Understanding how risk pathways lead to early substance use in youth using machine learning
Official title: Application of Mediation Mixture Modeling and Machine Learning to Understand Heterogeneity of Risk Pathways on Youth's Onset of Substance Use
Funded at Arizona State University-Tempe Campus · NIH-11454875 · Data updated 2026-11-12
A secondary data study using ABCD data to map diverse paths that could lead 9-year-olds toward starting substance use, with advanced analytics to find different subgroups.
Quick facts
| Grant type | R03 grant |
|---|---|
| Study type | NIH-funded research |
| Funding institution | Arizona State University-Tempe Campus NIH-funded |
| Lab location | 1 site (Scottsdale, United States) |
| Project start | September 1, 2026 |
| Project end | August 31, 2028 |
| Funding year | 2026 |
| Award (this year) | $314,000 |
| NIH project number | 1R03DA065201-01A1 |
| Project ID | NIH-11454875 on NIH RePORTER |
Who could benefit from this research
✓Could help
Children around age 9 who could potentially be followed through adolescence in future studies or trials.
✕Less likely to help
This specific grant analyzes existing data and may not immediately change care for individual patients.
What this research studies
This project analyzes existing longitudinal data from the ABCD study to uncover how factors like parent mental health and child behavior influence early substance use. It combines theory-based mediation models with machine learning to identify multiple, subgroup-specific pathways rather than a single dominant route. The approach uses advanced quantitative methods on a large, national dataset to illuminate how different mechanisms interact over time. The work aims to move beyond one-size-fits-all explanations and toward more nuanced understanding that could inform targeted prevention efforts.
Why it matters
Potential benefit: If successful, findings could inform personalized prevention strategies for at-risk youth and guide clinicians and families on which factors to monitor.
How similar studies have gone: Previous work has used machine learning for prediction; this project uniquely integrates theory-driven mediation with data-driven methods in a large national dataset.
Where this research is happening
-
Arizona State University-Tempe Campus
Scottsdale, United States Active
Researchers
- Principal investigator: Cheng, Cheuk Hei — Arizona State University-Tempe Campus
- Study coordinator: Cheng, Cheuk Hei
Related research
Other NIH-funded projects on the same topics.
-
Improving contraception decisions for women with substance use disorders after childbirth
A patient-centered decision tool called MyPath aims to help postpartum women with substance use disorders choose and keep effective contraception after giving birth.
-
Co-treated care for substance use and heart infection at UVMMC
A hospital-based project aims to improve how doctors and teams treat people with both substance use disorder and infective endocarditis during their hospital stay.
-
Preventing Suicide Among Teens Who Use Substances in School Settings
Aims to see if how schools respond to substance-use incidents (like alcohol or drugs) can lower suicide risk among high‑risk students.
-
People and Place: How geography affects HIV care and substance use in Los Angeles
Aims to understand how where you live in LA affects HIV care and substance use, and to adapt a proven intervention to help PLWH who use substances connect with better care.
-
A crowdsourced, survivor-led approach to boost HIV/STI/HCV testing and overdose-prevention service use for sex trafficking survivors in NYC
A NYC-based study tests a survivor-driven digital messaging approach and at-home tests to help sex trafficking survivors increase HIV/STI/HCV testing and use overdose-prevention resources.
-
Transforming prison care for HIV, HIV risk, and substance use disorders
A patient-friendly look at improving HIV care and prevention for people in state prisons, especially where substance use is common.
About this research
- This is an active NIH-funded research project — typically early-stage science, not a clinical trial accepting patient enrollment.
- Some NIH-funded labs run parallel clinical studies or seek volunteers for related work. To check, contact the principal investigator or institution listed above.
- For full project details, budget, and progress reports, visit the official NIH RePORTER page.