An ecological examination of early adolescent e-cigarette use: A machine learning approach to understanding a health epidemic.

E-cigarette use among adolescents is a national health epidemic spreading faster than researchers can amass evidence for risk and protective factors and long-term consequences associated with use. New technologies, such as machine learning, may assist prevention programs in identifying at risk youth and potential targets for intervention before adolescents enter developmental periods where e-cigarette use escalates. The present study utilized machine learning algorithms to explore a wide array of individual and socioecological variables in relation to patterns of lifetime e-cigarette use during early adolescence (i.e., exclusive, or with tobacco cigarettes). Extant data was used from 14,346 middle school students (Mage = 12.5, SD = 1.1; 6th and 8th grades) who participated in the Utah Prevention Needs Assessment. Students self-reported their substance use behaviors and related risk and protective factors. Machine learning algorithms examined 112 individual and socioecological factors as potential classifiers of lifetime e-cigarette use outcomes. The elastic net algorithm achieved outstanding classification for lifetime exclusive (AUC = .926) and dual use (AUC = .944) on a validation test set. Six high value classifiers were identified that varied in importance by outcome: Lifetime alcohol or marijuana use, perception of e-cigarette availability and risk, school suspension(s), and perceived risk of smoking marijuana regularly. Specific classifiers were important for lifetime exclusive (parent's attitudes regarding student vaping, best friend[s] tried alcohol or marijuana) and dual use (best friend[s] smoked cigarettes, lifetime inhalant use). Our findings provide specific targets for the adaptation of existing substance use prevention programs to address early adolescent e-cigarette use.

Pair Bond-Induced Affiliation and Aggression in Male Prairie Voles Elicit Distinct Functional Connectivity in the Social Decision-Making Network.

Complex social behaviors are governed by a neural network theorized to be the social decision-making network (SDMN). However, this theoretical network is not tested on functional grounds. Here, we assess the organization of regions in the SDMN using c-Fos, to generate functional connectivity models during specific social interactions in a socially monogamous rodent, the prairie voles (Microtus ochrogaster). Male voles displayed robust selective affiliation toward a female partner, while exhibiting increased threatening, vigilant, and physically aggressive behaviors toward novel males and females. These social interactions increased c-Fos levels in eight of the thirteen brain regions of the SDMN. Each social encounter generated a distinct correlation pattern between individual brain regions. Thus, hierarchical clustering was used to characterize interrelated regions with similar c-Fos activity resulting in discrete network modules. Functional connectivity maps were constructed to emulate the network dynamics resulting from each social encounter. Our partner functional connectivity network presents similarities to the theoretical SDMN model, along with connections in the network that have been implicated in partner-directed affiliation. However, both stranger female and male networks exhibited distinct architecture from one another and the SDMN. Further, the stranger-evoked networks demonstrated connections associated with threat, physical aggression, and other aversive behaviors. Together, this indicates that distinct patterns of functional connectivity in the SDMN can be detected during select social encounters.

Risk for substance use among adolescents at-risk for childhood victimization: The moderating role of ADHD.

BACKGROUND: Youth who are victimized by violence are at heightened risk for substance use (SU) during adolescence, a period characterized by elevated impulsivity and risk-taking behavior. This risk may be magnified by attention-deficit/hyperactivity disorder (ADHD). OBJECTIVE: To examine risk/protective factors for adolescent SU among adolescents at-risk for victimization and whether ADHD moderates these associations. PARTICIPANTS AND SETTING: Participants were 1058 caregiver-adolescent dyads in the U.S. who participated in the Longitudinal Studies of Child Abuse and Neglect (LONGSCAN). METHOD: Binary logistic regression analyses were conducted for each SU type. First-order effects of all variables were tested first and for each SU outcome, followed by tests of two-way interactions between ADHD group and each predictor, after controlling for first-order effects. RESULTS: More externalizing behavior (odds ratio [OR] = 1.38; 95 % confidence interval [CI]:1.12, 1.71) and less parental knowledge (OR = .75; 95 %CI: .60, .95) were associated with greater risk for subsequent tobacco use. Less positive peer affiliation was associated with greater risk for subsequent illicit SU (OR = .59; 95 %CI: .36, .96). More deviant peer affiliation were associated with greater risk for all forms of SU. ADHD moderated the association between deviant peer affiliation and marijuana use [b = .9, p < .05, 95 %CI: .03, 1.77), such that deviant peer affiliation was a significantly stronger predictor of marijuana use among adolescents with ADHD than those without. CONCLUSIONS: Findings suggest risk and protective factors for SU are largely consistent for adolescents at-risk for victimization with and without ADHD, but at-risk adolescents with ADHD may be more susceptible to deviant peer influences.