Print Friendly, PDF & Email

Submitted electronically via

November 3, 2014

Centers for Medicare & Medicaid Services
Department of Health & Human Services
Attention: CMS-6050-P
P.O. Box 8013
Baltimore, MD 21244-8013

Re: Request for Information regarding Data on Differences in Medicare Advantage (MA) and Part D Star Rating Quality Measurements for Dual-Eligible versus Non-Dual Eligible Enrollees

To Whom It May Concern:

The Center for Medicare Advocacy (Center) greatly appreciates the opportunity to provide information for the request for information regarding the relationship between enrollment of individuals dually eligible for Medicare and Medicaid, and MA and Part D quality measure scores.

The Center, founded in 1986, is a national, non-partisan education and advocacy organization that works to ensure fair access to Medicare and to quality healthcare.  We draw upon our direct experience with thousands of individuals and their families to educate policymakers about how their decisions affect the lives of real people.  Additionally, we provide legal representation to ensure that beneficiaries receive the health care benefits to which they are legally entitled, and to the quality health care coverage and services they need.


We are pleased that CMS is eager to improve the quality of care all patients receive, and is particularly concerned with improving care for disadvantaged populations, who are currently disproportionately afflicted by lower quality care, as reflected in this Request for Information.  It is essential that CMS address the existing differences in quality of care.  Therefore, the Center applauds CMS for seeking data on this important issue in order to identify existing disparities in quality, and for refraining from making changes to the Star Ratings program in the interim.  We urge CMS to approach changes to quality measurement and risk adjustment with caution.

The Center has serious concerns regarding incorporating socioeconomic status (SES) and sociodemographic status (SDS) like income, education, race and ethnicity, in quality performance measures, and performing risk adjustment for these factors in accountability applications.  Though we believe it is an error to conflate quality measures and payments, it is clear that as payments are increasingly tied to quality performance scores, the two areas are linked.  We agree with CMS that performance measures should aim to identify disparities in care and strive to eliminate these disparities.

There are data to suggest real disparities in care based on SES/SDS factors exist.  These differences are best resolved by identifying them and creating systems that work to improve care, rather than masking these differences through artificial inflation of performance scores through risk adjustment.  Quality measurements are designed to reveal disparities in care, and spur changes in order to address those disparities. 

We are concerned that risk adjustment will mask these disparities and disincentivize healthcare units from making the changes that could equalize care, making quality analysis and quality ratings useless.  Our response to this Request for Information will address these concerns, hold up an example of a high performing plan serving low income people, and point to alternative approaches.

Data Challenges

Disadvantaged Patients More likely to Receive Lower Quality Care

As stated previously, we are aware that data exist that suggest disadvantaged patients are more likely to receive poor care.  It appears that the data demonstrate correlation rather than causation.  For example, CMS' public comments on a draft report released by The National Quality Forum (NQF) regarding risk adjustment for SES/SDS, address the existence of a correlation of low income and poor plans, as opposed to the existence of a causal relationship between dually eligible enrollees and poor plan performance.  “Previous analyses have shown that in some cases patients with low SES do concentrate in providers, hospitals, and plans that provide lower quality of care to all patients, so adjusting for this patient characteristic could adjust away true differences in quality across plans.” [1]

Also, the 2012 National HealthCare Disparities Report released by the Agency for Healthcare Research and Quality (AHRQ) found that disadvantaged patients were more likely to receive poor quality care.  “We find that racial and ethnic minorities and poor people often face more barriers to care and receive poorer quality of care when they can get it.” [2]

The IOM Committee on Understanding and Eliminating Racial and Ethnic Disparities in Health Care had similar findings in its 2002 report regarding lower quality care for low income and minority patients.[3]  The report found that several factors may undergird this result, such as language and cultural barriers, potential provider bias, and the possibility that minorities are disproportionately enrolled in lower-cost health plans that place greater per-patient limits on healthcare expenditures and available services.[4]  “Three mechanisms might be operative in healthcare disparities from the provider’s side of the exchange: bias (or prejudice) against minorities; greater clinical uncertainty when interacting with minority patients; and beliefs (or stereotypes) held by the provider about the behavior or health of minorities.”[5]

CMS’ public comments on NQF’s draft report also addressed the importance of providers understanding cultural differences as a means to improve patient care.  “[T]he expertise of staff can and does affect outcomes, such as a provider who does not understand the cultural and social expectations of a patient when proposing/explaining a treatment option.  In Medicare, it is this very notion that sets apart Medicare Advantage and Prescription Drug plans that appear to be successful in achieving outcomes and those that do not.  Providers must gain additional expertise in understanding barriers to medical outcomes in order to address the disparities among groups.”[6] CMS concluded that NQF’s proposed approach would not incentivize providers to make the effort to understand cultural differences and tailor care accordingly.  Rather, CMS stated, it would “accept this inequality as reasonable and adjust away the difference.” [7]

As the research suggests, disadvantaged patients disproportionately receive poor care.  However, the data do not indicate that disadvantaged patients cause poor quality ratings.  As there is little publically available data demonstrating a causal link between SES/SDS factors and lower quality measure scores for MA and Part D plans, suggestions that dual enrollment in a plan causes low performance are anecdotal and do not reflect conclusions based on supporting data.  Without these data it is premature to make changes to the Star Ratings program.[8]

The National Quality Forum (NQF) Report

CMS’ Request for Information notice cites to an August 2014 final report released by The National Quality Forum (NQF) regarding risk adjustment for SES/SDS, Risk Adjustment for Socioeconomic Status or Other Sociodemographic Factors,[9] studying performance measures used in accountability applications.  Therefore, we address below the findings of NQF’s final report in our analysis of the issue of risk adjustment.  

NQF is a non-profit organization that establishes frameworks for quality measures and is the main quality measure endorsement group in the country.  NQF endorses performance measures that are intended for use both in performance improvement and in accountability applications like pay-for-performance payments and penalties, and public reporting. [10]  The overall performance measure score is used to make a conclusion about a “healthcare unit’s” quality (this can refer to an hospital, health plan, practice or other unit).[11]

NQF’s current Measure Evaluation Criteria expressly forbid using SES/SDS factors in risk adjusting quality measures for endorsement for accountability applications.  According to NQF’s Measure Evaluation Criteria, “[r]isk models should not obscure disparities in care for populations by including factors that are associated with differences/inequalities in care, such as race, socioeconomic status, or gender. . . . It is preferable to stratify measures by race and socioeconomic status rather than to adjust out the differences.”[12]

However, the NQF Panel released recommendations in its 2014 final report in support of risk adjustment.[13]  These included a recommendation that NQF criteria be revised to allow for the inclusion of SES/SDS factors in risk adjustment. [14]  This proposed policy would present a reversal of NQF’s quality measurement criteria.  This proposed change would mean that NQF would endorse measures that adjust performance scores for healthcare units where it is found that the sociodemographic factors theoretically affect an outcome or process of care reflected in the performance measure.  Up until this report, NQF has consistently refused to incorporate SES/SDS factors into its quality measure endorsement.  The NQF Panel’s proposed change has not yet been included in NQF’s Measure Evaluation Criteria; the current NQF criterion regarding SES/SDS still expressly forbids such factors.

Despite recommending such a tremendous shift from NQF’s current criteria, the NQF report provides insufficient evidence to support this recommendation.  The lack of data in the report regarding the causal relationship between these SES/SDS factors and poor healthcare unit quality undermines NQF’s methodology for suggesting the change.  The Consumer-Purchaser Alliance (C-P Alliance), a collaboration of labor, consumer and employment groups, also expressed similar concerns in their public comments on NQF’s draft report. [15]  “The recommendation to risk-adjust for patient-level sociodemographic factors is not accompanied by sufficient evidence showing that the current policy is resulting in harm to disadvantaged patients.  Further, the recommendations are not grounded in evidence that any of the proposed actions would prevent or preclude harm from occurring.  This is inconsistent with NQF’s standards of scientific acceptability.”[16]  The Center agrees with C-P Alliance’s analyses, and is also concerned that the recommendations presented in the NQF report do not comport with NQF’s usual standards of grounding measurement in scientific evidence.

CMS, in its public comments on NQF’s draft report, expressed concern regarding the lack of evidence that NQF used for its proposed change.  “We are concerned, however, that the recommendation that is being made to risk adjust for sociodemographic factors is premature, given the lack of evidence that has been generated to warrant such a recommendation.  Changing the criteria for NQF endorsement will have a sweeping impact, and should be based only on the strongest of evidence.”[17]

            Limited Data Collection

Another data concern in risk adjustment is that data collection for SES/SDS is quite limited.[18] Even the NQF report acknowledged this fact.[19]  NQF’s report also acknowledged that this lack of data coupled with the limitations in collecting this type of data, “may be the biggest barrier to adjustment for sociodemographic factors,” and “require[s] further initiatives to define standards and to implement data collection.”[20]  NQF’s report underscores this lack of data collection, by including a recommendation that strategies be developed to “identify a standard set of sociodemographic variables (patient and community-level) to be collected and made available for performance measurement and identifying disparities.”[21]  It is evident that even the variables used in collecting data have yet to be determined and standardized.

CMS’s public comments also addressed the challenges and likely costs involved in the standardized collection of SES/SDS data for patients.  “SES in particular can be difficult to measure, and typically requires multiple indicators, including employment status/history, income, and education level.  It is difficult to collect social support data, given that reports of social support tend to be confounded with the individual’s need for social support.”[22]

Without adequate data it is unclear how the NQF Panel could establish causality with enough confidence to reverse a core criterion in NQF’s quality assessment methodology.  Given these data challenges, we encourage CMS to review such recommendations with caution.  For CMS to make such a fundamental change in quality standards by using such factors in Part C and Part D plan performance ratings, there must be convincing data that have been rigorously collected and reviewed.  It is evident that these data challenges constrain risk adjustment from being an accurate or effective tool for quality improvement.

IMPACT Act and Data

On October 6, 2014, President Obama signed into law the “Improving Medicare Post-Acute Care Transformation Act of 2014” or IMPACT Act.[23]  The bill resulted from a bipartisan, bicameral effort by the House Ways & Means and Senate Finance Committees to address various issues relating to care Medicare beneficiaries receive once they are discharged from the hospital.  The IMPACT Act focuses primarily on standardizing post-acute care (PAC) assessment data relating to quality, payment and hospital discharge planning.  The Act standardizes data collection and assessment over the next few years with the aim of gathering enough information to make further, more fundamental changes to how Medicare approaches and pays for PAC.

The data that will result from the implementation of the IMPACT Act will be essential for CMS in making determinations regarding quality assessment.  CMS should utilize these data in its appraisal of the Star Ratings program.  Therefore, we urge CMS to review the results of the data that are collected and compiled as a result of this Act prior to making any changes to the Star Ratings program.

High Quality and Dual Eligible Beneficiaries

In analyzing the relationship between dual eligible status and quality performance, it is essential to examine successful programs treating a majority of disadvantaged patients.  Only through this analysis can CMS obtain a comprehensive understanding of the quality landscape.  We are encouraged that CMS has requested information regarding the achievement of high quality performance among plans serving dual eligible beneficiaries, and how those results were obtained.

A 2014 Center for Health Care Strategies study found that there are several key attributes of high performing integrated health plans that treat dually eligible Medicare and Medicaid enrollees.[24]  These attributes include: a culture of leadership and organizational structure that reflects a commitment to integrated care; infrastructure to "scale up" and "stretch out" without compromising quality and value; financial and nonfinancial incentives and related mechanisms that align plan, provider and member interests; coordinated care provided through comprehensive, accessible networks and person/family-centered care planning.[25]

A 2007 study conducted by the Commonwealth Fund’s Commission on a High Performance Health System, also found that it is in fact possible to attain high quality measure scores while treating low SES patients.[26]

Examples of high quality plans treating large populations of dually eligible individuals include: Commonwealth Care Alliance in Massachusetts, iCare in Wisconsin, UCare in Minnesota, and Denver Health in Colorado.  We will focus on Denver Health as a potential model of an effective healthcare unit with a large percentage of dual eligible enrollees that provides high quality care with excellent outcomes at length below.

Denver Health is Colorado’s largest health care safety-net provider and major Medicaid provider and has a national reputation as a high performance organization.[27]  One of Denver Health’s initiatives, Westside Family Health Center, demonstrates Denver Health’s effective approach to high quality patient care for low income populations.  The Westside Family Health Center’s innovations include:

  • "Open access" scheduling: 60 % of appointments are reserved so that patients can call in on the same day to be seen (has reduced no-show rates from about 30% to 15%).
  • Spanish fluency: 75 % of the providers at the clinic are fluent in Spanish.
  • Group visits for patients with certain conditions (i.e. diabetes) to help them with the management of their self-care, such as proper diet and regular blood-testing at home.
  • Convenient location next to a bus line and directly next door to the Department of Human Services, so interrelated social services can be addressed.
  • 24-hour call line to address questions or concerns regardless of the time of day or night (also available in Spanish).
  • Denver Health’s annual investment of $800,000 in interpreters and translation services provides assistance.

The Commonwealth Fund’s Commission on a High Performance Health System found that the innovations employed by Denver Health “are transferable to other settings.”[28]  The successes at Denver Health underscore the need for poor performing healthcare units with high dual enrollment to make changes in their processes to provide person-centered, culturally appropriate care.  The Commission found that “[Denver Health’s] best practices, and the lessons learned from the significant barriers it has overcome, can form a ‘learning laboratory’—a potential model—from which other states and the nation may benefit.”[29]

CMS’ public comments on NQF’s draft report also asserted that high performing health units treating disadvantaged patients can provide high quality care; CMS stated that their existence also undermines the need for risk adjustment.  “Further, the argument that not risk adjusting has the effect of driving providers/insurers away from low SES patients is directly contradicted by the growth in D-SNPs (322 plans in 2012 with 1,303,408 enrolled; 353 plans in 2014 with 1,576,291 enrolled) as the Star Rating program has been implemented.”[30]

The evidence of high quality healthcare units with large dual beneficiary populations undermines arguments that low SES/SDS and high quality are mutually exclusive without risk adjustment.  These data contradict low performing plans’ claims that due to high dual eligible enrollment, they are unable to achieve high quality without an artificial increase in their performance score from risk adjustment.  High quality care is possible for healthcare units with dual eligible beneficiaries; high quality care should be expected of all healthcare units, for all of their patients.

Unintended Consequences

We are concerned that risk adjustment for quality reporting and pay-for-performance programs based on SES/SDS factors will lead to several harmful unintended consequences for disadvantaged patients.  First of all, risk adjustment has the potential to mask existing disparities in care that low income patients receive, rather then expose and address these disparities.  Simply adjusting away these real differences only perpetuates the inequities.

Also, risk adjustment could create two divergent standards of care for healthcare units based on the wealth or poverty of the populations they serve.  Adjusting scores for healthcare units with significant proportions of disadvantaged patient populations would in effect lower the bar for healthcare units that treat these populations.  This type of adjustment would allow distinct and unequal quality standards for poor patients and wealthy patients. 

CMS’ public comments on NQF’s draft report expressed similar concerns regarding the risk of setting lower standards for disadvantaged patients through risk adjustment.  “CMS is concerned that the approach for addressing this problem as proposed by NQF may be inappropriate as it could result in an increase in health disparities, rather than a decrease.  Currently, CMS does not adjust quality outcome measures for patient socio-economic status (SES) because of the concern that doing so would establish a different standard of care for providers based on the socioeconomic status of the patients they care for, and can mask disparities in the quality of care provided.”[31]

Also, the root of the disparities in care is not likely to be addressed if the differences are concealed through the automatic and inaccurate inflation of performance scores.  For example, risk adjustment could raise the performance status of some healthcare units from an alarming “substandard” to the “average” level, or from “average” to “good,” without actually making any improvements in the quality of care.[32]  CMS must not allow a healthcare unit to receive the recognition and payment incentives attributed to an “average” healthcare unit when they are actually providing a “substandard” level of care.  

The NQF report addressed the concern of lessening incentives to improve care for low income patients.  “It is unknown whether such a change in labeling will have an impact on motivation to improve, but there is, of course, still an opportunity for such a unit to raise itself to a ‘superior’ level by implementing solutions to problems that affect outcomes for disadvantaged patients.”[33]

This response is alarming for several reasons.  It is troubling that the report views the resulting changes in healthcare unit status after risk adjustment as merely “a change in labeling.”  This significantly downplays the impacts on care that result from the Star Rating in terms of incentives, payment, reputation, and improvements in care.  It also reveals that the NQF Panel is aware that risk adjustment might lead to reverse incentives, and recommended the change in NQF’s criteria regarding risk adjustment nonetheless.  It does not allay the Center’s serious concerns regarding risk adjustment that the report states that healthcare units may, if they like, actually undertake changes to improve care if they are ambitious enough to seek an even higher quality score than their newly adjusted score.  Healthcare units should be focused on developing and implementing these solutions to problems that affect outcomes for disadvantaged patients. Healthcare units should not simply receive an adjusted score without any improvements in care quality, as the NQF report suggests.  This recommendation from the NQF report would allow healthcare units to view the development of these solutions as optional, rather than as an essential element of care.

It is significant to note that the National Committee for Quality Assurance (NCQA) does not adjust for SES in its development of quality measures, and is opposed to such risk adjustment.  NCQA is a nonprofit that serves as the largest developer of quality standards and performance measurements for a broad range of healthcare entities.  NCQA develops the widely used Healthcare Effectiveness Data and Information Set (HEDIS) tool; more than 90 percent of America's health plans measure performance on important dimensions of care and service through HEDIS.[34]

NCQA submitted public comments on the NQF draft report citing concerns similar to those expressed here by the Center regarding the various potential harms to low income patients.  NCQA stated in their comments that risk adjustment for SES “would unfairly lock in lower expectations for the very populations that most need better quality.”[35]

CMS’ public comments on NQF’s draft report expressed similar concerns regarding risk adjustment masking actual disparities, and thereby resulting in a reduction in incentives for healthcare units to make changes that could equalize care.  “Risk adjustment of quality measures for SES may reduce incentives to achieve high quality clinical goals for the economically disadvantaged.”[36]

Quality Improving

CMS’ current Star Ratings program for MA and Part D appears to be leading to improvements in quality of care.  A 2014 Avalere analysis of CMS quality data found that overall MA quality ratings are improving.[37]  The analysis found that, for plan year 2015, approximately 60 percent of MA enrollment is in four or five star plans, an increase from the 52 percent of MA enrollment in four or five star plans in 2014.[38]

A 2014 Kaiser Family Foundation analysis of entering and exiting plans also found that as quality improves, Medicare beneficiaries will see fewer plans with average or below average ratings in 2015.[39]

“Plans with relatively low enrollment and plans with average star quality ratings or below comprise the majority of plans exiting the markets.  These changes are primarily weeding out plans that did not attract many enrollees and plans that received relatively low quality ratings. The CMS quality-based bonus demonstration will draw to a close at the end of 2014, and only plans with above average ratings (4 or more stars) will receive bonuses in 2015 and future years. This could make it more challenging for plans with fewer stars to compete with higher-rated plans, unless they improve their ratings.”[40]  This analysis demonstrates that a disproportionate number of poor performing plans dropped out of the market.

Risk adjustment could reverse the tremendous progress that has been made in the quality arena. NCQA stated in their comments to the NQF proposal that their progress in quality improvement would be hindered through risk adjustment.  “Risk adjustment for SES and sociodemographic factors at the measure level will impede our progress by artificially reducing variation in performance measurement, putting a filter over the bright light of transparency, and lowering the bar of accountability.” [41]

CMS’ public comments on NQF’s draft report expressed similar concerns that risk adjustment for SES/SDS could impact the integrity of quality measures.  “[T]he quality gap between higher SES individuals and lower SES individuals may increase as a result of the proposed requirement [from NQF] to risk adjust outcome measures.  CMS is concerned that this recommendation could be a setback to the goal of equity within the healthcare system.”  Quality measurements are designed to reveal actual disparities in care, and serve as an impetus to change.  We are concerned that risk adjustment will mask these disparities and thwart changes that improve care. Without this process, quality analysis and quality ratings become meaningless.  We urge CMS to continue studying this issue and collecting the necessary data.


We agree that disparities exist between the quality of care delivered to low income populations and that delivered to higher income populations.  We also recognize that there are social and economic reasons that require alternative interventions or approaches by healthcare units in order to limit disparities in care for different populations.  It is critical to identify those factors, and develop systems to address the obstacles to high quality care for these populations.  

As evidenced by the successful efforts of healthcare units like Denver Health, healthcare units treating a high proportion of low income patients can provide quality care if they focus on person-centered care.  Healthcare units must recognize the role of non-medical factors, and address these realities in their care plans, particularly for low income populations who face additional challenges to treatment.  

CMS has indicated that healthcare units who receive low quality scores, and are treating a large proportion of disadvantaged patients, should address the needs of these patients, rather than receive an adjusted score.[42]  CMS stated in its public comments to the NQF draft report that plans that “enroll difficult to reach populations” should not receive “special allowances,” because “these plans have knowingly focused their enrollment efforts on this group and…their service delivery methods should be focused on the special needs of these groups.” [43]

Patients who receive person-centered care report higher quality care, and in turn have improved outcomes.  A Commonwealth Fund report from 2012 found that patients who reported high quality care were better able to manage chronic illnesses, and were more likely to receive preventive tests and services when they received reminders from their doctors. [44]

We are confident that because CMS seeks to utilize the Star Rating system to encourage continuous quality improvement in the MA and Medicare Prescription Drug programs, CMS will not accept any changes to performance measurement that would lead to masking disparities and harming disadvantaged patients.  We support CMS’ efforts to collect data and review this issue. We agree with CMS that changes to performance quality measures "should be based only on the strongest of evidence."[45]  We are certain that CMS will proceed with caution when reviewing any changes to the Star Rating system.

The Center greatly appreciates the opportunity to provide information for this request for information.  For further information please contact Center Policy Attorney Kata Kertesz at

Kata Kertesz
Policy Attorney

[1] Risk Adjusting for Sociodemographic Factors Draft Report-CMS Comments, Centers for Medicare & Medicaid Services 1, 5 (April 14, 2014) [hereinafter CMS Comments].
[2] National Healthcare Disparities Report, Agency for Healthcare Research and Quality (last modified Nov. 3, 2014), available at
[3]Unequal Treatment: What Healthcare Providers Need to Know About Racial and Ethnic Disparities in Healthcare, Institute of Medicine 1 (March 2002), [hereinafter Unequal Treatment].
[4] Id. at 3.
[5] Id.
[6] CMS Comments, supra note 1 at 5.
[7] Id. at 5.
[8] Id. at 1: “We are concerned, however, that the recommendation that is being made to risk adjust for sociodemographic factors is premature, given the lack of evidence that has been generated to warrant such a recommendation.”
[9] Risk Adjustment for Socioeconomic Effects or Other Sociodemographic Factors, National Quality Forum 1 (August 15, 2004), [hereinafter Risk Adjustment Report].
[10] Id. at 1.
[11] Id. at v.
[12]Measure Evaluation Criteria, National Quality Forum (last visited Nov. 3, 2014),; see also Risk Adjustment Report, supra note 9 at vii.
[13] Risk Adjustment Report, supra note 9 at iii .
[14] Id. at 9.
[15] Debra Ness, William Kramer, Consumer-Purchaser Alliance 2 (April 16, 2014)
[16] Id.
[17] CMS Comments, supra note 1 at 1-2.
[18] National Healthcare Disparities Report, supra note 2. AHRQ’s summary report states: “Identifying problems, targeting resources, and designing interventions all depend on reliable data. Unfortunately, data on underserved populations are often incomplete. Some data sources do not collect information to identify specific groups. Other data sources collect this information, but the numbers of individuals from specific groups included are too small to allow reliable estimates.” National Healthcare Disparities Report, supra note 2; see also Unequal Treatment, supra note 3.
[19] Risk Adjustment Report, supra note 9 at 40, “SES-related data are not widely collected.”
[20] Id. at 40.
[21] Id. at 12.
[22] CMS Comments, supra note 1 at 3.
[23] Improving Medicare Post-Acute Care Transformation Act of 2014, Pub. L. No. 113-185, 128 Stat. 1952,  available at
[24] Penny Hollander Feldman, Key Attributes of High-Performing Integrated Health Plans for Medicare-Medicaid Enrollees, Center for Healthcare Strategies, Inc. 1 (2014),
[25] Id. at 3-5.
[26] Rachel Nuzum et al., Denver Health: A High-Performance Public Health Care System, The Commenwealth Fund 1 (2007),–a-high-performance-public-health-care-system.
[27] Id.
[28] Id. at 3.
[29] Id. at 12.
[30] CMS Comments, supra note 1 at 6.
[31] Id. at 4.
[32] See also CMS Comments, supra note 1 at 4: “Risk-adjustment for SES or other demographic factors may result in the appearance of relatively improved outcomes for this hospital.”
[33] CMS Comments, supra note 1 at 4.
[34] NCQA HEDIS & Performance Measurement, (last visited Nov. 3, 2014) 
[35] Margaret O’Kane, NCQA 1 (2014),
[36] CMS Comments, supra note 1 at 5.
[37] Avalere, , (last visited Nov. 3, 2014).
[38] Id.
[39] What’s In and What’s Out? Medicare Advantage Market Entries and Exists  for 2015, The Henry J. Kaiser Family Foundation 1, 4 (2014),
[40] Id.
[41] O’Kane, supra note 35 at 2.
[42] CMS Comments, supra note 1 at 6.
[43] Id.
[44] Julia Brenenson et. al., Achieving Better Quality of Care Low Income Populations: The Roles of Health Insurance and the Medical Home in Reducing Health Inequities, The Commenwealth Fund 1, 5-6 (2012),
[45] CMS Comments, supra note 1 at 2.




Comments are closed.