Perceived risk of substance use and associations with early experimentation: A latent profile analysis using ABCD...

The current study aims to determine heterogenous latent profiles of substance use perceived harm and examine the concurrent associations between profiles and substance use experimentation.

We used data from the Adolescent Brain Cognitive Development^SM Study (5.1 data release) 1-year follow up. Participants (N = 11184) were 52.2% male between the ages of 9 and 14 (M _age = 10.48, SD = 0.65). Latent profile analysis was used with 11 perceived harm indicators. Multinomial logistic regression was used to determine the association between profile membership and substance use experimentation (i.e., alcohol sipping, tobacco puffing, and cannabis puffing).

A four-profile solution fit the data best. The four profiles included participants in 1) high harm perceptions across all substances, 2) low harm perceptions across all substances, 3) moderate harm perceptions across all substances, and 4) selective high-harm perceptions for some substances. Those in the low harm profile (AOR = 4.99, p < .01) and selective high-harm profile (AOR = 3.04, p < .05) were more likely to report puffing tobacco compared to the high harm profile. Also, those in the moderate harm profile (AOR = 1.36, p < .05) and those in the selective high-harm profile (AOR = 1.56, p < .001) were more likely to report sipping alcohol compared to the high harm group.

The four-profile solution illustrates meaningful heterogeneity in how youth perceive substance-related harm, suggesting the need for more tailored prevention approaches rather than one-size-fits-all messaging considering that even partial recognition of risk may not fully protect against experimenting with substances.