In this representative sample of middle school students (6th–8th grade), girls had about twice the prevalence of past 30-day EVP, alcohol, marijuana, and NMPPM use compared to boys. Our finding that girls reported more substance use than boys are consistent with literature among older adolescents (Centers for Disease Control and Prevention, 2024a; Miech et al., 2025). There is limited evidence on the sex differences among early adolescents, but our findings do align with national data among 8th graders (Miech et al., 2025). Our findings build upon this limited set of evidence by including a broader range of early adolescents (6th–8th graders) and including underreported substances, like NMPPM. Together, these findings highlight the importance of early intervention, as youth who initiate substance use at an early age have an increased risk of SUD (Garofoli, 2020; Nelson et al., 2022). The early adolescent developmental period is critical for substance use prevention efforts, and understanding the preventable and intervention-relevant factors that contribute to sex differences in substance use is critical.

Our decomposition framework used counterfactual simulation and allowed us to estimate what the sex disparity in substance use among middle school students would be if girls had the same prevalence of stressors and depressive symptoms as boys. We found that depressive symptoms statistically accounted for 44.9%–64.0% of the higher prevalence of substance use among middle school girls. Following depressive symptoms, household ACEs - such as emotional abuse, household mental illness, and household substance abuse – had the next strongest contribution (26.9%–44.7%). Our findings support the necessity of multilevel primary and secondary prevention efforts addressing these stressors.

There are multiple strategies for primary and secondary prevention of ACEs and electronic bullying. At the family level, parenting/co-parenting classes and home visiting programs have been shown to be promising at preventing ACEs (Kinsey et al., 2024). At the school level, school connectedness may act as a form of secondary prevention against stressors for both high school (McCabe et al., 2026) and middle school students (Clements-Nolle et al., 2022). ACE prevention may also occur at the community level, by leveraging community resources, such as increasing access to transportation, to mitigate the effects of lower incomes on ACEs (Blair et al., 2019). At the policy level, ACEs may be prevented through increased economic support, such as family-friendly work policies and tax credits (Ottley et al., 2022).

Additionally, while we did not explore substance use motivations, our findings align with the theory that the higher substance use among early adolescent girls is influenced by their motivation to use substances as a form of coping (Agoglia et al., 2020; Dir et al., 2017; Romano et al., 2021). Learning healthy coping behaviors is important for early adolescent girls, as they experience stress differently than boys in part due to neurobiological differences in stress responses. During adolescence, girls have a heightened stress reactivity compared to boys (Dir et al., 2017). So, early adolescent girls may especially benefit from assistance with finding healthy ways to deal with chronic stressors, instead of turning to self-medication with substance use. There is a critical need for substance use interventions that include a focus on addressing depressive symptoms and trauma among middle school girls. School-based interventions that utilize cognitive behavioral therapy (CBT) programs have been found to reduce depression among secondary-school populations (Zhang et al., 2023), and trauma-focused CBT interventions are beneficial to early adolescent girls (Auslander et al., 2017). These CBT programs may be especially beneficial for girls as they often include teaching healthy and adaptive coping skills (Dir et al., 2017), which may replace unhealthy coping behaviors, such as substance use.

We found it noteworthy that cyberbullying victimization was a significant contributor to understanding the higher prevalence of substance use among middle school girls, but bullying victimization on school property was not. Our findings align with prior research, which found that adolescent cyberbullying victimization has stronger associations with substance use (Rostam-Abadi et al., 2024) and emotional problems (Kim et al., 2018) than bullying victimization on school property does. The contribution of cyberbullying to the sex disparity in early adolescent substance use is concerning, given that half of teens have a daily screen time of four or more hours (Zablotsky et al., 2024).

A key strength of this paper is our focus on early adolescence (grades 6–8), a demographic that remains understudied despite evidence indicating that early initiation of substance use increases the risk of SUD (Garofoli, 2020; Nelson et al., 2022). Our sample of early adolescents is a large, representative sample of middle school students across a state, with 98.4% of middle schools participating. Our sample was also racially/ethnic diverse, with nearly half of adolescents reporting themselves as Hispanic. Intersectional analyses were beyond the scope of the current study, but should be explored in future research. Another strength of this study is our methodology; we used decomposition analysis to quantify how much of the sex disparity is statistically accounted for by intervention-relevant factors. Finally, we examined multiple substances, including less researched substances such as NMPPM.

However, this study is not without limitations. First, we analyzed cross-sectional data, so we cannot establish a temporal relationship between the intervention-relevant factors and past 30-day substance use and there is a possibility of reverse or bidirectional associations. Second, there are additional risk factors – such as social media use, relational stress, help-seeking behaviors, or anxiety – that may also play an important part in understanding why early adolescent girls have higher substance use than boys. Third, it is likely that many of the intervention-relevant factors we examined are interrelated and there may be mediated effects, but it was beyond the scope of this paper to examine multiple factors simultaneously. The percent contribution of each factor should be interpreted independent of other factors. Fourth, single items were used to measure depressive symptoms, which may affect the interpretation of the intervention-relevant factors. Future research should use validated, multi-item measures. Fifth, girls are more likely to self-report stressors (O'Gorman et al., 2024) and depressive symptoms (Shi et al., 2021) than boys, so if girls were also more likely to self-report substance use, there may be a bias away from the null. Sixth, boys had a higher prevalence of missing data than girls, as such, the imputed values may not fully capture the true distribution of these variables among boys. Seventh, our findings come from only the state of Nevada, which has a high population of Hispanic early adolescents. As such, our findings may not be generalizable to other states or countries. Finally, while we accounted for sociodemographics, we cannot rule out residual confounding.

In conclusion, we found that middle school-aged girls had almost twice the prevalence of past 30-day EVP, alcohol, marijuana, and NMPPM use compared to boys. These sex disparities may be statistically accounted for, in part, by a higher prevalence of depressive symptoms and chronic stressors – such as emotional abuse, household mental illness, and household substance abuse- among girls. There is an urgent need for primary and secondary prevention of stressors and for sex-specific substance use prevention efforts that are trauma-informed and includes a focus on healthy coping strategies.