REDUCING RISK TAKING IN ADOLESCENCE: EFFECTIVENESS OF A GIST-BASED CURRICULUM

The purpose of this study is to evaluate effectiveness of a new approach to reducing risk-taking behaviors and behavioral intentions in adolescents using a randomized control group design with three arms. One intervention was designed, and has been shown, to reduce sexual risk-taking behavior in adolescents (standard intervention). The second was an enhanced version of this intervention, based on fuzzy-trace theory, designed to increase efficacy by emphasizing the gist of the messages in the curriculum and increasing automatic retrieval of gist principles; the third intervention was a control group with comparable contact time.

Participants were 734 students aged 14-18 years (M = 15.5, SD = 1) from high schools in Arizona, Texas, and New York; 45% were Caucasian, 17% Hispanic, 27% African-American, and 11% other. Females were 58% and 37% reported having had intercourse. A variety of attitudinal and behavioral measures were obtained, including responses to a series of 5-item scales measuring intentions to have sex (ranging from very unlikely to very likely; α = .91) which was used as a dependent variable, and included such items as “Do you think you will have sex (or have sex again) before you turn 20?” A linear mixed-modeling approach was used, with random intercepts and slopes incorporated into the models to account for the correlated structure of longitudinal data. Other effects modeled were: score on the dependent variable at pretest, the intervention the participant was assigned to, the linear trend of time from post test to 12 month follow-up, and time-by-intervention and time-by-pretest score interactions. Variations on this model (e.g., with nonlinear time trends and alternative covariance structures) were compared with likelihood ratio tests.

Across time, intentions showed a positive linear trend. As predicted, adolescents assigned to the gist-enhanced intervention showed a significant decrease in intentions from pre to posttest when compared to the standard risk-reduction intervention and to the control group. Further, this difference did not fade across time; indeed initial differences were maintained over 12 months of follow-up. These results support fuzzy-trace theory in that increasing gist-based processing (enhanced intervention) had a stronger impact on reducing risky behavior than did facts and details (verbatim-based processing, standard intervention). Further, this impact was more durable and showed no fade-out effects, a criticism levied against current interventions.