Objective. Behavioral disinhibition is a highly heritable premorbid risk factor for drug use, including marijuana. We sought to model behavioral disinhibition as a latent factor and examine its heritability, factor structure, and predictive validity. It is unknown how behavioral disinhibition may be related to marijuana use trajectory. We sought to address this question using two longitudinal genetically informative samples.
Methods. We evaluated the factor structure and biometric variance decomposition of ADHD, Alcohol Dependence, and Conduct Disorder symptom counts in adolescence in two independent twin samples, one from Colorado (N=2333) and one from Minnesota (N=3630), with highly similar birth years and assessment waves. Next, we evaluated the relationship between adolescent behavioral disinhibition and marijuana use frequency in early adulthood (mean age = 26.2, SD = 2.5). Finally, we will evaluate marijuana use trajectories across adolescence and young adulthood and their relationship with behavioral disinhibition. To evaluate trajectories, we will run a growth model in which the observed variables will be marijuana use frequency. We will fix the factor loadings for the intercept and we will use a definition variable (participant age) as the factor loadings for the slope. We will evaluate differences in slope and intercept between states and we will examine the correlation between adolescent behavioral disinhibition and slope and intercept in each state.
Results. A single factor model fit the adolescent symptom counts well, under strong measurement invariance (χ2 = 5.11, df = 6, p = .53). Adolescent behavioral disinhibition was associated with young adult marijuana use frequency to the same extent in both states (CO r=.31, 95% CI=[.24, .38], MN r=.36, 95% CI= [.30, .42], χ2 = 2.10, df = 4, p=.71; Constrained Model AIC = -9,553.7, Base Model AIC = -9,547.8; Constrained Model BIC = -129,950.7, Base Model BIC = -129,920.8). Analyses evaluating marijuana use trajectories are underway.
Conclusions. The factor structure of behavioral disinhibition was highly similar between these two independently conducted studies. The large samples sizes mean that these analyses were well powered to detect differences in the latent factor and its predictive ability. The lack of significant differences between states means that behavioral disinhibition is a robust and replicable factor. The heritability of behavioral disinhibition and factor loadings for all observed variables are the same in each state. Adolescent behavioral disinhibition was positively associated with later marijuana use to the same degree in each cohort. Because adolescent behavioral disinhibition is the same in both states, future analyses can utilize this factor in predicting other adult outcomes associated with externalizing behaviors. Iacono, Malone, and McGue (2008) proposed that a common genetic vulnerability to behavioral disinhibition underlies the co-occurrence of substance use disorders and externalizing psychopathology. Our replication of this factor across states supports this and sets the stage for further research on specific factors to drug use development and expression of this underlying vulnerability over time.
Iacono, W.G., Malone, S.M., & McGue M. (2008) Behavioral disinhibition and the development of early-onset addiction: Common and specific influences. Annual Review of Clinical Psychology (4) 325-348. doi: 10.1146/annurev.clinpsy.4.022007.141157
Developmental Disorders (e.g. ADHD) , Psychopathology (e.g., Internalizing, Externalizing, Psychosis) , Statistical Methods , Substance use: Alcohol, Nicotine, Drugs
