Controlled Drinking: More Than Just a Controversy

Michael E. Saladin; Elizabeth J. Santa Ana


Curr Opin Psychiatry. 2004;17(3) 

In This Article

Behavioral Self-Control Training

Behavioral self-control training is a multi-component behavioral intervention for teaching skills that target controlled drinking as a treatment goal.[26] While variants of BSCT exist,[27] it typically consists of, but is not restricted to, the following treatment elements: (1) self-monitoring of drinking and urges to drink; (2) specific goal setting; (3) rate control of alcohol consumption and drink refusal; (4) behavioral contracting in which reward and consequences for goal adherence are specified; (5) identification and management of triggers for excessive drinking; (6) functional analysis of drinking behavior; and (7) relapse prevention training.[26] Earlier forms of BSCT attempted to train individuals to estimate their blood alcohol concentration (BAC), and thereby enhancing recognition of impending intoxication. However, this component is no longer considered an essential feature of BSCT because of difficulties associated with using subjective experience to estimate BAC.[28] Overall, BSCT employs a range of behavioral strategies (e.g. self-monitoring, rate control, drink refusal) to regulate both drinking behavior and urges in order to achieve and maintain adherence to specific drinking goals.

BSCT is by far the most intensely studied controlled-drinking treatment approach, with more than 30 studies published to date. Walters' meta-analysis[29] of randomized controlled trials published between 1984 and 1997 indicated that BSCT was superior, relative to alternative non-abstinence interventions and no intervention, on measures of alcohol consumption and drinking-related problems. The analysis also suggested that BSCT outcomes tended to be superior to abstinence-oriented interventions, albeit not significantly. Walters attributed this lack of significance to the relatively low number of studies comparing BSCT with an abstinence-oriented intervention (n=6). This study also noted that BSCT was equally effective for persons with moderate and severe impairment (either problem drinkers or alcohol dependent). Outcomes assessed at or after 1 year of follow-up were comparable between BSCT and abstinence interventions, indicating similar lasting treatment effects. In sum, Walters' review, together with that of other investigators,[30] indicates that BSCT is a highly effective treatment for moderate to severe alcohol-related problems.

While the effectiveness of BSCT has been well documented, more recent research has examined methods of enhancing either its accessibility and cost-effectiveness or its therapeutic effectiveness. With respect to accessibility and cost-effectiveness, a study by Hester and Delaney[31] found that 40 heavy drinkers who were administered a personal computer-based version of BSCT evidenced significant reductions in drinking behavior (i.e. standard drinks, mean peak BAC per week and mean drinking days per week) post-treatment and at a 10-week and 1-year follow-up assessment. To illustrate the size of the observed effects, the mean number of drinking days per week pre-treatment was 5.1 whereas at 12-month follow-up it was 3.5. This corresponded to a mean reduction of 20.1 standard drinks per week from pretreatment (mean, 38.7) to 12-month follow-up (mean, 18.6). It is also noteworthy that the use of other drugs (e.g. marijuana, cocaine) did not adversely affect treatment outcome. Whereas greater alcohol use occurred in participants who reported use of other drugs at intake, they exhibited a non-significant trend towards a greater reduction of alcohol use after treatment. The treatment-related decline in alcohol consumption was not offset by a compensatory increase in other drug use and, in fact, there was some evidence of a decrease. Finally, the authors note that the effect sizes observed with the personal computer-based version of BSCT were comparable in magnitude to the effect sizes achieved in studies in which BSCT was delivered by a therapist or via a self-help manual with minimal therapist supervision.

There has been little research on ways to bolster the therapeutic efficacy of BSCT. However, a recent study by Rubio and colleagues[32] examined the possibility that the opioid antagonist naltrexone could augment BSCT outcomes. Their study was based on two lines of research that suggested that naltrexone might be a beneficial adjunct to a controlled-drinking intervention. First, several studies in which naltrexone was used in the context of an abstinence focused therapy have variously reported reduced craving and drinking, decreased likelihood of heavy drinking, and increased latency to relapse.[33,34,35,36,37,38] Second, two uncontrolled open studies[39,40] that were not aimed at abstinence suggested that naltrexone and a brief intervention can reduce consumption in early problem drinkers.

In the Rubio et al. study, mild-to-moderate alcohol-dependent men without comorbid psychiatric disorder were randomly assigned to receive either naltrexone and BSCT (n=30) or BSCT alone (n=30). During treatment, both groups evidenced similar levels of drinking but individuals receiving the combined treatment reported lower craving. At 1-year follow-up, the number of heavy drinking days, number of drinks and craving was lower in the naltrexone treated group. An examination of outcomes for those men in each group that reported heavy drinking days (i.e. more than three drinks per day) during treatment indicated that men receiving naltrexone reported fewer drinking days, a lower total number of drinks and fewer total drinks on heavy drinking days. Regardless of treatment received, approximately 60% of men in the study reported no heavy drinking days at 1-year follow-up.

The literature on BSCT, old and new, is unequivocal with regard to its efficacy. BSCT produces robust reductions in drinking behavior among individuals with mild to severe alcohol problems. Importantly, these treatment gains do appear to persist over time and are no less stable than treatment gains attained via abstinence-focused treatment. Some very recent evidence indicates that a computer-based version of BSCT can produce substantial reductions in alcohol consumption, at least with mild-to-moderately impaired individuals. This particular mode of treatment delivery has several potential advantages over therapist-delivered BSCT including decreased cost, increased accessibility, enhanced convenience and efficiency afforded by the self-administration format, and lower perceived or real treatment-related stigma (the harm reduction section below discusses computer-based treatment further). Future studies should focus on the assessment of the relative therapeutic benefit, cost effectiveness and client satisfaction of computer- versus therapist-delivered BSCT. There is also some evidence that the treatment benefits of BSCT may be enhanced by concurrent administration of pharmacological agents that serve to moderate alcohol consumption (e.g. naltrexone). The positive findings with male problem drinkers are encouraging and should be extended to women, ethnic minorities and individuals with psychiatric comorbidities. Other related pharmacological agents should also be investigated (e.g. acamprosate).