Unwinding pandemic-era social programs: Effects on healthcare outcomes and disparities

点击翻译按钮获取中文摘要

Project Summary/Abstract This project will study the effects of the expiration of Families First Coronavirus Response Act (FFCRA) increases to Supplemental Nutrition Assistance Program (SNAP) benefit levels on healthcare outcomes and disparities by disability, race, and ethnicity. These groups experience disparities in disease prevention and management, unfavorable clinical events, and care expenditures, each of which are associated with food insecurity. Existing studies have found SNAP to reduce food insecurity and suggest SNAP may reduce unfavorable healthcare outcomes, especially for vulnerable groups, but these studies suffered key causal inference limitations and have focused on SNAP changes substantially different from the FFCRA unwinding. Meanwhile, state-level variation in approaches to unwinding the FFCRA SNAP increases (e.g., timing ranging from April 2021 to March 2023), individual-level variation in the size of benefit losses (ranging from $95 per month to over $1,000 per month), and a Massachusetts (MA) policy that broke the state’s SNAP rollbacks into two uneven steps all create natural experiments we can use to identify the causal impacts of SNAP. Further, we will use robust policy tracking to identify and adjust for non-SNAP policy unwinding patterns, and we will leverage both national T-MSIS Medicaid data and uniquely detailed linked data from MA. Ultimately, the MA and US analyses will complement each other and provide high internal and external validity. Specifically, we will create a registry of policies related to the unwinding of food, housing, and Medicaid access and then develop parsimonious sets of policy unwinding patterns to identify adjusters for later analyses (we will similarly attend to other healthcare, public health, and economic conditions) (Aim 1). We will then estimate the effects of SNAP unwinding on Medicaid event rates (preventive use, unfavorable event rates, and expenditures; Aim 2), and we will finally estimate the effects of SNAP unwinding on disparities in these event rates by disability, race, and ethnicity (Aim 3). For Aims 2 and 3 (2020-2026), we will measure differences in the levels and trends of healthcare outcomes and disparities before and after unwinding-related changes using a multilevel approach to growth curve modeling that is flexible and robust to many validity threats (while also utilizing eligible non-participants with past SNAP applications as a comparator with no unwinding exposure). Current policy proposals would reduce SNAP access, but there is little data to shed light on how such reductions may affect healthcare outcomes and disparities. Our study will reduce this critical knowledge gap. Studying the ongoing SNAP unwinding will extend past work that has focused on SNAP increases and/or changes uniform in their size and timing, and learnings from this project will also advance the NIH Minority Health and Health Disparities Strategic Plan’s Goal 3 to assess “interventions to reduce health disparities.”