Adolescent alcohol abuse has now reached epidemic proportions, carrying with it a toll of lost opportunity, suffering and death. Many studies document the correlation between early alcohol use and later alcohol dependence, and a widespread belief that delaying the onset of alcohol consumption until the end of adolescence will reduce the risk of pathological drinking. An alternate interpretation is that those who drink early are those who are especially vulnerable, either through biology or circumstance. There are major implications of these conflicting explanations for treatment, for research and for public policy. Yet, to disaggregate the influence of specific biobehavioral and sociocultural variables in any appropriate cross section of the human population is exceedingly difficult. We propose here a pilot study of the feasibility of addressing this question experimentally, capitalizing upon an animal model of spontaneous alcohol abuse. C. aethiops, a non-endangered African primate, is highly homologous with man, lives in social groups, has a distinct adolescent period in its ontogeny, and contains individuals who differ from one another with respect to alcohol consumption and to behavioral traits (sociability, excitability, etc.) In an 18 month cross-sectional study, 96 male and female vervet monkeys will be housed in groups balanced by sex, level of baseline alcohol consumption and temperamental behavioral profile. The manipulated factor will be exposure to ethanol and to drinking or non-drinking role models. The principal outcome measure (9 and 18 months later) will be quantity and pattern of ethanol consumption; weekly social behavioral measures will also be collected in standard primatological fashion. Evaluation of the independent or joint effects of gender, behavioral trait status, and initial propensity to drink and of exposure to ethanol or drinking role models will utilize multivariate regression methods. In a second phase, we will evaluate the utility of specific pharmacological interventions in modifying outcome. The exploratory aspect of this approach is the extent to which it might suggest methods to improve decomposition of the relevant factors in human samples.
