The MacAndrew Alcoholism Scale as a general measure of substance misuse.

Abstract

The MacAndrew Alcoholism Scale scores of 3 groups of men, self-identified as substance misusers [65 alcoholics (mean age 42), 48 heroin addicts (mean age 25) and 52 polydrug users (mean age 25)] were compared with those of matched control groups of psychiatric patients. Of the 165 controls, 64 were later found to have a history of substance misuse. Each of the 3 groups of substance misusers obtained higher mean scores than did the matched controls with no history of substance misuse; alcoholics and polydrug users also received higher scores than did controls with a history of substance misuse. The mean scores of the alcoholics, heroin addicts and polydrug users were 28.3, 28.6 and 29.3; their respective controls with a history of substance misuse scored 26.0, 26.7 and 25.4; controls with no history of substance misuse scored 23.0, 23.7 and 22.0. Control patients with a history of substance misuse obtained higher scores than did controls with no such history. There were no significant differences between the 3 groups of substance misusers. The generally accepted cutting score of > 23 on the MacAndrew scale correctly classified 85% of the self-identified substance misusers and 75% of the controls with a history of substance misuse; it misidentified 44% of the controls without a documented history of substance misuse. It was the only cutting score that resulted in an improvement over base rate prediction. Discussion of the factors which may relate to misclassification was presented which include the hypothesis that the MacAndrew Alcoholism Scale may identify a potential for substance misuse not yet manifest in overt behavior.