Accountable Health Communities: Advancing Health Equity with Data

Addressing SDoH is central to reducing health disparities and increasing health equity. Strategies for addressing SDoH can target underlying community-level social and economic conditions or individual-level health-related social needs (HRSNs), such as food insecurity and housing instability. For both community-level and individual-level strategies, data are critical to

• Documenting health disparities associated with SDoH
• Identifying individuals and communities that can benefit most from addressing the adverse impacts of SDoH
• Designing effective policies and interventions
• Measuring whether the policies and interventions increase health equity and reduce disparities.

However, the data needed for those strategies, including data to identify and document SDoH, are often lacking. The Accountable Health Communities (AHC) Model is a major Centers for Medicare & Medicaid Services (CMS) initiative that aims to fill this gap. The AHC Model tests whether screening Medicare and Medicaid beneficiaries for HRSNs and using navigators to connect qualified beneficiaries to community resources to address their needs can improve health outcomes and reduce costs. (Qualified beneficiaries have an HRSN and at least two emergency department visits in the year before screening.)

As the evaluator of the AHC Model, RTI International is analyzing rich data from model participants. These data provide information about (1) whether any of the five core HRSNs (i.e., food insecurity, housing instability, transportation problems, utility difficulties, and interpersonal violence) are prevalent among screened Medicare and Medicaid beneficiaries; (2) whether beneficiaries were successfully connected to community resources; and (3) whether their needs were resolved.

We are using a mixed methods evaluation design to

• Understand how the AHC Model was implemented
• Assess success in connecting beneficiaries to community resources and resolving their HRSNs
• Measure impacts on health care utilization and expenditures
• Understand how implementation and contextual factors are associated with model impacts.

The evaluation draws on various data types and sources—including model participant data, publicly available data to describe community context, fee-for-service claims and Medicare Advantage encounters for Medicare beneficiaries, Transformed Medicaid Statistical Information System (T-MSIS) data for Medicaid beneficiaries, key informant interviews, and beneficiary and organizational surveys. To evaluate health care utilization and expenditure impacts, we are taking advantage of random assignment of beneficiaries to the model intervention in some areas; in other areas, we are using quasi-experimental difference-in-differences analysis.

Our early findings illustrate how data can offer essential insights for addressing the upstream determinants of poor health and ensuring programs target those who can benefit most. For example, the data show that having an HRSN in combination with high emergency department use—but not having an HRSN alone—predicts persistently high health care use and expenditures. Therefore, initiatives that address HRSNs may choose to focus limited resources on those individuals because that is where the greatest opportunity exists for reducing health care utilization and costs.

The data also offer insights about what is required to address the needs of individuals with HRSNs. For example, almost 60% of navigation-eligible Medicare and Medicaid beneficiaries had more than one of the five core HRSNs. Addressing co-occurring HRSNs requires a multipronged approach.

CMS has identified data collection, analysis, and reporting as core components of its health equity framework. Improving data collection, analysis, and reporting on SDoH can provide powerful information to shape actions and policies—especially as the United States moves toward the goal of reducing disparities and advancing health equity.