Aggression as a Predictor of Early Substance Use Initiation among Youth with Family Histories of Substance Use Disorders.

Abstract

Early substance use initiation is a serious concern because it is associated with significantly increased risks for developing substance use disorders (Anthony & Petronis, 1995; Buchmann et al., 2009; Chen et al., 2009; Dawson et al., 2008; King & Chassin, 2007; Windle & Windle, 2012). For instance, early substance use initiation (operationalized as first use before age 14; King & Chassin, 2007) more than quadruples the odds of having a substance use disorder by young adulthood (OR = 4.21). Early substance use initiation may be an indicator of a broader vulnerability to substance misuse that includes family history of substance use disorder and behavioral disinhibition as risk factors (Iacono et al., 2008; Vanyukov et al., 2003). Although there is a broad literature on risk factors for developing substance use disorder, less is known about predictors of early substance use initiation (Fite et al., 2008). Aggression is one key risk factor associated with problem substance use. Adults with substance use disorder have greater histories of aggressive behaviors (Allen et al., 1997; Moeller et al., 2002; Morie et al., 2014), and aggression is significantly related to their substance use disorder age of onset (Gustavson et al., 2007). Among youth, aggression is significantly related to early substance use initiation (i.e., by age 14; Block et al., 1988; Jester et al., 2008; Leff et al., 2003). One prospective study of adolescents examined the relative contribution of aggressive behaviors versus other measures of externalizing symptoms, internalizing symptoms, and social adjustment in predicting substance use initiation (Ernst et al., 2006). Among youth at high risk for substance use initiation (due to ADHD diagnosis) who had not initiated substance use at study entry (mean age 12.7), aggression, impulsiveness, attention problems and social problems were related to prospective substance use initiation (mean follow-up 3.8 years). Of these, aggressive behavior was the most consistent and largest magnitude of the predictors of early substance use initiation (Ernst et al., 2006). The relationship of aggression with early substance use initiation has been interpreted within the context of Common Liability (Iacono et al., 2008; Vanyukov et al., 2003) and Affect Regulation (Ernst et al., 2006) theories. Based on the effect observed by Ernst and colleagues, we sought to extend this line of research by testing individual components of aggressive behavior to ascertain which are most associated with early substance use initiation. How aggression is assessed can impact how we understand its relationship with early substance use initiation (see meta-analysis, Miles & Carey, 1997). Various approaches to aggression assessment have appeared in the substance use initiation literature. Although these approaches are generally consistent in suggesting relatively greater aggression is related to early substance use initiation, questions remain about which aspects of aggression are responsible for this relationship. One measurement approach has been to examine aggressive personality traits rather than specific aggressive behavior(s). For example, nursery school teacher ratings of children's aggressive personalities predicted later onset of early substance use initiation (California Child Q-set method rating “hostility towards others” and “expression of hostility” Block et al., 1988). However, it is unclear if the teachers' ratings were based on specific aggressive acts observed in the classroom or more general impressions the teachers had on the emotional outlook or attitude of the children. This approach leaves questions about whether actual aggressive acts (i.e., intentionally harming another person) is related to early substance use initiation. The assessment of aggressive acts is typically done with checklist methods; however, checklists often contain a heterogeneous set of items and the use of summary scores that may obscure specific relationships with substance use initiation. Although checklists have shown relationships with early substance use initiation, it is not clear which specific items are responsible for these relationships. For instance, ratings of aggressive behavior on the Child Behavior Checklist are associated with relatively earlier substance use initiation (e.g. Jester et al., 2008). However, these aggressive behavior scores are derived from summing items ranging from “gets in fights” to “demands attention” and “sulks” (Achenbach & Rescorla, 2001), and plausibly not all these dimensions are equally related to substance use initiation. Similarly, the Life History of Aggression scale (LHA) typically is scored to reflect the frequency of 11 distinct classes of behaviors across three domains of aggression (aggression toward others, aggression towards self, and consequences of aggression at school, work or legally; Coccaro et al., 1997). These different classes of behaviors are driven by distinct biological and environmental mechanisms (Yeh, Coccaro, & Jacobson, 2010) and are also not likely to be equally related to early substance use initiation. Not surprisingly, valuable information can be lost when combining heterogeneous aggressive behaviors into composite, unitary scores (e.g., Farrell et al., 2000; Smith et al., 2009). Thus, tests of individual types of aggressive behaviors, rather than summary scores, may provide more precise information about their relationship with early substance use initiation. The purpose of this study was to examine the predictive relationship of distinct aggressive behaviors and early substance use initiation. This approach builds on previous work examining prospective relationships of aggressive behavior and early substance use initiation (i.e., Ernst et al., 2006) in three important ways. First, it extends previous literature by examining distinct classes of aggressive behaviors, as measured by the Life History of Aggression interview (Coccaro et al., 1997). Second, it extends the examination of aggressive behaviors to a sample at increased risk for early substance use initiation, children with family histories of substance use disorders. Third, it replicates the approach of Ernst by testing the relationship of aggression and substance use initiation within the context of other measures of externalizing, internalizing, and social adjustment. Measures of aggressive behaviors and other risk factors were collected before substance use onset, and assessment of early substance use initiation was conducted prospectively at 6-month intervals. Based on earlier findings, we hypothesized that aggressive behavior and impulsiveness would be predictive of early substance use initiation. Of the specific types of aggressive behaviors, we also expected that aggression towards others would be most related to early substance use initiation because it represents a more severe form of aggressive acts.