Adaptive Test Can Predict Future Suicidality in LGBTQ Youth

A computer adaptive test (CAT) developed to measure suicide risk can help predict future suicidal thoughts, plans, and attempts in LGBTQ youth, according to a new study published in the Journal of Consulting and Clinical Psychology by Northwestern University researchers. The paper was led by Brian Mustanski, Ph.D., director of the Institute for Sexual and Gender Minority Health and Wellbeing (ISGMH) at Northwestern, and used a CAT developed by the University of Chicago’s Robert Gibbons, Ph.D., who was also a collaborator on the study.

Previously found to accurately measure suicide risk with a brief series of around 10 questions answered using a computer, the CAT-SS was also found to measure suicide risk in lesbian, gay, bisexual, transgender, and queer (LGBTQ) young people without bias.

“LGBTQ youth are at significantly higher suicide risk compared to their heterosexual and cisgender peers. Because of this heightened risk, it’s imperative that any tools and assessments used to measure or predict suicidality in the general youth population are equally effective in the LGBTQ youth population. When we validated the CAT-SS suicide risk test developed by Robert Gibbons among two large cohorts of LGBTQ youth, we found that the CAT-SS accurately predicted suicidality among this overburdened population. The potential to reach and provide services to LGBTQ youth at risk of suicide by using the CAT-SS in environments like schools, community centers, and health clinics is incredibly promising,” said Mustanski.

Critically, the researchers found that the CAT-SS predicted an individual’s suicidality six months after taking the test more accurately than self-reports of past suicidal thoughts, which clinicians have long considered the best way to predict future suicidality.

“This paper demonstrates that the statistical calibration of the CAT-SS in the psychiatric sample exhibits no bias in a sexual and gender minority population. The paper also shows that the CAT-SS significantly predicts future suicidality ranging from ideation to attempt, and does so significantly better than traditional patient self-reports of suicidal thoughts, intention and attempts, previously thought to be the gold standard. This study is one of the first to study the ability of a suicide scale, adaptive or traditional, to predict future suicidal events,” said Gibbons.

The test’s level of predictive accuracy combined with the ease of administrating it both in a clinical setting and online make the CAT-SS a powerful tool for screening LGBTQ youth at risk of suicide, especially in the current moment when young people are increasingly online and the COVID-19 pandemic severely limits in-person interactions.

“The ability to predict future suicidal events on the basis of an adaptive measure like the CAT-SS, holds great promise for our nation’s ability to identify those at risk for suicide and potentially reduce that risk through treatment. These tools transcend the need for in-person assessments, and the need for structured clinical interviews that depend on the availability of trained and experienced clinicians,” said Gibbons.

“Predicting Suicidality Using a Computer Adaptive Test: Two Longitudinal Studies of Sexual and Gender Minority Youth” was authored by Brian Mustanski, Sarah W. Whitton, Michael E. Newcomb, Antonia Clifford, Daniel T. Ryan, and Robert D. Gibbons.