New Psychiatric Hospitals in the United States: Unequal Geographic Access to Inpatient Mental Health Care

Background: The availability and location of inpatient psychiatric services are important for timely treatment, crisis stabilization, and continuity of care. Studies of psychiatric hospital closure alone do not describe where new services become available. Objective. To characterize new psychiatric hospitals and examine whether their geographic distribution corresponds to communities that have lost psychiatric hospitals. Methods. We used a national census of Medicare-participating freestanding psychiatric hospitals recorded in Medicare cost reports for 2011–2023. Of 753 hospitals in the source files, 746 met reporting-period criteria and contributed 7,409 hospital-years. New hospitals were defined by a first report in 2014 or later. We compared new and established hospitals, examined changes after first reporting, and estimated county-level associations with a new psychiatric hospital opening. Results. Of 746 hospitals, 206 (27.6%) were newly reported during 2014–2023; 87.4% of new hospitals were for-profit. In the initial model restricted to counties with psychiatric hospitals, new hospitals were more likely in counties with larger populations (odds ratio [OR] 1.531, p = 0.009) and greater poverty (OR 1.321, p = 0.019), and less likely in counties with larger Black population shares (OR 0.762, p = 0.028) and in Medicaid expansion states (OR 0.574, p = 0.023). In the broader model of counties with general hospitals, population and poverty remained positively associated with new hospitals, while Black and Hispanic population shares and rural location were negatively associated; Medicaid expansion was not statistically significant (OR 0.66, 95% CI 0.39–1.12). Only 25 of 59 counties with closures also gained a psychiatric hospital, and new hospitals preceded closure in 15 of those 25 counties. Conclusions. This census-based analysis suggests that new freestanding psychiatric hospitals do not consistently appear where psychiatric hospitals have closed. Their distribution raises concerns about racial and geographic inequities in access. The Medicaid expansion finding was sensitive to model specification and should not be interpreted as evidence that expansion reduces psychiatric access.