A dataset of geocoded Medicaid office locations in the United States

Medicaid is the largest health insurance program in the United States, covering more than 86 million Americans as of early 2023, and is key for progress towards health equity. Although policy changes like Medicaid expansion have significantly expanded the number of people who are eligible for Medicaid, the administrative burdens of enrolling in and renewing coverage can be substantial. Although many applications are now submitted online, physical access to Medicaid offices still plays a critical role in understanding eligibility, getting help in applying, and navigating required documentation for both initial enrollment and redetermination of eligibility. However, as more government functions have moved online, in -person office locations and/or staff may have been cut to reduce costs, and gentrification has shifted where minoritized, marginalized, and/or low-income populations live, it is unclear if the key local connection point between residents and Medicaid has been maintained. To our knowledge, no single source of Medicaid office locations has been assembled and made available for research purposes. Our objective was to identify and geocode all public -facing Medicaid offices in the United States, which can then be paired with other spatial data (e.g., demographics, Medicaid participation, health care use, health outcomes) to explore policy-relevant research questions. We identified Medicaid office addresses in all 50 states and the District of Columbia by searching state government websites (e.g., Department of Health and Human Services or analogous state agency). Our corpus of Medicaid office addresses was then geocoded using the Census Geocoder with unresolved addresses investigated and/or manually geocoded using Google Maps. After deduplication (e.g., where multiple counties share a single office) and removal of mailing addresses (e.g., PO Boxes), our final dataset includes 3026 Medicaid office locations. (c) 2024 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ )