The Centers for Medicare & Medicaid Services (CMS) uses the CMS-Hierarchical Condition Category (HCC) risk-adjustment model to pay Medicare Advantage (MA) plans a capitated amount per member per month. The CMS-HCC model includes historic utilization, diagnoses, and costs for fee-for-service (FFS) Medicare to predict costs for MA plans. While the model accounts for certain demographic details, including age, sex, and dual and disability status, it does not account for beneficiaries’ race/ethnicity.

Stakeholders have raised concerns with various elements of the current CMS-HCC model, including the appropriate use of beneficiary demographics data and decisions on the inclusion of chronic diseases in the model. Previous research shows that, because the model predicts healthcare costs based on FFS Medicare spending, it can underpredict costs for beneficiaries who have had low spending as a result of systemic access barriers. In other words, the model coefficients may build inequities into the MA plan payment structure and perpetuate health disparities.

Approach

Avalere conducted an analysis of the CMS-HCC community model to assess its accuracy in predicting costs for groups of beneficiaries categorized by race/ethnicity. To do so, Avalere estimated predicted costs from the model compared to actual costs. After using claims data to first identify and then calculate the actual costs of FFS beneficiaries, Avalere compared the predicted costs for this sample using the CMS-HCC community model in use for plan year 2023.

The ratio of predicted costs to actual FFS Medicare spending is known as the predictive ratio. A predictive ratio of 1.0 means a beneficiary’s (or a group of beneficiaries’) expected healthcare costs are accurately predicted. The further the predictive ratio diverges from 1.0, the more significant the underprediction or overprediction (i.e., a predictive ratio greater than 1.0 means the model estimated more than actual costs and a predictive ratio less than 1.0 means estimated costs were less than actual costs).

Overall Results

Avalere’s analysis found that the CMS-HCC model, on average, closely predicts healthcare costs for beneficiaries who are Black and non-Hispanic White; it overpredicts for beneficiaries who are Asian/Pacific Islander and those who are Hispanic; and it underpredicts for beneficiaries who are American Indian/Alaska Native (Figure 1).

Race/Ethnicity and Dual Status

Avalere also analyzed the predictive ratio by beneficiaries’ race/ethnicity to stratify by beneficiaries’ dual-eligible status. Results for this analysis showed the CMS-HCC model underpredicts costs for full and partial dual beneficiaries who are Black compared to non-dual beneficiaries who are Black. Further, the model underpredicts across every category for American Indian/Alaska Native beneficiaries. Although the model more accurately predicts costs for dual-eligible beneficiaries who are Hispanic, the model demonstrates a greater overprediction of costs for non-dual beneficiaries who are Hispanic (Figure 2).

Disease/Condition Category

Avalere’s analysis showed the CMS-HCC model accurately predicted overall costs for beneficiaries who are Black. However, for beneficiaries with the most common HCCs, the model underpredicts for Black beneficiaries with eight of the top ten conditions. The model generally overpredicts for all beneficiaries who are Hispanic; however, the predictive ratios for specific conditions are generally lower than for the full group of beneficiaries (Table 1).

Table 1. Predictive Ratios by Race/Ethnicity and Disease Type

Top 10 HCCs*	Overall	Non-Hispanic White	Black	Asian/Pacific Islander	Hispanic	American Indian/Alaska Native
Overall Predictive Ratio	1.00	0.99	1.01	1.25	1.09	0.93
1. Diabetes with Chronic Complications	0.98	0.97	0.96	1.15	0.99	0.91
2. Vascular Disease	0.98	0.99	0.91	1.06	0.99	0.95
3. Specified Heart Arrhythmias	0.98	0.99	0.94	1.12	0.97	0.97
4. Chronic Obstructive Pulmonary Disease	0.98	0.99	0.91	1.12	0.95	0.96
5. Congestive Heart Failure	0.98	0.99	0.91	1.09	0.96	0.94
6. Major Depressive, Bipolar, and Paranoid Disorder	0.98	0.96	1.00	1.15	1.02	0.89
7. Diabetes without Complication	0.98	0.98	0.95	1.12	0.96	0.98
8. Rheumatoid Arthritis and Inflammatory Connective Tissue	0.98	0.98	0.97	1.13	1.00	0.95
9. Breast, Prostate, and Other Cancers and Tumors	0.98	0.97	1.00	1.14	1.03	0.95
10. Morbid Obesity	0.97	0.97	0.96	1.06	0.97	0.97

*Ranked by number of beneficiaries with the condition, largest to smallest.

The current model also underpredicts for enrollees who are Black and those who are Hispanic when they have at least 5 HCCs, and it overpredicts for beneficiaries who are non-Hispanic White who have at least five HCCs (Figure 3). For each of these groups, the finding for those with at least five HCCs is opposite of the finding for these groups of beneficiaries who have zero and up to four HCCs.

Discussion

An optimal risk-adjustment model accurately predicts costs overall and for subpopulations, including racial/ethnic minorities. Avalere’s research indicates that the CMS-HCC risk-adjustment model may incorrectly predict costs for certain subpopulations, which might perpetuate disparities by overpaying for some low-cost populations and underpaying for some high-cost groups of beneficiaries.

This study illustrates that, as beneficiaries have more diagnosed conditions, overprediction increases for those who are non-Hispanic White and underprediction emerges for those who are Black and those who are Hispanic. This finding may be due to the HCC count variable mandated by the 2016 21st Century Cures Act, which increased predicted costs for higher HCC counts. Further study may be warranted to better understand why trends in predictive accuracy differ across racial groups as the number as HCC counts go up, and whether the addition of the Cures Act may impact the accuracy of the model for specific subpopulations.

CMS enacted changes to the CMS-HCC risk-adjustment model as part of the 2024 MA and Part D Advanced Notice that would update the data years used for calibration, update the denominator year used to determine the average per capita predicted expenditures, and reclassify the HCCs using International Classification of Diseases, Tenth Revision, Clinical Modification codes. Further study would also be needed to evaluate the accuracy of this new model on payments for subpopulations of Medicare beneficiaries.

Funding for this research was provided by Arnold Ventures. Avalere retained full editorial control.

To learn more about social determinants of health, connect with us.

Methodology

Avalere identified all beneficiaries who were enrolled in both Medicare FFS Part A and Part B for at least one month in 2019. FFS beneficiaries were identified using the 100% files of Medicare FFS Parts A and B for 2019 and prior FFS enrollment in 2018. Medicare claims and enrollment data were accessed via a research collaboration with Inovalon Inc. under a CMS data use agreement.

Using 2018 and 2019 100% Medicare FFS claims data, Avalere identified all risk-adjustment eligible HCPCS codes in 2018 and claim expenditures, excluding hospice claims, in 2019. Avalere then estimated the CMS-HCC community model, which includes enrollees who have 12 months of Part B enrollment in 2018. The calibration included separate models for full duals, partial duals, and non-duals both under and over age 65 using 2018 diagnoses to predict Medicare Part A and Part B expenditures in 2019. For this model calibration Avalere used the CMS-HCC model that CMS used for 2023 payment (the V24 CMS-HCC model).

Many clinical trials underrepresent the diversity of the real-world population that may benefit from the product once marketed. Historically, factors such as race, ethnicity, age, gender, socioeconomic status, disability, and others, have been passive data points captured as a result of a clinical trial, and have not been used to align the enrollment to the intended real-world patient population. This has led to gaps in understanding of treatment safety, efficacy, and long-term outcomes in specific patient populations, and in part has motivated statutory interest to increase diversity of clinical trial programs and ensure representative enrollment in trials. The Food & Drug Administration (FDA) now has increased authority to monitor the diversity of late-stage clinical trials, which introduces new sponsor trial design and enrollment planning decisions that have unique impacts on stakeholders in the rare disease community.

Demographics of Clinical Trial Participation

In response to mandates from the 2012 FDA Safety and Innovation Act, the FDA’s Center for Drug Evaluation and Research has been reporting clinical demographic data for new drugs and biologics. The 2015–2019 Drug Trials Snapshots Summary Report provides aggregated data on trial participation across demographic groups.

Source: 2015–2019 Drug Trials Snapshots Summary Report

The 2020 Report highlighted that clinical trial participants were mostly white, non-Hispanic females under the age of 65, and the 2021 Report indicated that even though some improvements have been made, broad enrollment of diverse cohorts remains a challenge.

FDA Initiatives Designed to Increase Diversity

The FDA has focused on increasing clinical trial diversity over the past few years, but authority has been maintained via regulatory rather than statutory changes. In April 2022, the agency issued draft guidance for sponsors developing medical products outlining how to enroll adequate numbers of trial participants from various racial and ethnic groups. This guidance aligned to language on Diversity Action Plan (DAP) requirements in the Prescription Drug User Fee Act VII commitment letter. The DAP policy became a law as part of the 2023 Consolidated Appropriations Act.

The included diversity provision applies to sponsors of a new drug or device and must include enrollment goals, rationale for those goals, and an explanation of how a sponsor will achieve them. This plan must be submitted to the FDA by the beginning of a phase 3 or pivotal study. Relevant information could include prevalence and incidence of the disease, current pharmacokinetic and pharmacogenomic data, or patient demographics.

However, the Health and Human Services Secretary can waive the DAP requirement for instances such as difficulty identifying patients due to the small size of the testable population, multifactorial diagnosis that complicates eligibility, or the potential for a disease to affect a specific subset of the general population. Sponsors requesting a DAP waiver must have it granted by the FDA, whereas submitting a DAP does not require acceptance or rejection.

DAP information will become part of a new mandate that will begin by 2025, whereby updated demographic information will be aggregated to demonstrate whether applicants met their DAP-established enrollment goals. This information will be made available in annual reports published by the FDA.

Impact on Stakeholders in Rare Disease Community

The DAP provision impacts rare disease community stakeholders in unique ways.

While payers may not be directly affected by the policy, they will need to have a broader understanding of rare disease life-long patient journey and patients’ needs when transitioning between care systems (e.g., pediatrics to adults). Plans may also need to identify diverse data points from clinical trials that could be helpful in creating patient-centric framework which would lead to cost reduction.

How Avalere Can Help

Avalere’s subject matter expertise across product lifecycle in the rare disease landscape lends in-depth perspectives to support a variety of stakeholders interested in product that target rare diseases. For more information on how Avalere can support your goals, connect with us.

Background

In January 2023, the Centers for Medicare & Medicaid Services (CMS) released an update to the Hospital Change of Ownership (CHOW) dataset, which provides information on hospital ownership changes, acquisitions/mergers, and consolidations from 2016 to 2022. To explore recent consolidation and acquisition trends among certain hospitals, Avalere reviewed characteristics of hospitals found in the CHOW dataset by linking these data to hospital characteristics available in the CMS Provider of Services data. Avalere’s analysis was limited to transactions that were characterized as an acquisition or merger in the CHOW dataset.

Avalere Findings

Avalere identified 52 unique hospitals that were sold (purchased) and 49 unique hospitals that bought another hospital (buyers). All but one of these hospitals were short-term acute-care hospitals (STACH), and one purchased hospital was a critical access hospital. Characteristics of buyer and purchased hospitals were compared to STACH hospitals overall (3,450 in 2021).

While the overall count of STACHs that serve as buyers and purchased in hospital mergers and acquisitions (M&A) is low, certain characteristics, including bed size, ownership, location, and 340B status, were significantly different than the average across all STACHs (Table 1). For example, the analysis found that hospitals involved in M&A were more likely to be located in the northeast and in urban areas compared to the national average. Buyers and purchased hospitals were also more likely to be not for profit than the national average (69.4% of buyers and 59.6% of purchased compared to 53.4% overall).

*This category includes tribal and physician-owned hospitals.

In several characteristics, the buyers and purchased hospitals deviated both from the national average and from each other. For example, buyers were more likely than the national average to be 340B covered entities (81.6% vs. 59.9%), teaching hospitals (49.0% vs. 19.7%) and have a 500+ bed capacity (38.8% vs. 10.5%). In contrast, purchased hospitals were less likely than the national average to be 340B covered entities (28.8% vs. 59.9%), teaching hospitals (11.5% vs. 19.7%), and have a 500+ bed capacity (1.9% vs. 10.5%).

These findings help to describe the types of hospitals involved in M&A activities and provide important context for the healthcare ecosystem. Previous research has identified a relationship between hospital consolidation and increasing prices for both payers and patients. Past research has also documented that 340B hospitals are more likely to engage in vertical consolidation (e.g., hospital acquisition of physician practices). More research is needed to understand the impact of different hospital characteristics on the likelihood of M&A behaviors as well as the broader impact these transactions have on patients and the healthcare system.

Methodology

Avalere used CMS’s CHOW dataset to identify hospitals undergoing mergers or acquisition from 2016 to 2022 and examined the “buyer” and “purchased” CMS certification numbers, linking them to ownership, bed size, teaching status, urban location, and geography. 340B sites were identified using the Health Resources and Services Administration’s 340B Office of Pharmacy Affairs Information System. While all 340B hospitals are not for profit, not all not-for-profit hospitals are 340B sites.

Funding for this research was provided by the Pharmaceutical Researchers and Manufacturers of America. Avalere maintained full editorial control.

To learn more about healthcare related acquisitions and mergers, connect with us.

Avalere analyzed adherence to a combination oral and intravenous (IV) kidney cancer treatment regimen among Medicare fee-for-service (FFS) and Medicare Advantage (MA) patients enrolled in Medicare Part D. The analysis stratified patients based on whether they qualified for the Part D low-income subsidy (LIS). Qualifying patients receive cost-sharing assistance for the oral therapy. Because most of these patients are dually eligible for Medicaid, they also pay less in coinsurance for the IV therapy than patients who do not qualify for the LIS. Avalere found that LIS and non-LIS patients adhered to the oral treatment at similar rates, whereas LIS patients, despite their lower cost burden, adhered to the IV treatment at lower rates than non-LIS patients. This disparate rate of adherence may signal the presence of non-financial risk factors (e.g., lack of access to transportation or childcare) affecting LIS patients’ utilization of the IV therapy, resulting in potential health outcome disparities.

Analysis

Avalere’s analysis included Medicare FFS and MA patients who in 2019 were enrolled in Medicare Part D and were prescribed an oral/IV combination regimen to treat kidney cancer. The analysis used a metric called proportion of days covered (PDC)—a metric of the rate at which a prescription is refilled—to measure adherence to the oral therapy, and measured adherence to the prescribed IV schedule relative to the oral PDC metric: specifically, if a patient record showed an IV administration within each period covered by one oral fill—as clinical dosing guidelines specify—then that patient was considered adherent to the prescribed IV utilization schedule. Patients were considered adherent if the PDC was greater than 80%.

 Adherence to the Oral Component of the Combination Regimen

Results showed that LIS and non-LIS patients adhered to the oral component of the combination regimen at similar rates, with median PDCs of 72.4% and 72.1%, respectively. Median monthly out-of-pocket (OOP) costs, however, varied between groups: non-LIS patients paid $283 in FFS or $272 in MA, whereas LIS patients paid little to nothing in cost sharing for the oral therapy.

To investigate further how OOP burden may be associated with reduced adherence, Avalere stratified the non-LIS patients by whether they were enrolled in an employer group waiver plan (EGWP). This stratification isolated financial variables by enabling comparison of patients with similar demographic profiles who may experience different financial burden associated with treatment (i.e., because EGWP patients typically incur lower OOP costs because their plans have lower deductibles and fixed copays through supplemental benefits). Results showed a higher median proportion of days covered among EGWP patients (75.2%) than non-EGWP patients (71.7%), suggesting that cost may be one component of differential adherence.

Adherence to the IV Therapy Schedule

Results also showed that LIS and non-LIS patients differed in their rates of adherence to IV treatment schedules, as measured relative to their PDC metric for oral adherence. A higher percentage of LIS patients (76.8%) adhered less to IV treatment compared to non-LIS patients (83.2%). Avalere also stratified the non-LIS population by enrollment in an EGWP plan, finding that 81.3% of EGWP beneficiaries adhered to schedule compared to 82.1% of non-EGWP beneficiaries. Non-LIS patients in FFS paid a median monthly OOP cost of $652 for IV treatment, whereas those in MA paid a lower median monthly OOP cost of $558. LIS patients, most of whom are also eligible for Medicaid, paid lower OOP costs because of cost-sharing assistance via Medicaid’s medical benefit. Despite this reduced financial burden, these patients adhered to the IV treatment schedule at a lower rate than non-LIS beneficiaries, signaling the potential influence of non-financial factors such as social determinants of health (SDOH) that may pose barriers to access.

Discussion

Findings show that although patient OOP cost may impact medication adherence, other variables including social risk factors may impact adherence as well. Adherence barriers may also vary across treatments based on variables such as route of administration and coverage under the medical versus the pharmacy benefit.

LIS and non-LIS patients adhered to the oral treatment regimen at similar rates, despite differing OOP cost burdens. Had cost been the key adherence barrier, LIS patients may have adhered to the oral treatment regimen at higher rates. The similar adherence rates between LIS and non-LIS patients may suggest that  SDOH factors may have affected oral adherence among both patient cohorts. Conversely, EGWP patients paid lower OOP costs and adhered to the oral treatment regimen at higher rates than non-EGWP patients, signaling differences in adherence between these two groups related more directly to differences in OOP costs.

This analysis revealed a pattern of adherence challenges for both oral and IV regimens among LIS patients, suggesting that non-financial social risk factors may create barriers for low-income patients seeking treatment access. Such social risk factors may include:

• Health literacy challenges
• Lack of proximity to providers
• Difficulty identifying provider sites
• Lack of transportation to provider sites
• Lack of time for IV infusion appointments

As manufacturers plan for commercialization, especially for combination regimens, they should develop comprehensive patient support and access strategies, accounting for all factors likely to influence patient choice and access to effective treatment.

Funding for the research was provided by Janssen Pharmaceuticals. Avalere Health retained full editorial control.

To learn more about health equity and social determinants of health, connect with us.

Methodology

This analysis used 100% Medicare FFS and Part D prescription drug event (PDE) data, accessed via a research collaboration with Inovalon under a Centers for Medicare & Medicaid (CMS) data use agreement. It also leveraged Inovalon’s MORE² registry to access a convenience sample of payer-sourced claims for MA patients. Avalere identified Medicare FFS and MA patients who, in 2019, were enrolled in Medicare Part D and were prescribed an oral/IV combination regimen to treat kidney cancer. The analysis included all such patients who submitted both a Part D claim for the oral drug and a Part B claim for the IV drug within a 30-day period.

To measure adherence to the oral therapy, Avalere used PDE data to calculate PDC, a measure of medication adherence based on CMS’s method of calculating its Part D star ratings measure. Avalere divided the number of days within the measurement period for which patient records showed coverage with treatment, by the total number of days in the measurement period​. Treatment coverage period was determined based on service date and days of drug fill supply​. The analysis adjusted for overlapping prescription fills such that each fill began when coverage for the preceding fill ended​. To calculate OOP costs for the oral therapy, Avalere used the patient pay amount reported on the PDE.

Avalere used Medicare Part B claims and a sample of MA claims to measure utilization and OOP costs for the IV therapy. A patient was considered adherent to the IV regimen if an IV administration for that patient was logged within the expected timeframe of each oral fill per clinical dosing guidelines. For dually eligible beneficiaries, Avalere calculated coinsurance for the IV therapy before application of any liability reduction via Medicaid.

The objective of this research was to uncover any potential disparities across race/ethnicity that Black Americans face in wait times and, subsequently, in survival rates for a kidney transplant. For individuals with end-stage renal disease (ESRD), kidney transplantation may be the optimal treatment option compared to dialysis due to increased life expectancy, greater quality of life, and improved clinical outcomes. Given the increased incidence of ESRD among Black Americans compared to White Americans, Avalere analyzed 100% Medicare Fee-for-Service (FFS) claims data from 2018–2021 to determine whether Black Medicare patients face disproportionately longer wait times for kidney transplants.

Key Findings from This Analysis

• The average wait time for Medicare FFS beneficiaries from time of initial diagnosis of chronic kidney disease (CKD) or ESRD to receipt of a kidney transplant was 4.46 years. Compared to all patients, Black Medicare FFS beneficiaries had a longer average wait time from diagnosis to transplant (5.09 years; P < .0001).
• Mortality for all patients 4 years after initial diagnosis of CKD/ESRD was 5.03%, which increased to 6.03% for 5 years and 27.05% for 6 years or greater.

*Found to be statistically significant with a p value <.05.

Read the abstract.

To learn more about the evolving kidney care market and opportunities to identify and address disparities in care, connect with us.

Avalere analyzed claims for Medicare beneficiaries receiving care in skilled nursing facilities (SNF) at the onset of the COVID-19 pandemic (fiscal year (FY) 2019 – FY 2021) and found that patient acuity and social risk were greater among these beneficiaries than for beneficiaries who received care in SNFs before the pandemic. Although some patient acuity and social risk factors returned to pre-pandemic patterns by FY 2021, the rapid shifts, particularly during the first few months of the pandemic (April–June 2020), highlight the potential for significant changes in the characteristics of patients treated during a public health emergency.

Avalere used Medicare claims data to identify SNF stays and beneficiary demographics and used beneficiary ZIP codes to link to data on characteristics of beneficiary communities (e.g., income, poverty status). Stays with a COVID-19 diagnosis code and stays admitted under the 3-day acute hospital stay waiver were excluded from the analyses to inform changes relative to the population typically treated in SNFs before the pandemic. Avalere examined characteristics of the SNF population during several periods from FY 2019–FY 2021 to track changes in beneficiary characteristics relative to both the start of the COVID-19 pandemic and the introduction of changes to the Centers for Medicare & Medicaid Services’ (CMS) Skilled Nursing Facility Prospective Payment System with the Patient Driven Payment Model (PDPM) in FY 2020.

Key Findings

The findings of Avalere’s analyses indicated higher patient acuity among SNF stays at the onset of the COVID-19 pandemic. Medicare beneficiaries receiving care in SNFs had higher case-mix indices (CMI; Figure 1) for nursing as well as for speech-language and pathology (SLP), and a higher proportion of stays were for beneficiaries living with End-Stage Renal Disease (ESRD) and for beneficiaries dually eligible for both Medicare and Medicaid (dual-eligible), indicating higher rates of SNF stays for beneficiaries with high medical complexity and low income status (Figure 2). The average nursing and SLP CMIs remained above pre-pandemic levels into FY 2021. In contrast, the proportion of SNF stays for beneficiaries with ESRD and for dual-eligible beneficiaries returned to pre-pandemic levels in FY 2021.

Avalere also analyzed differences in beneficiary demographics including age, gender, and race/ethnicity in beneficiaries receiving care in SNFs at the start of the COVID-19 pandemic (data not shown) and found small but notable shifts in demographics. For example, the proportion of stays among beneficiaries aged 85 and older decreased slightly between Q1/Q2 and Q3/Q4 of FY 2020 (from 30.5% to 28.7%), while the proportion of stays among male beneficiaries increased slightly during the same time period (from 41.6% to 43.1%). Regarding race, the proportion of SNF stays among Black beneficiaries increased from the first to the second half of FY 2020, from 11.4% to 12.1%.

Avalere also identified shifts in the proportion of beneficiaries residing in rural or economically disadvantaged neighborhoods at the start of the COVID-19 pandemic. However, these shifts were minor and returned to pre-pandemic proportions in FY 2021. In the second half of FY 2020, the proportion of SNF stays among beneficiaries living in rural areas increased by 1.3 percentage points (Figure 3). The average net worth in the ZIP codes of Medicare beneficiaries receiving care in SNFs decreased in the second half of FY 2020 before returning to pre-pandemic levels in FY 2021 (Figure 4). Consistent with this finding, the proportion of SNF beneficiaries living in neighborhoods with higher levels of poverty increased slightly in the second half of FY 2020 before returning to pre-pandemic levels in FY 2021 (Figure 4).

Conclusions

Medical complexity and social risk levels increased among Medicare beneficiaries who received care in SNFs toward the beginning of the COVID-19 pandemic. Although the SNF payment system changed significantly with the introduction of the PDPM in FY 2020, observed changes in these clinical and social risk factors were correlated with the timing of the pandemic rather than with the introduction of the payment system. Trends in demographic and socioeconomic factors largely returned to FY 2019 levels after the first wave of the pandemic, although changes in patient acuity, as measured by CMI, persisted into FY 2021. These changes in beneficiary characteristics suggest that the impacts of the pandemic on SNFs was broader than the additional burden of managing patients diagnosed with COVID-19 and understanding changing patient characteristics will be increasingly important as we move to the later stages of the COVID-19 pandemic.

Funding for this research was provided by the American Health Care Association. Avalere retained full editorial control.

To learn more about social determinants of health, connect with us.

Methodology

Avalere used the 100% Medicare fee-for-service claims data to analyze SNF stays from FY 2019 through FY 2021. Claims data were accessed via a research collaboration with Inovalon, Inc. and governed by a research-focused CMS data use agreement.

SNF stays were grouped by admission date (i.e., FY 2019, first half of FY 2020, second half of FY 2020, and FY 2021). The analysis excluded stays with an International Classification of Diseases, 10th Revision diagnosis code of U071 (COVID-19) or with a condition code signifying that the 3-day acute stay waiver (condition code = DR) had been applied.

Avalere used the Medicare claims to identify sex, race/ethnicity, age, disability status, dual eligibility, and rural residence for beneficiaries with SNF stays. Avalere also used Medicare claims to identify trends in the CMIs based on the Health Insurance Prospective Payment System codes. Each of the five CMI components (i.e., nursing, SLP, occupational therapy, physical therapy, and non-therapy ancillaries) is a measure of patient acuity based on diagnoses and patient assessment data and is used to adjust payments.

For socioeconomic characteristics such as mean net worth and percentage of residents below poverty by ZIP code, Avalere linked Medicare claims data to data from Acxiom and the 2019 American Community Survey at the 5-digit ZIP code level.

Avalere analyzed how area-level measures of sociodemographic factors relate to Medicare patients’ adherence to multiple myeloma (MM) treatment regimens. To understand how demographic factors and social determinants of health (SDOH) affect patients with varying out-of-pocket costs, Avalere assessed these sociodemographic factors both for standard Medicare Part D beneficiaries and for Part D beneficiaries who receive low-income subsidies. Adherence challenges may differ between oral and intravenous therapies, and patients prescribed combination treatments that include both modalities may face particular challenges.

Avalere calculated a proportion-of-days-covered metric to quantify the percentage of days in the measurement period in which a patient received coverage for one of three selected oral oncolytic therapies used to treat MM. To analyze adherence according to patient characteristics, Avalere analyzed household SDOH factors, by ZIP-code level averages, including ratio of income to federal poverty level, educational attainment, percentage of households containing a personal computer, and census division. Sociodemographic characteristics at the ZIP-code neighborhood level can act as proxies for SDOH characteristics at the member level and contextualize adherence and outcomes for MM patients.

Key Findings

Results show inconsistencies in how SDOH and demographics correlate to therapy adherence for Medicare beneficiaries with MM. Key findings include:

Avalere’s analysis classified patients into three groups. For both multi-adult households and households by PC ownership, the results consider analysis between the highest (≥75 percentile) and lowest (≤25 percentile) group for each SDOH factor.

These results indicate that SDOH are a factor in patients’ treatment adherence. However, results revealed substantial variation, with some factors—such as living in a multi-adult household and having a personal computer in the home—more influential in impacting adherence in Medicare patients diagnosed with MM. This may indicate that socioeconomic issues such as transportation, childcare, and familial support systems may be key contributing factors to lower adherence rates. Furthermore, since there was inconsistency in the findings, there might be additional factors such as a patient’s health literacy level, transportation, access to technology, and specific treatment regimen impacting how adherent patients are to their therapies, which could be further explored in additional analyses. Future analyses also could consider how these measures may be benchmarked to the broader Medicare population to understand if health disparities are more prevalent among patients with MM and how SDOH variables contribute to adherence relative to other measures such as access to care (e.g., pharmacy access) and out-of-pocket costs.

To learn more about addressing health equity, connect with us.

Methodology

Fee-for-service (FFS) beneficiaries were identified using the 100% file of Medicare FFS Parts A, B, and D data, accessed via a research-focused data use agreement with the Centers for Medicare & Medicaid Services. Medicare Advantage members were identified using medical and pharmacy claims from the Inovalon MORE² Registry®, a large scale, real-world, multi-payer dataset comprising medical, pharmacy, lab claims, and clinical data on more than 332 million de-identified patients. Avalere leveraged the Acxiom database to assess socioeconomic factors in the relevant geographic areas in which each individual resides at the ZIP-code level.

Both the public and private sectors have shown a renewed focus on expanding the diversity of clinical trial participants.  To date, most research on medical treatments, interventions, and cures has not appropriately represented all communities of patients, especially racial and ethnic minority populations (Figure 1). Clinical trials that do not include a diverse range of participants may overlook relevant patient perspectives and risk unpredictable post-market product results in populations that were not included in initial studies.

* Note: Of the 20,692 US-based trials represented, only 43% (8,898) reported any race/ethnicity data.

1. US Census, Quickfacts (date accessed: May 18, 2022).
2. E.B. Turner et al., “Race/Ethnicity Reporting and Representation in US Clinical Trials: A Cohort Study,” The Lancet Regional Health: Americas 8.100252 (2022).

Legislative and Regulatory Outlook

In addition to past decades of policymaking to tackle ethical issues and increase equity in clinical trials, the Biden administration, federal agencies, and professional organizations are prioritizing diversity by highlighting specific initiatives to improve equity in clinical trials. For example, the Food & Drug Administration (FDA) released guidance for including older adults in cancer clinical trials, while the National Institutes of Health released a strategic plan on minority health and health disparities, which includes ensuring appropriate representation of minority populations. Additionally, PhRMA released an equity initiative that includes clinical trial diversity.

Currently being considered at the federal level are FDA draft guidance and multiple bills to support enhancing equity in clinical trials. The FDA released draft guidance in April on how medical product sponsors can develop race and ethnicity diversity plans to enroll clinical trial participants from underrepresented racial and ethnic groups. Additionally, several bills have been proposed to enhance diversity in clinical trial populations, including the House FDA User Fee Amendment (UFA) Package (H.R.7667).

Public comments were collected for the FDA’s draft guidance in June, and it is expected final guidance will be released after review. Conversely, little movement has occurred with the bills proposed to enhance diversity in clinical trial participation. Additionally, the Senate’s version of the FDA UFA package does not include the House’s clinical trial diversity provisions. While these proposed bills aim to increase diversity in clinical trials, this lack of advancement creates an opportunity for life sciences companies and other stakeholders to lead efforts to identify and implement solutions.

Barriers and Approaches to Improving Equity

Improving equity in clinical trials requires considering barriers to patient participation at multiple levels (Figure 2) and identifying strategies to help address specific issues at each level. For example, at the institutional level, leaders can increase diversity of research teams and expand resources for research (e.g., staffing, financial supports).

BIPOC: Black, Indigenous, and People of Color. Sources: D.H. Bodicoat, A.C. Routen, A. Willis, et al., “Promoting Inclusion in Clinical Trials: A Rapid Review of the Literature and Recommendations for Action,” Trials 22.880 (2021); N. Thakur, S. Lovinsky-Desir, et al., “Enhancing recruitment and retention of minority population for clinical research in pulmonary, critical care, and sleep medicine: An official American Thoracic Society research statement,” Am J Respiratory and Critical Care Medicine 204.3 (2021); 204(3).

Life sciences companies may consider both qualitative and quantitative approaches to increasing clinical trial diversity. For example, manufacturers can lead qualitative research with patients and providers to understand how their needs vary among different subgroups or assess patient data to understand social needs within potential study populations. As stakeholders consider future clinical trial design, including diverse populations will be important to ensuring that products are effective for a broad range of patients.

To talk with our experts about identifying barriers to trial participation and options for addressing those barriers, connect with us.

The information featured in this Insight comes from Avalere Federal Policy 360™. For information on subscription pricing or for in-depth context and recommendations on advancing equity in clinical trials, contact us.

Avalere focused on 3 Pharmacy Quality Alliance (PQA)-developed proportion of days covered (PDC) measures: diabetes all class, renin angiotensin system agonists (RASA), and statins.¹  The analysis shows that those who qualify for the Low-Income Subsidy (LIS) program more often experience lower medication adherence across the 3 measures than their counterparts who do not have LIS. Additionally, medication adherence trends within LIS groups are worse in MUAs compared to non-MUAs. Similarly, the analysis finds that those who are racial minorities more often experience lower medication adherence across 3 chronic diseases than their counterparts who are not from a racial minority group. Notably, medication adherence trends within racial groups are consistent in both MUAs and non-MUAs.

Low medication adherence is linked to out-of-pocket costs,² clinical side effects,³ and certain social risk factors.⁴ To explore how these factors converge, Avalere analyzed rates of medication adherence across race and socioeconomic groups of Part D beneficiaries living both in and out of MUAs. MUAs are geographic areas designated as having a lack of access to primary care. The analysis examined groups of beneficiaries by race—Asian, Black, Hispanic, North American Native, and White—following the Medicare Part D claims data structure. The analysis also assessed beneficiaries by level of LIS, including full LIS, partial LIS, and non-LIS.

Avalere utilized the PQA PDC scores as a measure of adherence. According to the PQA, a PDC of at least 80% is recognized as the threshold above which a beneficiary is reasonably likely to receive the most clinical benefit from a medication. Avalere refers to achievement of a PDC score of 80% or more as “sufficient adherence.”⁵

LIS Status

Regardless of MUA status, beneficiaries with LIS have lower rates of sufficient adherence for all 3 disease groups compared to those without LIS. For example, in MUAs, a 10-percentage point difference in adherence exists between non-LIS (78%) and partial-LIS (68%) beneficiaries who take diabetes medications. Beneficiaries in MUAs with full LIS who take diabetes medications are at the mid-point (73%) between these 2 groups.

Further, for those taking diabetes medications, living in an MUA correlates to lower rates of sufficient adherence for each of the income groups. The group of beneficiaries taking diabetes medications with the lowest rate of sufficient adherence is the partial-LIS beneficiaries. In fact, across all 3 medication groups, those with partial LIS have the lowest rates of sufficient adherence across income groups, though rates are closer to the full LIS for those taking RASA medications.

Race

Regardless of MUA status or medication group, Black, Hispanic, and North American Native beneficiaries have the lowest rates of sufficient adherence compared to Asian and White beneficiaries. North American Native beneficiaries have the lowest rates of sufficient adherence across all analytic groups. The greatest disparity in adherence rates is for those taking RASA medications; North American Native beneficiaries have the lowest rate of sufficient adherence in MUA (60%) and non-MUAs (68%) compared to White beneficiaries in MUAs (83%) and non-MUAs (85%; Figure 5).

Discussion

This analysis offers insight into the intersection of race, LIS status, MUA, and medication adherence for beneficiaries taking 3 groups of medications. These findings indicate that social determinants and income could correlate with rates of sufficient adherence.

The results suggest that MUA status may be linked to adherence rates for some conditions (e.g., diabetes) more than others and for some racial minority groups (e.g., North American Native) more than others. This analysis leveraged Part D claims data only; thus, prescriptions being filled by the Indian Health Service would not be reflected in this analysis. This limitation might offer some explanation for rates of sufficient adherence for North American Natives included herein.

Further, the lower rates of sufficient adherence found for partial-LIS beneficiaries suggest that the cost-sharing requirements for those with partial LIS may not subsidize out-of-pocket costs to the extent necessary to achieve adherence rates in parity with other groups of beneficiaries.

Funding for this research was provided by The Pharmaceutical Research and Manufacturers of America. Avalere retained full editorial control.

To receive Avalere updates, connect with us.

Methodology

After identifying MUAs⁶ and urban and rural areas utilizing HRSA data files, Avalere cross walked these designations to ZIP codes.⁷ To measure adherence, Avalere analyzed a randomly selected 20% sample of the 2019 Part D Drug Event data from the Centers for Medicare and Medicaid Services to determine the PDC as defined by PQA as a measure of medication adherence for 3 selected medication categories: diabetes, RASA, and statins.⁸ This allowed Avalere to assess drugs that treat 3 different chronic conditions: diabetes, hypertension, and hyperlipidemia.

The analysis is subject to some limitations. Because MUAs are designated at the county, county subdivision, and census tract levels, the ZIP code analysis may be capturing portions of an area that may not be part of an MUA and could potentially overstate or dilute the differences in the results between MUAs versus non-MUAs. Avalere also excluded beneficiaries residing outside the 50 states and DC, as well as those enrolled in the Limited Income Newly Eligible Transition Program.

Per the data use agreement’s requirement, Avalere identified a random sample representing less than 20% of the total Part D population. Avalere did not include beneficiaries in the US territories, and beneficiaries who were enrolled in the Limited Income Newly Eligible Transition Program were excluded from this analysis

Notes

1. Technical Specifications for PQA-Endorse Measures, February 2018.
2. Eaddy MT, Cook CL, O’Day K, Burch SP, Cantrell CR. How patient cost-sharing trends affect adherence and outcomes: a literature review. PT. 2012; 37(1): 45-55.
3. Vignon Zomahoun HT, de Bruin M, Guillaumie L, Moisan J, Grégoire JP, Pérez N, Vézina-Im LA, Guénette L. Effectiveness and Content Analysis of Interventions to Enhance Oral Antidiabetic Drug Adherence in Adults with Type 2 Diabetes: Systematic Review and Meta-Analysis. Value Health. 2015 Jun; 18(4): 530-40.
4. Gast A, Mathes T. Medication adherence influencing factors—an (updated) overview of systematic reviews. Syst Rev 8, 112 (2019). https://doi.org/10.1186/s13643-019-1014-8.
5. Based on the PQA Manual, a measurement period includes treatment periods that are at least 91 days.
6. The MUA designation is based on county, county subdivision, and census tract.
7. Avalere utilized the Department of Housing and Urban Development US Postal Service ZIP code crosswalk files.
8. Technical Specifications for PQA-Endorsed Measures, February 2018.

Background on Malnutrition, Food Insecurity, and Health Equity

Malnutrition places an immense burden on the healthcare system and poses significant risks to patient health outcomes, hospital performance, and broader community health. A major risk factor behind the clinical presentation of malnutrition is food insecurity, which is “a household-level economic and social condition of limited or uncertain access to adequate food.” Food insecurity may be influenced by a number of factors, including income, employment, race/ethnicity, and disability.

Despite substantial awareness of the burden of malnutrition, rates of diagnosis and identification of malnutrition are low, leaving many malnourished patients potentially undiagnosed and untreated. Malnutrition affects more than 30% of hospitalized patients, but only 9% of discharged patients had a coded malnutrition diagnosis in 2018. Patients require resources or assistance related to nutrition after discharge, but without adequate connections to community-based organizations and programs, patients’ needs may not be met—presenting a missed opportunity to improve their health. This phenomenon is magnified in underserved communities, contributing to even larger health disparities.

Roundtable Purpose

The Malnutrition Quality Improvement Initiative (MQii) aims to advance evidence-based, high-quality, patient-driven care for those who are malnourished or at risk of being malnourished. Figure 1 represents a recommended care pathway put forth by the MQii leadership team to jointly address nutrition and food insecurity risk. It accounts for the broad array of related risk factors that contribute to health disparities and how use of quality measurement can identify patients at risk and create accountability for implementing appropriate interventions to improve care.

To advance the work of the MQii and various organizations tackling these issues, a multi-stakeholder group of 23 leaders and experts in healthcare and community health came together for a roundtable entitled “Advancing Health Equity Through Malnutrition Quality Measurement” on March 3, 2022, to discuss connections between health equity, hospital malnutrition care, and food insecurity. Their directive was to identify and share solutions to these issues that can be readily implemented or replicated.

Roundtable Output

These identified solutions, and the understanding of existing barriers and challenges that the healthcare system already faces when treating malnutrition gathered from qualitative interviews with national experts prior to the Roundtable, brought the participants to discuss how solutions could be replicated and expanded. Participants focused on pathways to manage and address malnutrition, food insecurity, and health disparities across transitions of care while leveraging malnutrition composite measure performance data, hospital and community partnerships, and relevant experts all working toward the same goal. The conversation made it clear that these pathways and proposed solutions would require coordinated and collaborative efforts from all stakeholders. The results of this discussion are the basis for these proceedings, which outline a roadmap of actionable tactics for different stakeholder groups to advance malnutrition care and reduce food insecurity. Beginning with identification of malnutrition in the acute care setting and continuing through interventions implemented at the time of patients’ discharge and solutions integrated within the community, stakeholders can come together to advance health equity among underserved populations.

For more information, please direct any questions to Christina Badaracco or call 202-355-6094.

View the proceedings document and the executive summary.

To receive more expert insights on the latest healthcare news, connect with us.

About the MQii

The MQii is a project of the Academy of Nutrition and Dietetics, Avalere Health, and other stakeholders who provide guidance and expertise through a collaborative partnership. The Academy of Nutrition and Dietetics provides resources and services to support the MQii. General support is provided by Abbott.

Over the past few years, policymakers have considered various Part D benefit redesign proposals. Although key parameters of proposals differ, all would implement a beneficiary out-of-pocket (OOP) spending cap, create a new manufacturer discount throughout the benefit, and decrease Medicare’s liability while increasing plan liability in the catastrophic phase. Under the Part D benefit redesign proposal considered as part of the Build Back Better Act (BBBA), beneficiary OOP costs would be capped at $2,000 beginning in 2024.¹

Avalere analyzed characteristics of beneficiaries who would have OOP costs of $2,000 or more in 2024 (based on 2019 spending²) and who would therefore likely benefit from a $2,000 OOP spending cap under the BBBA Part D benefit redesign proposal. The analysis finds that certain underrepresented populations, including Black and Hispanic enrollees, would have higher annual average OOP costs in 2024 compared to the Part D population overall and other subpopulations (e.g., White beneficiaries). Similar results were also found for beneficiaries entitled to Medicare on the basis of disability compared to those entitled to Medicare due to age. Additionally, a disproportionate share of beneficiaries who are disabled or from a racial/ethnic minority group would have OOP spending of $2,000 or more.

For example, while disabled beneficiaries represent only 24% of the Part D patient population, they make up almost half (47%) of beneficiaries estimated to have OOP spending of at least $2,000 in 2024 (Figure 1). Among all beneficiaries who would have OOP spending of $2,000 or more in 2024, average annual OOP spending for these beneficiaries would be more than 30% higher for those entitled to Medicare on the basis of disability compared to those entitled by age ($6,007 vs. $4,555, respectively).

Note: Beneficiaries with End-Stage Renal Disease were not included in the age or disability category and represented less than 1% of all Part D enrollees and of enrollees with $2,000 or more in OOP costs.

Similarly, while Black and Hispanic beneficiaries represent 11% and 3% of the overall Part D population, respectively, Black beneficiaries make up 15% and Hispanic enrollees comprise 5% of all beneficiaries who were estimated to have OOP spending of $2,000 or more in 2024 (Figure 2).

Among all beneficiaries who would have OOP costs of at least $2,000, those who are White were estimated to have average annual OOP costs of just over $5,000 in 2024 compared to average annual OOP costs of about $5,500 or more for beneficiaries who are Asian, Black, Hispanic, or North American Native (Figure 3).

While debate on the BBBA is currently stalled in Congress, Part D benefit redesign remains a key priority among lawmakers, raising the possibility of a scaled-down legislative version, standalone Part D redesign legislation, or other actions to implement drug pricing reforms. As discussion continues, stakeholders with an interest in Part D policy should consider the impact that changes to the Part D benefit parameters would have on disparate subsets of beneficiaries, such as the impact that various OOP cap levels could have on different patient populations.

To receive Avalere updates, connect with us.

Methodology

Avalere used 2019 Medicare Part D Drug Event data accessed under a research-focused data use agreement (DUA) with the Centers for Medicare & Medicaid Services to simulate the Part D benefit for 2024 under current law.

Per the requirement of the DUA, Avalere identified a random sample representing less than 20% of the total Part D population. The sample was scaled to represent the full Medicare Part D population. Avalere excluded the following populations from the analysis: those for whom benefit design data was missing, including those with claims paid by an Employer Group Waiver Plan; those residing outside the 50 states and DC; and those enrolled in the Limited Income Newly Eligible Transition program.

To model the current law benefit design, Avalere deflated 2024 benefit parameters to 2019 dollars using annual percentage increases in Part D per capita benefits to account for policy changes to the calculation of the catastrophic threshold under current law. Avalere assumed 2019 plan-specific formulary coverage, tier placement, and cost sharing. Avalere also used 2019 negotiated prices and assumed scripts were filled at in-network preferred pharmacies in determining beneficiary OOP costs.

Notes

1. Analysis does not reflect the most recent Part D benefit redesign bill introduced by Sen. Warnock on April 6, 2022.
2. All dollar amounts in this document represent costs in 2024, inflated from 2019 dollars. Avalere assumed that the 2019 distribution of beneficiaries and their spending would remain the same in 2024.