Substance use as a longitudinal predictor of the perpetration of teen dating violence.

The prevention of teen dating violence is a major public health priority. However, the dearth of longitudinal studies makes it difficult to develop programs that effectively target salient risk factors. Using a school-based sample of ethnically diverse adolescents, this longitudinal study examined whether substance use (alcohol, marijuana, and hard drugs) and exposure to parental violence predicted the perpetration of physical dating violence over time. 1,042 9th and 10th grade high schools students were recruited and assessed in the spring of 2010, and 93 % of the original sample completed the 1-year follow-up in the spring of 2011. Participants who had begun dating at the initial assessment and who self-identified as African American (n = 263; 32 %), Caucasian (n = 272; 33 %), or Hispanic (n = 293; 35 %) were included in the current analyses (n = 828; 55 % female). Slightly more than half of the adolescents who perpetrated dating violence at baseline reported past year dating violence at follow-up, relative to only 11 % of adolescents who did not report perpetrating dating violence at baseline. Structural equation modeling revealed that the use of alcohol and hard drugs at baseline predicted the future perpetration of physical dating violence, even after accounting for the effects of baseline dating violence and exposure to interparental violence. Despite differences in the prevalence of key variables between males and females, the longitudinal associations did not vary by gender. With respect to race, exposure to mother-to-father violence predicted the perpetration of dating violence among Caucasian adolescents. Findings from the current study indicate that targeting substance use, and potentially youth from violent households, may be viable approaches to preventing the perpetration of teen dating violence.