Although social support has been repeatedly identified as a strong correlate of recovery from alcohol problems, enhancing social support has seldom been a focus of treatment research. Married problem drinkers who were willing to have their spouses involved in their treatment were randomly assigned among two brief outpatient treatment conditions: directed social support (DS, n=28) and natural social support (NS, n=28). In both conditions the treatment in which the problem drinkers participated was an identical program of guided self-change, a cognitive-behavioral motivational intervention involving an assessment and four individual treatment sessions. Problem drinkers’ spouses each attended two individual counseling sessions where they were informed about the counseling procedures. The conditions differed in that spouses in the DS group were encouraged to play an active role in helping their partner by being supportive and particularly by reacting to relapse episodes in a manner consistent with a relapse prevention model. One year follow-up found that participants in both groups improved significantly from pretreatment to the end of treatment, and that the gains were maintained over follow-up. The two groups did not differ significantly from one another at any point in time. Possible explanations for the results include that (a) the baseline level of social support in this population may have created a ceiling effect, and (b) the prognosis for this population may be so positive that it is difficult to demonstrate significant enhancement of outcomes.