Undercounts of people with serious mental illness using the Washington Group Short Set questions

Front Psychiatry. 2025 Jun 25;16:1606154. doi: 10.3389/fpsyt.2025.1606154. eCollection 2025.

ABSTRACT

BACKGROUND: Accurately counting Americans with mental health conditions is essential to support program development and appropriate resource allocations, which are often based on prevalence data. Multiple federal surveys use the Washington Group Short Set (WG-SS) questions to identify people with disabilities, including those with mental health conditions. However, the WG-SS questions miss many people with mental illnesses, under-representing this population in US federal survey data. Hence, we sought to explore the degree to which people with serious mental illness are missed.

METHODS: We used data from the 2020 National Survey on Health and Disability to assess the rates that respondents with self-reported serious mental illness (SMI) conditions, i.e., major depression, bipolar disorder, schizophrenia, and schizoaffective disorder (n=263), were missed as disabled by the WG-SS questions.

RESULTS: Using the three WG-SS questions suggested by the Washington Group to capture people with mental illnesses, 66.2%, 88.6%, and 96.6% of respondents with SMI were characterized as non-disabled; 58.2% were characterized as non-disabled across the three questions combined.

DISCUSSION: Previous research demonstrated that the WG-SS questions missed almost 60% of respondents with any mental illness. However, the Washington Group states that its question set better captures people with more severe disabilities, so this study focused only on respondents with serious mental illnesses and only on questions that the Washington Group suggests capture people with psychosocial disabilities.

CONCLUSION: Results indicate that the WG-SS questions miss large percentages of even those with the most severe mental illnesses, who therefore may be substantially undercounted in US federal surveys using these questions. In turn, public mental health programs may be substantially underfunded.

PMID:40636437 | PMC:PMC12238763 | DOI:10.3389/fpsyt.2025.1606154