Discriminative validity of the MacAndrew Scale in settings with a high base rate of substance abuse

The discriminative validity of the MacAndrew Alcoholism Scale (MAC) was examined with regards to race and sex in a clinic with a high base rate of substance abusers. Subjects included 168 men (17 blacks, 151 whites) and 32 women (6 blacks, 26 whites). All subjects were assigned to either a psychoactive substance disorder group (n = 172) or no psychoactive substance disorder group (n = 28) based on DSM-III-R criteria. Using the optimal cutting score that provided the best overall accuracy, chi-square analyses indicated the MAC scale significantly discriminated substance abusers among samples of blacks and white men but did not significantly discriminate substance abusers among samples of whites, men, women, or white women. When the standard MAC scale cutting score of 24 was used, the overall accuracy decreased for each sample but resulted in significant chi-square analyses for the samples of whites, men, women and white men but not for blacks or white women. Results indicated the MAC scale is not an efficacious discriminator in settings with a high base rate of substance abusers.