Predictive Analytics Tool Accurately Assesses Teen Suicide Risk

Using predictive analytics algorithms, a universal screening tool can accurately determine an adolescent’s suicide risk and alert providers of which patients are in need of follow-up interventions, according to a study published in JAMA Psychiatry.

Researchers stated that suicide is the second-leading cause of death among teens in the US, and the suicide rate among adolescents in the US has grown by 62 percent since 2000. In 2018, the US reported its highest annual number of adolescent suicide deaths that included 1,750 young people aged 12 to 17.

While the number of suicide deaths among teens are clearly on the rise, many at-risk individuals aren’t receiving the mental health services they need. It can be very challenging for providers to detect which of their young patients need the most help.

"Too many young people are dying by suicide and many at high risk go completely unrecognized and untreated," said lead author Cheryl King, PhD, a professor, clinical child psychologist, and director of the Youth and Young Adult Depression and Suicide Prevention Research Program in the Department of Psychiatry at Michigan Medicine.

"About half of the youth who die by suicide have never received any mental health services and some die on their first suicide attempt. We saw an urgent need to improve proactive, universal suicide screening of young people."

Screening tools do currently exist, the team noted. However, previous research has shown that many young people who are high-risk still aren’t detected or too many are detected as being at risk, including many who are false positives.

"Different combinations of risk factors can place youth at risk. If we screen only for suicidal thoughts, we will miss some high-risk adolescents,” said King, who is also a child and adolescent psychologist at Michigan Medicine C.S. Mott Children's Hospital and member of the University of Michigan Injury Prevention Center.

"There are many reasons young people may not share suicidal thoughts, possibly because they're ashamed, they aren't experiencing the thoughts at the time of screening, or someone reacted in a way they didn't feel was helpful when they shared suicidal thoughts or sensitive information in the past."

Researchers set out to develop a personalized system to better detect youths at high risk of suicide. The screening tool, called the Computerized Adaptive Screen for Suicidal Youth (CASSY) is designed to be used in emergency rooms through a brief and efficient system that doesn’t disrupt care.

When an adolescent or teen is admitted for any reason, whether it’s a psychiatric complaint or something unrelated, they complete a questionnaire on a digital device. Follow-up questions and the number of questions are based on their answers so that the screening is tailored to the individual patient.

Unlike existing screening tools, CASSY asks adolescents not only about suicidal thoughts but also about other factors that may put them at risk. These include sleep disturbance, trouble concentrating, agitation, depression and hopelessness, and issues with family and school connectedness. The score for their suicidal risk level is determined by the combination of risk factors.

The CASSY system provides healthcare clinicians in emergency services with the information about the probability of a future suicide attempt. The tool offers thresholds for identifying different levels of risk ranging from mild to high.

The study included two cohorts of adolescents aged 12-17 that visited emergency departments. Researchers developed the CASSY screening tool in the first cohort with 2,075 adolescents and validated in a second independent cohort with data from 2,754 adolescents.

In this second cohort, a total of 165 adolescents, or six percent, made at least one suicide attempt over the three-month period, and the CASSY predicted risk for suicide attempts with over 88 percent accuracy over the next three months.

Researchers expect that emergency departments across the country will implement this personalized screening tool into their care models to improve suicide risk identification and treatment planning.

"This screening tool has the potential to be a step forward in our effort to improve clinical care models to adequately meet the needs of youth mental health," King said.

The predictive analytics algorithm was based on data from multiple centers that participated in the Emergency Department Screen for Teens at Risk for Suicide, funded by the National Institutes of Mental Health.

Researchers noted that emergency departments are ideal for suicide risk screening, as nearly 19 percent of US adolescents visit the ED over the course of a year. ED visits for youth suicide risk and self-harm have also recently doubled and are a common point of access for health services.

As the COVID-19 pandemic continues to keep individuals isolated and put additional strain on teens with mental health issues, the CASSY tool could help reduce the rate of suicide among young people.

"Improving suicide risk detection through effective screening has the potential to facilitate treatment, reduce morbidity and prevent death among teens and young people," King said.