Probing the National Increase in U.S. Adult Mental Health Care Use, 2019–2023
OBJECTIVE: The authors aimed to estimate the extent to which changes in anxiety and depression symptoms and in insurance coverage were associated with the national increase in mental health care use from 2019 to 2023. METHODS: A total of 147,488 respondents were identified from the 2019-2023 National Health Interview Surveys. The authors used item response theory to model longitudinal anxiety and depression scores and Oaxaca-Blinder decompositions of trends to compare data from 2019 with those from 2023. Associations between mental health care use, mental health symptoms, and insurance coverage were modeled among demographic subgroups with penalized logistic regression. RESULTS: From 2019 to 2023, mental health care use increased from 19.2% of individuals reporting use in the past year to 23.8%, a 24.0% increase. Psychotherapy use increased from 9.5% to 13.4%, a 41.1% increase, and mental health medication use increased from 15.7% to 18.8%, a 19.7% increase. Decomposition analysis suggested that 36.7% of the increase in mental health care use was associated with changes in depression and anxiety, whereas 11.3% was associated with changes in health insurance coverage. CONCLUSIONS: Increasing anxiety and depression symptoms and changes in health insurance status were associated with significant shares of the increase in mental health care use from 2019 to 2023. In the context of a shifting U.S. policy landscape, outlining strategies to maintain accessible, affordable, and high-quality mental health care is important.
Authors
- Mark Olfson (ORCID: https://orcid.org/0000-0002-3958-5662)
- Y. Nina Gao (ORCID: https://orcid.org/0000-0003-4058-607X)
- Kara E. Rudolph (ORCID: https://orcid.org/0000-0002-9417-7960)
Institutions
- Columbia University (US)
Publication Details
- Journal
- Psychiatric Services
- Published
- 2026-09-11
- DOI
- https://doi.org/10.1176/appi.ps.20260060
- Primary Topic
- Mental Health Treatment and Access
- Type
- article
- Field-Weighted Citation Impact
- 0.00