities the rule affects, and an 
explanation of any meaningful options that achieve the objectives with 
less significant adverse economic impact on the small entities. The 
vast majority of MFN participants are considered to be small entities, 
based upon the SBA standards. There are over twenty thousand MFN model 
participants that will be included or affected by the MFN Model. 
Because many of the affected entities are small entities, the analysis 
and discussion provided in this section, as well as elsewhere in this 
IFC is intended to comply with the RFA requirements regarding 
significant impact on a substantial number of small entities.
    The RFA requires that a Regulatory Flexibility Analysis (RFA) be 
prepared if an IFC will have a ``significant impact on a substantial 
number'' of small entities. HHS interprets the statute as mandating 
this analysis only if the impact is adverse, though there are differing 
interpretations. For purposes of the RFA, most practitioners, 
hospitals, and other providers and suppliers are small entities, either 
by nonprofit status or by having annual revenues that qualify for small 
business status under the Small Business Administration standards 
(having revenues of less than $7.5 million to $38.5 million in any 1 
year). For details, see the Small Business Administration's ``Table of 
Small Business Size Standards'' at https://www.sba.gov/document/
support--table-size-standards. The rule of thumb used by HHS for 
determining whether an impact is ``significant'' is an adverse effect 
equal to 3 percent or more of total annual revenues. Because the 
majority of providers/suppliers in the U.S. qualify as ``small,'' and 
this model includes all eligible providers/suppliers that submit claims 
for separately payable Medicare Part B drugs, we expect the majority of 
MFN participants to be small entities. However, some of these small 
entities may not administer Medicare Part B drugs and will not be MFN 
participants.
    There are a number of providers and suppliers, including various 
physician specialties, that will see reduced drug component payments of 
3 percent or more in performance year 1. Please refer to Table 3 to see 
the number of entities impacted, as well as the types of providers and 
suppliers that will be most likely impacted by the rule. Lower MFN 
Model drug payments will likely be a fraction of these entities' total 
revenues, taking into account non-Medicare patients and all other 
services provided. Moreover, the alternative add-on payments could 
offset such reductions to some extent, as described in section III.F. 
of this IFC. We considered potential impacts on small entities; we 
expect that the model's impact on an MFN participant's revenue will be 
driven by the proportion of Medicare payments to the MFN participant 
that is related to administering Medicare Part B drugs rather than its 
size. Further, to provide financial protection for MFN participants, we 
are including a financial hardship exemption for MFN participants 
(regardless of size) that experience significant financial hardship as 
a result of the model test, as described in section III.I.2. of this 
IFC. It is likely that many, if not all, included providers and 
suppliers will see an overall decrease in revenue for MFN Model drugs 
of 3 percent or more over the course of the model. Accordingly, we have 
determined that a Regulatory Flexibility Analysis (RFA) is required. 
This RIA, together with the preamble, constitutes the required 
analysis.
    As a result of the model, we expect total allowed charges for 
Medicare Part B drugs for small entities to go down commensurate with 
the phase-in of the MFN Price in the calculation of the MFN Drug 
Payment Amount (Year 1: 75 percent applicable ASP and 25 percent MFN 
Price; Year 2: 50 percent applicable ASP and 50 percent MFN Price; 
etc.). Although the alternative add-on payment was designed to hold MFN 
participants harmless based on current revenue to the greatest extent 
possible, as shown in Table 8, some specialties will benefit from a 
higher aggregate add-on payment amount, while for other specialties 
some portion of such specialties will have a decrease in aggregate add-
on payment. We estimate that MFN participants, on average, will see an 
approximate 40 percent increase in historical revenue related to the 
alternative add-on portion of the MFN Model payments, which will total 
approximately $4.4 billion in

[[Page 76246]]

the OACT estimate and $2.2 billion in the ASPE estimate over the 7-year 
model. In these estimates, the total Medicare FFS impact, as indicated 
in Tables 12 and 16, would be a reduction of approximately $85.5 
billion in Medicare FFS spending in the OACT estimate and a majority of 
the $52.1 billion in reduced federal spending in the ASPE estimate over 
the 7-year model, and will apply mainly to urban and non-340B MFN 
participants. We note that there is much uncertainty around the 
assumptions for these estimates. Finally, we have and will continue to 
take steps to minimize the impact of this IFC on administrative and 
reporting burdens for small businesses. We welcome comments on our 
estimate of significantly affected providers and suppliers and the 
magnitude of estimated effects. We also welcome comments on adjustments 
to the MFN Model that could be considered for future rulemaking while 
preserving the innovative approach to payment in the MFN Model.
8. Effects on Small Rural Hospitals
    Section 1102(b) of the Act requires CMS to prepare a regulatory 
impact analysis if a rule may have a significant impact on the 
operations of a substantial number of small rural hospitals. This 
analysis must conform to the provisions of section 604 of the RFA. For 
purposes of section 1102(b) of the Act, we define a small rural 
hospital as a hospital that is located outside of a metropolitan 
statistical area and has 100 or fewer beds. We estimate that this IFC 
will have a significant impact on small rural hospitals.
    As described in section III.C. of this IFC, we will exclude CAHs 
from the types of providers and suppliers that will be MFN 
participants. Slightly less than 10 percent ($3.35 billion) of total 
Medicare Part B drug allowed charges in 2019 are associated with rural 
providers and suppliers (other than CAHs) based on claims with ZIP 
codes associated with areas that are not assigned to metropolitan core 
based statistical areas (CBSA) identified by the Office of Management 
and Budget; of that amount, less than 0.015 percent ($4.87 million) is 
for drugs furnished in the U.S. territories outside of the metropolitan 
areas of Puerto Rico. These rural entities will experience drug payment 
reductions and overall payment reductions similar to urban entities 
under the MFN Model.
9. Unfunded Mandates Reform
    Section 202 of the Unfunded Mandates Reform Act of 1995 (UMRA) also 
requires that agencies assess anticipated costs and benefits before 
issuing any rule whose mandates require spending in any 1 year of $100 
million in 1995 dollars, updated annually for inflation. In 2020, that 
threshold is approximately $156 million. This IFC does not mandate any 
spending by State, local, or tribal governments, or by the private 
sector, and hence an UMRA analysis is not required.
10. Federalism
    Executive Order 13132 establishes certain requirements that an 
agency must meet when it promulgates a final rule that imposes 
substantial direct costs on State and local governments, preempts state 
law, or otherwise has Federalism implications. We have examined the 
provisions in the MFN Model included in this IFC in accordance with 
Executive Order 13132, and have determined that they will not have a 
direct effect on state, local or tribal governments, preempt state law, 
or otherwise have a Federalism implication.

D. Reducing Regulation and Controlling Regulatory Costs

    Executive Order 13771, titled Reducing Regulation and Controlling 
Regulatory Costs, was issued on January 30, 2017, and requires that the 
costs associated with significant new regulations ``shall, to the 
extent permitted by law, be offset by the elimination of existing costs 
associated with at least two prior regulations.'' This IFC is 
considered an E.O. 13771 regulatory action. Details on the estimated 
costs of this IFC can be found in the preceding and subsequent 
analyses.

E. Alternatives Considered

    This IFC contains a range of policies. It also provides 
descriptions of the statutory provisions that are addressed, identifies 
the final policies, and presents rationales for our policies and, where 
relevant, alternatives that we considered in section III of this IFC.
    Several alternatives we considered included: (1) The parameters 
included in this IFC; (2) variations of certain parameters included in 
this IFC, such as lengthening the phase-in of the MFN Price (described 
in section III.E.8. of this IFC) to occur over 5-7 performance years, 
limiting the model performance period to 5 performance years, expanding 
or limiting the Medicare Part B drugs that would be eligible for 
inclusion in the MFN Model and a different geographic area; (3) the 
parameters in the October 2018 ANPRM for a potential IPI Model for 
Medicare Part B Drugs; \89\ and (4) not implementing the model. In 
addition, when developing the parameters for the October 2018 ANPRM and 
this IFC, we noted that there are a range of methods to implement 
external reference pricing, and these different approaches would affect 
the impact of the model.90 91 In examining potential 
variations of certain parameters included in this IFC, we considered 
potential differences such variations would have on the impacts 
presented in sections VI.C.1. and VI.C.2. of this IFC. We note that a 
potential model design with a longer MFN Price phase-in would have a 
lower estimate of overall Medicare savings; for example, a 7-year 
phase-in of the MFN Price over a 7-year model performance period would 
reduce estimates of Medicare savings in the OACT estimate by 
approximately 25 percent. As noted in section III.E.5. of this IFC, our 
policy is to phase-in the MFN Price more quickly during the initial 
years to allow CMS to test the full phase-in of the MFN Price. In 
considering the scope of the model, we actively assessed whether to 
pursue a smaller geographic scope. As we discuss in section III.C.3. of 
this IFC, we reviewed the comments that we received on the October 2018 
ANPRM, where we considered 50 percent of the country in a model. We 
weighed whether the ability to have a research design where we would 
compare changes in drug spending and utilization relative to a 
comparison group, a design that CMS uses frequently in its models, 
would outweight the concerns we highlight in section III.C.3. of this 
IFC. We ultimately concluded that operational concerns such as 
administrative complexity as well as the risk to model integrity 
associated with a limited geographic scope, as described in section 
III.C.3. of this IFC, necessitate a test with a nationwide scope using 
a different evaluation design.
---------------------------------------------------------------------------

    \89\ https://www.federalregister.gov/documents/2018/10/30/2018-23688/medicare-program-international-pricing-index-model-for-medicare-part-b-drugs.
    \90\ https://jasmin.goeg.at/432/1/EURIPID_GuidanceDocument_V8.1_310718.pdf.
    \91\ https://ec.europa.eu/health/sites/health/files/systems_performance_assessment/docs/pharmaproductpricing_frep_en.pdf.
---------------------------------------------------------------------------

    The estimates for the impact of this IFC show a substantial 
reduction in Medicare Part B spending over a 7-year model.

[[Page 76247]]

In comparison, the parameters considered in the October 2018 ANPRM were 
estimated to result in a less substantial reduction in Medicare Part B 
spending over a 5-year model.\92\ The alternative of not implementing 
the model would not have an impact compared to existing policy.
---------------------------------------------------------------------------

    \92\ https://www.federalregister.gov/documents/2018/10/30/2018-23688/medicare-program-international-pricing-index-model-for-medicare-part-b-drugs.
---------------------------------------------------------------------------

F. Accounting Statements and Tables

    As required by OMB Circular A-4 under Executive Order 12866 
(available at https://obamawhitehouse.archives.gov/omb/circulars_a004_a-4/) in Tables 17 and 18 we have prepared two 
accounting statements, based on the OACT and ASPE estimates 
respectively, showing the classification of transfers, benefits, and 
costs associated with the provisions in this IFC. The transfer from 
beneficiaries to providers and MA plans represents the premium change 
attributable to the drug price, i.e., the difference between the gross 
impact and the net impact in the drug price section of Table 12. The 
accounting statement in Table 17 is based on estimates provided in this 
regulatory impact analysis in Table 12 and the accounting statement in 
Table 18 is based on estimates in Table 16. Tables 17 and 18 include 
the estimated effect and burden estimates on beneficiaries outlined in 
section VI.C.4. of this IFC and on participants and manufacturers in 
section VI.C.5. of this IFC. The costs shown in Table C18 reflect 
additional medical expenses incurred as a result of the potential loss 
of access to certain drugs for some beneficiaries in the ASPE estimate.
[GRAPHIC] [TIFF OMITTED] TR27NO20.010


[[Page 76248]]


[GRAPHIC] [TIFF OMITTED] TR27NO20.011

G. Conclusion

    The changes in this IFC will affect providers and suppliers that 
furnish separately payable Medicare Part B drugs in the outpatient 
setting for which annual Medicare FFS spending is high. These providers 
and suppliers are mostly physicians (including physician practices), 
non-physician practitioners, supplier groups, HOPDs (including on- and 
off-campus outpatient provider-based departments, but excluding cancer 
hospitals, children's hospitals and CAHs), and ASCs. We estimate that 
the effect of the MFN Model on providers and suppliers willvary, 
depending on their type, location, what drugs they furnish, their 
clinical patterns, and the alternative add-on payment for the MFN 
Model. We estimate that eligible providers and suppliers will 
experience a decrease in overall payment related to the MFN Model. We 
estimate that beneficiaries who receive included drugs from MFN 
participants will experience a decrease in cost-sharing, however, some 
beneficiaries' providers and suppliers may choose not to offer access 
to the MFN Model drugs, causing these beneficiaries to seek alternative 
providers, treatment alternatives, or forgo access. The financial 
hardship exemption is designed to mitigate this risk.
    The changes in this IFC will also affect MA organizations, drug 
manufacturers, primary and secondary payers, and potentially non-
Medicare patients. MA organizations will experience lower payments as a 
result of the MFN Model because the MA ratebook calculations will 
reflect changes in actual FFS spending due to the impact of the model. 
Drug manufacturers may have lower revenue, depending upon their 
behavioral response to the MFN Model. Other payers, including State 
Medicaid Programs, and patients who take prescription drugs may 
experience direct or indirect spillover effects that may increase or 
decrease their costs. In addition, as shown in Tables 12 and 16, the 
changes we are adopting in this IFC will reduce state and federal 
Medicaid spending and beneficiary spending on Medicare premiums.
    In accordance with the provisions of E.O. 12866, this IFC was 
reviewed by the Office of Management and Budget.

VII. Waiver of Proposed Rulemaking and Delay in Effective Date

    Under 5 U.S.C. 553(b) of the Administrative Procedure Act (APA), 
the agency is required to publish a notice of the proposed rule in the 
Federal Register before the provisions of a rule take effect. 
Similarly, section 1871(b)(1) of the Act requires the Secretary to 
provide for notice of the proposed rule in the Federal Register and 
provide a period of not less than 60 days for public comment. Section 
553(b)(B) of the APA provides for exceptions from the notice and 
comment requirements; in cases in which these exceptions apply, section 
1871(b)(2)(C) of the Act provides for exceptions from the notice and 
60-day comment period requirements of the Act as well. Section 
553(b)(B) of the APA and section 1871(b)(2)(C) of the Act authorize an 
agency to dispense with

[[Page 76249]]

normal rulemaking requirements for good cause if the agency makes a 
finding that the notice and comment process is impracticable, 
unnecessary, or contrary to the public interest.
    High drug prices in the U.S. have serious economic and health 
consequences for beneficiaries in need of treatment. Increasing 
premiums, out-of-pocket costs in both Part B and Part D, and increases 
in drug prices are causing beneficiaries to divert scarce resources to 
pharmaceutical treatments and away from other needs, or prompting them 
to skip doses of their medications, take less than the recommended 
doses, or abandon treatment altogether.93 94 In Medicare 
Part B, drug spending increased by over 9 percent between 2009 and 
2017. Over two thirds of that increase in spending was based on 
increases in drug prices alone, and only one third due to increases in 
utilization.\95\ Prices of certain drugs have increased by double-digit 
percentages over time.\96\ These dramatic increases are on prices where 
the U.S. already pays significantly more than other countries.\97\ When 
CMS announced the 2020 Part B Premiums and Deductibles, we noted that 
the increases in Part B premiums and deductibles was largely due to 
rising spending on physician-administered drugs.\98\
---------------------------------------------------------------------------

    \93\ Kirzinger A, Neuman T, Cubanski J, Brodie M. Data Note: 
Prescription Drugs and Older Adults, Aug 09, 2019, Kaiser Family 
Foundation https://www.kff.org/health-reform/issue-brief/data-note-prescription-drugs-and-older-adults/.
    \94\ Kirzinger A, Lopes L, Wu B, Brodie M. KFF Health Tracking 
Poll--February 2019: Prescription Drugs, Kaiser Family Foundation, 
Mar 01, 2019, https://www.kff.org/health-costs/poll-finding/kff-health-tracking-poll-february-2019-prescription-drugs/.
    \95\ Medicare Payment Advisory Commission, Report to Congress 
``Chapter 3: Medicare payment strategies to improve price 
competition and value for Part B drugs,'' June 2019, http://www.medpac.gov/docs/default-source/reports/jun19_ch3_medpac_reporttocongress_sec.pdf?sfvrsn=0.
    \96\ Hernandez I, San-Juan-Rodriguez A, Good CB, Gellad WF. 
Changes in List Prices, Net Prices, and Discounts for Branded Drugs 
in the US, 2007-2018. JAMA. 2020;323(9):854-862. doi:10.1001/
jama.2020.1012.
    \97\ Comparison of U.S. and International Prices for Top 
Medicare Part B Drugs by Total Expenditures'' https://aspe.hhs.gov/pdf-report/comparison-us-and-international-prices-top-medicare-part-b-drugs-total-expenditures.
    \98\ CMS Newsroom. 2020 Medicare Parts A & B Premiums and 
Deductibles. https://www.cms.gov/newsroom/fact-sheets/2020-medicare-parts-b-premiums-and-deductibles.
---------------------------------------------------------------------------

    With more than 25 million Medicare beneficiaries living at or below 
200 percent of the Federal Poverty Line (FPL),\99\ high drug prices 
could lead to improper medication adherence or skipped treatment. The 
consequences of these behaviors can result in poor clinical outcomes 
for chronic disease management.\100\ The COVID-19 pandemic has rapidly 
exacerbated these problems. The risk of severe illness from COVID-19 
increases with age and the presence of chronic illnesses, putting many 
older adults at the highest risk levels.101 102 This is of 
particular concern given that 84 percent of individuals over the age of 
65 having at least one chronic health condition, and more than 53 
million adults over the age of 65 are enrolled in 
Medicare.103 104 With adults 65 and older comprising 8 out 
of 10 COVID-19 deaths reported in the U.S., COVID-19 has 
disproportionately impacted Americans 65 or older.\105\
---------------------------------------------------------------------------

    \99\ Schoen C, Davis K, Willink A, Medicare Beneficiaries' High 
Out of Pocket Costs: Cost Burdens by Income and Health Status, 
Commonwealth Fund, May 12, 2017, https://www.commonwealthfund.org/publications/issue-briefs/2017/may/medicare-beneficiaries-high-out-pocket-costs-cost-burdens-income?redirect_source=/publications/issue-briefs/2017/may/medicare-out-of-pocket-cost-burdens.
    \100\ https://www.cdc.gov/mmwr/volumes/66/wr/mm6645a2.htm.
    \101\ https://www.cdc.gov/coronavirus/2019-ncov/need-extra-
precautions/older-
adults.html#:~:text=Risk%20for%20Severe%20Illness%20Increases%20with%
20Age&text=The%20greatest%20risk%20for%20severe%2cas%20having%20under
lying%20medical%20conditions.
    \102\ https://www.cdc.gov/coronavirus/2019-ncov/need-extra-precautions/people-with-medical-conditions.html.
    \103\ National Council on Aging, https://www.ncoa.org/economic-security/money-management/debt/senior-debt-facts/.
    \104\ MMCO Statistical & Analytic Reports, Managed Care 
Enrollment Trends (2006-2018 Data), https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medicaid-Coordination-Office/Analytics.
    \105\ https://www.cdc.gov/coronavirus/2019-ncov/need-extra-
precautions/older-
adults.html#:~:text=Risk%20for%20Severe%20Illness%20Increases%20with%
20Age&text=The%20greatest%20risk%20for%20severe%2cas%20having%20under
lying%20medical%20conditions.
---------------------------------------------------------------------------

    Furthermore, the COVID-19 pandemic has led to historic levels of 
unemployment in the U. S., with both the unemployment rate and number 
of unemployed persons remaining nearly twice their Feburary (pre-
pandemic) numbers.\106\ The COVID-19 pandemic has also led to an 
increase in food prices, straining budgets for many of America's 
seniors, particularly those who live on fixed incomes,\107\ such as the 
6 million Medicare fee-for-service beneficiaries without supplemental 
coverage and over 12 million beneficiaries dually eligible for Medicare 
and Medicaid.108 109 110 Already facing increased financial 
burden, this population is in need of urgent relief from high drug 
prices in order to prevent stinting on care and alleviate general 
financial instability worsened by the COVID-19 pandemic. This need is 
exacerbated in communities of color and among women, wherein Black, 
Latino, and Hispanic adults face higher economic insecurity than their 
white counterparts.\111\ The economic disruptions caused by the COVID-
19 pandemic have increased the burdens placed on America's seniors and 
other Medicare Part B beneficiaries and given rise to an urgent need 
for swift action to reduce drug prices. Though we have seen some 
positive economic and employment trends since the initial peak in 
April,\112\ we are currently seeing a new surge in COVID-19 cases that 
may lead to additional hardship and requires immediate action.\113\ As 
such, we find that there is good cause to waive the notice and comment 
requirements under sections 553(b)(B) of the APA and section 
1871(b)(2)(C) because of the particularly acute need for affordable 
Medicare Part B drugs now, in the midst of the COVID-19 pandemic. 
Implementation of this model will provide immediate relief to Medicare 
beneficiaries through reduced copays for MFN drugs due to lower drug 
payments and no beneficiary cost-sharing on the alternative add-on 
payment.
---------------------------------------------------------------------------

    \106\ U.S. Bureau of Labor Statistics. Economic News Release.The 
Employment Situation--October 2020. November 06, 2020. https://www.bls.gov/news.release/empsit.nr0.htm.
    \107\ 54 million people in America face food insecurity during 
the pandemic. It could have dire consequences for their health. 
AAMC. October 15, 2020. https://www.aamc.org/news-insights/54-million-people-america-face-food-insecurity-during-pandemic-it-could-have-dire-consequences-their.
    \108\ MMCO Statistical & Analytic Reports, Managed Care 
Enrollment Trends (2006-2018 Data), https://www.cms.gov/Medicare-Medicaid-Coordination/Medicare-and-Medicaid-Coordination/Medicare-Medicaid-Coordination-Office/Analytics.
    \109\ An Overview of Medicare, Kaiser Family Foundation, Feb 12, 
2019, https://www.kff.org/medicare/issue-brief/an-overview-of-medicare/.
    \110\ Note: The number of Medicare beneficiaries without 
supplemental insurance is from 2016 while the dual eligible numbers 
are from 2018.
    \111\ Center on Budget and Policy Priorities, Tracking the 
COVID-19 Recession's Effects on Food, Housing, and Employment 
Hardships. November 9, 2020. https://www.cbpp.org/research/poverty-and-inequality/tracking-the-covid-19-recessions-effects-on-food-housing-and.
    \112\ U.S. Bureau of Labor Statistics. Economic News Release.The 
Employment Situation- October 2020. November 06, 2020. https://www.bls.gov/news.release/empsit.nr0.htm.
    \113\ Center for Disease Control. COVID-19 Forecasts: Cases. 
https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/forecasts-cases.html.
---------------------------------------------------------------------------

    We also usually provide for a delay in effective date under section 
553(d) of the APA and section 1871(e)(1)(B) of the Act. However, such 
delay in effective date may be waived for good cause,

[[Page 76250]]

when such delay is impracticable, unnecessary, or contrary to the 
public interest, and the agency incorporates a statement of the finding 
and a brief statement of the reasons therefore in the notice. We find 
that delaying implementation of this IFC is contrary to the public 
interest for the same reasons that we find good cause to waive prior 
notice and comment.

List of Subjects in 42 CFR Part 513

    Administrative practice and procedure, Health facilities, Medicare, 
Reporting and recordkeeping requirements.


0
For the reasons set forth in the preamble and under the authority at 5 
U.S.C. 301, the Centers for Medicare & Medicaid Services amends 42 CFR 
Chapter IV by adding part 513 to read as follows:

SUBCHAPTER H--HEALTH CARE INFRASTRUCTURE AND MODEL PROGRAMS

PART 513--Most Favored Nation (MFN) MODEL

Sec.
Subpart A--General Provisions
513.1 Basis, scope, and duration.
513.2 Definitions.
Subpart B--Inclusion in the Model
513.100 MFN Model payments and MFN participants.
513.120 MFN Model geographic area.
513.130 MFN Model drugs, updates, categories and exclusions.
513.140 Included international data.
Subpart C--Payment Process and Methodology
513.200 Payment process and beneficiary cost-sharing.
513.210 Model payment methodology for MFN Model drugs.
513.220 Model alternative add-on payment.
513.230 Financial Hardship Exemptions, Request Process, and 
Reconciliation Payment.
Subpart D--[Reserved]
Subpart E--Quality Strategy, Beneficiary Protections, and Compliance 
Activities
513.400 Quality measures.
513.410 Beneficiary protections.
513.420 Monitoring and compliance activities.
513.430 Audits and record retention.
513.440 Enforcement authority.
513.450 Limitations on review.
Subpart F--Waivers
513.500 Waivers of Medicare program requirements for purposes of 
testing the MFN Model.
Subparts G through J--[Reserved]
Subpart K--Model Termination
513.1000 Termination of the MFN Model.

    Authority: 42 U.S.C. 1302, 1315(a), and 1395hh.

Subpart A--General Provisions


Sec.  513.1   Basis, scope, and duration.

    (a) Basis. This part implements the test of the Most Favored Nation 
(MFN) Model under section 1115A of the Act. Except as specifically 
noted in this part, the regulations under this part do not affect 
payment, coverage, program integrity, or any other requirements that 
otherwise apply to providers of services and suppliers under this 
chapter.
    (b) Scope. This part sets forth the following:
    (1) The types of providers and suppliers required to participate in 
the MFN Model and applicable requirements.
    (2) The beneficiaries included in the MFN Model.
    (3) The drugs included in the MFN Model.
    (4) The methodologies for establishing Medicare payment amounts for 
and making payments for MFN Model drugs, including an alternative add-
on payment.
    (5) Beneficiary protections.
    (6) Beneficiary cost-sharing.
    (c) Duration. The MFN Model has a performance period of 7 
performance years. The first performance year (performance year 1) 
begins on January 1, 2021, and the final performance year ends on 
December 31, 2027, unless sooner terminated in accordance with Sec.  
513.1000.


Sec.  513.2   Definitions.

    For the purpose of this part the following definitions are 
applicable unless otherwise stated:
    Add-on percentage means the percentage above 100 percent.
    Alternative add-on payment means the payment described in Sec.  
513.220.
    Applicable ASP means the payment amount determined in accordance 
with section 1847A of the Act for a quarter minus the applicable add-on 
percentage.
    ASP stands for average sales price.
    ASP calendar quarter means the period that is two calendar quarters 
prior to the calendar quarter to which the MFN Drug Payment Amount will 
apply.
    CCN stands for CMS Certification Number.
    Country-level price means the unadjusted country-level price for an 
MFN Model drug at the HCPCS code level as calculated in accordance with 
Sec.  513.210(b)(2).
    CPI-U stands for Consumer Price Index for All Urban Consumers based 
on all items in U.S. city average and not seasonally adjusted.
    Days means calendar days.
    DME stands for Durable Medical Equipment.
    FDA stands for Food and Drug Administration.
    GDP stands for gross domestic product.
    GDP-adjusted country-level price means the country-level price 
adjusted by the GDP adjuster as calculated in accordance with Sec.  
513.210(b)(4).
    GDP adjuster means the country-specific adjuster as calculated in 
accordance with Sec.  513.210(b)(3).
    HCPCS stands for Healthcare Common Procedure Coding System.
    HCPCS code level means the specified drug and amount described in 
the HCPCS code long descriptor.
    MAC stands for Medicare Administrative Contractor.
    Manufacturer's average sales price has the same meaning as under 42 
CFR Subpart J.
    MFN stands for most favored nation.
    MFN beneficiary means an individual who is furnished an MFN Model 
drug by an MFN participant and who, on the date of service, is enrolled 
in Medicare Part B, has Medicare as his or her primary payer, and is 
not covered under Medicare Advantage or any other group health plan, 
including a United Mine Workers of America health plan.
    MFN Drug Payment Amount means the portion of the total allowed 
payment amount for an MFN Model drug determined in accordance with 
Sec.  513.210.
    MFN Model drug means a separately payable Medicare Part B drug or 
biological described by a HCPCS code included on the MFN Model Drug 
HCPCS Codes List.
    MFN Model Drug HCPCS Codes List means the list of drugs included in 
the MFN Model for a given calendar quarter of a performance year 
established under Sec.  513.130.
    MFN participant means a Medicare participating provider or 
supplier, identified by its CCN or TIN, that is required to participate 
in the MFN Model in accordance with Sec.  513.100(b).
    MFN Model Payment means the total payment to an MFN participant for 
an MFN Model drug in accordance with subpart C of this part, inclusive 
of the MFN Drug Payment Amount and the Alternative Add-on Payment.
    MFN Price means the lowest GDP-adjusted country-level price of the 
countries specified in Sec.  513.140(b) for an MFN Model drug.
    Model performance period means the 7-year period of time beginning 
on January 1, 2021, through December 31, 2027.
    NOC stands for not otherwise classified.

[[Page 76251]]

    OIG stands for the Department of Health and Human Services Office 
of Inspector General.
    Outpatient prospective payment system (OPPS) means the payment 
system for designated hospital outpatient items and services and 
certain Medicare Part B services furnished to hospital inpatients when 
Part A payment cannot be made as defined by section 1833(t) of the Act.
    Performance year means each 12-month period beginning on January 1 
and ending on December 31 during the performance period for the MFN 
Model specified in Sec.  513.1(c).
    Provider means a ``provider of services'' as defined under section 
1861(u) of the Act and codified at Sec.  400.202 of this chapter.
    Supplier means a supplier as defined in section 1861(d) of the Act 
and codified at Sec.  400.202 of this chapter.
    TIN stands for taxpayer identification number.
    WAC means wholesale acquisition cost as defined at section 
1847A(c)(6)(B) of the Act.

Subpart B--Inclusion in the Model


Sec.  513.100   MFN Model payments and MFN participants.

    (a) General. Subject to the exceptions specified in paragraph (d) 
of this section, the MFN Model payments specified under this part apply 
only to claims for an MFN Model drug furnished to an MFN beneficiary by 
an MFN participant.
    (b) MFN participants. Subject to the exclusions specified in 
paragraph (c) of this section, the MFN Model requires participation by 
each Medicare participating provider and supplier that submits a claim 
(except for claims specified in paragraph (d) of this section) for a 
separately payable drug that is an MFN Model drug furnished to an MFN 
beneficiary.
    (c) Excluded providers and suppliers. The following are excluded 
from participation in the MFN Model:
    (1) Children's hospitals (defined under section 1886(d)(1)(B)(iii) 
of the Act).
    (2) PPS-exempt cancer hospitals (defined under section 
1886(d)(1)(B)(v) of the Act).
    (3) Critical access hospitals (CAHs) (defined under section 1820 of 
the Act).
    (4) Indian Health Service (IHS) facilities (as described in section 
1880 of the Act)), except when MFN Model drugs are furnished and such 
service is described in section 1880(e)(2)(B) of the Act.
    (5) Federally Qualified Health Centers (FQHCs) (defined under 
section 1861(aa)(4) of the Act).
    (6) Rural Health Clinics (RHCs) (defined under section 1861(aa)(2) 
of the Act).
    (7) Hospitals that are not subsection (d) hospitals (as defined in 
section 1886(d)(1)(B) of the Act) and are paid on the basis of 
reasonable costs subject to a ceiling under section 1886(b) of the Act.
    (8) Extended neoplastic disease care hospitals (defined in section 
1886(d)(1)(B)(vi) of the Act).
    (9) For the first quarter and second quarter of performance year 1, 
acute care hospitals that participate in any model authorized under 
section 1115A of Act for which payment for outpatient hospital services 
furnished to Medicare FFS beneficiaries, including MFN Model drugs, is 
made under such model on a fully capitated or global budget basis under 
a waiver of section 1833(t) of the Act.
    (10) Beginning with the third quarter of performance year 1, acute 
care hospitals that participate in any model authorized under section 
1115A of Act for which payment for outpatient hospital services 
furnished to Medicare FFS beneficiaries, including MFN Model drugs, is 
made under such model on a fully capitated or global budget basis under 
a waiver of section 1833(t) of the Act, where the parameters of such 
model adjust for the difference in payment for MFN Model drugs between 
the MFN Model and non-MFN Model drug payments such that savings under 
the MFN Model are incorporated into the other CMS Innovation Center 
model's parameters (for example, the annual global budget) for the 
duration of the MFN Model.
    (d) Exceptions. The MFN Model payments specified under this part do 
not apply to any of the following:
    (1) Claims for MFN Model drugs furnished in the inpatient hospital 
setting under those circumstances where Part A would not pay for 
hospital services.
    (2) Claims for MFN Model drugs administered during an inpatient 
hospital stay or included on an inpatient hospital claim.
    (3) Claims administered by the DME MACs as described in Sec.  
421.404(c)(2) of this chapter.
    (4) Claims paid under the End-Stage Renal Disease Prospective 
Payment System, including claims paid using the transitional drug add-
on payment adjustment.
    (e) MFN participant requirements during the MFN Model. During the 
model performance period described in Sec.  513.1(c), MFN participants 
must do all of the following:
    (1) Adhere to the beneficiary protections requirements under Sec.  
513.410.
    (2) Adhere to the MFN Model-specific billing instructions 
requirements established by CMS and the MAC responsible for processing 
the MFN participant's claims, including without limitation those 
described in Sec.  513.200.
    (3) Participate in MFN Model monitoring and evaluation activities 
in accordance with Sec.  403.1110(b) of this chapter, including 
collecting and reporting information as the Secretary determines is 
necessary to monitor and evaluate the MFN Model, including without 
limitation ``protected health information'' as that term is defined at 
45 CFR 160.103.
    (f) MFN participant requirements after the MFN Model. For 2 years 
after termination of the MFN Model, MFN participants must participate 
in MFN Model monitoring activities as described in Sec.  513.420.


Sec.  513.120   MFN Model geographic area.

    The MFN Model geographic area is all states and U.S. territories.


Sec.  513.130   MFN Model drugs, updates, categories and excluded 
drugs.

    (a) MFN Model drugs. CMS creates and periodically updates the MFN 
Model Drug HCPCS Codes List as described in this section. The MFN Model 
Drug HCPCS Codes List designates the MFN Model drugs, which are subject 
to the MFN Model payments specified in subpart C of this part.
    (1) Initial MFN Model Drug HCPCS Codes List. For the beginning of 
performance year 1, CMS identifies the top 50 drugs by HCPCS code with 
the highest aggregate 2019 Medicare Part B total allowed charges after 
making the exclusions specified in paragraphs (b)(1) and (b)(2) of this 
section, and adds the remaining HCPCS codes, after updating such HCPCS 
codes for any applicable changes, to the MFN Model Drug HCPCS Codes 
List. Final action claims with dates of service within calendar year 
2019 and allowed charges greater than $0 are used to determine 
aggregate 2019 Medicare Part B total allowed charges.
    (2) Annual Update of the MFN Model Drug HCPCS Codes List. For the 
start of each subsequent performance year, using Medicare Part B total 
allowed charge from the next subsequent calendar year, CMS identifies 
the top 50 drugs by HCPCS code with the highest aggregate Medicare Part 
B total allowed charges, after making the exclusions specified in 
paragraphs (b)(1) and (b)(2) of this section, for the most recent full 
calendar year, and adds any remaining

[[Page 76252]]

HCPCS codes not already on the MFN Model Drug HCPCS Codes List to the 
MFN Model Drug HCPCS Codes List, after updating such HCPCS codes for 
any applicable changes, effective on the first day of the performance 
year.
    (3) Removal. No more frequently than quarterly, CMS removes HCPCS 
codes from the MFN Model Drug HCPCS Codes List when CMS becomes aware 
that all of the National Drug Codes assigned to the HCPCS code have 
been permanently withdrawn from the U.S. market and the drug has been 
permanently withdrawn from the U.S. market, the specific HCPCS code 
included on the MFN Model Drug HCPCS Codes List is terminated with no 
replacement code available or planned, or an exclusion in paragraph 
(b)(1) of this section applies.
    (4) Maintenance. No more frequently than quarterly, CMS revises 
HCPCS codes on the MFN Model Drug HCPCS Codes List as necessary to 
reflect quarterly HCPCS code updates that are applicable to the HCPCS 
codes on the MFN Model Drug HCPCS Codes List, including adding 
replacement codes for HCPCS codes that were terminated.
    (b) Exclusions. (1) The following are excluded from the MFN Model:
    (i) Vaccines specified in section 1861(s)(10) of the Act 
(influenza, pneumococcal pneumonia, coronavirus disease 2019 (COVID-
19), and Hepatitis B vaccines).
    (ii) Radiopharmaceuticals.
    (iii) Oral anticancer chemotherapeutic agents described in section 
1861(s)(2)(Q) of the Act.
    (iv) Oral anti-emetic drugs described in 1861(s)(2)(T) of the Act.
    (v) Oral immunosuppressive drugs described in section 1861(s)(2)(J) 
of the Act.
    (vi) Compounded drugs.
    (vii) Intravenous immune globulin products.
    (viii) Drugs billed with HCPCS codes that describe a drug product 
that was approved under an abbreviated new drug application under 
section 505(j) of the Federal Food, Drug, and Cosmetic Act;
    (ix) Drugs for which there is an Emergency Use Authorization (EUA) 
from FDA, or FDA approval, to treat patients with suspected or 
confirmed COVID-19; or
    (x) Drugs billed using a not otherwise classified (NOC) or not 
otherwise specified (NOS) billing and payment code.
    (2) The following claims are excluded from the determination of 
whether a drug is to be included on the MFN Model Drug HCPCS Codes 
List:
    (i) Professional claims with a place of service code indicating a 
home setting, including home, homeless shelter, assisted living 
facility, group home, temporary lodging, and custodial care facilities.
    (ii) Claims administered by the DME MACs as described in Sec.  
421.404(c)(2) of this chapter.


Sec.  513.140   Included international data.

    (a) General. (1) CMS uses drug pricing information from 
international data sources, available to CMS at least 20 business days 
prior to the start of a calendar quarter, meeting the requirements in 
paragraph (c) of this section for MFN Model drugs from countries 
included in paragraph (b) of this section.
    (2) For purposes of selecting a data source for each MFN Model drug 
for a calendar quarter, CMS identifies available international drug 
pricing information data sources for the MFN Model drug, by aligning 
the MFN Model drug's HCPCS code long description (including dosage 
form) with the data sources' standardized method for identifying 
scientific names or nonproprietary names and dosage formulations, as 
applicable.
    (b) Non-U.S. member countries of the Organisation for Economic Co-
operation and Development (OECD). (1) CMS uses available international 
sales, volume, and pricing data for countries that were non-U.S. OECD 
member countries as of October 1, 2020 with a GDP per capita that is at 
least 60 percent of the U.S. GDP per capita as determined by CMS in 
accordance with this paragraph (b).
    (2) Each country's GDP per capita is assessed using data available 
at the end of the applicable ASP calendar quarter.
    (3) Subject to the limitation specified in paragraph (b)(4) of this 
section, the GDP per capita for a country is the the most recent 
estimate of GDP per capita based on purchasing power parity for that 
country available in the U.S. Central Intelligence Agency (CIA) World 
Factbook.
    (4) The country's GDP per capita and U.S. GDP per capita selected 
from the CIA World Factbook must be for the same year.
    (5) CMS identifies countries with a GDP per capita that is at least 
60 percent of the U.S. GDP per capita by dividing the GDP per capita 
for a country by the U.S. GDP per capita and assessing the results.
    (c) Identification of international data sources. (1) CMS obtains 
data from one or more international drug pricing information data 
sources for purposes of identifying available international drug 
pricing information for the countries specified in paragraph (b) of 
this section.
    (2) Such data sources must, as determined by CMS--
    (i) Utilize a standardized method for identifying drugs across 
countries within that data source, such as using internationally 
recognized scientific and nonproprietary product names;
    (ii) Utilize a standard method for identifying drug forms that at a 
minimum distinguishes among injectable, oral, and other forms of a 
drug; and
    (iii) Be maintained by an organization that seeks to limit the lag 
inherent in data to no more than 180 days from the end of the calendar 
quarter for which drug pricing information is compiled to the time that 
the organization makes such updates available to users of the database.
    (iv) Contains international drug pricing information stated in U.S. 
currency, such as the following:
    (A) Sales data, which may be based on ex-manufacturer prices 
(sometimes called ex-factory prices) that represent actual or 
calculated prices paid to the manufacturer by wholesalers and other 
distributors, or retail prices that represent actual or calculated 
sales for retail purchasers, or prices paid by other purchasers in the 
distribution channels.
    (B) Volume data (for example, number of packages or units sold).
    (C) List prices.
    (v) Have mechanisms in place to maintain, update, and correct, if 
necessary, the information on international drug pricing in the data 
source on at least a quarterly basis.
    (3) For each MFN Model drug for a calendar quarter, CMS selects a 
data source using the following hierarchy.
    (i) The data source contains sales and volume data for the 
applicable ASP calendar quarter from at least one country described in 
paragraph (b) of this section.
    (ii) The data source does not have sales and volume data for the 
applicable ASP calendar quarter, but contains sales and volume data for 
any prior ASP calendar quarter beginning on or after October 1, 2019 
from at least one country described in paragraph (b) of this section. 
If sales and volume data from a prior ASP calendar quarter are used, 
CMS uses sales and volume data from the most recent ASP calendar 
quarter for which both sales and volume data are available.
    (iii) The extracted data used by CMS to determine the most recent 
MFN Price used to calculate an MFN Drug Payment Amount posted on the 
MFN Model website.

[[Page 76253]]

    (iv) The data source contains ex-manufacturer price data for the 
applicable ASP calendar quarter from at least one country described in 
paragraph (b) of this section.
    (v) The data source contains list price data for the applicable ASP 
calendar quarter from at least one country described in paragraph (b) 
of this section.
    (vi) If there is more than one data source for an ASP calendar 
quarter, for each MFN Model drug, CMS selects the data source at the 
highest level of the hierarchy that contains information from the 
highest number of countries described in paragraph (b) of this section 
and, if available, incorporates discounts and rebates into its drug 
pricing information, and uses this data source to calculate the MFN 
Price as described in Sec.  513.210(b).

Subpart C--Payment Process and Methodology


Sec.  513.200  Payment process and beneficiary cost-sharing.

    (a) General. For purposes of the MFN Model, the allowed MFN Drug 
Payment Amount does not exceed the billed amount on the claim for the 
MFN Model drug.
    (b) Model-specific billing instructions. MFN participants submit 
claims for MFN Model drugs to the applicable MAC in the form and manner 
specified by CMS in model-specific billing instructions.
    (c) Beneficiary cost-sharing. Beneficiary coinsurance does not 
apply to the portion of the allowed payment amount for an MFN Model 
drug that is determined under Sec.  513.220.


Sec.  513.210   Model payment methodology for MFN Model drugs.

    (a) Payment amount. The total allowed payment amount for an MFN 
Model drug furnished to an MFN beneficiary by an MFN participant on a 
given date of service within a calendar quarter is determined in 
accordance with this section. The total allowed payment equals--
    (1) For each billing unit in the HCPCS code descriptor of the MFN 
Model drug, the MFN Drug Payment Amount determined in accordance with 
paragraphs (b), (c) and (d) of this section, as applicable, where the 
allowed MFN Drug Payment Amount does not exceed the billed amount on 
the claim for the MFN Model drug as described in Sec.  513.200(a); and
    (2) The alternative add-on payment determined under Sec.  513.220.
    (b) Calculation of the MFN Drug Payment Amount with Available 
International Drug Pricing Data. CMS selects an available international 
drug pricing information data source described in Sec.  513.140(c) for 
at least one country specified in Sec.  513.140(b) for an MFN Model 
drug, and calculates, in advance of each calendar quarter for a 
performance year, the applicable MFN Drug Payment Amount for one 
billing unit of an MFN Model drug using the following steps:
    (1) Available international drug pricing data. (i) For the MFN 
Model drug, using the data source selected in accordance with Sec.  
513.140(c)(3) (except for a data source described in Sec.  
513.140(c)(3)(iii)), CMS identifies available international drug 
pricing data for the MFN Model drug, by aligning the MFN Model drug's 
HCPCS code long description (including dosage form) with the data 
sources' standardized method for identifying scientific names or 
nonproprietary names and dosage formulations, as applicable. CMS 
extracts available drug pricing data for the countries specified in 
Sec.  513.140(b) from the selected international drug pricing 
information data source. CMS uses the extracted data that have complete 
package size information and only for dosage formulations that could be 
described by the MFN Model drug's HCPCS code descriptor, as determined 
by CMS. If a data source described in Sec.  513.140(c)(3)(iii) is 
selected, CMS uses such extracted data.
    (ii) When international drug pricing data with sales and volume 
data are available, CMS excludes from the calculation of the unadjusted 
country-level price under paragraph (b)(2) of this section 
international drug pricing data without both sales and volume data, 
with less than $1,000 in quarterly sales (expressed as U.S. currency), 
or with less than 1,000 units in quarterly volume.
    (iii) CMS converts the extracted volume data to the MFN Model 
drug's HCPCS code billing unit level, as applicable.
    (iv) CMS adjusts the extracted volume data, as applicable, before 
converting the extracted volume data to the MFN Model drug's HCPCS code 
billing unit level when the data source shows the package size of a 
drug product that is inconsistent with the manufacturer's information 
about that product, as determined by CMS.
    (v) CMS limits the number of HCPCS code billing units when--
    (A) The package labeling indicates a limited amount of drug is to 
be used from the package; and
    (B) The HCPCS code dosage is per dose.
    (2) Calculate the unadjusted country-level price for the MFN Model 
drug by country.
    (i) Using the drug pricing data extracted and adjusted in 
accordance with paragraph (b)(1) of this section, CMS calculates the 
unadjusted country-level price for the MFN Model drug by country, using 
the calculation that is applicable.
    (ii) If an international drug pricing information data source with 
sales and volume data is used, the applicable calculation is as 
follows:
    (A) CMS sums the adjusted volume data (as specified in paragraph 
(b)(1)(iii) of this section) for the drug.
    (B) CMS sums the total sales for the drug (that remain after 
performing the exclusions in paragraph (b)(1)(ii) of this section).
    (C) CMS divides the sum determined in paragraph (b)(2)(ii)(B) of 
the section by the sum determined in paragraph (b)(2)(ii)(A) of this 
section, resulting in an average price per unit of drug, where the unit 
of drug is the same as the HCPCS code billing unit.
    (iii) If an international drug pricing information data source with 
ex-manufacturer or list prices is used, the applicable calculation is 
as follows:
    (A) For each extracted ex-manufacturer or list price, CMS 
calculates the number of HCPCS billing units in the package by dividing 
the amount of drug in the package by the amount of drug represented in 
one HCPCS billing unit.
    (B) CMS divides the ex-manufacturer or list price, as applicable, 
by the number of HCPCS billing units in the package, resulting in a 
price per unit of drug where the unit of drug is the same as the HCPCS 
code billing unit.
    (C) CMS sums the price per unit of drug calculated in paragraph 
(c)(3)(iii)(B) of this section.
    (D) CMS divides the sum calculated in paragraph (c)(3)(iii)(C) of 
this section by the number of ex-manufacturer or list prices that were 
summed in paragraph (c)(3)(iii)(C) of this section, resulting in an 
average price per unit of drug where the unit of drug is the same as 
the HCPCS code billing unit.
    (iv) CMS performs the applicable calculation for each country 
specified in Sec.  513.140(b) for which international drug pricing 
information is available in the selected data source.
    (3) Calculate the GDP adjuster for each country. (i) CMS calculates 
the GDP adjuster by dividing the country's GDP per capita by the U.S. 
GDP per capita for the same year.
    (ii) In cases where the resulting ratio exceeds 1.0, the GDP 
adjuster is set to 1.0.

[[Page 76254]]

    (iii) Subject to the limitations specified in paragraph (b)(3)(iv) 
of this section, the GDP per capita for a country is the most recent 
estimate of GDP per capita based on purchasing power parity for that 
country available in the CIA World Factbook at the end of the 
applicable ASP calendar quarter.
    (iv) Limitations. (A) The country's GDP per capita and U.S. GDP per 
capita must be for the same year.
    (B) The GDP per capita used must be for the same year as the data 
used to calculate the unadjusted country-level price, if available, or 
the most recent earlier year available.
    (4) Apply the GDP adjuster to calculate the GDP-adjusted country-
level price. CMS applies the applicable GDP adjuster identified in 
paragraph (b)(3) of this section to each unadjusted country-level price 
identified in paragraph (b)(2) of this section to calculate the GDP-
adjusted country-level price by dividing each unadjusted country-level 
price by the applicable GDP adjuster.
    (5) Identify the lowest GDP-adjusted country-level price. CMS 
identifies the lowest GDP-adjusted country-level price for the MFN 
Model drug. Except as provided in paragraph (b)(7) of this section, the 
price identified is the MFN Model drug's MFN Price.
    (6) Identify Applicable ASP. CMS identifies the applicable ASP for 
the applicable quarter.
    (7) Compare the MFN Price to the applicable ASP. CMS compares the 
price determined in paragraph (b)(5) of this section to the applicable 
ASP identified in paragraph (b)(6) of this section. The MFN Price 
equals the applicable ASP if the applicable ASP is less than the price 
determined in paragraph (b)(5) of this section.
    (8) Phase-in. CMS identifies the applicable phase-in formula based 
on the applicable performance year as follows:
    (i) Performance year 1: 75 percent applicable ASP and 25 percent 
MFN Price.
    (ii) Performance year 2: 50 percent applicable ASP and 50 percent 
MFN Price.
    (iii) Performance year 3: 25 percent applicable ASP and 75 percent 
MFN Price.
    (iv) Performance year 4: 100 percent MFN Price.
    (v) Performance year 5: 100 percent MFN Price.
    (vi) Performance year 6: 100 percent MFN Price.
    (vii) Performance year 7: 100 percent MFN Price.
    (9) Final calculation steps. (i) CMS applies the applicable phase-
in formula to the applicable ASP and the MFN Price. Subject to any 
applicable adjustments as provided in paragraph (d) of this section, 
the amount determined in this paragraph is the MFN Drug Payment Amount.
    (ii) Subject to the limitation in paragraph (b)(iii) in this 
section, CMS recalculates the MFN Drug Payment Amounts for prior 
quarters when revised international drug pricing information is 
available in the data source that was used to calculate the MFN Price 
and applicable ASP updates are available from CMS. CMS prospectively 
applies the recalculations in the quarterly update following the 
availability of revised international drug pricing information and ASP 
updates.
    (iii) MFN Drug Payment Amounts may be recalculated for the prior 
four calendar quarters of the model.
    (c) Frequency of MFN Drug Payment Amount updates. CMS updates the 
MFN Drug Payment Amounts on a calendar quarter basis. CMS publishes the 
quarterly MFN Drug Payment Amounts on the MFN Model website in advance 
of the calendar quarter in which the MFN Drug Payment Amounts apply, 
along with any recalculated MFN Drug Payment Amounts for prior 
quarters.
    (d) Exceptions. (1) Payment for MFN Model drugs for which no 
international drug pricing data are available. If, as of the first 
calendar quarter during which an MFN Model drug has been included in 
the MFN Model Drug HCPCS Codes List in accordance with Sec.  513.130, 
no international sales, volume or pricing information meeting the 
requirements described in Sec.  513.140(c)--including data used by CMS 
to determine the most recent MFN Price used to calculate an MFN Drug 
Payment Amount posted on the MFN Model website--is available from any 
country described in Sec.  513.120(b) for any calendar quarter 
beginning on or after October 1, 2019 through the applicable quarter, 
the MFN Drug Payment Amount is the applicable ASP.
    (2) Payment for MFN Model drugs that are in short supply. If an MFN 
Model drug is reported as ``Currently in Shortage'' by FDA, beginning 
with the first day of the next calendar quarter after the date on which 
it is reported in shortage, the MFN Drug Payment Amount is the 
applicable ASP. CMS calculates payment in accordance with paragraph (b) 
of this section as of the first day of the calendar quarter after the 
date upon which the drug is no longer reported as ``Currently in 
Shortage'' by FDA.
    (3) Adjustment to phase-in formula. (i) CMS accelerates the phase-
in of the MFN Price by 5 percentage points at the next quarterly update 
to calculate the MFN Drug Payment Amount for the MFN Model drug where 
both of the following conditions are met:
    (A) There is a greater cumulative percentage increase in either the 
applicable ASP or any of the monthly U.S. list prices for the NDCs 
assigned to the MFN Model drug's HCPCS code compared to the cumulative 
percentage increase in the CPI-U.
    (B) There is a greater cumulative percentage increase in either the 
applicable ASP or any of the monthly U.S. list prices for the NDCs 
assigned to the MFN Model drug's HCPCS code compared to the cumulative 
percentage increase in the MFN Price.
    (C) For purposes of paragraphs (d)(3)(i)(A) and (B) of this 
section, the cumulative percentage increase means the cumulative 
percentage change from the end of the baseline to the end of the 
applicable ASP calendar quarter.
    (D) The baseline in paragraph (d)(3)(i)(C) of this section for an 
MFN Model drug is the ASP calendar quarter for the applicable ASP for 
the first quarter of performance year 1. If there is not an applicable 
ASP for the first quarter of performance year 1 for an MFN Model drug, 
the baseline for that MFN Model drug is the ASP calendar quarter for 
the first applicable ASP based on the manufacturer's average sales 
price for that MFN Model drug that occurs after the ASP calendar 
quarter for the applicable ASP for the first quarter of performance 
year 1.
    (ii) For purposes of paragraph (d)(3)(i) of this section, if the 
cumulative percentage increase in CPI-U or MFN Price is negative, CMS 
uses zero as the cumulative percentage increase in CPI-U or MFN Price, 
as applicable.
    (iii) The application of an acceleration of the phase-in formula 
continues for the duration of the model performance period.
    (iv) CMS applies an additional acceleration of the phase-in formula 
for each calendar quarter where the conditions specified in paragraph 
(i) are met.
    (4) Adjustment for rapid increases in the applicable ASP or any 
monthly U.S. list prices beyond inflation and MFN Price after the full 
phase-in of the MFN Price. If the conditions described in paragraphs 
(d)(3)(i)(A) and (B) of this section are met after the full phase-in of 
the MFN Price for an MFN Model drug, for each calendar quarter 
thereafter, CMS decreases the MFN Drug Payment Amount equal to largest 
difference in the cumulative percentage increase in the applicable ASP 
or any of the monthly U.S. list prices for the NDCs

[[Page 76255]]

assigned to the MFN Model drug's HCPCS code compared to the cumulative 
percentage increase in the CPI-U and in the MFN Price, respectively, 
determined quarterly.
    (5) Limitation on MFN Drug Payment Amount. The MFN Drug Payment 
Amount cannot exceed the non-model drug payment amount for claim lines 
submitted with the JG modifier (or any successor modifier used to 
identify drugs purchased under the 340B program) after removing any 
add-on amount, if applicable.
    (e) Blood clotting factor furnishing fee. When applicable, the 
blood clotting furnishing fee under Sec.  410.63(c) of this chapter is 
payable along with the MFN Drug Payment Amount.


Sec.  513.220   Model alternative add-on payment.

    (a) Payment amount. (1) The total allowed alternative add-on 
payment amount for a separately payable dose of an MFN Model drug 
furnished to an MFN beneficiary by an MFN participant on a given date 
of service within a calendar quarter is determined in accordance with 
this section.
    (2) The total allowed alternative add-on payment amount for a claim 
line does not exceed the billed amount on that claim line.
    (b) Calculation of the per-dose alternative add-on payment amount. 
CMS calculates the per-dose alternative add-on payment for performance 
year 1, quarter 1 for MFN Model drugs using the following steps:
    (1) CMS identifies available Medicare Part B fee-for-service final 
action claims lines, with dates of service in 2019, for drugs on the 
initial MFN Model HCPCS Codes List described in Sec.  513.130(a)(1), 
excluding claims for providers and suppliers specified in Sec.  
513.100(c), and claims specified in Sec.  513.100(d), that were 
furnished by Medicare-participating providers and suppliers, have a 
separately paid allowed charge greater than $0, and for which Medicare 
Part B was the primary payer. If a HCPCS code on the initial MFN Model 
HCPCS Codes List was not in use during any calendar quarter in 2019, 
CMS uses the HCPCS code that was applicable for the MFN Model drug 
during 2019.
    (2) CMS identifies the applicable ASP for each calendar quarter of 
2019 for the drugs (by HCPCS code as specified in paragraph (b)(1) of 
this section) included on the initial MFN Model HCPCS Codes List. In 
the case of a biosimilar biological product, the applicable ASP for the 
reference biological product is identified and used in paragraph (b)(3) 
of this section.
    (3) CMS multiplies the number of units billed for each claim line 
described in paragraph (b)(1) of this section by 6.1224 percent of the 
applicable ASP identified in paragraph (b)(2) of this section for the 
HCPCS code on the claim line and date of service.
    (4) CMS sums the products calculated in paragraph (b)(3) of this 
section for all claim lines for each MFN Model drug to calculate the 
total add-on spending amount for each MFN Model drug.
    (5) CMS sums the amounts calculated in paragraph (b)(4) of this 
section to calculate the total pooled add-on spending amount for all 
MFN Model drugs.
    (6) CMS divides the amount calculated in paragraph (b)(5) of this 
section by the total number of claim lines retained in paragraph (b)(1) 
of this section, excluding claim lines billed with the JW modifier.
    (7) CMS trends the amount calculated in paragraph (b)(6) of this 
section forward to the applicable ASP calendar quarter for quarter 1 of 
performance year 1 using the percentage change in CPI-U from July 2019 
through October 2020.
    (c) Frequency of alternative add-on payment amount updates. For 
each calendar quarter after quarter 1 of performance year 1, CMS 
updates the alternative add-on payment by applying a cumulative 
inflation factor based on the cumulative percentage change in CPI-U 
from October 2020 through the first month of the prior calendar 
quarter. If the cumulative percentage change in the CPI-U is negative, 
CMS uses an inflation factor of 1.
    (d) Limitation on the alternative add-on payment. The alternate 
add-on payment is not payable for claim lines billed--
    (1) With the JW modifier; or
    (2) By MFN participants that receive an alternative add-on payment 
for an MFN Model drug under any other model authorized by section 1115A 
of the Act that tests an alternative approach to the add-on portion of 
payment for Medicare Part B drugs.


Sec.  513.230   Financial hardship exemptions, request process, and 
reconciliation payment.

    (a) General. For purposes of the MFN Model, a financial hardship 
exemption for a performance year may be granted to an MFN participant 
by CMS, in its sole discretion and not subject to appeal, when the 
provisions in this section are met.
    (b) Request for financial hardship exemption. To be eligible for a 
financial hardship exemption, the MFN participant must submit a request 
for financial hardship exemption in the form and manner and with the 
content specified by CMS, including without limitation the requirements 
of this paragraph (b).
    (1) Timing and form of request. The MFN participant must submit its 
request for a financial hardship exemption to CMS in accordance the 
submission process posted on the MFN Model website and such request 
must be submitted within 60 calendar days following the end of the 
performance year for which the MFN participant seeks a financial 
hardship exemption.
    (2) Request content. The MFN participant's request a financial 
hardship exemption must include, at a minimum, all of the following:
    (i) Evidence of methods used to obtain each MFN Model drug that was 
furnished by the MFN participant during the performance year to any 
patient.
    (ii) Average net acquisition cost for each MFN Model drug 
(inclusive of all on- and off-invoice discounts or adjustments, and any 
other price concessions related to the purchase of the MFN Model drug) 
that was furnished by the MFN participant during the performance year 
to MFN beneficiaries.
    (iii) Average net acquisition cost for each MFN Model drug 
(inclusive of all on- and off-invoice discounts and adjustments, and 
any other price concessions related to the purchase of the MFN Model 
drug) that was furnished by the MFN participant during the performance 
year to patients who were not MFN beneficiaries.
    (iv) Statement of any remuneration received by the MFN participant 
from manufacturers of MFN Model drugs, wholesalers, and distributors 
that is not reflected in the MFN participant's average net acquisition 
costs with a justification of why such remuneration should not be 
treated as a price concession related to the purchase of an MFN Model 
drug.
    (v) Administrative information, including: MFN participant's name, 
TIN or CCN (as applicable), contact name, phone number, and email 
address.
    (vi) The MFN participant's attestation that:
    (A) The MFN participant experienced a reduction in Medicare Part B 
FFS payments for separately payable drugs on a per beneficiary basis 
during the performance year as compared to the prior year (that is, the 
four calendar quarters immediately preceding the performance year) due 
to its inability to obtain one or more of the MFN Model drugs at or 
below the MFN Model Payments for such drugs during the performance 
year;

[[Page 76256]]

    (B) The MFN participant has not received and will not receive any 
remuneration from manufacturers of MFN Model drugs, wholesalers, and 
distributors related to the purchase of an MFN Model drug that was 
furnished by the MFN participant during the performance year that is 
not reflected in the MFN participant's submission; and
    (C) The MFN participant submission is true, accurate and complete.
    (c) Standard of review. (1) Incomplete requests for a financial 
hardship exemption, as determined by CMS, are not reviewed.
    (2) CMS grants a financial hardship exemption to an MFN participant 
for a performance year, if the agency in its sole discretion determines 
the following requirements have been met:
    (i) The MFN participant submits a timely, complete request for 
financial hardship exemption in accordance with the requirements of 
this section which in the sole discretion of CMS demonstrates all of 
the following:
    (A) The MFN participant exhausted all reasonable methods to obtain 
MFN Model drugs at or below the MFN Model Payment for such drugs during 
the performance year.
    (B) The MFN participant's average net acquisition cost for each MFN 
Model drug (including invoices and off-invoice discounts or 
adjustments) that was furnished by the MFN participant during the 
performance year to patients who were not MFN beneficiaries was not 
less than the MFN participant's average net acquisition costs for such 
MFN Model drug (including invoices and off-invoice discounts or 
adjustments) that was furnished by the MFN participant during the 
performance year to MFN beneficiaries.
    (C) Any remuneration the MFN participant received from 
manufacturers of MFN Model drugs, wholesalers, and distributors that 
was not reflected in the MFN participant's average net acquisition 
costs was not a price concession related to the purchase of an MFN 
Model drug.
    (ii) The MFN participant's excess reduction amount per beneficiary 
(as determined in paragraph (d)(6) of this section), is greater than 
zero.
    (d) Excess reduction amount per beneficiary. CMS calculates the MFN 
participant's excess reduction amount per beneficiary using available 
final action claims data where Medicare was the primary payer that is 
estimated to be more than 90 percent complete in accordance with the 
following steps:
    (1) CMS calculates, separately for dates of service within the 
performance year and prior year, the MFN participant's total allowed 
charges for separately payable Medicare Part B drugs, and the total 
number of beneficiaries that had at least one claim for a service 
furnished by the MFN participant with a Medicare Part A or Medicare 
Part B allowed charge greater than $0.
    (2) CMS divides the MFN participant's total allowed charges for 
separately payable Medicare Part B drugs for dates of service within 
the performance year by the total number of beneficiaries that had at 
least one claim for a service furnished by the MFN participant with a 
Medicare Part A or Medicare Part B allowed charge greater than $0 with 
a service date within the performance year, to calculate the MFN 
participant's average per beneficiary total allowed charges for 
separately payable Medicare Part B drugs for the performance year.
    (3) CMS divides the MFN participant's total allowed charges for 
separately payable Medicare Part B drugs for dates of service within 
the prior year by the total number of beneficiaries that had at least 
one claim for a service furnished by the MFN participant with a 
Medicare Part A or Medicare Part B allowed charge greater than $0 with 
a service date within the prior year, to calculate the MFN 
participant's average per beneficiary total allowed charges for 
separately payable Medicare Part B drugs for the prior year.
    (4) CMS subtracts the MFN participant's average per beneficiary 
total allowed charges for separately payable Medicare Part B drugs for 
the performance year (as calculated in paragraph (d)(2) of this 
section) from the MFN participant's average per beneficiary total 
allowed charges for separately payable Medicare Part B drugs for the 
prior year (as calculated in paragraph (d)(3) of this section).
    (5) CMS calculates 25 percent of the MFN participant's total 
allowed charges for all Medicare Part A and Part B claims with dates of 
service within the prior year and divides that amount by the total 
number of beneficiaries that had at least one claim for a service 
furnished by the MFN participant with a Medicare Part A or Medicare 
Part B allowed charge greater than $0 with a service date within the 
prior year, to calculate 25 percent of the MFN participant's average 
per beneficiary total allowed charges for all Medicare Part A and Part 
B claims with dates of service within the prior year.
    (6) CMS subtracts 25 percent of the MFN participant's average per 
beneficiary total allowed charges for all Medicare Part A and Part B 
claims with dates of service within the prior year (as calculated in 
paragraph (d)(5) of this section) from the difference calculated in 
paragraph (d)(4) of this section, to calculate the MFN participant's 
excess reduction amount per beneficiary.
    (e) Reconciliation payment. (1) If CMS in its sole discretion 
grants a financial hardship exemption to an MFN participant for a 
performance year, CMS provides such MFN participant a reconciliation 
payment for the performance year that equals the amount calculated by 
multiplying the excess reduction amount per beneficiary specified in 
paragraph (d)(6) of this section by the total number of beneficiaries 
that had at least one claim for a service furnished by the MFN 
participant with a Medicare Part A or Medicare Part B allowed charge 
greater than $0 with a service date within the performance year.
    (2) The amount of a reconciliation payment provided in accordance 
with this section is--
    (i) Not subject to appeal;
    (ii) Not subject to beneficiary cost-sharing, including any 
deductible or coinsurance; and
    (iii) Made by CMS (or a CMS contractor) as soon as practical.

Subpart D--[Reserved]

Subpart E--Quality Strategy, Beneficiary Protections, and 
Compliance Activities


Sec.  513.400   Quality measures.

    (a) General. Quality measures do not adjust model payments to MFN 
participants and are used for monitoring purposes.
    (b) Collection of quality measures. (1) CMS administers a patient 
experience survey to a sample of beneficiaries who receive an MFN Model 
drug. A sample of non-MFN beneficiaries may also be surveyed.
    (2) If during the MFN Model CMS determines that the quality 
measures specified in paragraph (b) of this section are not sufficient 
to adequately monitor the quality of care that MFN beneficiaries are 
receiving from MFN participants or that MFN participants are providing, 
CMS may specify additional measures. CMS applies the following criteria 
when specifying additional quality measures:
    (i) Additional measures are among one or more of the following 
categories:
    (A) Patient experience of care.
    (B) Patient activation
    (C) Shared decision making.
    (D) Adherence.
    (E) Utilization.
    (F) Process measures.

[[Page 76257]]

    (ii) Additional measures will not add significant burden to MFN 
participants or beneficiaries.
    (iii) Additional measures utilize an instrument that CMS has used 
previously in a model to adjust payment or for monitoring or 
evaluation.


Sec.  513.410   Beneficiary protections.

    (a) Beneficiary choice.
    (1) MFN participants must not restrict beneficiaries' ability to 
choose to receive care from any Medicare participating provider or 
supplier or any provider or supplier who has opted out of Medicare.
    (2) The MFN participant must not commit any act or omission, nor 
adopt any policy that inhibits a beneficiary from exercising his or her 
freedom to choose to receive care from any Medicare participating 
provider or supplier or any provider or supplier who has opted out of 
Medicare. Notwithstanding the foregoing, MFN participants may 
communicate to beneficiaries the benefits of receiving care from an MFN 
participant, if otherwise consistent with the requirements of this part 
and applicable law.
    (b) Appeals. An MFN beneficiary and his or her assignees retain 
their right to appeal claims in accordance with part 405 subpart I of 
this chapter.
    (c) Availability of services. MFN participants must not take any 
action to select or avoid treating beneficiaries based on their 
diagnoses, care needs, income levels or other factors that would render 
the beneficiary an ``at-risk beneficiary'' as defined at Sec.  425.20 
of this chapter.


Sec.  513.420   Monitoring and compliance activities.

    (a) Compliance with laws. (1) Agreement to comply. The MFN 
participant must comply with all applicable laws and regulations.
    (2) Notification. The MFN participant must notify CMS within 15 
days after becoming aware that the MFN participant is under 
investigation or has been sanctioned by the federal, state, or local 
government, or any licensing authority (including, without limitation, 
the imposition of program exclusion, debarment, civil monetary 
penalties, corrective action plans, and revocation of Medicare billing 
privileges).
    (b) CMS monitoring and compliance activities. (1) CMS conducts 
monitoring activities to ensure compliance by MFN participants with the 
terms of the MFN Model, to obtain timely information about the effects 
of the MFN Model on MFN beneficiaries, providers, suppliers, and on the 
Medicare program and to facilitate real time identification and 
response to potential issues. Such monitoring activities may include, 
without limitation, the following:
    (i) Documentation requests sent to the MFN participant including, 
without limitation, surveys and questionnaires.
    (ii) Audits of claims data, medical records, and other data from 
the MFN participant.
    (iii) Interviews with any individual or entity participating in the 
MFN Model including members of the MFN participant's leadership, 
management, and staff.
    (iv) Interviews with beneficiaries and their caregivers.
    (v) Site visits to the MFN participants, performed in a manner 
consistent with Sec.  513.420(c).
    (vi) Tracking patient complaints and appeals.
    (2) In conducting monitoring and oversight activities, CMS or its 
designees may use any relevant data or information including without 
limitation, all Medicare claims submitted for items or services 
furnished to beneficiaries in the MFN Model.
    (3) The MFN participant must cooperate with evaluation and 
monitoring activities as may be necessary to enable CMS to evaluate the 
MFN Model in accordance with section 1115A(b)(4) of the Act and to 
conduct monitoring activities under this section.
    (c) Site visits. (1) To the extent practicable, CMS or its designee 
provides the MFN participant with no less than 15 days advance notice 
of any site visit. To the extent practicable, CMS attempts to 
accommodate a request for particular dates in scheduling site visits. 
However, the MFN participant may not request a date that is more than 
60 days after the date of the initial site visit notice from CMS.
    (2) The MFN participant must ensure that personnel with the 
appropriate responsibilities and knowledge associated with the purpose 
of the site visit are available during all site visits.
    (3) Notwithstanding the foregoing, CMS may perform unannounced site 
visits at all physical locations of the MFN participant at any time to 
investigate concerns about the health or safety of beneficiaries or 
other patients or other program integrity issues.
    (4) Nothing in this part must be construed to limit or otherwise 
prevent CMS from performing site visits permitted or required by 
applicable law.
    (d) Right to correct. If CMS discovers that it has made or received 
an incorrect model-specific payment under the terms of the MFN Model, 
CMS may make payment to, or demand payment from, the MFN participant.


Sec.  513.430   Audits and record retention.

    (a) Right to audit. The Federal Government, including CMS, HHS, and 
the Comptroller General, or their designees, has the right to audit, 
inspect, investigate, and evaluate any documents and other evidence 
regarding implementation of the MFN Model.
    (b) Access to records. MFN participants must maintain and give the 
Federal Government, including CMS, HHS, and the Comptroller General, or 
their designees, access to all such documents and other evidence 
sufficient to enable the audit, evaluation, inspection, or 
investigation of the implementation of the MFN Model, including without 
limitation, documents and other evidence regarding the following:
    (1) The MFN participant's compliance with the terms of the MFN 
Model, including this subpart.
    (2) Quality measure information and the quality of services 
performed under the terms of the MFN Model, including this subpart.
    (3) Patient safety.
    (4) The accuracy of model-specific payments made under the MFN 
Model.
    (5) Utilization of items and services furnished under the MFN 
Model.
    (6) Other program integrity issues.
    (c) Record retention. The MFN participant must maintain the 
documents and other evidence described in paragraph (b) of this section 
and other evidence for a period of 6 years from the last payment 
received by the MFN participant under the MFN Model or from the date of 
completion of any audit, evaluation, inspection, or investigation, 
whichever is later, unless--
    (1) CMS determines there is a special need to retain a particular 
record or group of records for a longer period and notifies the MFN 
participant at least 30 days before the normal disposition date; or
    (2) There has been a termination, dispute, or allegation of fraud 
or similar fault against the MFN participant, in which case the records 
must be maintained for an additional 6 years from the date of any 
resulting final resolution of the termination, dispute, or allegation 
of fraud or similar fault.


Sec.  513.440   Enforcement authority.

    (a) Remedial action--(1) Grounds for remedial action. In addition 
to any other grounds for remedial action that are permitted under the 
terms of this part, CMS may take one or more of the remedial actions 
set forth in paragraph (a)(2) of this section if CMS determines,

[[Page 76258]]

in CMS' sole discretion, that an MFN participant:
    (i) Has failed to comply with any applicable Medicare program 
requirement, rule, or regulation.
    (ii) Has failed to comply with any of the terms of the MFN Model, 
including applicable requirements of this part.
    (iii) Has systematically engaged in the under delivery or over 
delivery of an MFN Model drug.
    (iv) Has taken any action that threatens the health or safety of an 
MFN beneficiary or other patient.
    (v) Has undergone a change of control that presents a program 
integrity risk.
    (vi) Has submitted false data or made false representations, 
warranties, certifications or attestations in connection with any 
aspect of the MFN Model.
    (vii) Has avoided at-risk beneficiaries, as this term is defined in 
Sec.  425.20 of this chapter.
    (viii) Has avoided patients on the basis of payer status.
    (ix) Is subject to sanctions or final actions of an accrediting 
organization or Federal, State, or local government agency.
    (x) Takes any action that CMS determines for program integrity 
reasons is not in the best interests of the MFN Model or the Medicare 
program, or fails to take any action that CMS determines for program 
integrity reasons should have been taken to further the best interests 
of the MFN Model or Medicare program.
    (xi) Is subject to investigation by HHS (including the HHS Office 
of Inspector General (OIG)) or the Department of Justice due to an 
allegation of fraud or significant misconduct, including being subject 
to the filing of a complaint, filing of a criminal charge, being 
subject to an indictment, being named as a defendant in a False Claims 
Act qui tam matter in which the Federal Government has intervened, or 
similar action;
    (xii) Is the subject of administration enforcement action imposed 
by CMS; or
    (xiii) Has failed to demonstrate improved performance following any 
remedial action imposed under this section.
    (2) Taking remedial actions. If CMS determines that one or more 
grounds for remedial action described in paragraph (a)(1) of this 
section exist, CMS make take one or more of the following remedial 
actions:
    (i) Notifying the MFN participant of the violation.
    (ii) Requiring the MFN participant to provide additional 
information to CMS or its designees.
    (iii) Requiring the MFN participant to develop and implement a 
corrective action plan in a form and manner and by a deadline specified 
by CMS.
    (iv) Subjecting the MFN participant to additional monitoring, 
auditing, or both.
    (v) Removing the MFN participant from the MFN Model.
    (vi) Recouping model-specific payments.
    (vii) Other action as may be permitted under the terms of this 
part.
    (b) OIG authority. Nothing contained in the terms of the MFN Model 
or this part limits or restricts the authority of the HHS Office of 
Inspector General or any other Federal Government authority or agency, 
including its authority to audit, evaluate, investigate, or inspect 
model participant for violations of any statutes, rules, or regulations 
administered by the Federal Government.


Sec.  513.450   Limitations on review.

    There is no administrative or judicial review under sections 1869 
or 1878 of the Act or otherwise for any of the following:
    (a) The selection of models for testing or expansion under section 
1115A of the Act.
    (b) The selection of organizations, sites, or participants, 
including MFN participants, to test the MFN Model, including a decision 
by CMS to remove an MFN participant from the MFN Model.
    (c) The elements, parameters, scope, and duration of such MFN Model 
for testing or dissemination, including without limitation all of the 
following:
    (1) The selection of the model geographic area for the MFN Model by 
CMS.
    (2) The selection of MFN Model drugs by CMS.
    (3) The selection of included international data, including 
selection of countries, international drug pricing databases, and 
international drug pricing data.
    (d) Determinations regarding budget neutrality under section 
1115A(b)(3) of the Act.
    (e) The termination or modification of the design and 
implementation of an MFN Model under section 1115A(b)(3)(B) of the Act.
    (f) Determinations about expansion of the duration and scope of the 
MFN Model under section 1115A(c) of the Act, including the 
determination that the MFN Model is not expected to meet criteria 
described in paragraphs (c)(1) or (2) of such section.

Subpart F--Waivers


Sec.  513.500   Waivers of Medicare program requirements for purposes 
of testing the MFN Model.

    CMS waives the Medicare program requirements in the following 
provisions that are necessary solely for purposes of testing the MFN 
Model:
    (a) Sections 1833(t)(6) and 1833(t)(14) of the Act and Sec. Sec.  
419.62 and 419.64 of this chapter related to Medicare payment amounts 
for drugs and biologicals under the hospital outpatient prospective 
payment system (OPPS) as necessary to permit testing of an alternative 
payment amount for MFN Model drugs.
    (b) Section 1833(i)(2)(D) of the Act related to Medicare payment to 
ASCs for drugs and biologicals as necessary to permit testing of an 
alternative payment amount for MFN Model drugs.
    (c) Sections 1847A(b) and 1847A(c) of the Act and Sec. Sec.  
414.904 and 414.802 of this chapter related to use of the ASP-based, 
WAC-based, or other applicable payment methodology and calculation of 
manufacturers' ASP as necessary to permit testing of an alternative 
payment for MFN Model drugs and to exclude certain units of MFN Model 
drugs from manufacturers' ASPs.
    (d) Section 1833(a)(1) of the Act related to Medicare payment 
portion of the allowed payment amount for an included MFN Model drug 
that is determined under Sec.  513.220 as necessary to permit testing 
of an innovative payment approach for the alternative add-on payment 
amount.
    (e) Section 1833(a)(1)(S) of the Act related to Medicare payment 
for drugs and biologicals is 80 percent of the lesser of the actual 
charge or the payment amount established in section 1842(o) of the Act 
as necessary to permit testing of an innovative payment approach for 
the total allowable MFN Model payment as determined under subpart C.
    (f) Section 1833(a)(1)(G) of the Act related to the amounts paid 
with respect to facility services furnished in connection with certain 
surgical procedures and with respect to services furnished to an 
individual in an ASC must be 80 percent of the lesser of the actual 
charge for the services or the amount determined by the Secretary under 
such revised payment system as necessary to permit testing of an 
innovative payment approach for the total allowable MFN Model payment 
as determined under subpart C.
    (g) Section 1833(t) of the Act related to how beneficiary copayment 
is calculated under the OPPS as necessary to permit testing of an 
innovative

[[Page 76259]]

payment approach for the total allowable MFN Model payment as 
determined under subpart C of this part.
    (h) Section 1833(t)(9)(B) of the Act related to the requirement 
that Medicare account for adjustments to ensure that the amount of 
expenditures under the OPPS for the year does not increase or decrease 
from the estimated amount of expenditures under the OPPS that would 
have been made if the adjustments had not been made.

Subparts G through J--[Reserved]

Subpart K--Model Termination


Sec.  513.1000   Termination of the MFN Model.

    (a) CMS may terminate the MFN Model for reasons including, but not 
limited to, the following:
    (1) CMS determines that it no longer has the funds to support the 
MFN Model.
    (2) CMS terminates the model in accordance with section 
1115A(b)(3)(B) of the Act.
    (b) As specified in section 1115A(d)(2) of the Act, termination of 
the model in accordance with section 1115A(b)(3)(B) of the Act is not 
subject to administrative or judicial review.

    Dated: November 18, 2020.
Seema Verma,
Administrator, Centers for Medicare & Medicaid Services.
    Dated: November 18, 2020.
Alex M. Azar II,
Secretary, Department of Health and Human Services.
[FR Doc. 2020-26037 Filed 11-20-20; 4:15 pm]
BILLING CODE 4120-01-P