policy

March 6, 2023

The Honorable Chiquita Brooks-LaSure
Administrator
Centers for Medicare and Medicaid Services
U.S. Department of Health and Human Services
200 Independence Avenue SW
Washington, DC 20201

RE: Advance Notice of Methodological Changes for Calendar Year (CY) 2024 for Medicare Advantage (MA) Capitation Rates and Part C and Part D Payment Policies

Dear Administrator Brooks-LaSure:

The Association for Community Affiliated Plans (ACAP) is a national trade association representing 78 not-for-profit Safety Net Health Plans. Collectively, ACAP plans serve more than 23 million people enrolled in Medicaid, Medicare, the Marketplaces, and other publicly supported coverage programs. Thirty-three of our member plans are D-SNPs. Our mission is to support our member plans’ efforts to improve the health and well-being of people with low incomes and with significant health care needs.

ACAP appreciates this opportunity to respond to the 2024 Advance Notice to inform the best possible policy for beneficiaries enrolled in Medicare Advantage, particularly those dually eligible for Medicare and Medicaid.
Below is a summary of ACAP’s comments:

CMS-HCC Risk-Adjustment Model and Frailty Adjuster:

• CMS should also not implement changes to the risk model that undermine the sustainability of D-SNPs, HIDE SNPs, and FIDE SNPs. ACAP has significant concerns that the proposed changes to the CMS-HCC risk-adjustment model – including the proposed reduction in the frailty adjuster – will disproportionately impact D-SNPs, HIDE SNPs, and FIDE SNPs operated by Safety Net Health Plans. Although CMS’ modeling of the 2024 CMS-HCC model may suggest that the model accurately predicts costs across all MA plans in aggregate, the impacts on individual D-SNPs, HIDE SNPs, and FIDE SNPs may vary significantly. We urge CMS to not implement any changes that will reduce the predictive accuracy of the model for D-SNPs.
• CMS should be aware that the actual costs and risk profiles of full-benefit dually eligible individuals do not change significantly from year-to-year. Proposed changes to D-SNP risk scores that result in large risk score reductions from year-to-year, are a strong signal to CMS that the proposed changes to the risk model may reduce the predictive accuracy of the risk model for full-benefit dually eligibles.
• We strongly urge CMS to:
  • Delay implementation of the proposed changes to the CMS-HCC risk model and the reduction in the frailty adjuster.
  • Publish an impact assessment of the impact of the proposed changes to the risk model (including the reduction of the frailty adjuster) on D-SNPs, HIDE SNPs, and FIDE SNPs, and publish the underlying data used in CMS’ analyses so that plans can use the underlying data in their own modeling.
  • Give stakeholders more time to assess and understand the proposed changes and to discuss their potential impact with CMS. Convene a Technical Advisory Panel of D-SNPs and other stakeholders to help CMS identify solutions to any risk model changes that will have a disproportionate impact on certain subpopulations of Medicare beneficiaries.
  • Phase-in any significant, future changes to the CMS-HCC risk model over 3 years.

Universal Foundation of Quality Measures:

• ACAP supports the concept and development of the Universal Foundation of Quality Measures.
• We recommend that measures being considered for inclusion in the measure set (1) be proposed for stakeholder comment; (2) be included on the Star Ratings display page; (3) be included in the Star Ratings for at least 3 years; and (4) have undergone rigorous analysis and testing to ensure that the measure accurately reflects the quality of care for subgroups of beneficiaries, including those dually eligible for Medicare and Medicaid.

Star Ratings Outlier Removal Policy:

• As noted in our comments to the 2024 MA-PD Proposed Rule, ACAP has significant concerns about the impact of CMS’ proposed outlier removal policy on D-SNPs. An analysis performed by ZAHealth found that the outlier policy will result in 14 percent of D-SNP contracts losing their QBPs compared to 7 percent of non-D-SNPs, and 27 percent of D-SNPs losing rebate dollars compared to 20 percent of non-D-SNPs.
• Due to the disproportionate impact the outlier policy will have on D-SNPs and their dually eligible enrollees, ACAP strongly recommends that CMS delay implementation of the outlier removal policy and reevaluate the methodology of this policy, the purpose of this policy, and the potentially disproportionate impact this policy will have on D-SNPs.

Star Ratings CAPHs Web-Based Mode:

o ACAP appreciates CMS’ efforts to increase the response rate of the CAHPs survey.
However, before implementing a web-based mode, we request that CMS conduct more
rigorous testing of the impact of inclusion of the web-based mode on CAHPs results for
subgroups of beneficiaries, including those dually eligible for Medicare and Medicaid,
and those who lack access to broadband.
• Potential New Measure Concepts: Social Connection Screening and Intervention:
o ACAP agrees with CMS on the importance of focusing on social connections screening
and intervention. However, we feel that this concept is not ready to be a quality measure
at this time.
• Potential New Measure Concepts: Assessing Mental Health Conditions, Mental Health Care
Access, and Health-Related Social Needs through the HOS Survey:
o ACAP agrees with CMS on the importance of measuring mental health care – both
prevalence of conditions and access to mental health services – as well as beneficiaries’
health-related social needs. However, we feel strongly that the HOS survey is not the
correct vehicle to collect data on mental health conditions, mental health care access, or
health-related social needs.
The following is a more detailed discussion of ACAP’s comments:
CMS-HCC Risk-Adjustment Model and Frailty Adjuster
ACAP has significant concerns that the proposed changes to the CMS-HCC risk-adjustment model
will disproportionately impact D-SNPs – including HIDE and FIDE SNPs – operated by Safety Net
Health Plans. First, although CMS’ modeling of the 2024 CMS-HCC model may suggest that the
model accurately predicts costs across all MA plans in aggregate, the impacts on individual D-SNPs,
HIDE SNPs, and FIDE SNPs may vary. Some of our Safety Net Health Plan members’ internal
modeling shows that risk scores and risk-adjusted payments for their D-SNP products (including
HIDE and FIDE SNPs) will decrease significantly due to the proposed changes to the risk model (initial
estimates range from 4.2 to 6 percent decrease in risk scores; one ACAP D-SNP estimates about a
$600 per person per year decrease in risk-adjusted payments for their plan). We note that both
Congress and CMS have taken major steps in recent years to improve the predictability of the Part C
risk model for D-SNPs and dually eligibles in general. We urge CMS to not implement any changes
that will negate CMS’ prior efforts and reduce the predictive accuracy of the model for D-SNPs.
CMS should also not implement changes to the risk model that undermine the sustainability of DSNPs, HIDE SNPs, and FIDE SNPs.
Moreover, FIDE SNPs will be experiencing a two-prong negative impact from the Advance Notice: a
reduction in risk-adjusted payments from the proposed changes to the CMS-HCC risk model, and a
4
reduction in risk-adjusted payments from a reduction of the amount of the frailty adjuster. One
ACAP FIDE SNP member preliminarily estimates a 10 percent combined reduction in risk scores from
both of these changes. CMS should be aware that the actual costs and risk profiles of full-benefit
dually eligible individuals do not change significantly from year-to-year. Changes to D-SNP risk
scores and risk-adjusted payments that result in large reductions from year-to-year, are a strong
signal to CMS that the proposed changes to the risk model may reduce the predictive accuracy of
the model for full-benefit dually eligibles enrolled in D-SNPs.
Given the potential for a significantly disproportionate impact on D-SNPs, we strongly request the
following:
• CMS must delay implementation of the proposed changes to the CMS-HCC risk model and the
reduction in the frailty adjuster.
• CMS must publish an impact assessment of the effect of the proposed changes to the risk model
(including the reduction of the frailty adjuster) on D-SNPs, HIDE SNPs, and FIDE SNPs. In addition
to publishing the impact assessment, CMS should publish the underlying data used in CMS’
analyses so that plans can use the underlying data to model the impact of the proposed risk
model changes on their plan. CMS should also give stakeholders more time to assess and
understand the proposed changes and to discuss their potential impact with CMS. We
recommend CMS establish a Technical Expert Panel (TEP) consisting of representatives from DSNPs and other stakeholders. The purpose of the TEP would be to assist CMS with the
development of policy solutions in instances where proposed risk model changes will have a
disproportionate negative impact or unintended consequences on subgroups of Medicare
beneficiaries.
• Any significant, future changes to the CMS-HCC risk model should be phased in over 3 years.
Consistent with CMS’ phase-in of the change from RAPs to EDS, CMS should understand that
removal of codes from the risk model and other significant changes will take time for plans to
work with their providers to understand the differences. A three-year phase-in also helps with
plan stability in instances where risk model changes will result in decreases to risk scores and
risk-adjusted payments.
ACAP has long emphasized the importance of payment accuracy in the CMS-HCC risk model. We
welcome further discussions with CMS on the impact of changes to the CMS-HCC risk model on DSNPs operated by community-based Safety Net Health Plans.
Universal Foundation of Quality Measures
ACAP is supportive of the concept and development of the Universal Foundation of Quality
Measures. A universal measure set would help ameliorate the reporting burden experienced by
plans and providers and would facilitate comparisons across quality programs. To improve the
5
ability of the Universal Foundation of Quality Measures to accurately measure quality of care, we
recommend that measures be thoroughly vetted across programs and populations before being
considered for inclusion in the measure set. Specifically, we recommend that any potential measure
(1) be proposed for stakeholder comment; (2) be included on the Star Ratings display page; (3) be
included in the Star Ratings for at least 3 years; and (4) have undergone rigorous analysis and testing
to ensure that the measure accurately reflects the quality of care for subgroups of beneficiaries,
including those dually eligible for Medicare and Medicaid. Finally, we urge CMS to collaborate with
stakeholders to ensure that measures selected for the Universal Foundation of Quality Measures
promote equity and avoid penalizing plans who enroll dually eligible individuals or other subgroups
of Medicare beneficiaries.
Star Ratings Outlier Removal Policy
As noted in our comments to the 2024 MA-PD Proposed Rule, ACAP has significant concerns about
the impact of CMS’ proposed outlier removal policy on D-SNPs. Specifically, we are concerned that
the outlier policy will have a disproportionately negative impact on D-SNPs. An analysis performed
by ZAHealth found that the outlier policy will result in 14 percent of D-SNP contracts losing their
QBPs compared to 7 percent of non-D-SNPs, and 27 percent of D-SNPs losing rebate dollars
compared to 20 percent of non-D-SNPs. The decline in Star Ratings and associated rebate dollars will
negatively impact the dually eligible beneficiaries enrolled in D-SNPs by reducing their access to
supplemental benefits.
Moreover, ZAHealth also estimates that the average Star Ratings loss from the outlier removal
policy will be larger than the average gain D-SNPs are expected to receive from both the existing CAI
and the proposed HEI (which will not be available for 4 more years) combined (see graph above).
We do not believe that CMS intended for the outlier removal policy to have a disproportionate
impact on D-SNPs, or for the impacts of this policy to be so skewed towards one specific type of
plan. However, our analyses do show this disproportionate negative impact on D-SNPs relative to
non-D-SNP plans. We believe this to be an unintended consequence of the outlier removal policy,
but one that is now realized, must be addressed.
6
Due to the disproportionate impact the outlier policy will have on D-SNPs and their dually eligible
enrollees, ACAP strongly recommends that CMS delay implementation of the outlier removal
policy and reevaluate the methodology of this policy, the purpose of this policy, and the
potentially disproportionate impact this policy will have on D-SNPs.
Star Ratings CAHPS Web-Based Mode
ACAP appreciates CMS’ efforts to increase the response rate of the CAHPs survey. However, before
implementing a web-based mode, we request that CMS conducts more rigorous testing of the
impact of inclusion of the web-based mode on CAHPs results for subgroups of beneficiaries,
including those dually eligible for Medicare and Medicaid, and those who lack access to broadband
(we note that lack of access to broadband is a problem for beneficiaries in both rural and urban
areas). After thoroughly testing the impact of the web-based mode on CAHPs results for subgroups
of beneficiaries, CMS should then publish the results of this analysis for stakeholder review and
comment. Thus, we request that CMS hold off on implementing a CAHPs web-based mode until
these impact analyses have been conducted and published for stakeholder comment.
Potential New Measure Concepts: Social Connection Screening and Intervention
ACAP agrees with CMS on the importance of focusing on social connections screening and
intervention. While we agree with the concept of measuring social connections and interventions,
we do not feel that this concept is ready to be a quality measure at this time.
First, more detail from CMS is needed on how data collection for this measure would be
operationalized, and specifically more detail on how CMS expects data for this measure to be
included in electronic health records. Additionally, while developing this measure, CMS should
consider that social isolation and loneliness can be transient, with individuals not necessarily
experiencing these feelings consistently throughout the year. CMS should also consider that a
provider documenting that an intervention occurred (e.g. a referral was made), does not guarantee
that an intervention occurred. Moreover, we note that other data sources, such as Health Risk
Assessments, could be more reliable sources of data for this information. And finally, we
recommend that the denominator for the potential measure be limited to individuals with
depression or anxiety so that the proposed survey measure is targeted to those members at higher
risk for social isolation.
Potential New Measure Concepts: Assessing Mental Health Conditions, Mental Health Care
Access, and Health-Related Social Needs through the HOS Survey
7
ACAP agrees with CMS on the importance of measuring mental health care – both prevalence of
conditions and access to mental health services – as well as beneficiaries’ health-related social
needs. As CMS knows, both mental health care and health-related social needs are key challenges
experienced by dually eligible individuals. However, we feel strongly that the HOS survey is not the
correct vehicle to collect data on mental health conditions, mental health care access, or healthrelated social needs. The HOS is fielded on a random sample of beneficiaries and does not result in
actionable data for a plan or the ability for CMS to track outcomes. Rather than adding these data
elements to the HOS survey, we suggest that CMS re-evaluate the goals of collecting this data and
consider other vehicles for collection of this information.
Additional Comments on Quality Measures
We also offer the following recommendations on specific quality measures for CMS’ consideration:
• Regarding the Identifying Chronic Conditions in HEDIS Measures, we request that CMS provide
additional information on how chronic conditions would be identified using clinical data.
• We recommend that CMS reduce the Getting Appointments and Care Quickly measure to the
proposed two items.
• For the Chronic Pain Assessment and Follow-up Measure, we request that CMS provide
additional details to help stakeholders better understand how multidimensional pain will be
defined.
• We request that CMS clarify whether the Timely Follow-up After Acute Exacerbations of Chronic
Conditions measure will replace the Follow-Up After Emergency Department Visit for People
with Multiple High-Risk Chronic Conditions (FMC), which is currently a Star Ratings measure.
• We note that, for consistency, any changes made by CMS to the Care for Older Adults measure
should reflect any changes or updates adopted by NCQA.
• Regarding the Care for Older Adults – Functional Status Assessment and Medication Review
measure, we recommend that CMS wait to report this measure for a wider population until the
measure has been fully tested and is ready to move from the Display Page to Star Ratings.
• For the Care for Older Adults – Pain Assessment measure, we agree that retirement of this
measure is appropriate. However, CMS should not retire this measure from the HEDIS
measurement set until an appropriate replacement measure has been developed and is ready to
move from the Display Page to Star Ratings.
• With respect to the Part D Medication Adherence for Diabetes Medication/Medication
Adherence for Hypertension (RAS Antagonists)/ Medication Adherence for Cholesterol (Statins)
measures, we are concerned that removing the IP/SNF stay adjustment from the adherence
measures inappropriately penalizes plans by making members appear non-adherent during an
inpatient stay. CMS could more accurately reflect inpatient stays by excluding this population
from the denominator.
• For the Part D MTM Program Completion Rate MTM Program Completion Rate for
Comprehensive Medication Review (CMR) measure, ACAP stated our concerns with the
proposed expansion of the MTM program eligibility in our comment letter on the 2024 MA-PD
8
Proposed Rule. We note here that CMS’ proposed MTM program eligibility changes would
constitute a significant change to the eligible population, which we believe necessitates the
MTM Comprehensive Medication Review (CMR) completion rate measure being moved to the
Display Page.
• With respect to Diabetes Care – Eye Exam and Diabetes Care – Blood Sugar Controlled, we
request that CMS provide additional details about the specifications on use of GMI so that plans
can more meaningfully assess whether the GMI might be a reasonable alternative to A1c. We
also note that any changes adopted by NCQA to how eligible members with diabetes are
identified, should also be applied by CMS.
• We feel that the Initiation and Engagement of Substance Use Disorder (SUD) Treatment
measure should not be moved to the Star Ratings. This is not an appropriate measure of the
adequacy for substance use disorder treatment and follow-up care as the specification coding
requirements to identify actual new exacerbations often results in the inclusion of non-events in
the denominator, leading to artificially low initiation and engagement rates.
• Regarding the Timely Follow-up After Acute Exacerbations of Chronic Conditions measure, we
recommend that NQF further define what is considered an “acute exacerbation” prior to issuing
proposed rulemaking, as this detail is not currently publicly available.
• For the Adult Immunization Status measure for Parts C and D, we recommend that CMS update
the measure specifications to include consideration of vaccine refusals (i.e., for religious
exemptions).
• With respect to Concurrent Use of Opioids and Benzodiazepines (COB), Polypharmacy Use of
Multiple Anticholinergic Medications in Older Adults (Poly-ACH), and Polypharmacy Use of
Multiple Central Nervous System Active Medications in Older Adults (Poly-CNS) Part D
measures, we are concerned about the impact of including these Part D measures in 2026 Star
Ratings on medically complex individuals and plans serving those populations. We recommend
that CMS not include these three measures in Star Ratings. However, if they are included in the
Star Ratings in 2026, then the measures should be case-mix adjusted for enrollee SDS
characteristics.
• We disagree that the Antipsychotic Use in Persons with Dementia, Overall (APD)/Antipsychotic
Use in Persons with Dementia, in Long-Term Nursing Home Residents (APD-LTNH) (Part D)
measure, is an appropriate mechanism for assessing treatment of mood and behavior changes
in individuals with dementia. Additionally, similar to our concerns with the COB, Poly-ACH, and
Poly-CNS measures, we are concerned about the impact of including these Part D measures on
medically complex individuals. Rather than move forward with the measure, we recommend
CMS reconsider how treatment pathways for dementia patients with behavior issues are
assessed and whether another measure, specifically one that focuses on treatment of the
population, could be more beneficial.
9
We thank you for the opportunity to comment and for your time and attention. If you have any
questions regarding the above comments, please do not hesitate to reach out to Christine Aguiar
Lynch, Vice President of Medicare and MLTSS Policy (clynch@communityplans.net).
Sincerely,
/s/
Margaret A. Murray
Chief Executive Officer
Association for Community Affiliated Plans

View the full article »