university researchers Hawaii At Manoa, we use artificial intelligence and machine learning (AI/ML) to better understand what improves treatment outcomes for individuals receiving treatment for substance use disorders.
A recent study published in The Journal of Prevention Science analyzed more than 7.9 million publicly available treatment records across the United States to identify patterns in services, recovery, and outcomes. The study was led by Trina Becker, an assistant research scientist at the University of California Center on Aging, and Alberto Gonzalez Martinez, a computer scientist at the University of California.
“We believe our findings will help states and local governments better understand how to treat substance use disorders and support people in the long-term recovery process at a time when drug overdose deaths remain a major public health concern across the United States,” Becker said.
Key factors predicting good treatment outcome
“We developed and used an ensemble machine learning model called a random forest model to predict the 10 most important features that increase the likelihood of a good treatment outcome,” Becker said.
The analysis found that the most important factor associated with positive outcomes was the length of time an individual remained in treatment, regardless of setting. According to Becker, longer involvement significantly increases the chances of reducing or stopping drug use.
Other important factors include ease of access to treatment according to clinical need, type of treatment at admission and discharge, housing status, participation in self-help groups, employment status, and referral source.
Mapping disparities in treatment services
AI/ML tools have also allowed researchers to map and visualize data, revealing patterns that are difficult to detect using traditional methods. Using a machine learning random forest model, the research team found that states with the highest overdose death rates tended to have fewer clinically appropriate treatment services available.
“Without the help of AI/ML, it would have been virtually impossible to analyze so many treatment records,” Becker said.
Based on the findings, Becker recommends that states prioritize behavioral health services and work together to expand access to long-term, clinically appropriate treatment programs. Increased availability, especially in states with limited treatment infrastructure, could significantly improve recovery outcomes nationwide.
Professor Becker, who recently received a Pilot Project Award from PIKO (Pacific Center for Innovation, Knowledge and Opportunities), plans to further develop his research by examining local data on addiction treatment and recovery for Native Hawaiians and Pacific Islanders.
The article Using AI to identify key components of drug use recovery was first published in University of Hawaii System News.
/Open to the public. This material from the original organization/author may be of a contemporary nature and has been edited for clarity, style, and length. Mirage.News does not take any institutional position or stance, and all views, positions, and conclusions expressed herein are solely those of the authors. Read the full text here.
