..
..
..
..
..
..
..
..
..
..
..
.

425 I Street, NW • Suite 701
Washington, DC 20001
202-220-3700 • Fax: 202-220-3759
www.medpac.gov
Francis J. Crosson, M.D., Chairman
Jon B. Christianson, Ph.D., Vice Chairman
James E. Mathews, Ph.D., Executive Director

January 3, 2018
Seema Verma, MPH
Administrator
Centers for Medicare & Medicaid Services
Department of Health and Human Services
200 Independence Avenue, SW
Washington, DC 20201
RE: CMS-4182-P
Dear Ms.Verma:
The Medicare Payment Advisory Commission (MedPAC) welcomes the opportunity to comment
on the Centers for Medicare & Medicaid Services (CMS) notice of proposed rulemaking entitled
“Medicare Program; Contract Year 2019 Policy and Technical Changes to the Medicare
Advantage, Medicare Cost Plan, Medicare Fee-for-Service, the Medicare Prescription Drug
Benefit Programs, and the PACE Program,” published in the Federal Register, vol. 82, no. 227,
pages 56336 to 56527. We appreciate your staff’s work on the notice, particularly considering the
competing demands on the agency.
This proposed rule includes provisions that would revise regulations for both the Medicare
Advantage program (Part C) and the Prescription Drug Benefit program (Part D). Our comments
focus on the following provisions:
•
•
•
•
•
•
•
•
•
•

Flexibility in the Medicare Advantage (MA) uniformity requirements
Meaningful differences in MA bid submissions and bid review
Coordination of enrollment and disenrollment through MA organizations and effective
dates of coverage and changes of coverage
Passive enrollment flexibilities to protect continuity of integrated care for dually eligible
beneficiaries
Establishing limitations for the Part D Special Election Period (SEP) for dually eligible
beneficiaries
MA and Part D Prescription Drug Plan (PDP) quality rating system
Changes to the days’ supply required by the Part D transition process
Expedited substitutions of certain generics and other midyear formulary changes
Treatment of follow-on biological products as generics for non-low-income subsidy (LIS)
catastrophic and LIS cost sharing
Eliminating the requirement to provide PDP Enhanced Alternative (EA) offerings with
meaningful differences

 Seema Verma
Administrator
Page 2
•
•
•

Request for information regarding the application of manufacturer rebates and pharmacy
price concessions to drug prices at the point of sale
Lengthening adjudication timeframes for Part D payment redeterminations and
independent review entity (IRE) reconsiderations
Removal of Quality Improvement Project (QIP) for MA organizations

Flexibility in the MA Uniformity Requirement
The Medicare statute requires MA organizations to offer their plans “at a uniform premium, with
uniform benefits and level of cost sharing throughout the plan's service area.”1 CMS has
historically interpreted the statute as requiring MA plans to offer all enrollees access to the same
benefits at the same level of cost sharing. Plan sponsors have asserted that this interpretation of the
statute was too narrow and precluded them from offering value-based insurance design (VBID)
products that could be used to better address chronic conditions.
In this rule, CMS is providing notice of a change to its previous interpretation of the statute. CMS
has determined that these statutory provisions permit MA organizations to reduce cost sharing for
certain covered benefits, offer specific tailored supplemental benefits, and offer lower deductibles
for enrollees that meet specific medical criteria, provided that similarly situated enrollees (that is,
all enrollees who meet the identified criteria) are treated the same. Reviews of plan designs would
still ensure that plan cost-sharing does not discriminate against high-cost beneficiaries. Also, CMS
notes that the new flexibility would still prohibit an MA plan from denying, limiting, or
conditioning the coverage or provision of a service or benefit based on health status-related
factors.
Comment
The Commission recommended in our March 2013 report to the Congress that the Secretary permit
MA plans greater flexibility to tailor benefits, so that benefits can vary based on the medical needs
of individuals with chronic conditions.2 Specifically, “under this new flexibility, MA plans could
vary the supplemental benefits, cost sharing for services and drugs, and provider networks for
chronically ill enrollees.” The proposed regulation uses similar language to argue for the new
interpretation of benefit uniformity. We agree that plans should be able to lower cost sharing to
help encourage beneficiaries to seek the most effective care, as long as the flexibility does not
result in excessive cost sharing for beneficiaries without special conditions.
Meaningful differences in MA bid submissions and bid review
CMS will approve bids submitted by an MA organization for multiple plans in an area only if the
plan benefit packages are substantially different with respect to key plan characteristics such as
premiums, cost sharing, or benefits offered. CMS promulgates annual meaningful difference
evaluation standards to determine substantial differences in benefit packages. The current
1

42 C.F.R. 422-100(d)(2)
Medicare Payment Advisory Commission. 2013. Report to the Congress: Medicare payment policy. Washington,
DC: MedPAC.

2

 Seema Verma
Administrator
Page 3
meaningful difference evaluation uses the results of a model that estimates enrollee out-of-pocket
(OOP) costs (the OOP model). For 2018, similar plans (same plan sponsor, same plan-type)
needed to have an OOP model estimated difference of $20 per month to meet the meaningful
difference requirement.
CMS proposes to eliminate this meaningful difference requirement beginning with MA bid
submissions for contract year (CY) 2019. CMS believes that the meaningful difference
requirement based on the OOP model is flawed, resulting in plans altering benefits in complicated
and confusing ways, and limiting the introduction of innovative benefits. CMS expects that
eliminating the meaningful difference requirement will improve the plan options available for
beneficiaries. CMS does not believe the number of similar plan options offered by the same MA
organization in each county will increase significantly or create confusion in beneficiary decisionmaking.
Comment
The Commission supports CMS’s goal of encouraging competition and plan flexibility. However,
while the current use of the OOP model as the only measure of meaningful differences between
MA plans is flawed, the Commission has been concerned that beneficiaries have more trouble
choosing plans when there are many similar plans offered.3 CMS acknowledges our concern, but
asserts that the removal of the meaningful difference standards will not lead to more similar plans
being offered. We nevertheless believe that CMS should maintain a quantifiable meaningful
difference standard for plan bids, but allow plans to seek waivers by providing alternate evidence
of meaningful differences. For example, CMS could require that if the OOP model meaningful
difference standard were not met, plan sponsors would have to provide stronger evidence that
beneficiaries would be able to easily distinguish between the sponsor’s offerings. Applying the
meaningful difference standard as leverage would provide CMS with tools to address any
confusion. We also suggest that more detailed information about differences among plans be
provided in Medicare Plan Finder and beneficiary information outreach (mailings and 1-800MEDICARE). These decision aids are important in helping beneficiaries evaluate plans.
Coordination of enrollment and disenrollment through MA organizations and effective dates
of coverage and changes of coverage
Individuals who have been enrolled in a non-Medicare health plan—such as a Medicaid managed
care plan or a commercial plan—can be enrolled in an MA plan operated by the same organization
when those individuals first become eligible for Medicare through a process known as “seamless
conversion.” Such enrollment is automatic unless the beneficiary enrolls in fee-for-service (FFS)
Medicare or a different MA plan. In November 2016, CMS suspended approval of new requests to
use seamless conversion in order to reevaluate its policies in this area.
In this notice of proposed rulemaking, CMS proposes to resume the use of seamless conversion,
but only for organizations that offer both Medicaid managed care plans and MA dual eligible
3

Medicare Payment Advisory Commission. 2017. Report to the Congress: Medicare and the health care delivery
system. Washington, DC: MedPAC.

 Seema Verma
Administrator
Page 4
special needs plans (D–SNPs). These organizations would be able to automatically enroll members
of the Medicaid plan in the affiliated D–SNP when they become eligible for Medicare (and thus
become dual eligibles).
Comment
The Commission has been supportive of efforts to better integrate Medicare and Medicaid for dual
eligibles.4 For individuals in Medicaid managed care plans, the start of Medicare eligibility can
lead to more fragmented care, because their coverage of Part A and B services and Part D drugs
may now be provided separately, either through FFS Medicare or through MA plans or Part D
plans offered by other organizations. In these instances, automatic enrollment into an affiliated D–
SNP can promote the use of integrated care (e.g., shared provider networks) by encouraging these
beneficiaries to receive both their Medicare- and Medicaid-covered services from the same
organization. We therefore support CMS’s proposal to resume the use of seamless conversion for
dual eligibles.
Passive enrollment flexibilities to protect continuity of integrated care for dually eligible
beneficiaries
Currently, CMS can passively enroll beneficiaries in a new MA plan if the contract for their
current plan is being terminated in the middle of a plan year or if there is potential harm to the
beneficiaries. CMS proposes expanding the use of passive enrollment to include dual eligibles who
are in certain types of integrated D–SNPs (where the parent MA organization also provides
Medicaid coverage) and face an involuntary disruption in their Medicare or Medicaid coverage.
For example, such a disruption could occur when a state re-procures its Medicaid managed care
plans and one of the incumbent plans is not selected. Under this proposal, CMS would passively
enroll dual eligibles in these plans in other integrated D–SNPs.
Comment
Without the use of passive enrollment, dual eligibles in the situations covered by CMS’s proposal
would likely see their care become more fragmented. We therefore support this proposal as a way
to encourage dual eligibles to receive their Medicare- and Medicaid-covered services from the
same organization.
Establishing limitations for the Part D Special Election Period (SEP) for dually eligible
beneficiaries
Dual eligibles and other beneficiaries who receive Part D’s low-income subsidy (LIS) currently
have a special enrollment period (SEP) that allows them to switch Part D plans, which include MA
plans that offer Part D coverage, on a monthly basis. CMS proposes limiting the SEP so that dual
eligibles and other LIS recipients could only switch Part D plans once a year (outside of the
standard opportunities to switch plans that apply to all Part D enrollees, such as the annual open
enrollment period and SEPs for events such as permanently moving to a new residence). Dual
4

Medicare Payment Advisory Commission. 2013. Report to the Congress: Medicare payment policy. Washington,
DC: MedPAC.

 Seema Verma
Administrator
Page 5
eligibles can currently switch Part C (MA) plans on a monthly basis as well; CMS would use subregulatory guidance to make changes to the Part C SEP to conform with the Part D changes.
Comment
In 2008, the Commission recommended that the SEP for dual eligibles should be eliminated,
although we also recommended that dual eligibles should be able to return to FFS Medicare at any
time during the year. CMS’s proposal to limit the SEP for dual eligibles is generally consistent
with this recommendation. Commission staff have made a series of site visits to states that are
testing the use of Medicare-Medicaid Plans for dual eligibles as part of CMS’s financial alignment
demonstration, and we have heard from numerous stakeholders that the ability of dual eligibles to
switch plans on a month-to-month basis makes it difficult for plans to provide care coordination.
Therefore, we support this proposal.
MA and Part D Prescription Drug Plan quality rating system
Each year CMS evaluates the quality of care and contract performance of the private plans
participating in MA (Part C) and the Part D prescription drug program. A variety of clinical quality
measures, patient experience measures, and contract performance measures form the basis of
overall star ratings for Part C and Part D, and an overall combined rating for MA–PD contracts
(MA contracts that include Part D). Under the quality bonus program for MA plans, the applicable
overall star rating determines whether or not a contract is eligible for bonus payments, which take
the form of an increase in the benchmark for contracts with an overall rating of four stars or higher
in the five-star rating system. The contract-level star rating applies to all plans under the contract,
resulting in increased plan payments for every enrollee under the contract. The star rating also
determines the share of any difference between the MA benchmark and a plan bid below the
benchmark that is available for the provision of extra benefits to plan enrollees. There is not a
similar quality bonus program in Part D, but for both Part D and MA, the star ratings are publicly
announced through the Medicare Plan Finder at Medicare.gov. The published star ratings, and
details about the individual measures included in the star ratings, enable beneficiaries to compare
the level of quality of the contracts available in their geographic area.
The proposed regulations codify the policy and methods for determining star ratings that CMS
formulated through sub-regulatory guidance. Heretofore, interested parties have been able to
comment on the star rating system through the MA and Part D annual call letter, and through a
separate announcement of changes and updates to the method of assigning star ratings. Based on
feedback from plans and other stakeholders, the proposed rule calls for any future major changes
in the star system to be subject to the formal notice-and-comment rulemaking process.
Because the star rating provisions of the proposed regulations are intended primarily to codify
current policy, there are generally no new proposed policies. One exception is the policy regarding
the effect of contract consolidations on star ratings, for which CMS is proposing a change to
current policy. The proposed rule also includes a discussion of CMS’s consideration of alternative
reporting units for quality measurement, with consideration of the plan as a reporting unit. CMS is
also inviting comment on the current method of determining star ratings.

 Seema Verma
Administrator
Page 6
Star ratings after contract consolidations
A contract consolidation occurs when a company decides to combine two or more MA contracts
into one “surviving” contract. The surviving contract absorbs the membership of the contracts that
will be discontinued and which are referred to as “consumed” contracts. Under current policy, the
consumed contracts immediately acquire the star rating of the surviving contract. As a result, a
company has the opportunity to increase its bonus revenue because of the timing of when
• Stars are determined (in October, 14 months prior to the payment year),
• MA bids are submitted (in June, 6 months prior to the payment year), and
• Companies make decisions about contract consolidation (coinciding with bidding, in June).
Even though bids are for the following payment year, the star ratings used for bonus
determinations are the ratings publicized in October of the preceding year. Each company knows
which of its contracts are in bonus status for the coming payment year. This advance knowledge
allows companies to have contracts that are in bonus status consume other contracts that have a
star rating below four stars. At the end of 2017, 1.4 million beneficiaries in consumed contracts
will be moved from contracts below four stars to a contract in bonus status. (The Commission has
called attention to this issue in past reports.5)
CMS proposes to change the approach to assigning stars when there is a contract consolidation.
Rather than have the consumed contract(s) acquire the star rating of the surviving contract, CMS
would determine an enrollment-weighted average star rating.
Comment
CMS’s proposed averaging method will address the situation in which a relatively small, bonuseligible contract is the surviving contract that consumes a large non-bonus contract, as has
happened in the last two rounds of consolidations. However, the averaging method would only
give an accurate picture of quality in a given geographic area if the two or more contracts involved
in a given consolidation shared exactly the same service area.
In the current cycle of contract consolidations (the end of 2017), there were 17 contract
consolidations in which a contract below 4 stars was consumed by a contract at or above 4 stars. In
only one of the cases was there any overlap of service areas (one company, which purchased
another company, undertook a consolidation in which 3 of 13 counties were in the service areas of
both contracts). Other combinations of areas included state combinations such as Missouri and
Virginia; Wisconsin and Kentucky; and Kentucky and New Hampshire. If there are differences in
the quality of care under the contracts undergoing consolidation—which would be the case in the
current cycle, given that the consolidation raises the star rating of the consumed contract(s)—the
proposed approach of averaging star ratings does not give an accurate picture of the quality of care
in the different market areas.

5

Medicare Payment Advisory Commission. 2017. Report to the Congress: Medicare payment policy. Washington,
DC: MedPAC.

 Seema Verma
Administrator
Page 7
In addition, the averaging method would continue to provide an incentive for organizations to use
contract consolidation as a means of obtaining unwarranted bonus payments. For example, two
contracts with equal enrollment, one with a 4.5-star rating and one with a 3.5-star rating, could be
combined to result in what would likely be a 4-star rating of the consolidated contract. The
averaging method forecloses certain types of combinations that have occurred in the past, but it
does not fully address the concern about unwarranted program expenditures or inaccurate
information provided to beneficiaries.
As mentioned in the proposed rule, the Commission has expressed concern about the practice of
consolidating contracts for the purpose of artificially boosting star ratings. The Commission
continues to examine the question of how to address the issues that arise with such consolidations.
Both of the issues that are of concern to the Commission—the need to have accurate information
on quality in a geographic area, and the avoidance of incentives to consolidate in order to increase
bonus payments—can be addressed by the Commission’s long-standing recommendation that
CMS move to reporting of quality, and determination of star ratings, by local market areas. The
Commission continues to consider what options are feasible prior to the use of local market areas
as the reporting units for quality measures and star ratings.
Lastly, CMS should provide additional detail underlying its estimate in the regulatory impact
analysis of the proposed rule (page 56486 of the Federal Register notice) that 160,000
beneficiaries were affected by consolidations in 2018. The Commission’s analysis indicates that
1.4 million beneficiaries were moved from non-bonus status to bonus status for the 2018 payment
year.
Reporting quality at the plan level rather than the contract level
While CMS is not proposing changes to the current policy of having quality measures reported at
the MA contract level with stars computed at the contract level, CMS is considering alternative
reporting units for quality measurement. One particular approach that CMS is evaluating is to have
quality reporting at the plan level rather than the contract level. The agency is also requesting
comments on “alternative reporting units…requiring reporting at different levels (for example,
parent organization, contract, plan, or geographic area) by measure.”
Comment
The Commission has stated its preference for reporting at a local market area level rather than at
the contract level, as we discuss below. A drawback of reporting at the contract level is that some
contracts cover large geographic areas, in part due to contract consolidations. For example, one
contract that has consumed 20 other contracts covers 35 states, with one star rating across all states
for its 850,000 enrollees. The contract will have 125 separate plans in 2018 (in addition to
employer group waiver plans). While it is often true that when a contract has multiple plans each
plan covers a different geographic area, that is not always the case.
Reporting at the plan level would not resolve the problems that occur with contract-level reporting.
One reason is that plan sponsors define the service area themselves. Its defining feature is
essentially that the plan is the bidding unit, and all beneficiaries under the plan generally have a

 Seema Verma
Administrator
Page 8
uniform benefit package. Like contracts, plans can span wide geographic areas. In 2017, there
were 30 HMO plans with a service area that included 10 or more Metropolitan Statistical Areas
(MSAs), and 35 local PPO plans with 10 or more MSAs. There is no requirement that plan service
areas be contiguous. If the plan is the reporting unit for quality and the determination of stars, MA
organizations could construct plans in such a way that the combination of counties under the plan
maximizes star ratings for the greatest number of enrollees. Then, in place of contract
consolidations to boost star ratings, organizations could instead use plan geographic consolidations
and plan configurations to boost star ratings in a situation in which market-level quality reporting
would have resulted in a star rating below four stars.
If CMS does decide to change the reporting unit for quality indicators as a replacement for
contract-level reporting, other options are better than plan-level reporting (though separate
reporting by special needs plans would still have value). The Commission supports reporting of
quality indicators for health plans and for fee-for-service Medicare by local areas “in a way that is
consistent with the organization of local health care delivery markets.”6 The Commission
recommended specific geographic configurations in its 2010 report on how to compare quality
among MA plans and between MA and FFS Medicare.7 CMS should bear in mind the
Commission’s position on the importance of having information on quality at the local geographic
level as the agency explores alternatives to contract-level reporting.
Minimum number of measures needed to have a star rating
Currently, a contract will receive an overall star rating only if the contract has results for at least
half the measures used to calculate star ratings (up to 22 of 43 distinct measures for MA–PD
plans). CMS proposes no changes to this policy.
Comment
The current rule dates from the period during which all measures had a weight of 1, rather than the
current weights of 1, 1.5, or 3 for measures other than the two measures of improvement across all
measures (Part C and Part D). Given that measure weighting is now used, it would be appropriate
to allow a contract to receive a star rating if the available measures for the plan are at least half of
the weighted value of the full measure set. Rather than needing 22 of 43 measures, the contract
would have to have measures with a total weight of at least 34 (one-half of the total weight of
67.5).
Cut points below those of the prior year
Currently, for most measures, the distribution of star ratings across contracts is based on a yearly
evaluation of relative contract performance (rather than, for example, setting a pre-determined
threshold for getting a 4-star rating on a given measure, as was previously the case for many
6

Medicare Payment Advisory Commission. 2015. Report to the Congress: Medicare and the health care delivery
system. Washington, DC: MedPAC.
7
Medicare Payment Advisory Commission. 2010. Report to the Congress: Medicare payment policy. Washington,
DC: MedPAC.

 Seema Verma
Administrator
Page 9
measures). To determine relative performance for a measure, CMS employs a clustering/grouping
analysis to group all contracts into five levels of performance corresponding with each level of the
5-star scale. The lowest level of performance within each group is the cut-off point for a result to
be classified as falling within the group. For example, if for a given measure the cluster of highest
performing contracts includes rates of 90, 91, 93 and 95 percent, a contract has to have a measure
result of at least 90 percent to earn a 5-star rating. If the 4-star group included rates of 81, 85, 86,
and 86 percent, the cut-off point for 4-star performance would be 81 percent. Each year, the
clustering is recomputed to establish the current cut points. Under this approach, it is possible that,
in a situation in which overall quality in MA declines, the cut points are lower than in the previous
year. For example, the 5-star cluster in the example may become 88, 90, 91 and 93—making the
new 5-star cut point 88 percent.
Comment
From year to year, the star cut points should not be allowed to drop below the level of the
preceding year so that Medicare would not be in the position of financially rewarding plans even
though quality had declined in absolute terms. In the example, the cut point for a 5-star rating
would remain at 90 rather than drop to 88. Alternatively, given that the star rating system rewards
plans for improvement through a heavily weighted improvement measure, CMS should consider
developing a measure that reduces star ratings when there is a decline in performance for a given
contract.
Data integrity
Currently, CMS reduces a contract’s star rating if the agency determines that a contract’s measure
data are incomplete, inaccurate, or biased. CMS proposes specific rules to apply in these
circumstances. First, CMS would continue to reduce Health Effectiveness Data Information Set
(HEDIS®) measures to 1 star when audited data are submitted to the National Committee for
Quality Assurance (NCQA) with an audit designation of “biased rate” based on an auditor’s
review of the data if the plan chooses to report, or with a designation of “non-report” if the plan
does not.8 Second, CMS would continue to reduce Part C and D reporting requirements data to 1
star when a contract does not score at least 95 percent on data validation for the applicable
reporting section or was not compliant with data validation standards for data used to calculate the
measure. In addition, CMS proposes a new policy that would reduce a contract’s Part C or Part D
appeal measures’ star ratings when submitted independent review entity (IRE) data are not
complete or otherwise lack integrity. The proposed methodology would result in reductions of 1
star to 4 stars depending on the degrees of data issues identified.
Comment
The Commission is concerned about the extent to which Part C and Part D reported data used to
calculate star measures are incomplete or invalid, as such data are essential for beneficiaries’
ability to compare plans and for the program to reward good performance as assessed by quality
measures. We support the proposals to continue and to expand policies to reduce star ratings when
contracts’ data are not reported or do not meet data validation requirements.
8

HEDIS® is a registered trademark of NCQA.

 Seema Verma
Administrator
Page 10
Changes to the days’ supply required by the Part D transition process
Currently, Part D sponsors must provide a transition policy for certain enrollees who receive
ongoing therapy with prescribed drugs that are not on their plan’s formulary or that are on the
formulary but require prior authorization or step therapy. Within the first 90 days of a beneficiary
enrolling in a new plan, sponsors must provide the enrollee with a one-time, 30-day supply of
drugs or, for enrollees who are long-term care (LTC) residents, at least a 91-day supply and up to a
98-day supply, consistent with the dispensing increment.
CMS proposes to shorten the transition supply in the LTC setting to be consistent with that for
enrollees in the community setting. The agency also proposes a technical change to require “a
month’s supply” rather than 30 days of medication, to better reflect how certain drugs are typically
dispensed (e.g., in 28-day increments).
Comment
The intent behind Part D’s transition policy is to provide enrollees with needed medications while
they request a formulary exception or obtain a prescription for an alternative formulary drug. At
the start of Part D, CMS was concerned that LTC residents might have limited access to
prescribers and would therefore need a more generous transition supply. However, there has been
little or no evidence that bears out this concern. LTC facilities typically have prescribers on staff or
under contract to attend to the circumstances of residents, and most are serviced by LTC
pharmacies that can provide 24-hour delivery and emergency drug supplies. Given the lack of
evidence that the current transition process for enrollees residing in LTC facilities requires any
more time than for other enrollees, we support CMS’s proposal to shorten the transition supply
length in the LTC setting. We are persuaded that such a measure could help to reduce waste and
costs for the Part D program.
Expedited substitutions of certain generics and other midyear formulary changes
When plan sponsors use formularies, Part D law lays out certain requirements for how plans must
develop and operate them. For example, Part D formularies must include some (but not necessarily
all) drugs in all therapeutic categories and classes. For a few “classes of clinical concern,”
formularies must cover all drugs. Plan sponsors may not change the structure of therapeutic
categories in the middle of a benefit year. The law also states that plans must provide appropriate
notice before removing a drug from their formularies or changing a drug’s preferred or tiered costsharing status.
Part D guidance aims to ensure that beneficiaries maintain access to drugs that were offered by
their plan at the time they enrolled. Nevertheless, there may be circumstances in which new
clinical information or the entrance of a new competing therapy may warrant changes to a
formulary in the middle of a benefit year. Currently, plan sponsors must request and receive CMS
approval before carrying out most “negative” formulary changes, such as removing a drug from a
formulary or setting new utilization management requirements. Plans must also give affected
enrollees 60 days’ advance notice or provide a 60-day refill upon request of an affected enrollee.

 Seema Verma
Administrator
Page 11
CMS proposes to make changes to its formulary review and notice processes. Plan sponsors would
be allowed to add a newly approved generic and remove or change the tier status of a
therapeutically equivalent brand-name drug at any point during the benefit year without prior
approval from CMS. The new generic would have to be offered at the same or lower cost sharing,
and at the same or less restrictive utilization management criteria that applied to the brand
alternative. Plan sponsors would be required to provide general notice that such substitutions could
occur without additional advance notice. Sponsors would also be required to provide 30 days’
advance direct notice to affected enrollees, CMS, and other entities. If requested, the plan must
provide a month’s supply refill to an affected enrollee. The Part D transition process would not
apply to the situation in which the sponsor substitutes a new generic for the brand-name drug.
Comment
As part of a broader package of proposed improvements to Part D, the Commission recommended
in its June 2016 report that CMS streamline the agency’s process for reviewing formulary
changes.9 The Commission has specifically stated that CMS could provide plan sponsors with
greater flexibility to make certain midyear formulary changes, such as allowing plans to add a
generic drug and remove the brand-name version without first receiving agency approval. We
commend CMS for examining its formulary procedures and we strongly support the proposed
changes. The Commission also encourages CMS to continue to review its procedures and look for
other opportunities where plans might be given greater flexibility to operate formularies without
detrimentally affecting beneficiaries’ access to needed medications.
Treatment of follow-on biological products as generics for non-LIS catastrophic and LIS cost
sharing
Follow-on biologics (also called biosimiliars) are highly similar to an originator biologic. As with
generic drugs, use of biosimilars may be an important means for improving access to medicines
and restraining spending through lower prices. Currently, because biosimilars do not meet the
definition of a generic or multi-source drug, enrollees who receive Part D’s LIS pay the same
maximum cost-sharing amounts for a biosimilar that they would for its originator biologic. As a
result, if a plan sponsor covered both products on its formulary but placed the biosimilar on a tier
with lower cost sharing, LIS enrollees would not see any financial incentive to use the biosimilar.
Similarly, Medicare law specifies cost-sharing amounts for non-LIS enrollees who reach Part D’s
OOP threshold as the greater of 5 percent coinsurance or, for 2018, $3.35 for generics and
preferred multi-source drugs, and $8.35 for other prescriptions. Because CMS has not previously
included biosimilars in the definition of generics or preferred multi-source drugs, the higher of the
two copay amounts would apply.
CMS now proposes to revise its definition of generics to include follow-on biologics approved
under section 351(k) of the Public Health Services Act—the section of law by which most followon biologics are licensed. CMS states that this revised definition of generics applies only to Part D
9

Medicare Payment Advisory Commission. 2016. Report to the Congress: Medicare and the health care delivery
system. Washington, DC: MedPAC.

 Seema Verma
Administrator
Page 12
cost sharing for LIS enrollees and for non-LIS enrollees who reach the OOP threshold. FDA
approval of a biosimilar product is distinct from a designation of interchangeability; the latter
indicates that the follow-on product may be substituted for an originator product without a new
prescription. Because of this difference, the agency will not treat biosimilars as generics for other
purposes, such as for midyear formulary changes (e.g., substitution of a newly approved biosimilar
while removing an originator biologic).
Comment
The Commission strongly supports CMS’s proposed revision of the definition of generics to
include biosimilars for purposes of cost sharing. Encouraging the use of biosimilars among LIS
beneficiaries and non-LIS enrollees with very high spending could spur greater price competition
among biological products, expand access for beneficiaries, and help to restrain growth in program
spending. The change would be consistent with our June 2016 recommendation for the Congress
to modify Part D’s LIS copayments to encourage the use of generics, preferred multi-source drugs,
and biosimilars.10 (CMS now interprets its authority as allowing the agency to set maximum costsharing amounts, which would make a change in statute unnecessary.) We further suggest that
CMS also include follow-on biological products licensed under Section 505(b)(2) of the Federal
Food, Drug, and Cosmetic Act in the years before 2020 (after which they will be deemed to be
licensed under section 351) within the revised definition of generics for purposes of Part D cost
sharing.
Eliminating the requirement to provide PDP Enhanced Alternative (EA) offerings with
meaningful differences
Currently, a sponsor of a stand-alone PDP may offer up to two enhanced alternative (EA) plans in
a PDP region once it has offered one basic benefit plan. EA plans must provide more generous
coverage than the basic plan offered by the same sponsor in the same region. Similarly, if a
sponsor offers two EA plans, the second EA plan must be more generous than the first. To ensure
that plan offerings are “no more numerous than necessary to offer beneficiaries choices” of plan
options, each year CMS has set a “meaningful difference” standard as measured by target
differences in enrollee out-of-pocket (OOP) costs (the “OOP cost model”). For example, in 2018, a
sponsor’s basic and enhanced plans in the same region must have a difference of at least $20 in
expected monthly OOP costs, while two enhanced plans must have at least a $30 differential.
For 2019, CMS is proposing to eliminate the meaningful difference requirement between the two
EA plans offered by a sponsor in a region. Under the proposed policy, a sponsor could offer two
EA plans that “vary with respect to their plan characteristics and benefit design, such that they
might appeal to different subsets of Medicare enrollees,” but have similar expected enrollee OOP
costs. In the future, CMS intends to reexamine how it defines the meaningful difference
requirement between basic and EA plans offered by a PDP sponsor within a region. The agency
also recognizes that the current OOP cost methodology is only one method for evaluating whether
the differences between plan offerings are meaningful, and intends to investigate whether the
10

Medicare Payment Advisory Commission. 2016. Report to the Congress: Medicare and the health care delivery
system. Washington, DC: MedPAC.

 Seema Verma
Administrator
Page 13
current OOP cost model or an alternative methodology should be used to evaluate meaningful
differences between PDP offerings.
Comment
The Commission supports CMS’s goal of encouraging competition and plan flexibility in Part D.
However, we are concerned about the removal of the meaningful difference requirement in Part D.
Similar to our comments for Part C, we believe the use of an objective, quantifiable measure can
provide valuable information to beneficiaries when evaluating plan options that have different
benefit designs (and to the agency during the bid review process). In the absence of an alternative
measure, we encourage CMS to continue to use the OOP cost model in the near term, but we
suggest that the agency investigate improvements to the model.
CMS could consider a waiver of the meaningful difference requirement only in cases where plan
sponsors can show that there are significant differences in value between their EA offerings even
when the difference in expected OOP costs do not exceed the minimum threshold. Even in those
cases, we strongly encourage CMS to remain vigilant in ensuring that “differences in plan
characteristics and benefit designs” reflect significant differences in value and that beneficiaries
can evaluate and compare their options in an informed manner. Ensuring that Medicare Plan
Finder tool allows beneficiaries to understand the differences among plan options will be
especially important.
We support CMS’s proposal to continue use of the meaningful difference requirement between
basic and EA plans. Eliminating this requirement could result in sponsor behaviors that could
adversely affect the program, such as offering EA plan options to engage in risk segmentation.
Risk segmentation is counter to the notion of insurance policy and could be a concern for the
program, particularly if it involves avoiding LIS enrollees.11 By enrolling healthier beneficiaries
into certain EA plans, plan sponsors may segment higher cost enrollees into plans with higher
premiums for basic benefits. To the extent that the basic plan qualifies as an LIS benchmark plan,
it could increase the amount Medicare pays for the low-income premium subsidy. Given this
potentially adverse impact on the program, we believe it is important to continue to distinguish
between basic and EA plans.
Further, we share CMS’s concern with the current OOP cost methodology and encourage the
agency to investigate its accuracy and effectiveness. In particular, current pricing trends have been
driving a wedge between gross prices used in the OOP cost model and prices net of rebates that
affect the final costs of providing the Part D.12 As a result, the current OOP cost model may place
an unreasonable constraint on sponsors as they design plan benefits.
11

In implementing the meaningful difference rule, CMS noted that “it was urgent that we adopt the proposed policy as
soon as possible so that we could bring an end to this bidding practice” that allowed some plan sponsors to offer “low
value enhanced plans” that had premiums below the sponsors’ basic plans due to favorable risk selection (Medicare
Program; Contract Year 2015 Policy and Technical Changes to the Medicare Advantage and the Medicare Prescription
Drug Benefit Programs; Final Rule published on May 23, 2014.)
12
Centers for Medicare & Medicaid Services, Department of Health and Human Services. 2017. Medicare Part D:
Direct and indirect remuneration (DIR). Press release. January 19.
https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2017-Fact-Sheet-items/2017-01-19-2.html.

 Seema Verma
Administrator
Page 14
Request for information regarding the application of manufacturer rebates and pharmacy
price concessions to drug prices at the point of sale
In drug classes with competing therapies, pharmaceutical manufacturers often provide post-sale
rebates to plan sponsors and their pharmacy benefit managers in return for preferred placement on
a plan formulary or for successfully encouraging enrollees to use their drugs. Similarly, plan
sponsors often negotiate with pharmacies for post-sale discounts based on measures of cost,
quality, or medication adherence. Price concessions that sponsors negotiate with pharmaceutical
manufacturers and pharmacies are referred to collectively as direct or indirect remuneration (DIR).
At the point of sale (POS), such as a pharmacy counter, Part D sponsors generally do not reflect
these post-sale rebates and discounts in the prices that beneficiaries pay when their plans apply a
deductible or coinsurance. Instead, plans tend to use DIR to offset the expenditures of all plan
enrollees, which can lower enrollee premiums and premium subsidies paid by Medicare.
However, in recent years, the amount of DIR that sponsors received consistently has exceeded the
amount that sponsors projected in their bids. As CMS has noted, under Part D’s risk corridors, any
DIR received above the projected amount contributes primarily to plan profits. Further, rapid
growth in DIR has resulted in a widening disparity between gross Part D drug costs, based on POS
prices, and costs net of all DIR.13 CMS notes that this “gross-to-net” disparity shifts costs from
plan sponsors “to beneficiaries who utilize drugs in the form of higher cost sharing, and to the
government through higher reinsurance and low-income cost-sharing subsidies.” That is, the
current DIR construct provides a financial advantage to plan sponsors. As a result, CMS is
concerned that sponsors may have a weak or no incentive to lower prices at the POS and may
prefer “high cost-highly rebated drugs” when available over alternative drug that have lower net
costs.
To mitigate these incentives, CMS is proposing to require that sponsors reflect at least a minimum
percentage of manufacturer rebates and all pharmacy price concessions in prices at the POS. The
agency is seeking comment on specific parameters of the policy such as the minimum percentage
of rebates that must be reflected at the POS, the methodology for calculating the applicable
average rebate amount, and potential approaches for reflecting all pharmacy price concessions in
POS prices.
Comment
The Commission shares CMS’s concerns and strongly agrees that a change is needed to ensure that
Part D sponsors face incentives that are aligned with the program and its enrollees. Because
manufacturer rebates account for the largest share of all DIR and have accounted for much of the
growth in DIR, our comments will focus on the proposed approach to moving a portion of
manufacturer rebates to the POS.

13

Centers for Medicare & Medicaid Services, Department of Health and Human Services. 2017. Medicare Part D:
Direct and indirect remuneration (DIR). Press release. January 19.
https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2017-Fact-Sheet-items/2017-01-19-2.html.

 Seema Verma
Administrator
Page 15
The Commission has previously discussed how Part D’s benefit design can create incentives for
plan sponsors to include certain high-cost drugs on their formulary over others, which can increase
beneficiary cost sharing and Medicare spending for reinsurance.14 Rising prices have contributed
to the expansion of DIR, which in turn has contributed to both slow growth in premiums and fast
growth in reinsurance payments. Premiums and fixed-dollar copayments are lower for all enrollees
when plan sponsors offset their benefit costs with DIR. However, enrollees who pay coinsurance
for their prescriptions do not see any reduction to their cost sharing because coinsurance is based
on the price before any DIR is applied. Also, Part D’s OOP threshold is based on the higher (gross)
price. This contributes to more beneficiaries exceeding that OOP threshold and increases
Medicare’s payments for reinsurance. As a consequence, growth in DIR has mixed effects on
beneficiaries: All beneficiaries experience lower premiums, but some incur higher cost sharing.
In 2016, the Commission recommended changes to Part D that would phase in a reduction of
Medicare’s reinsurance from 80 percent to 20 percent, among other changes.15 Those
recommendations could reduce Medicare spending under Part D by providing plans with better
financial incentives to include lower-priced drugs on their formularies. Beneficiaries would also
benefit from lower cost sharing if they selected those lower-priced drugs.
The Commission agrees with the principle behind the requirement for plan sponsors to share at
least a portion of manufacturer rebates with enrollees who use those drugs. At the same time, we
note that any policy that shifts some or all DIR to lower prices at the POS rather than premiums
would increase Medicare’s costs through its effects on premium subsidies and manufacturer
discounts. We share a concern for enrollees who pay coinsurance on high-priced specialty drugs;
even 5 percent coinsurance during the catastrophic phase of the Part D benefit can be burdensome
when a beneficiary’s medication costs thousands of dollars per prescription. To the extent that
there are rebates associated with expensive medications, some of the rebate amount could be used
to reduce beneficiary cost sharing. However, we are concerned that CMS’s proposed approach
would be complex to implement, administratively burdensome and, for drug classes with few
competing therapies, would risk disclosure of confidential rebate information. Further, the policy
would not help beneficiaries who take expensive drugs with no post-sale rebates or discounts. We
strongly encourage CMS to search for alternative policies that are less complex but could help to
achieve similar aims. For example, CMS may want to consider requiring plan sponsors to reflect a
portion of expected DIR in cost sharing amounts when they submit their bids.
In general, given the growth in the disparity between gross and net prices, a model based on gross
prices may no longer be appropriate for demonstrating actuarial equivalence to the defined
standard benefit. This is because cost sharing amounts calculated to be actuarially equivalent to the
defined standard benefit based on gross prices would, in effect, have beneficiaries pay a higher
share of the actual costs than is set by the defined standard benefit. Therefore, the agency may
want to evaluate whether that actuarial model (i.e., the current bid-pricing tool) could incorporate a
portion of expected DIR so that cost sharing reflects some of the rebates.
14

Medicare Payment Advisory Commission. 2017. Report to the Congress: Medicare payment policy. Washington,
DC: MedPAC.
15
Medicare Payment Advisory Commission. 2016. Report to the Congress: Medicare and the health care delivery
system. Washington, DC: MedPAC.

 Seema Verma
Administrator
Page 16
Lengthening adjudication timeframes for Part D payment redeterminations and IRE
reconsiderations
Currently, both Part D plan sponsors and the independent review entity (IRE) must issue decisions
within 7 days in response to enrollee standard (non-expedited) requests for plan redetermination
and IRE reconsideration. CMS proposes to extend these periods to 14 days in cases where the
enrollee has received the drug, paid for the prescription out of pocket, and is requesting
reimbursement. CMS would keep existing time periods the same in cases where the enrollee did
not receive the drug and is requesting that it be provided. CMS asserts that limiting the change to
cases where reimbursement is sought would limit the health effects of the proposed policy. CMS
notes that in cases resolved in the enrollee’s favor, the plan sponsor would still be required to make
payment no later than 30 days from receipt of the request for redetermination, or the IRE
reconsideration notice, respectively.
Comment
We recognize that collecting information sufficient to adjudicate a decision within 7 days can be a
challenge in some cases; this is a shorter time period than that allowed in Part C, and plan sponsors
report difficulties in contacting prescribers. However, given our concerns about the effect of this
change on beneficiaries, we encourage CMS to keep the existing deadline for plan sponsors and
the IRE. In cases that are appealed to the IRE, existing deadlines provide enrollees with a decision
within a total of 17 days from initial appeal. The proposed policy would add 14 days for a total of
31 days. Given that many Medicare beneficiaries are on limited budgets (e.g., on average, Social
Security benefits account for more than 60 percent of income for seniors; for more than one-fifth
of seniors, Social Security benefits account for 100 percent of income), we are concerned about the
increased financial burden this proposal would place on enrollees. Enrollees who wait up to a
month to then learn that their case has been decided against them, would have to either pay for the
drug out of pocket again or get a prescription for an alternative drug within a short time period.
These options jeopardize enrollees’ access to needed drugs, either initially prescribed or
alternative, in a timely manner.
Removal of Quality Improvement Project (QIP) for MA organizations
CMS proposes to remove the Quality Improvement Project (QIP) attestation from MA quality
improvement requirements because it is duplicative of activities plans are already doing to meet
other requirements, such as reporting clinical quality and patient experience measures for the star
ratings program. CMS proposes to continue implementation of the Chronic Care Improvement
Program (CCIP), so that MA organizations can focus on one project that supports improving the
management of chronic conditions.
Comment
The Commission believes that quality measurement should not be unduly burdensome for
providers, and Medicare quality programs should focus on population-based measures such as
outcomes, patient experience, and value (cost/low-value) measures. Because plans are provided
incentives to focus on aspects of clinical quality and patient experience in the star ratings program,

 Seema Verma
Administrator
Page 17
we support CMS’s removing the duplicative QIP attestation. We also encourage CMS to remove
the duplicative CCIP attestation because measures around prevalent chronic conditions are already
measured in the star rating program (e.g., diabetes, hypertension).
Conclusion
The Commission values the ongoing cooperation and collaboration between CMS and our staff on
technical policy issues. We look forward to continuing this productive relationship. If you have
any questions, or require clarification of our comments, please feel free to contact James E.
Mathews, the Commission’s Executive Director, at 202-220-3700.
Sincerely,

Francis J. Crosson, M.D.
Chairman