This study explores whether giving families a choice of family-based prevention programs to prevent adolescent alcohol use will make a difference in program recruitment, retention, completion, as well as adolescent outcomes. Half of the families are assigned to a traditional random control trial condition and half are assigned to a choice condition. Further, this effectiveness study is being implemented by Kaiser Permanente Health Care system, and explores the issues of implementing such programs within such settings.

This study examines the efficacy and effectiveness of two theory based, universal family prevention programs that have shown efficacy for reducing adolescent alcohol and other drug use: Family Matters (FM) (Bauman, 1996) and (Iowa) Strengthening Families Program (Spoth, 1999) when implemented with families randomly assigned to one of two different conditions: a family "choice" condition (two-groups--FM or SFP) and a traditional random control trial condition (three groups--FM, SFP, and control) called the "assigned" condition. The two conditions (choice vs. assigned) are being compared for differences in: (a) adolescent outcomes related to alcohol use and related behaviors; (b) family characteristics for those who participate; (c) family recruitment, retention and completion rates; d) costs for program implementation. The sample is drawn from families (N=614) with a child age 11 or 12 currently enrolled in Kaiser Permanente Health Care Plan (KP), in one of four medical centers in Northern California. An initial face-to-face interview was conducted (separately) with parent and adolescent prior to program delivery, and two follow-up interviews (12 and 24 months later) are being conducted over the telephone with parent and adolescent (separately). The health care system represents an important new mode for delivering adolescent alcohol use prevention programs to families. The specific aims of this project will provide a real world test of implementation issues and feasibility. Finally, the choice condition represents an innovative, realistic condition under which families make participation decisions outside of the traditional experimental study design protocols. Choosing a treatment based on personal preference may increase patient's sense of autonomy and self-efficacy for behavior change thereby improving outcomes (Williams, 1998; Clarke, 1999). Social cognitive theory and principles of self-regulation provide a connection between these psychological constructs and behavioral choices/health outcomes (Bandura, 1986; Clark & Zimmerman, 1990).

Inclusion Criteria:

• Families who were members of one of four Kaiser Permanente medical centers in Northern California (Oakland, Vallejo, San Francisco, and Walnut Creek) at the time the sample was drawn who had an 11-12 year old child.

Exclusion Criteria:

• Child in alcohol or drug treatment,
• Did not speak English fluently,
• parents or adolescents with mental disorders which would hamper their ability to participate effectively or would cause them to be disruptive to the group process or for whom the comprehension of the reading material in Family Matters would be a problem.

Families were assigned either to a group with no choice or with choice but were aware that they would be randomly assigned but knew which option they received.