The major objective of this proposal is to conduct a randomized controlled trial of an ultra-brief, personalized feedback intervention (a pamphlet) for problem drinkers. Subjects will be recruited via a telephone survey which will collect baseline data. The households of half of the subjects will receive the pamphlet as unaddressed ad mail shortly thereafter. Follow-up interviews will be conducted, by telephone, three and six months after the mailing of the pamphlets.

Hypothesis 1: Respondents from households who receive the pamphlet will display significantly improved drinking outcomes at the three-month and six-month follow-ups as compared to respondents from households in the no intervention control condition.

Hypothesis 2: More calls will be received on a help-line listed on the pamphlet (and advertised elsewhere) from residents of households who receive the pamphlet as compared to residents from households who do not receive the pamphlet.

Hypotheses 3 - 6 deal with mediator and moderator hypotheses, exploring the role of perceived risk, perceived drinking norms, and drinking for social reasons.

How do we help those problem drinkers who will never seek treatment? This is a challenging issue because of the large number of problem drinkers and the limited amount of resources available. Public health initiatives employing educational materials have met with little or no success. However, clinical research has developed effective brief interventions to help problem drinkers. This project will merge these two worlds, modifying a clinically developed intervention and producing it in an ultra-brief format that is suitable for use as a public health intervention. The major objective of this proposal is to conduct a randomized controlled trial of an ultra-brief, personalized feedback intervention for problem drinkers. The advantages of the personalized feedback pamphlet are that it is low cost and can be widely distributed to the population of problem drinkers who never seek treatment. Because the materials are based on some of the best of research-based interventions, such an ultra-brief normative feedback pamphlet has the potential of helping problem drinkers. An effective intervention of this type would yield significant public health benefit. 1830 problem drinkers will be recruited on a baseline population telephone survey and randomized to one of three conditions - personalized feedback pamphlet condition, control pamphlet condition (to test if it is the specific content of the pamphlet that leads to the change or just the receipt of any pamphlet) and a no intervention control condition (sent intervention pamphlet after the six-month follow-up). In the week after the baseline survey, all households in the postal code areas that contain respondents in the two pamphlet conditions will be sent their respective pamphlets. Changes in drinking will be assessed on post intervention three-month and six-month follow-ups. Drinking outcomes will be compared between experimental conditions. The primary hypothesis is that respondents from households who receive the personalized feedback pamphlet intervention will display significantly improved drinking outcomes at three and six-month follow-ups as compared to respondents from households in the no intervention control condition. Secondary hypotheses will test the impact of the intervention on help seeking, and explore the mediating or moderating role of perceived drinking norms, perceived risk and the problem drinker's social reasons for drinking.

Inclusion Criteria:

• Age 19 years or more
• Audit score of 8 or more (the AUDIT is a validated screener, developed by WHO, to identify problem drinkers in the community and health-care settings)