Meet CASSY: the suicide prediction algorithm targeting US youth
Join Our Newsletter - Get important industry news and analysis sent to your inbox – sign up to our e-Newsletter here
X

Meet CASSY: the suicide prediction algorithm targeting US youth

By Chloe Kent 19 Mar 2021

Suicide attempts among US adolescents have increased dramatically over the past two decades, but almost half of adolescents who die by suicide have never been treated for any mental health problem. Now, researchers at Michigan Medicine have developed an algorithm that could help predict the risk of suicide attempts in teen patients.

Meet CASSY: the suicide prediction algorithm targeting US youth
Only 41.2% of US adolescents who die by suicide have ever been treated for a mental health problem. Credit: Shutterstock

Over the past two decades the suicide rate among US adolescents has drastically increased, shooting up by a dramatic 62% since the year 2000. In 2018, 1,750 suicide deaths were recorded among US adolescents aged between 12 and 17 years of age, the highest annual number of suicide deaths in this age group to date. Results from a 2019 survey into youth behavioural risks found that 18.8% of high school students seriously considered attempting suicide and 8.9% had attempted suicide – yet only 41.2% of adolescents who die by suicide have ever been treated for a mental health problem.

Researchers at Michigan Medicine, the University of Michigan’s academic medical centre in Ann Arbor, have now developed a universal screening tool designed to predict the suicide risk of a teenage patient during emergency care visits. Whenever an adolescent is admitted to hospital for any reason, they’re asked to complete the questionnaire on a digital device.

According to data published in JAMA Psychiatry, the Computerized Adaptive Screen for Suicidal Youth (CASSY) algorithm has demonstrated a sensitivity of 82.4% and a specificity of 80% in predicting a suicide attempt.

The researchers behind the tool say it has the potential to facilitate linkage to mental health services for suicidal youth – potentially preventing future suicide attempts from occurring.

What is CASSY?

The CASSY questionnaire is tailored to the individual, with the number of questions and follow-up inquiries contingent on their answers.

The CASSY algorithm evaluates their responses according to how strongly they present with suicidal ideation and behaviour, psychopathology, post-traumatic stress disorder, social adjustment, sleep, anger/ aggression and substance use. It then expresses a final 0% to 100% score to determine their suicide risk level.

Michigan Medicine professor Dr Cheryl King says: “[CASSY] provides a continuous measure of risk for suicide attempt (level of risk), rather than just yes/no. It has excellent classification accuracy for use in general medical emergency departments.”

Because the tool presents its final score as a risk percentage, it enables emergency departments to pick a follow-up threshold compatible with the resources at their disposal.

“This is important because the two challenges to suicide risk screening are:  firstly, screening tool accuracy, and secondly, the system’s capacity to follow up with all youth who screen positive,” says King. “Thus, a clinic or emergency department could choose a threshold with high sensitivity (identifying more youth at risk), even though that means more false positives and youth to follow up.  Or, if they have fewer resources, they could choose a threshold where they still identify many youth at risk (not as many), and have fewer false positives – they could follow up with all of these [if there’s a capacity issue].”

How was the tool tested?

The study of the tool’s efficacy included two cohorts of adolescents aged 12 to 17 that visited emergency departments.

CASSY was developed in the first cohort with 2,075 youths and validated in a second, independent cohort with data from 2,754 youths. In this second cohort, a total of 165 adolescents (6%) made at least one suicide attempt over the three-month period, and CASSY predicted risk for suicide attempt with more than 88% accuracy over the next three months.

King and her colleagues are now hoping that US emergency departments will incorporate the tool into their care models to identify and treat youth who are at risk of suicide.

“The tool is being distributed by Adaptive Testing Technologies and they expect it to be ready within the next two to four weeks for distribution,” says King. “We have worked with them to set threshold and ‘warning signs’, which the clinicians also get in the screen report.”

As the Covid-19 pandemic continues, the mental health of young people continues to be placed under additional strain – there’s every chance an algorithm like CASSY could, for some, end up being lifesaving.