Researchers develop AI algorithm to personalize treatment of substance use in homeless youth

Researchers develop AI algorithm to personalize treatment of substance use in homeless youth

Researchers at Penn State University have developed an artificial intelligence algorithm they think can not only identify homeless youth who are most likely to develop a substance use disorder, but also help behavioral health providers offer personalized solutions to them for treatment.

The algorithm, developed by lead researcher Amulya Yadav, assistant professor of information sciences and technology in the school’s College of Information Sciences and Technology, looks at not just whether a homeless youth is likely to develop an addiction, but why they turn to drugs in the first place.

Finding the “why,” he said, can help social workers and others not only identify at-risk youth before they become addicted.

“Proactive prevention of substance use disorder among homeless youth is much more desirable than reactive mitigation strategies such as medical treatments for the disorder and other related interventions,” he said. “Unfortunately, most previous attempts at proactive prevention have been ad-hoc in their implementation.”

Researchers interviewed more than 1,400 homeless youth between the ages of 18 to 26 in six states, Yadav said.

“They were given questions about their education status, their sexual behaviors, their employment status, if they had any history of mental illness or whether they were currently suffering from anxiety or depression,” he said in an interview with Health Crisis Alert. “And we asked if they had any past adverse childhood experiences including abuse. And right at the beginning, it asked about the reason they’re homeless and how long they have been homeless.”

Many of them, he said, ended up homeless as they left the foster care system.

When researchers looked at the information, they found that negative childhood experiences and physical street violence were more associated with substance use disorder (SUD) than other types of victimization, such as sexual assault, and that mental health issues like PTSD and depression were more likely to be associated with SUD than other mental health disorders.

“We wanted to understand what the causative issues are behind people developing opiate addiction,” said Yadav. “And then we wanted to assign these homeless youth to the appropriate rehabilitation program.”

By focusing on the causes of the SUD, he said, not only would treatment providers be able to identify which youth are more likely to develop problems, but they would also be able to personalize treatment for those individuals.

“For example, if a person developed an opioid addiction because they were isolated or didn’t have a social circle, then perhaps as part of their rehabilitation program they should talk to a counselor,” explained Yadav. “On the other hand, if someone developed an addiction because they were depressed because they couldn’t find a job or pay their bills, then a career counselor should be a part of the rehabilitation plan.”

Yadav added, “If you just treat the condition medically, once they go back into the real world, since the causative issue still remains, they’re likely to relapse.”

Although not in use in any real-world settings right now, Yadav said the algorithm could be used in a homeless youth shelter. By interviewing the youths as they come to the shelter for food and clothing, he said, social workers could input that data into a computer and let the model determine the likelihood of the youth developing a SUD. If the youth already had an addiction issue, the model would allow the social worker to personalize that treatment specifically for them, using its information on the most likely causes of substance use disorder for them, allowing the behavioral health provider to address the root cause of the drug use.