BEHAVIORAL AND COGNITIVE MEASUREMENTS PREDICT SCORES ON THE MAST - A 3-YEAR PROSPECTIVE-STUDY

This experiment examined the ability of neuropsychological and behavio ral tests of anterior brain functioning to predict changes in alcohol- related behaviors 3 years after the initial evaluation. One hundred fo ur young adult subjects, sixty-six of whom had a positive family histo ry of alcoholism, filled out the Michigan Alcoholism Screening Test (M AST), self-rated the frequency with which they consumed alcohol, and c ompleted a retrospective test of childhood hyperactivity/impulsivity/c onduct problems (Wender Behavioral Checklist). Eighty-three of the sub jects also completed a neuropsychological battery consisting of cognit ive tests of executive functioning (i.e., Trails, Categories Test, Sim ilarities) and motor tests (i.e., TPT, hand dynamometer, finger tappin g). Subjects returned to complete the alcohol-related measures during a second testing session 3 years later. Changes in MAST scores, and in alcohol consumption, between the first and second test session were c omputed, and median splits classified subjects into ''high'' versus '' low'' change groups. After factor analysis of the cognitive and behavi oral data, hierarchical logistic regression equations assessed the abi lity of the cognitive and behavioral variables, as well as the presenc e of a positive family history of alcoholism, to predict future change s in the alcohol outcome measures. Scores from the WENDER behavioral c hecklist predicted future changes in the MAST scores (p = 0.0026), wit h more impaired scores associated with higher MAST scores. Tests of ex ecutive functioning, in subjects with a positive family history of alc oholism, predicted alcohol consumption (p = 0.033). None of the other predictor variables showed any relationship to the alcohol-related out come measures.