Sociodemographic Factors May Predict Substance Use Initiation in Kids

NEW YORK CITY -- Sociodemographic factors such as religion and race were the most robust predictors of substance use initiation in early adolescence, according to an analysis of the Adolescent Brain Cognitive Development (ABCD) study.

Of the included kids who were substance-naive at baseline, those raised in Mormon families were less likely to start using substances by age 12 (coefficient -0.87), as were Black kids (coefficient -0.32), while those raised in Jewish families were more likely to initiate use (coefficient 0.32), reported ReJoyce Green, PhD, of the Medical University of South Carolina in Charleston, during a presentation at the American Psychiatric Association annual meeting.

The study was also published in the American Journal of Psychiatry.

"In terms of thinking about what can we do clinically with this, I think including a comprehensive assessment of different sociodemographic measures or mental health factors can be particularly informative for providers in a variety of different settings," Green said during her presentation.

Substance use during early adolescence has been associated with development of substance use and mental health disorders in adulthood, Green said. She and colleagues wanted to use the ABCD study, a large longitudinal dataset, to identify potential early predictors for the start of substance use among adolescents by age 12.

The researchers used three hierarchical penalized logistic regression models to assess 420 predictors, including demographic characteristics, parenting behaviors, culture and environment, structural neuroimaging, and physical and mental health.

While sociodemographic factors were the most robust predictors, Green noted that cultural and environmental factors, physical health factors, and parenting behaviors were strong predictors as well.

The models found that other top predictors for not initiating substance use included being raised in the Muslim faith (coefficient -0.25) and living in a household with an annual income range of $12,000-$15,999 (coefficient -0.22).

Meanwhile, top predictors for initiating substance use included a history of detention or suspension (coefficient 0.20) and prenatal exposure to substance use (coefficient 0.15).

Green pointed out that they found that the model that analyzed only self-report predictors had a similar prediction performance to the models that also analyzed hormones, neurocognitive factors, and neuroimaging predictors.

"We essentially found that the more resource-intensive measures -- the neurocognitive functioning, neuroimaging variables -- actually didn't improve our ability to predict initiation," she said.

"This was somewhat surprising for us because you would think that the more variables we would add to the model, the better we might be able to actually predict initiation, but that wasn't the case," she added.

Ned Kalin, MD, of the University of Wisconsin School of Medicine and Public Health in Madison, said the study showed the potential promise of identifying very early predictors for substance use, and the models' potential use in other conditions as well.

"This is really critical to think about in relation to all psychiatric illnesses, including substance use disorders," said Kalin, who moderated the presentation. "Getting better and smarter at predicting risk early in life, and then hopefully being able to intervene in ways that are going to change someone's risk trajectory into something that will be much more positive from the standpoint of outcomes."

To conduct the analysis, Green and colleagues collected data from 6,829 substance-naive children in the ABCD study at 21 sites across the U.S. The children were 9 or 10 years old at baseline and were followed for 3 years.

By age 12, 14.4% of the kids reported substance initiation, with alcohol being the most common substance. These kids tended to be older on average at baseline compared with those who did not initiate substance use (10.08 vs 9.95 years). The two groups had similar distribution across sexes; of those who did initiate use, 54% were boys and 46% were girls.

While the study had a large, diverse sample size, the researchers noted that a primary limitation of their study was the imbalanced outcome groups of those who initiated substance use.

"Further studies are needed to reexamine whether these risk factors for initiation remain consistent or change over time during mid to late adolescence, when we expect to see greater rates of substance use initiation," they wrote.

Comment