Alcohol craving predicts drinking during treatment: an analysis of three assessment instruments.

OBJECTIVE: The purpose of this investigation was to examine the utility of thee craving instruments to predict drinking during treatment. The three assessments used were the Penn Alcohol Craving Scale (PACS), the Alcohol Urge Questionnaire (AUQ) and Items 1-6 of the Obsessive subscale (OBS) of the Obsessive Compulsive Drinking Scale (OCDS). METHOD: The three instruments were administered during the course of a 9-month, double-blind, placebo-controlled trial of 100 mg/day of naltrexone, and a manual-based psychosocial intervention using the BRENDA manual conducted at the University of Pennsylvania's Treatment Research Center. Participants (133 men and 50 women at the initiation of the study) used these instruments to self-report craving on a weekly or biweekly basis. The weekly number of drinks was reported using the Timeline Followback interview. The data were analyzed with generalized estimating equations using craving scores at 1 week as the independent variable and number of drinks in the subsequent treatment week as the dependent variable. RESULTS: Each of the three scales predicted drinking during the subsequent treatment week. The PACS was the strongest predictor followed closely by the OBS and then the AUQ. Most important, craving as measured by the three scales was a stronger predictor of subsequent drinking than was drinking during the prior treatment week. CONCLUSIONS: Craving assessment provides a useful means of predicting drinking during treatment. Such information would be helpful in designing clinical trials and for many treatment modalities.