Health Care’s First Attempt at Social Risk Adjusting: How the ACO REACH Program Will Test a Financing Model Intended to Incentivize Equity
Post Written by Ken Robin, Chief Data Analyst
CMMI released its first new value-based model of care on February 24, 2022, replacing the population Global and Professional Direct Contracting Model with the ACO REACH program. Last week, HSG’s Chief Solutions Officer Juliette Price outlined ACO REACH, detailed similarities and differences between the two models, and explained that the new approach seeks not only to improve quality of care and health outcomes, but also to “promote health equity among all model participants.”
New to the financial methodology is a Health Equity Benchmark Adjustment (HEBA) that adjusts beneficiary-level premiums based on a composite measure of neighborhood and individual need. The HEBA is one of what will likely be a rapidly growing number of strategies to address equity across health and social sectors.
The Biden Administration’s Stance on Equity
On his first day in office, President Biden signed an Executive Order on “Advancing Racial Equity and Support for Underserved Communities Through the Federal Government.” The order is essentially a call for fairness to be embedded in the day-to-day thinking and practices of government. True to the distinction between equality and equity, the goal is not to direct the same services to all people, but rather to as efficiently as possible direct the right resources to the right people. To help chart an initial course, the Office of Management and Budget submitted a report detailing how government can and has, both inadvertently and purposefully, stratify and silo populations, creating barriers to the establishment of equity. Two important points emerged from this report that are relevant to the ACO REACH model:
Data sets to provide accurate, detailed information on where and how inequity manifests itself are often incomplete or unavailable. Measures of access or impact vary across sectors, and though the assessment of equity is a developing field, much work is yet to be done to create a shared, actionable knowledge base.
The OMB addresses the concept of “Administrative Burden,” which represents the wide array of roadblocks or difficulties individuals can run into when trying to access a service. Examples include paperwork, transportation, required documentation, and navigating complex networks of providers. It is well established that impacts of administrative burdens fall most heavily on disadvantaged populations, meaning they are roadblocks to access particularly for people most in need of services.
The HEBA is a mechanism through which ACO REACH attempts to address these challenges by specifying a standardized and universally accessible demographic dataset and by streamlining the definition of enhanced need to little more than an address.
In her piece last week, Juliette touted the policy of requiring beneficiary-reported demographic and social need data. Rather than run the risk of reinforcing stereotypes by painting with broad brushes and defining need formulaically, CMMI’s push for ACOs to be data-driven and strategic, as well as to tailor service plans based on beneficiary-reported need, is admirable. Much like identified special needs in education are addressed with Individualized Education Plans, a case could be made for health systems to strive for similar levels of customization and personalization.
Selecting an Approach to Adjust for Equity
The logic of using individualized, beneficiary-reported data to build service plans does not however hold when applied to policies that impact financial modeling. A “just ask the patient” approach lacks standardization, is not reliable, and can easily be gamed. The HEBA purposefully builds in shared definitions and is purely mathematical based on established metrics. There are two components that determine an individual’s HEBA score: The Area Deprivation Index (ADI) of the Census Block Group where the beneficiary resides, and Dual Medicaid Status of the individual. To better understand the calculation and application of the composite score, the two model components are discussed in greater detail below.
The Area Deprivation Index was developed at the University of Wisconsin-Madison based on a measure created by the Health Resources and Services Administration more than 30 years ago. It allows neighborhoods to be ranked within a given geography of interest (state or national) and includes considerations of income, education, employment, and housing. The ADI uses U.S. Census data, specifically American Community Survey (ACS) 5-Year Estimates, and thus inherits strengths and limitations of ACS data. The geographic unit of construction for the ADI is the Census Block Group, meaning that application to any other Census geography is invalid. Each Block Group represents a neighborhood and is assigned a percentile score (0-99). CMS will rank Census Blocks nationally based on ADI, with higher numbers representing more deprived neighborhoods. For example, a hypothetical average neighborhood, where income, education, housing, and employment were typical of national norms would receive an ADI score of about 50. The most deprived neighborhoods in the nation would be in the 99th percentile.
A word about Census Blocks and why they were selected as the ADI reference unit: The Block Group is the second smallest unit in the Census core geographic hierarchy. It is a combination of Census Blocks and a subdivision of a Census Tract. Importantly, the Block Group is the smallest entity for which the Census tabulates and publishes sample data and is thus included in the annual ACS. As shown in the chart below, because it is part of the core hierarchy (nation to block), the Block Group is a stable unit for which comprehensive demographic data are generally available. Units outside of the center column, such as Zip Code Tabulation Areas (ZCTA), can be problematic as they can overlap any unit to which they are not connected in the chart. ZCTAs, for example, can cross regional, state, and county lines, and should really be reserved for issues of efficient mail delivery.
Dual Medicaid-Medicare Status is a simple yes/no component in the ACO REACH HEBA calculation, with points added based on full or partial dual eligibility. Categories of dual eligibility are complex, and more than 3 million beneficiaries meet one of the “partial” definitions. Based on income and asset cut-offs, some beneficiaries qualify to have, for example, Medicare Part B premiums covered by Medicaid, but not other costs. This beneficiary-level indicator is included to capture economic factors that impact access to high quality care.
The two measures that feed the HEBA are not equally weighted - the ADI is a continuous variable based on a percentile and Dual Medicaid Status is binary, so CMS developed an approach to combine them. A composite score will be calculated starting with the ADI value of a beneficiary’s Census Block Group of residence, then adding 25 points if the beneficiary is dually eligible. For example, an individual living in a highly deprived neighborhood (90th percentile ADI), who is also dually eligible (+25), would receive a composite score of 115 (90+25). Based on this composite score, CMS will then take all aligned beneficiaries and organize them into deciles (10 equal size groups) ranked from low to high with the top decile theoretically representing the highest need population. Benchmark adjustments will ultimately be made based on the decile into which each beneficiary is placed.
Both upward and downward adjustments will be made. An upward adjustment of $30 per beneficiary per month (PBPM) will be applied to beneficiaries in the top decile, and a smaller downward adjustment of $6 PBPM will be made for beneficiaries in the bottom five deciles (representing the lower risk 50% of the population). No adjustment will be applied to individuals in deciles 6-9. Mathematically, the HEBA is set up to be 1:5 net neutral, meaning if an ACO signed up one beneficiary from the highest decile and five from the bottom five, there would be no impact on the ACOs overall benchmark. The terminology of “adjustment” rather than “bonus” is meaningful because it conveys the idea that no extra service is being offered. Rather, it reinforces that equal levels of access and quality will cost more to deliver to certain populations. CMS does offer the caveat that other variables may yet be explored as additions to the HEBA, and applicants will be notified of any changes before the start of Performance Year 1 (2023).
Is HEBA Enough to Change Behavior?
This, in essence, is the million dollar (or perhaps, $30 PBPM) question. For now, we have a payment mechanism built into a health care contracting model that is intended to promote equity by incentivizing service to high need populations–whether this approach will actually do so or not, is yet to be seen. Remember, new models from CMMI are all “test” models, they exist to test new hypotheses about how we can change the way we pay for and deliver care to reach our collective goals.
The only information we have right now about how HEBA will affect the model is based on simulations that CMMI has run and alludes to in the concept documents. Assuming these models were run on Direct Contracting Entities, simulations of the model as currently designed suggest quite modest impact on the vast majority of ACOs. Most would see an impact on their Performance Year Benchmark of +/- 0.2%, with the greatest benefit being about +1.0% and the greatest deficit being about -0.5%.
However, it is important to note that any CMMI simulations were done to previously-organized ACOs. Remember, Direct Contracting Entities had to apply for the designation and to participate in the program. They did so by evaluating the program outlines of Direct Contracting, looking at their own beneficiary spread, and making the decision that this was a good fit. Now that CMMI has opened the application process for a new program and new ACOs can apply, will we see new applicants to ACO REACH that are unlike Direct Contracting Entities? We cannot assume that all REACH ACO applicants will be similar in nature to former Direct Contracting applicants–that would be like walking into a basketball game and assuming all attendees also like baseball, just because both games include round objects.
Ultimately, it will be interesting to see if the HEBA, as a component of ACO REACH, will affect any significant behavior change. It represents an undeniably important philosophical stance, but time will tell how convincing a case it makes and whether health care access truly expands in neighborhoods that need it most.
About the Author: Dr. Ken Robin is the Chief Data Analyst at Helgerson Solutions Group. Connect with him on LinkedIn.
