Abstract
BACKGROUND: This study explores whether internalizing and externalizing behaviors mediate the association between adolescent social media use and novel psychoactive substance (NPS) use in Australia, addressing a gap where counterfactual mediation methods and NPS use have been rarely applied. METHODS: The study sample included 1605 adolescents from the Longitudinal Survey of Australian Children. We assessed adolescents’ internalizing and externalizing problems using the Strengths and Difficulties Questionnaire. Adolescent NPS use was evaluated by self-reported ever use of synthetic cannabis and other NPS from longitudinal surveys. Social media use was also assessed using self–reported weekday and weekend time use. A causal inference–informed mediation framework was applied to estimate the extent to which internalizing and externalizing problems explained associations between social media use and NPS use. RESULTS: Frequent weekday or weekend social media use was associated with a higher likelihood of NPS use compared with infrequent use. Externalizing problems accounted for approximately 8% of the association for weekday social media use and 9% for weekend social media use. Internalizing behaviors did not significantly mediate these associations. CONCLUSIONS: This study provides insights into the association between adolescent social media use and novel psychoactive substance (NPS) use. Externalizing behaviors statistically accounted for part of this association (potential mediation) for both weekday and weekend use, whereas internalizing behaviors did not. Findings should be interpreted cautiously due to concurrent measurement and potential unmeasured confounding. Despite modest effect sizes, these associations may be relevant at the population level. Externalizing behaviors and digital literacy may be important correlates within broader prevention and risk-reduction frameworks. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s40359-026-04407-7.