This study focuses on whether intervention effects on substance use are mediated by sensation seeking, and whether DRD4 genotype moderates this meditational process. Sensation seeking increases risks for substance use (Bates, White, & Labouvie, 1994) and has been found to mediate intervention effects on the same (Clayton, Cattarello, & Walden, 1991). Sensation seeking is regulated by dopamine within the limbic system due to its associations with reward and risk taking (Breiter & Rosen, 1999). More specifically, DRD4 variability has been linked to sensation seeking (Benjamin et al., 1996, Ebstein et al., 1996), executive attention (Fossella et al., 2002), and substance use risk (Conner et al., 2010).
The analyzed sample (n= 943, 60% female; 56% intervention) participated in the PROSPER (PROmoting School-community-university Partnerships to Enhance Resilience) Project, an evidence-based substance use intervention program. A moderation-mediation model was used to test whether intervention effects on late adolescence substance use (Mean Age=18.14) would be mediated by sensation seeking for individuals who possessed the DRD4 7-repeat variant, compared to those lacking this variant. Sensation seeking was assessed with three-items (α = .84; M = 1.17, SD = .97), and substance use was assessed with five dichotomous lifetime substance use items (α = .90; M = .31, SD = 39).
Multi-group SEM examined the effect of the intervention (invention vs. control) on late adolescent substance use across DRD4 genotypes (7+ vs. 7-). The path from intervention status to substance use was significant and similar for both DRD4 genotypes (7-:β=-.13, p <.01; 7+ β = -.14, p< .05). Sensation seeking was added as a mediator, and the a model allowing different parameters across DRD4 genotypes showed good fit to the data (χ2( 25) = 65.57, p<.05, CFI=.99 TLI=.99; RMSEA = .04). Constraining the path from intervention to sensation seeking to be equivalent across DRD4 genotypes (7+ vs. 7-) significantly reduced model fit (Δχ2(1) = 4.48, p<.05). Single-group results revealed individuals who possess the 7+ variant were more sensitive to the intervention’s effects on sensations seeking (7-:β=-.01, ns; 7+:β= -.16, p< .01), which contributed to an indirect effect on substance use via sensation seeking for 7+ youth (7+:βID= -.06, p< .05).