A striking finding in Figure 6 is the large impact of self-efficacy on outcome. Self-efficacy at posttreatment was the largest associate of the PDA Intercept (PDA at posttreatment), and self-efficacy at 15 months was the largest determinant of PDA slope over the 2-year follow-up period. Self-efficacy at 15 months, in turn, was largely a function of self-efficacy measured at posttreatment, and the number of abstinent friends in the social network at posttreatment. The structural model showed effects for treatment on posttreatment self-efficacy, despite the fact that this did not appear in the direct test of treatment (Figure 5A). This suggests that other variables (namely baseline PDA and AA meetings attended) played a role in eliciting this effect.
