The type A/type B distinction. Subtyping alcoholics according to indicators of vulnerability and severity

Multidimensional typologies of alcoholics are reviewed, including Cloninger's neurobiological learning model, Morey and Skinner's hybrid model, and Zucker's developmental model. The more recent Type A/Type B typology proposed by Babor and colleagues is reviewed in more depth, as is a previous replication and extension by Litt and colleagues. Both the original study and the replication indicate this typology is a useful tool in classifying alcoholic inpatients into two groups and in matching alcoholics to the most suitable treatment. The present study replicates the typology using outpatient samples of male alcoholics. The resulting two clusters are very similar to those identified by the two earlier studies. As expected, the relative proportion of Type A alcoholics is higher in the outpatient samples than in the previously studied inpatient samples. Preliminary analysis of the typology's clinical efficiency suggests that the variables used to classify subjects might be appreciably reduced, thus effecting a considerable time savings in assessment. A discriminant function analysis indicates that using only 5 of the original 16 clustering variables results in a correct classification rate of almost 95%. Future research directions are addressed.