Can Medicaid Managed Care mitigate race/ethnic health disparities in diabetes?

ABSTRACT This study represents a timely investigation that addresses race/ethnic disparities in type 2 diabetes (T2DM) care over a period that included a major pandemic shock. T2DM is burdensome and disproportionately impacts vulnerable and disenfranchised populations; of note, there are stark race/ethnic disparities in T2DM care goals, emergency department (ED) visits, and hospitalizations. Medicaid covers 25% of Americans with T2DM. More than 80% of Medicaid beneficiaries nationally receive at least some of their care from Medicaid managed care organizations (MMCO). States contract with private (non-profit or for-profit) MMCOs to lower costs, increase quality, and pass on financial risks of covering Medicaid beneficiaries. Heterogeneity across and within state programs can have implications for quality of T2DM care and, specifically, race/ethnic disparities through benefit generosity or by affecting MMCO entry and post-entry behavior. State policymakers also have significant influence over marketplaces in which MMCOs compete, which can have consequences for race/ethnic disparities, given that Medicaid disproportionately covers non-white populations. Little is known about whether and how MMCOs and the state programs they operate in influence disparities in T2DM care and, if or how the COVID-19 pandemic changed the trajectory of health disparities. We propose to answer these unknowns using a convergent mixed-methods study: we will compile a database of MMCO/state program features that could influence care using a health disparities conceptual framework (Aim 1); we will empirically explore race/ethnic disparities among adults with T2DM and whether these vary by MMCO/state features and pre-/post-COVID-19 using comprehensive data from the Transformed Medicaid Statistical Information System over 2016-2025 (Aims 2 and 3); and we will collect and analyze qualitative data from Medicaid stakeholders to triangulate and contextualize the quantitative findings (Aim 4). We focus on non- disabled, non-pregnant 18-64-year-old adults with T2DM who tend to remain stably covered by Medicaid over time. To reduce selection bias, we focus our analyses on 12 states and the District of Columbia that mandate enrollment in comprehensive MMCOs. We will use panel data models to examine race/ethnic and sex-specific receipt of key T2DM services and ED visits and hospitalizations, overall and by MMCO/state features. We will also follow a continuously enrolled cohort over 2020-2025 to assess if and how MMCO/state program features moderate the pandemic's effects on T2DM disparities. Sensitivity analyses will explore the influence of churn. Further, our preliminary analyses identify Kentucky and Florida as having the lowest and highest disparities in T2DM care, respectively; we will conduct interviews in these states to examine what MMCO/state features and implementation might explain these disparities. This policy-relevant work will provide critical insights into how Medicaid managed care programs can be designed to reduce disparities in chronic disease burdens.