BACKGROUND – Mental health disorders are a growing problem in the U.S. today and has only increased since the social isolation and stress brought by the COVID-19 pandemic in 2020. Due to the care fragmentation and difficulty in obtaining upstream mental health services, emergency departments (ED) have been utilized for treating severe mental health crises, despite their high costs and limited capacity to provide appropriate care.

METHODS – The National Hospital Ambulatory Medical Care Survey (NHAMCS) from 2016-2022 was analyzed by initially conducting three trends’ analyses: 1) descriptive statistics to evaluate prevalence of mental health-related ED visits by socio-structural predictors, 2) descriptive statistics of mental health-related ED visits by diagnostic category, 3) changes in mental health-related ED use between metropolitan and nonmetropolitan statistical areas. Next, logistic regression analyses were conducted to identify associations between socio-structural predictors and likelihood of a mental health-related primary ED diagnosis between 2019 and 2022, stratified by insurance type. Finally, linear probability models, also stratified by insurance type, assessed differences in rates of mental health ED use in metropolitan and non-metropolitan areas.

RESULTS – Overall, the proportion of all ED visits that were mental health-related have increased from 5.0% in 2016 to 5.8% in 2022, peaking at 6.6% in 2020. Within all mental health-related visits, suicidal behavior has increased since 2020. Logistic regression models conveyed that, in general, individuals living in non-metropolitan statistical areas had significantly lower odds of a mental health-related ED visit in 2019 (aOR = 0.47, 95% CI: 0.37-0.58, p <0.001), an association no longer present in 2022 (aOR = 0.80, 4 95% CI: 0.63-1.00, p = 0.057). Trend analyses indicated stagnant or decreasing Lin proportions of mental health-related ED visits in metropolitan areas and increasing proportions in non-metropolitan statistical areas, with linear probability models depicting that the difference in rates of mental health-related ED visits between metropolitan and non-metropolitan status areas may be diminishing from 2019 to 2022 for all insurance types except for Medicaid or CHIP enrollees. CONCLUSION – While this observational study cannot pinpoint the reason for the narrowing gap in mental health-related ED visits between metropolitan and nonmetropolitan statistical areas, it provides exploratory insights into possible unmet mental health needs in non-metropolitan statistical areas that may have worsened due to the COVID-19 pandemic.