sectional, and so the results cannot distinguish factors associated with alcohol use from true risk factors that precede the onset of alcohol abuse or dependence. Because early and frequent use may lead to alcohol abuse, these factors are included in this review. Almost all of the studies on psychosocial risk factors fail to control for parental alcoholism or other mental disorder. Finally, many studies do not differentiate between alcohol use and other substance use.
There is little evidence that individuals at risk differ from controls on baseline measures of personality. The lack of evidence may at this time reflect methodological limitations in the studies that have been done or a true lack of association. However, there is one major exception to the lack of conclusiveness of personality measures —antisocial behavior. Antisocial behavior during childhood has been consistently related to alcohol problems in adulthood.
Difficulty in achievement-related activity among adolescents also has proved important, with studies documenting the following problems among those who later became alcoholic: poorer school performance, less productivity in high school, greater truancy, and greater incidence of dropping out (IOM, 1989). Aggressive behavior and the combination of aggressive behavior and shyness in the first grade have been found to predict heavy alcohol use at ages 16 and 17 (Kellam, Brown, and Fleming, 1982). Another study yielded similar results, indicating that males who were judged to be shy as children were least likely to become heavy alcohol users, but those who were rated as both shy and aggressive as children were most likely to develop drinking problems (McCord, 1988a,b). The relative risk of alcoholism has been found to be higher among males in their thirties who previously used alcohol with their peers when they were younger than 14 years of age (Hagnell, Isberg, Lanke, Rorsman, and Ohman, 1986). Males who later became alcoholic also had weaker interpersonal ties, ranging from being less considerate and less accepting of dependency to having a greater likelihood of leaving home early (IOM, 1989).
Cloninger (1987) has proposed an association between severe alcohol-related difficulties (in the type II alcoholism discussed earlier) and personality characteristics of high levels of novelty seeking and low levels of survival dependence and harm avoidance. These associations are all the more intriguing because the two disorders—alcohol abuse and dependence and antisocial personality disorder—probably have separate genetics.