leads to ambivalent attitudes toward underage drinking and to easy opportunities for young people to drink, it is impossible as a practical matter to drive underage drinking to zero. Increasing the rate of abstention cannot be the sole measure of effectiveness.
Thus, it is necessary to develop different standards of effectiveness. In this light, it is important to recognize that some types of underage drinking are especially likely to be associated with harmful consequences, given the age of the drinkers, the characteristics of the drinking, and the contexts in which it occurs. Accordingly, effectiveness can be sensibly measured by reductions in these bad consequences, or in the intensity and dangerousness of underage drinking.
The committee has identified five goals that are pertinent to evaluating the effectiveness of a comprehensive strategy for preventing and reducing underage drinking.
delaying onset (e.g., increasing the average age of first use or of first episode of heavy use);
reducing the prevalence of (current) alcohol use;
increasing the proportion of youths who are current abstainers and intend to continue to abstain until they meet the legal drinking age;
reducing the intensity (frequency and quantity) of drinking (e.g., heavy drinking); and
reducing the harmful consequences of alcohol use.
Delaying onset (meaning delaying the first episode of drinking, however measured) is an important outcome goal because of the documented relationship between early onset and adverse consequences, and because the average age of onset has been falling in recent years (see Chapter 2). Rates of prevalence (of use) and abstention are typically regarded as reciprocals of one another; however, in the present context, the committee believes that reducing prevalence and increasing abstention should be regarded as distinct objectives. In most surveys, prevalence of “current use” is operationalized as use within the last 30 days. As so measured, prevalence is not the reciprocal of abstention because individuals who are not abstaining and have no intention of doing so in the future may not have used alcohol within the last 30 days. This situation is particularly pertinent to underage drinking because many nonabstaining youths may not be current users (as measured by 30-day prevalence). As discussed in Chapter 2, young people who drink tend to drink heavily. One of the guiding assumptions of this report is that the most plausible goal for teenagers is to prevent or