One of the ways in which negative mood may lead to substance use is through its effect on craving for alcohol and other drugs. Several current models of substance use disorder etiology or relapse have offered explanations for the interplay among negative mood, craving, and substance use. For example, learning-based models postulate that substance abusers learn that use of alcohol and drugs can temporarily alleviate painful emotions. After such learning has occurred, negative emotions trigger the desire for alcohol or drugs, which in the immediate absence of alcohol or drugs is experienced as craving. The craving in turn then drives the substance abuser to seek out and use alcohol or drugs in an effort to reduce the negative affect [5].

Analyses demonstrated that greater decreases in negative mood during the 16-week treatment phase predicted lower frequency of heavy drinking. Moderation analyses indicated that the relation between negative mood and heavy drinking was much stronger in patients who did not receive the craving module than in those who did receive the module. Dose–response analyses showed that the moderation effect increased with each session of the craving module that was received. Specifically, the correlation between negative mood and heavy drinking was strongest in those who did not receive the module, and became progressively smaller as the number of module sessions received increased from one to five. Patients who received the craving module had lower craving scores over time than those who did not receive the module. Furthermore, mediation analyses indicated that the moderating effect of the craving module on the relation between negative mood and heavy drinking was mediated, or explained, by reductions in craving. The moderating effect of the craving module on the relation of negative mood to heavy drinking was still significant at the 1-year follow-up, although the magnitude of the effect had diminished.

The primary limitation of the study is that participants were not randomly assigned to receive or not receive the coping with craving module, or to different numbers of craving module sessions. Therefore, the effects that were observed, including the dose–response relation between the number of craving sessions received and the degree of moderation, could reflect other factors that were not assessed and controlled for in the study. The effect of the coping with craving module was shown to be mediated by decreases in craving, as was hypothesized. However, the study did not address whether this was accomplished through increases in coping skills, or whether some other therapeutic process yielded reductions in craving. Therefore, further work is needed to help us more fully understand how the coping with craving module exerts a therapeutic effect.