Investigators will use a Solomon 4-group design to test the effectiveness of a brief intervention to promote health behaviors for homeless youth who are 18-23 years of age. We will recruit 300 youths from Austin, TX and 300 from Columbus, OH. We will gather quantitative data on health behaviors and life satisfaction 3 and 6 months following the intervention, which will be delivered on an individual basis. Data will be analyzed using structural equation modeling and ANCOVA statistics.

The specific aims of this proposal are to:

1. determine the effectiveness of a self-reflective preventive intervention to promote responsible substance use and safe sex behaviors (prevent HIV/AIDS) in homeless youths, ages18-23 years;
2. determine whether psychological capital factors mediate the relationship between background risk factors (sexual orientation, duration of homelessness, sex abuse history, history of substance use, past sexual risk behaviors, and service utilization) and improved proximal outcomes (self-efficacy and intentions) and distal outcomes (responsible substance use, safer sex, life satisfaction);
3. determine the effect of pre-testing on intervention proximal outcomes (immediately following intervention) and distal outcomes (at an interval of 3 and 6 months following intervention); and an exploratory aim
4. to explore general and specific differences in proximal and distal outcomes of the intervention between participants in two geographically different sites (Austin, TX and Columbus, OH). We will recruit a total of 600 homeless youths from 2 drop-in centers in 2 geographic areas of the central US: Austin, TX and Columbus, OH. Using a Solomon four-group design with repeated measures, we will conduct a randomized controlled trial of a brief possible selves intervention (PSI) consisting of 6 brief individualized sessions that focus on generating images of one's possible self as feared or as best. Intervention participants will also receive weekly electronic messages to encourage them to move toward a goal of becoming their best possible self. Quantitative data will be collected before, immediately after the intervention, and at intervals of 3 and 6 months following intervention to test hypotheses about the effectiveness of the intervention on health behaviors and life satisfaction as well as the effects of pre-testing on these outcomes using structural equation modeling and ANCOVA statistics. The findings are expected to have a public health impact on further development of individualized services to promote healthy behaviors, healthy lifestyles and life satisfaction for vulnerable youths.

Inclusion Criteria:

• does not have current secure place of residence,
• receives services from a homeless street outreach center, and
• is able to read and write English

Exclusion Criteria:

• Currently enrolled in Alcoholics Anonymous or other substance treatment program

Data collectors were blinded to who received the pre-test and/or intervention.