Controlled Drinking: More Than Just a Controversy

Michael E. Saladin; Elizabeth J. Santa Ana


Curr Opin Psychiatry. 2004;17(3) 

In This Article

Moderation-Oriented Cue Exposure

MOCE is a variant of cue exposure treatment for alcohol dependence[41] and is specifically designed to train moderation of alcohol consumption. In general, cue exposure is based on the assumption that cues associated with alcohol consumption acquire the capacity to elicit conditioned responses that bear a functional relationship to craving for alcohol. Cue exposure treatment presumably exerts it therapeutic effects via unreinforced exposures to these cues (i.e. Pavlovian extinction) in an effort to reduce conditioned responses associated with the motivation to drink. The treatment involves systematically exposing patients to cues, such as the sight and smell of their preferred beverage, without being allowed to consume the beverage. While controlled clinical trials of cue exposure aimed at abstinence have not demonstrated a clear superiority over comparison treatments, they have generally shown cue exposure to decrease alcohol consumption and increase the latency to relapse during follow-up.[42,43] Since cue exposure does confer some moderation benefits, it seemed logical that it might be profitably combined with a non-abstinence treatment goal.

Three controlled trials comparing MOCE and BSCT have been conducted. In the first trial,[44] problem drinkers were randomly assigned to receive 6, 90-min sessions of either MOCE (n=22) or a modified version of BSCT (n=20) in group format. MOCE incorporated therapist-guided consumption of priming doses of alcohol (i.e. three standard drinks for men and two for women) to enhance learning of moderate alcohol use and to diminish desire to drink excessively. It also included directed homework practice in which patients (1) identified one or two 'natural' situations (occurring between treatment sessions) in which controlled drinking would be an achievable challenge, and (2) were instructed to have the same number of priming drinks as in a session and then refrain from further drinking. The results at 6-month follow-up indicated that MOCE was superior to BSCT in reducing the frequency of drinking bouts and amount consumed per drinking bout. The authors also noted that 6-month follow-up drinking was not predicted by pretreatment measures of drinking frequency, thereby indicating that treatment responsiveness did not vary as a function of pretreatment drinking severity. Remarkably, drinking outcomes at 6-month follow-up were strongly predicted (mean r=-0.8) by a self-efficacy measure[45**] that assessed confidence in ability to control drinking in difficult situations. This latter finding parallels previous research identifying an association between drinking-related self-efficacy and treatment outcome.[46,47]

Two more recent studies comparing MOCE and BSCT found substantial but similar reductions in drinking outcome.[48,49] In these two trials, MOCE and BSCT were delivered in an individual therapy format, a difference that may account for the differential treatment effects observed in the first study. While it is possible that the group format employed in the first study increased the efficacy of MOCE relative to BSCT, there is no obvious and compelling explanation why this might be true. Similar to the first study, these two also failed to identify any association between pretreatment drinking severity and treatment outcome. In one of the studies,[49] a sub-sample of individuals with pre-treatment levels of dependence above the commonly accepted cut-point for a moderation goal evidenced drinking outcomes comparably favorable to those below the cut-point. This same subsample of more dependent drinkers reported significantly less drinks per drinking day and greater percent days abstinent at 6-month follow-up if they received BSCT rather than MOCE, thereby conferring a therapeutic advantage to BSCT with more severely dependent persons.

The results of three controlled trials comparing MOCE with BSCT suggests that both treatments produce substantial reductions in drinking behavior at 6-8-month follow-up. These studies also uniformly indicate that individuals with a broad range of drinking problem severity benefit from both MOCE and BSCT. With regard to differential treatment effects, it appears that MOCE may offer some advantage over BSCT when the treatments are offered in a group format. Conversely, severely dependent individuals may benefit more from individual BSCT than MOCE. Confidence in the differential treatment effects observed in these three studies should remain modest until the effects have been replicated in studies with longer follow-up. Future studies should examine the relative merit of a combined MOCE and BSCT treatment relative to either treatment alone. It might also be beneficial to develop a computer-based version of MOCE that could be compared with the existing PC version of BSCT described above. Such efforts would substantially increase the accessibility of empirically validated controlled-drinking interventions.