ACO REACH And Advancing Equity Through Value-Based Payment, Part 1

Given the wide expansion of value-based payment (VBP) models in the United States, VBP can be a useful tool for reducing health inequities and advancing equity goals. VBP models provide more flexibility to address social drivers of health (SDoH); they provide services and care coordination that are not typically covered but may be particularly important for people experiencing worse outcomes due to socioeconomic status, race, ethnicity, or other factors. 

However, some evidence suggests that the uptake of VBP models has lagged in marginalized populations, and that in some circumstances, VBP could exacerbate disparities. Consequently, many VBP models are now adopting a more intentional focus on equity, such as the Oregon Health Authority’s Coordinated Care Organizations 2.0 model and Supporting Health for All through Reinvestment Initiative. But addressing equity through VBP remains a nascent field with diverse approaches in use—with limited evidence on which approaches work or help avoid undesirable outcomes.

In this context, the Center for Medicare and Medicaid Innovation (the Innovation Center) recently announced that its Global and Professional Direct Contracting pilot will be redesigned into a new model called the Accountable Care Organization (ACO) Realizing Equity, Access, and Community Health (REACH). Reflecting the Innovation Center’s strategic focus on health equity as a top priority—and on achieving the broader equity goals of the Department of Health and Human Services and the Centers for Medicare and Medicaid Services (CMS)—ACO REACH incorporates many new health equity-focused elements affecting upfront payments, data collection, provider selection, community governance, and delivery and benefit design. These health equity-focused VBP design elements are on top of ACO REACH’s main advanced alternative payment structure of monthly population-based payments for either primary care services or for all covered services with strong links to quality of care.

ACO REACH is a substantial pilot program. It is aimed at improving equity but also at assessing which payment reform design elements can best do so while also achieving the “Triple Aim”—better care experience, better population outcomes, and lower health care costs. ACO REACH and the other equity-focused VBP models consequently provide an unprecedented opportunity for augmenting the meager evidence we have on how to improve equity. 

To inform health policy makers and stakeholders on approaches to achieve equity through VBP, we scanned the literature and reviewed various policy initiatives. We spoke with experts implementing equity-focused VBP models or with expertise on these models. In this article, we present a framework for categorizing the spectrum of approaches we identified for improving equity through VBP model design. Below and in part 2 (to be published tomorrow), we discuss the key design elements for explicitly addressing equity in ACO REACH; highlight examples (and evidence or early lessons, where available) of how some states and commercial payers have approached those same design elements; and identify additional opportunities for the Innovation Center and other parts of CMS, states, and commercial payers to continue to test ways to embed equity in VBP models.

A Framework Of Equity-Focused VBP Design Elements, From Early Stages To More Advanced Stages

A range of frameworks have been developed to support assessments of VBP models and the potential impacts of policy reforms on equity. The Institute for Healthcare Improvement (IHI)’s Framework for Achieving Health Equity focuses on equity, including some VBP considerations. The Health Care Payment Learning and Action Network’s (HCPO-LAN’s) alternative payment model framework is a widely used framework for describing VBP models, and the HCP-LAN Health Equity Advisory Team’s recent guidance outlines categories of advancing health equity through VBP.

In exhibit 1, which is also a roadmap of this article, we synthesize elements of existing frameworks and real-world examples into a new framework.

Exhibit 1: Summary of equity-focused VBP design elements in action, comparing ACO REACH to other public and private payer approaches (arranged from early stages to more advanced stages)

Source: Authors’ analysis (from literature and policy document scans, interviews with equity-focused VBP model implementors and experts, and concise synthesis of concepts from the IHI’s Framework for Achieving Health Equity, and the HCP-LAN on alternative payment model design and guidance for advancing health equity through VBP).

The new framework is designed to accomplish two goals. First, the new framework more concisely than existing frameworks categorizes the intersection of equity and VBP design. (See exhibit 2 for how each VBP design element maps to corresponding elements from the three external frameworks.) Second, the new framework gathers specific examples from the field of how these design elements were operationalized in practice. We arrange design elements (table rows) from less to more advanced, and within each design element, also arrange key examples from less to more advanced.

For brevity, this article frames the discussion mainly around ACO REACH’s design elements, so we do not discuss in the text every single design element in the framework. In future work, we will continue to explore these options—such as infrastructure to connect accountable providers to social and community services for social needs—and using VBP attribution methods (ways to use data, providers nomination, or patient selection) specifically for equity (a topic in which we found no examples in the field). At the end of the article, we will call back to the framework in a summary table showing how ACO REACH, and other public and private payers, have approached each design element.

Exhibit 2: Equity-focused VBP design elements, from early stages to more advanced stages, mapped to external framework components

Source: Authors’ analysis (from literature and policy document scans, interviews with equity-focused VBP model implementors and experts, and concise synthesis of concepts from the IHI’s Framework for Achieving Health Equity, and the HCP-LAN on alternative payment model design and HCP-LAN Health guidance for advancing health equity through VBP).

VBP Design Element: Using Request-For-Proposal And Contract Language To Foster Equity Design And Codify Accountability

VBP request for proposals (RFPs) and contracts are key instruments that determine and establish activities implemented within VBP programs. Within the ACO REACH model RFP, CMS makes clear in the model’s title and background section that the focus of ACO REACH is on equity through payment reform (where few VBP models focus), and operationalized the specific equity elements we detail below and in part 2.

Similar to the ACO REACH model’s RFP, state contracts and RFPs commonly require Medicaid managed care organizations (MCOs) to implement activities designed to support improvements in health equity. Examples include stratifying quality measures by member demographic data (including race and ethnicity); implementing equity training for staff; creating health equity-related staff positions; reporting demographic data of contracted providers; and implementing payment models that incentivize equity goals. As one notable example, California’s Medi-Cal managed care plan’s RFP requires the implementation of health equity performance improvement projects with the aim of achieving health equity-specific goals—and a health equity officer to oversee this work. Notably, Pennsylvania’s MCO contract requires MCOs to implement a maternal-care bundled payment model that incentivizes providers to reduce racial and ethnic inequities in birth-related outcomes.

The contracting process can then be used to codify these RFP requirements and incentives. However, more evidence is still needed on how RFP and contracting approaches can best be used to improve equity through VBP.

VBP Design Element: Encouraging, Selecting, And Retaining A Diverse Set Of Providers

When they apply to the ACO REACH model, providers will be evaluated using new scoring criteria that considers their track record of direct patient care and their record of providing high-quality care to historically underserved communities. North Carolina’s COVID-19 Support Services Program to address pandemic-related social needs and health equity—while not a VBP approach—takes a similar provider selection approach: The North Carolina program prioritizes selection of grantee organizations staffed by or serving historically marginalized populations. Getting new providers to be successful in VBP and in equity work, however, also requires competency building, and there could be a role for “VBP helper” organizations, like such as Aledade or Caravan, to create models to help newer safety-net provider groups succeed in equity-focused VBP programs.

VBP Design Element: Engaging The Community Through Governance

A key way to promote equity in payment models is to elevate community and beneficiary voices in governance structures. CMS requires ACO REACH governing bodies to have representation from at least one Medicare beneficiary and at least one other consumer advocate. In addition, these individuals must hold voting power on the governing bodies.

These requirements have precedent in other state and federal health policies and can be further strengthened with not only community representation but also community partnerships. For instance, the Health Resources and Services Administration requires federally qualified health centers (FQHC) to maintain a board majority of patients who are demographically representative of the community served, indicating opportunity to foster community governance through more thoughtfully including FQHCs in VBP. The Oregon Health Authority (OHA) requires each coordinated care organization (CCO) to have at least one community advisory council (CAC) to share experiences in accessing care and to make recommendations to improve services in their communities. The CCO CACs require a composition of at least 51 percent consumers and inviting a tribal representative, suggesting that ACO REACH governing boards might increase their community member requirements to better capture must different lived experiences within and across marginalized population groups.

While there has not been any comprehensive impact assessment of Oregon’s CAC program, case study interviews have yielded examples in which CACs have led to millions of dollars in community reinvestment, community conversations about the health system, and community-focused wellness programs. Furthermore, as part of their current 1115 waiver proposal, the OHA has proposed taking the CAC concept further with community investment collaboratives (CICs); this would establish community-led partnerships with decision-making power to direct Medicaid funding to communities experiencing the greatest harm from health inequities due to structural racism and discrimination.

VBP Design Element: Providing Upfront Financial Resources Or Shifting Resources To Support Equity

A common VBP challenge is providing upfront capital so that organizations can build infrastructure and start redesigning care delivery and other health services. The ACO REACH model accomplishes this through a “health equity benchmark” in which ACOs receive an additional $30 per beneficiary per month (PBPM) payment for serving more medically underserved beneficiaries. (Technically, the RFP defines “underserved” as those beneficiaries in the top decile of the model using a score based on whether the beneficiary is dually eligible for both Medicare and Medicaid combined with the score of the beneficiary’s census block group on HRSA’s Area Deprivation Index (ADI), which measures social vulnerability.) To balance these payments in the model, ACOs receive $6 less PBPM for serving beneficiaries less likely to be “underserved” (technically, those in the bottom five deciles [lower half] of the model by the above score).

While this policy uses the language of “benchmark,” and the Innovation Center describes the policy in conjunction with risk adjustment of the benchmark, it is different than the usual benchmark adjustments based on expected spending for an attributed population. Instead, each month, ACO REACH’s “equity benchmark” shifts upfront financial resources from organizations caring for less marginalized patients to those caring for more. These transfers are not tied to financial or quality performance; thus, they are comparable to HCP-LAN’s Category 2A payments—payments for infrastructure investments, such as care coordination nurses or electronic health records, that can improve the quality of patient care—and not considered an alternative payment model design feature. (This is in juxtaposition to the overall ACO REACH model’s payment approach of PBPM payments for either just primary care services or for all covered services, which is considered an HCP-LAN Category 4 model.)

The health equity benchmark differs from other methods of adjusting for differing social risk. In traditional risk adjustment, each risk adjustment factor is correlated with overall health care expenditures. One concern with including social factors in risk adjustment methods is that social risk factors may be negatively correlated with overall expenditures and thereby could unintentionally decrease resource allocation to traditionally underserved populations. The reason for this may be limitations in access for people with these social risk categories. Further work is needed to generate evidence on indices and their association with cost and use.

Thus, one open question is this: What is the best metric to use for identifying people in underserved areas? ACO REACH is using ADI in the health equity benchmark calculation. There are other deprivation indices, too, and many debates around which approach is most appropriate. Another popular equity-focused risk adjustment approach is to use the Centers for Disease Control and Prevention’s Social Vulnerability Index (SVI). Both ADI and SVI use a combination of census sociodemographic measures to create a geographic score, but with two key differences: SVI includes race and ethnicity, whereas ADI does not; and ADI is calculated at more granular geographic levels—census block group—whereas SVI is calculated at the county or census-tract level. Evidence comparing use of the two methods in health services delivery is sparse (and needed), but some early evidence shows that they have similar associations with COVID-19 outcomes and vaccine allocation.

A second open question is whether the additional amount for serving traditionally underserved beneficiaries in ACO REACH is large enough to garner attention from organizations that care for large numbers of such people, for example, safety-net organizations. Exhibit 3 illustrates the scale of resources shifting through the health equity benchmark using a rough, back-of-the-envelope approach. Furthermore, the other side of the coin is whether the amount is sufficient to encourage ACO REACH participants that have not focused on underserved people to date to shift resources into attracting more beneficiaries who are dually eligible or reside in areas with a higher deprivation index.

To estimate the scale of resources provided through the health equity benchmark by organizational size, we drew on data from the Medicare Shared Savings Program (MSSP), such as number of dual-eligible beneficiaries per ACO and the area deprivation index for attributed beneficiaries for different sized ACOs. From analyzing the most recent MSSP year performance data, 8 percent of the average ACO’s patients were dual-eligible beneficiaries, with the top decile having an average of 26 percent duals (ranging from 15 percent to 77 percent). For reference, recent data show that the average community health center’s Medicare patient base is 39 percent dual-eligibles, a number well into the top MSSP ACO decile. Similarly, the average MSSP ACO had approximately 12 percent of their attributed beneficiaries living in a high ADI area (assumed as over 90^th percentile ADI ranking—comparable to the health equity benchmark definition), with the ACOs in the highest decile having 21 percent to 72 percent of their beneficiaries in such areas. The wide ranges in percent duals and percent living in high ADI per ACO are due to both measures having a skewed distribution and the top decile of each measure capturing the widest margins of that distribution.

Exhibit 3: Estimates of potential equity payments for REACH ACOs that are in the top decile of traditionally underserved beneficiaries

Source: Authors’ analysis of 2020 Medicare Shared Savings Program public data, examining Area Deprivation Index and duals status at the county level. The average MSSP ACO has around 14,000 to 25,000 beneficiaries (based on recent years’ data). Estimates were calculated based on the specifications in CMS’s ACO REACH RFA and financial methodology webinar materials, such that those beneficiaries in the top decile (highest scores on composite of ADI and duals status) were given a $30 pbpm payment and those in the bottom 5 deciles (lowest scores on composite of ADI and duals status) received a -$6 pbpm payment adjustment. Both upward and downward adjustments can be applied to an ACO in ACO REACH; for every one beneficiary in the upper decile of the duals+ADI score, this could be zeroed out by having five beneficiaries in the bottom 5 deciles of the score. Any ACO where this ratio is higher than 1:5 (for example, 1:4) would receive a net positive payment that increases as the ratio increases; vice versa, any ACO where this ratio decreases below 1:5 (for example, 1:6), would receive a net negative penalty that increases as the ratio decreases. 

Is this benchmark adjustment large enough to have beneficial effects, not offset by undesirable effects? To put exhibit 3 in perspective, we compare the scale of this PBPM financial resource shift (for organizations that care for many underserved populations) to the bonuses in other VBP programs. To reiterate, safety-net organizations would be more likely to have positive financial shifts due to caring for substantially more dual-eligible beneficiaries and other beneficiaries living in resource-deprived areas. We arrange the following examples from less- to more-advanced design in terms of equity-spending requirements and ties to cost and quality performance:

Maryland’s HEART Payment Program

Maryland’s Primary Care Program is currently testing a similar approach to ACO REACH—adjusting PBPM payments (not tied to equity spending requirements or performance) by ADI score and dual population through the Health Equity Advancement Resource and Transformation payment program. However, we do not yet have information on the size of the payments to compare them to the ACO REACH incentives.

FQHCs

To put ACO REACH in an FQHC perspective, Section 330 funding for FQHCs was expanded by the Affordable Care Act to include grants for capital to expand FQHC capacity. These grants, while not tied to equity spending requirements or performance, paid for FQHCs to offer more services, to increase the number of service delivery sites, and to build health information technology capabilities. A recent Government Accountability Office report totaled the value of these capacity-building grants over their first eight years (through fiscal year 2017) at $3.2 billion. There are roughly 1,400 FQHCs, meaning roughly $2,300,000 of potential capacity-building grants were available per FQHC on average. This amount is comparable in size to the upper end of the average health equity benchmark payment we estimated for potential REACH applicants serving more underserved patients. 

Massachusetts MassUP Program

The Moving Massachusetts Upstream (MassUP) Investment Program provided upfront grants to four provider organizations partnering with community-based organizations to address upstream social determinants of health and health equity. The grants ranged from roughly $550,000 to $650,000, which is similar to the lower bound size of our ACO REACH equity payment estimates for smaller ACOs. These grants have community SDoH and equity initiative spending requirements but are not tied to cost and quality performance.

The MSSP Program

The average MSSP annual bonus for ACOs that achieved shared savings (which is based on financial and quality performance) roughly ranged from $4,000,000 to $6,000,000 (for example, over the first five years of the program). The average health equity benchmark payment we calculated is roughly a quarter to half of MSSP ACO shared savings payments, depending on the size of the ACO.

The ACO AIM Program

The ACO Investment Model (AIM) program provided savings upfront to MSSP ACOs in rural or underserved areas in the form of a $250,000 fixed upfront payment, a $36 per beneficiary upfront payment (up to 10,000 beneficiaries), and a $6–$8 PBPM payment (up to 10,000 beneficiaries). Since many of these ACOs were smaller (closer to the small ACOs in exhibit 3), our health equity benchmark estimates are close to the upfront payments in AIM for smaller ACOs (which were expected to be paid back with later savings). 

Examples from other CMS models and commercial payers show that upfront funds can be used to either attract new organizations to VBP models or to specifically build equity-related infrastructure. As discussed above, the AIM program encouraged providers to form new ACOs in rural areas, underresourced areas, or areas with low VBP uptake by providing up-front payments to ACOs for investments in infrastructure and staffing. Results from AIM showed that new ACOs formed from the program using advanced savings generated savings to the system and improved some performance outcomes, although downside risk for these smaller, less-experienced organizations created concerns over program exit. Other payers may consider upfront health equity-focused infrastructure building funds, such as the provider-focused equity grant program funded by Blue Cross Blue Shield of Massachusetts and administered by the Institute for Healthcare Improvement.

Authors’ Note

Support for this brief was provided by the Robert Wood Johnson Foundation. The views expressed here do not necessarily reflect the views of the Foundation. We would like to acknowledge specific support and interest from Katherine Hempstead, PhD. We would like to thank the following individuals for completing interviews with us for this article and helping to review content. They provided crucial insight and answered key questions we had, and we greatly appreciate their time and contributions to this work. The viewpoints expressed in this article do not necessarily reflect the viewpoints of the individuals below nor their organizations. Chris DeMars, MPH, interim director, delivery systems innovation office and director, Transformation Center, Oregon Health Authority; Mark Friedberg, MD, MPP, senior vice president, performance measurement and improvement, Blue Cross Blue Shield of Massachusetts; Christian Lassonde, senior director of negotiation financial management, Blue Cross Blue Shield of Massachusetts; Valerie Lewis, PhD, MA, associate professor, Department of Health Policy and Management, UNC Gillings School of Global Public Health; and Alissa Robbins, MPA, deputy director, Transformation Center, Oregon Health Authority. Dr. Bleser has previously received consulting fees from Merck for research for vaccine litigation unrelated to this work, from BioMedicalInsights, Inc., for subject matter expertise on value-based cardiovascular research unrelated to this work, from Gerson Lehrman Group, Inc., on health policy subject matter expertise unrelated to this work, and from StollenWerks, LLC, on health policy delivery system change unrelated to this work. He also serves as board vice president (uncompensated) for Shepherd’s Clinic, a clinic providing free health care to the uninsured in Baltimore, Maryland. Dr. Saunders has a consulting agreement with Yale-New Haven Health System for development of measures and development of quality measurement strategies for the Innovation Center Alternative Payment Models under CMS contract No. 75FCMC18D0042/Task Order No. 75FCMC19F0003, “Quality Measure Development and Analytic Support,” Option Year 2. Dr. McClellan is an independent director on the boards of Johnson & Johnson, Cigna, Alignment Healthcare, and PrognomIQ; co-chairs the Guiding Committee for the Health Care Payment Learning and Action Network; and receives fees for serving as an adviser for Arsenal Capital Partners, Blackstone Life Sciences, and MITRE. The Duke-Margolis Center for Health Policy values academic freedom and research independence, and its policies on research independence and conflict of interest are available here.