Elucidating the "active ingredients" of CBT treatments remains an important and challenging goal. Consistent with the RP model, changes in coping skills, self-efficacy and/or outcome expectancies are the primary putative mechanisms by which CBT-based interventions work [126]. However, few studies support these presumptions. One study, in which substance-abusing individuals were randomly assigned to RP or twelve-step (TS) treatments, found that RP participants showed increased self-efficacy, which accounted for unique variance in outcomes [69]. In a recent study, Witkiewitz and colleagues (under review) found that individuals in the combined behavioral intervention of the COMBINE study who received drink-refusal skills training as part of the behavioral intervention had significantly better outcomes than those who did not receive the drink-refusal skills training, particularly African American clients [127]. Further, there was strong support that increases in self-efficacy following drink-refusal skills training was the primary mechanism of change. In another study examining the behavioral intervention arm of the COMBINE study [128], individuals who received a skills training module focused on coping with craving and urges had significantly better drinking outcomes via decreases in negative mood and craving that occurred after receiving the module.
