Preference for ethanol and diazepam in light and moderate social drinkers: a within-subjects study.

Preference for ethanol (versus placebo) and diazepam (versus placebo) was assessed in light and moderate social drinkers. The study was designed to investigate the relationship of habitual alcohol use to the subjective and behavioral effects of the two drugs. A secondary purpose of the study was to investigate relationships within subjects in their responses to ethanol and diazepam. Light drinkers (n = 13) who consumed one to five drinks per week and moderate drinkers (n = 14) who consumed seven or more drinks per week participated in two seven-session choice experiments, one assessing preference for 0.5 g/kg ethanol versus placebo and the other assessing preference for 20 mg diazepam versus placebo. Drugs were administered double-blind and double-dummy, and the order of participation in the two experiments was counterbalanced. Sessions were conducted during the evenings in a comfortable laboratory environment. The primary dependent measure was the number of times each subject chose the drug (i.e., ethanol or diazepam) over placebo. Subjective and objective measures of the drugs' effects were obtained using standardized self-report questionnaires and psychomotor tests. Whereas both groups preferred the ethanol over placebo (i.e., 63% and 83% ethanol choice for light and moderate drinkers, respectively), only the moderate drinkers preferred the diazepam over placebo (i.e., 40% and 73% diazepam choice for light and moderate drinkers, respectively). Subjective responses to the drugs were generally similar across the groups, although on some measures the light drinkers reported more marked drug effects. The number of times each subject chose ethanol was positively correlated with the number of times he or she chose diazepam (r = 0.57), and on subjective measures, responses to ethanol and diazepam were positively correlated. Thus, subjective and behavioral responses to diazepam and ethanol were related to habitual alcohol consumption, and most notably, moderate drinkers were more likely than light drinkers to prefer diazepam over a placebo.