Cover Image


View/Hide Left Panel

reduce drinking altogether, rather than focusing on reducing drinking intensity. Accordingly, rates of prevalence and abstention are particularly important outcomes for children and teens.1

Comparing Outcomes

Assessed independently, the effectiveness of specific policies depends on the aspect of the problem they are designed to address. Some policies aim to discourage initiation by young teens or preteens; others aim to reduce the prevalence of any drinking in a high school population; and others aim to reduce the number of occasions when high school students engage in heavy drinking or when they drive after drinking. For the most part, the current policy evaluation literature does not compare the effectiveness of different policies or interventions. Instead, a given intervention is evaluated in terms of one or more particular outcomes.

Ultimately, however, a sophisticated assessment of cost-effectiveness requires a common metric for comparing the outcomes of policies that address different components of the problem of underage drinking. For example, preventive interventions for disease or injury are often evaluated in terms of such outcomes as deaths prevented, years of potential life lost before age 65, or the quality-adjusted years “saved” by the intervention. Such consequence-based assessments of effectiveness are rarely possible for underage drinking. The dots cannot now be connected in any rigorous way between an incremental reduction in the prevalence, intensity, or age-of-onset of underage drinking and any “ultimate” outcome.

The committee considered what metric would be best for comparing the value of upward shifts in the age of onset, downward shifts in current use (prevalence) of drinking among 15-year-olds, reductions in levels of heavy drinking among high school students, or reductions in the prevalence of driving after drinking among underage drinkers. It seems clear that the most important factor in identifying and ranking outcomes is the harms or negative consequences associated with particular patterns of consumption. Just as different components of the problem might need separate targeting,


Educational programs and media campaigns aimed at young adults (18 to 21) often must grapple with the reality of pervasive drinking, and they must decide whether and how to formulate a “harm reduction” message—i.e., one that says, in effect: “It’s illegal to use alcohol, and you shouldn’t do it at all, but if you do, do it responsibly …” Though such approaches might be useful for young adults, such a “harm reduction” or “responsible drinking” message is wholly inappropriate for children and young teens. Nonetheless, exploring such options was inconsistent with the committee’s charge and the committee did not consider interventions with this objective.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement