Landmark Study of 3.5 Million Patients Reveals Surprising Psychosis Patterns Among Asian Americans

Asian Americans have long been described as the population least likely to use mental health services in the United States, a pattern so consistent that it has become a fixture of psychiatric literature and public health planning. Yet a new large-scale analysis suggests that when Asian individuals do make it through the doors of the mental health system, the clinical picture that greets providers may be strikingly different from what many clinicians expect. In one of the largest help-seeking samples ever assembled for this question, researchers at Stanford University School of Medicine found that Asian adults seeking outpatient mental health care were more likely than many of their non-Asian peers to carry a recorded diagnosis of psychosis, a finding that challenges assumptions about who develops severe mental illness and who reaches the treatment system.

The study, published in BMC Psychiatry by Kimberly Smith, Bohye Kim, Jane Paik Kim and Sripriya Chari, examined records from 3,521,905 US-based adults who received mental health services provided or funded through state mental health agencies in 2018. The data came from the Substance Abuse and Mental Health Services Administration’s Mental Health Client-Level Data Annual Report, a federal dataset that aggregates client-level information from publicly funded mental health programs across the country. Because the dataset is publicly available and fully de-identified, the researchers were able to conduct a secondary analysis without direct contact with human participants, and formal ethics approval and informed consent were not required under institutional standards for such research.

The central question was deceptively simple: among people who are already seeking outpatient mental health care, do rates of diagnosed psychosis differ between Asian individuals and people of other racial groups? This framing matters enormously. Most previous research on racial disparities in psychosis has compared rates across the general population, where differences in service use can mask or distort true differences in illness. By restricting the analysis to a help-seeking sample, the Stanford team attempted to hold constant, at least partially, the enormous barrier that mental health service use represents for many Asian Americans, and to look instead at what happens once people are inside the system.

The comparison group was carefully constructed. The researchers contrasted psychosis diagnoses against what they call DAPS conditions: depression, anxiety, personality disorders and substance use disorders. These four diagnostic categories represent the bulk of outpatient psychiatric practice, and they serve as a benchmark for how frequently different racial groups receive non-psychotic diagnoses once they seek care. Prevalence ratios and odds ratios were calculated to quantify the differences, the standard epidemiological tools for comparing the probability of a diagnosis across groups within a large cross-sectional dataset.

The headline result was that, relative to Asian individuals, the proportion of people with a recorded psychosis diagnosis was 49 percent lower among non-Asian/non-Black individuals, corresponding to a prevalence ratio of 0.51. In other words, within this enormous help-seeking population, Asian clients were roughly twice as likely as clients from other racial groups, excluding Black individuals, to have a psychosis diagnosis on record. Black and African American individuals showed a different pattern: the proportion with a recorded psychosis diagnosis was 6 percent higher than among Asian individuals, a prevalence ratio of 1.06, a modest elevation that the authors characterize as relatively small in magnitude.

The odds ratio analysis added further nuance. Asian Americans had higher odds of having a recorded psychosis diagnosis than non-Asian/non-Black individuals, reinforcing the prevalence findings. Black and African American individuals had modestly higher odds of a recorded psychosis diagnosis than Asian individuals, although the authors emphasize that the size of this difference was small. The results for the DAPS conditions were more varied, with different racial groups showing different patterns across depression, anxiety, personality disorders and substance use diagnoses. This heterogeneity is itself informative, because it suggests that the elevated psychosis signal among Asian help-seekers is not simply a mirror image of lower rates of other diagnoses, but part of a more complicated diagnostic landscape that differs group by group.

Why might Asian individuals who seek mental health care be more likely to carry a psychosis diagnosis? The authors are careful to note that their study represents an initial effort to understand these differential rates, and they frame the identification of culture- and race-specific factors as a crucial direction for future research. Several non-exclusive explanations are consistent with the data. One possibility is a help-seeking selection effect: because stigma surrounding mental illness remains pronounced in many Asian communities, individuals with milder conditions such as depression or anxiety may be far less likely to present to publicly funded services, while those experiencing severe psychotic symptoms may be pushed into care by families regardless of stigma. If the mildest cases never enter the dataset, the proportion of severe cases among those who do enter will appear inflated.

Diagnostic and systemic factors offer additional layers of explanation. Clinicians may interpret symptoms differently across racial and cultural boundaries, and language barriers, culturally specific idioms of distress, and limited access to culturally competent assessment can all shape which diagnosis ends up in the record. Prior research has documented that Black individuals in the United States are disproportionately diagnosed with psychotic disorders relative to mood disorders, a pattern widely attributed to a combination of presentation differences, diagnostic bias and pathways into care, and the modest elevation seen for Black individuals in this study is consistent with that literature. Whether similar mechanisms operate for Asian clients, or whether different ones do, is precisely the kind of question the authors argue future work must disentangle.

The scale of the dataset lends the findings considerable weight, but it also imposes constraints that the authors acknowledge. The Mental Health Client-Level Data capture only people who received services through state mental health agencies, meaning the many Americans who obtain care through private insurance, primary care or never seek care at all are invisible to the analysis. Diagnostic records are recorded by agencies and may vary in accuracy and completeness across states and systems. The cross-sectional design cannot establish whether differences reflect true differences in illness rates, differences in diagnostic practice, or differences in who reaches these particular services. The authors also note that the study used ChatGPT during preparation to improve readability in the introduction and discussion, with the authors reviewing and editing the content and taking full responsibility for the publication.

Even with those caveats, the implications for clinical practice and health policy are significant. If Asian Americans who do reach publicly funded outpatient care carry a heavier burden of psychotic illness than previously appreciated, then early intervention services, culturally tailored outreach and diagnostic vigilance in Asian communities become urgent priorities, rather than afterthoughts. The finding also reframes the familiar narrative of low service use: the problem may not simply be that Asian Americans use fewer services, but that the services they reach are encountering them at later, more severe stages of illness. The authors conclude that future studies aimed at identifying culture- and race-specific factors that impact people seeking mental health services in the United States could be crucial in improving functional outcomes, a research agenda that could ultimately determine whether the country’s largest mental health datasets can be turned into a tool for equity rather than a mirror of its blind spots.

Subject of Research: Racial differences in diagnosed psychosis and other mental disorders among Asian and non-Asian adults seeking US outpatient mental health services

Article Title: Exploring rates of psychosis in Asians and non-Asians seeking outpatient services in the United States

Article References: Smith, K., Kim, B., Kim, J. P., & Chari, S. (2026). Exploring rates of psychosis in Asians and non-Asians seeking outpatient services in the United States. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08432-z

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08432-z

Keywords: psychosis, Asian Americans, mental health services, outpatient care, health disparities, SAMHSA, diagnosis, schizophrenia, race and ethnicity, psychiatric epidemiology, BMC Psychiatry, Stanford University