Associations between substance use and oppositional defiant disorder among U.S. adolescents: A national study of high-prevalence subgroups
This national study of over 226,000 U.S. adolescents found that cannabis use has the strongest association with oppositional defiant disorder (ODD), while machine learning analysis identified male adolescents aged 12–17 who use alcohol but not cannabis as the subgroup with the highest predicted probability of ODD, supporting the need for targeted screening in school settings.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine a massive library containing the health records of over 226,000 American teenagers. Researchers from the University of South Florida decided to sift through these records to answer a specific question: How are rebellious behaviors (like arguing with authority or being defiant) connected to teenagers using drugs or alcohol?
They weren't just looking for a simple "yes or no" answer. They wanted to find out which specific substances were linked to these behaviors and, more importantly, to spot specific groups of teenagers who were at the highest risk.
Here is the story of what they found, broken down into simple terms:
The Detective Work: Cleaning the Data
Before they could draw any conclusions, the researchers had to play detective. In the real world, teenagers who use drugs might also be more likely to have other problems (like family issues or different school environments) that could confuse the results.
To fix this, they used a statistical tool called Propensity Score Matching. Think of this like a high-tech matchmaking service. They took a teenager who used a specific substance (say, cannabis) and found a "twin" from the database who was almost identical in every other way (same age, same gender, same background) but didn't use that substance. By comparing these matched pairs, they could be much more confident that any difference in behavior was actually linked to the substance use, not some other hidden factor.
The Main Findings: The "Big Two"
Once the data was cleaned and matched, two substances stood out as having a strong connection to Oppositional Defiant Disorder (ODD)—a condition characterized by angry, irritable, and defiant behavior.
- Cannabis (Marijuana): This had the strongest link. Teenagers who used cannabis were significantly more likely to have ODD than their non-using "twins." The researchers calculated that cannabis users were about 60% more likely to have these behavioral issues.
- Alcohol: This also showed a link, though it was smaller than cannabis. Alcohol users were slightly more likely to have ODD compared to non-users.
The "Silent" Substances:
The study also looked at cocaine and opioids. However, very few teenagers in this massive dataset reported using these. Because the numbers were so tiny, the researchers couldn't say for sure if there was a connection or not. It's like trying to predict the weather by looking at only two raindrops; there just isn't enough data to make a clear pattern.
The Machine Learning "Crystal Ball"
To go deeper, the researchers used Machine Learning (a type of computer intelligence that finds hidden patterns). Instead of just looking at one thing at a time, the computer built a "decision tree" to see how different factors combined to create the highest risk.
Imagine a flowchart that asks a series of questions to sort teenagers into different risk buckets. The computer found that the most important questions were:
- Are you a boy or a girl? (Boys were at higher risk).
- Do you use cannabis?
- Do you use alcohol?
- How old are you?
The Highest Risk Group:
The computer identified one specific group as having the highest "predicted probability" of having ODD (about 20% chance):
- Male teenagers aged 12–17 who reported using alcohol but not cannabis.
The second highest risk group was:
- Male teenagers aged 12–17 who reported using cannabis (but didn't mention alcohol).
Interestingly, the group with the lowest risk (only about 1–2% chance) was female teenagers aged 18–20 who didn't use either alcohol or cannabis.
What the Researchers Concluded
The study found that:
- Cannabis has the strongest statistical link to defiant behavior among US teens.
- Alcohol is also linked, especially in boys aged 12–17 who use alcohol but not cannabis.
- Machine learning is a useful tool for spotting these specific high-risk groups that traditional math might miss.
The researchers noted that because this study looked at a snapshot in time (a single year of data), they cannot say that the drugs caused the behavior or that the behavior caused the drug use. They also admitted that because the data came from people already in the mental health system, it might not represent every single teenager in the country perfectly.
In short: The study suggests that if you are looking for teenagers who might need extra support for behavioral issues, paying close attention to boys in their early teens who are drinking alcohol or using cannabis is a good place to start.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.