Suicide prediction models may not work as well with American Indian, Alaskan Native, and Black patients compared with White, Hispanic, Asian, and unreported ethnicities, a recent study in JAMA Psychiatry shows. These prediction models may even exacerbate existing racial disparities.
Because suicide death rates vary widely according to race and ethnicity and because mental health diagnosis and treatment patterns vary by race, the researchers wanted to see how well the risk algorithm works in patients of various races.
All 7 health systems that participated in the study provide mental health and general medical care. The health systems also had access to electronic health records and insurance claims data, among other criteria. The study sample included more than 1.4 million patients aged 13 years and older and more than 13 million visits.
Prior suicide attempts and inpatient hospitalizations with a mental health diagnosis were most common among visits for American Indian and Alaskan Native patients (5% suicide attempts, 26% hospitalizations) and White patients (4% suicide attempts and 26% hospitalizations).
The number of suicide attempts was higher among White, Hispanic, unreported race, and Asian patients, in that order. Fewer suicide attempts were reported among Black, American Indian, Alaskan Native, Native Hawaiian, and Pacific Islander patients.
Area under the curve (AUC) values were highest for White, Hispanic, and Asian patients and lowest for Black, American Indian, and Alaskan Native patients.
The researchers say the prediction models to target suicide prevention would provide less benefit for Black, American Indian, Alaskan Native patients, and unreported racial groups compared with other races because the models identify a smaller number of suicides.
The researchers cite several reasons for the lackluster prediction models. People of color typically face barriers to accessing affordable, culturally competent mental health care. Practitioner bias may also play a role.
The study is not representative of all people who need mental health care. Too few suicides among minority patients were observed to assess model performance.
“[Black, Indigenous, and people of color] populations already face significant barriers to accessing mental health care and, as a result, have poorer outcomes,” the researchers conclude. “In this context, deploying a prediction model that provides less benefit to already underserved populations will widen this care gap. Health care systems should prioritize resources to reduce race/ethnicity-based inequities in care.”
Disclosures: One study author declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.
Coley RY, Johnson E, Simon GE, Cruz M, Shortreed SM. Racial/ethnic disparities in the performance of prediction models for death by suicide after mental health visits. JAMA Psychiatry. Published online April 28, 2021. doi:10.1001/jamapsychiatry.2021.0493