Violence Exposure and Substance Use throughout Adolescence and Young Adulthood: Timing and Duration of Effects By Gender

Introduction. Adolescents who are exposed to violence in their homes and communities are at increased risk for substance use problems. However, little research has paid attention to the timing of this relationship across adolescence and young adulthood. In the current study, we address this by using the time-varying effect model (TVEM) to identify the critical ages at which violence exposure is most strongly associated with substance use across adolescence and young adulthood (ages 14-30).

Method. This study used data from the National Longitudinal Study of Adolescent to Adult Health (Add Health), a nationally representative sample of adolescents in grades 7-12 in the United States. The first assessment (Wave 1) took place during 1994-1995; follow-up assessments took place during 1995-1996 (Wave 2), 2001-2002 (Wave 3), and 2007-2008 (Wave 4). Violence exposure was a dichotomous marker that captured whether the adolescent had (a) witnessed others engaging in violence or (b) been a victim of violence since the last assessment. The within-person, time-varying association between violence exposure and substance use (count of binary indicators for heavy drinking, cigarette smoking, and marijuana use) was assessed using TVEM. Adolescents’ own engagement in violence was included as a covariate with a time-varying effect. Gender was included as a moderator of the dynamic effect.

Results. For males, rates of violence exposure increased from ages 14-18, peaking at age 18 (26% of males exposed), declining from ages 18-23, and remaining steady thereafter. For females, violence exposure peaked at age 16 (14% of females exposed), steadily declined from ages 16-24, and remained steady after. TVEM with males showed that violence exposure was significantly associated with substance use during ages 14-18 only, with the strongest effects appearing from ages 14-15. TVEM with females showed that association between violence exposure and substance use emerged later and lasted longer compared to males, as the association for females was significant from ages 15-22, with the strongest effects appearing from 18-21.

Conclusion. The findings suggest that (1) peak rates of violence exposure were higher among males versus females, and (2) the timing and duration of the significant link between violence exposure and substance use differed by gender, with males affected earlier, and for a shorter period of time, than females. This study demonstrates an application of TVEM to uncover the developmental timing of violence exposure’s association with substance use throughout adolescence and young adulthood. Findings such as these could inform our knowledge of the optimal timing and targeting of interventions aimed at preventing negative outcomes associated with violence exposure.