Effects of Universal Preventive Intervention Delivered during Early Adolescence on Substance Initiation, High School Engagement, School Success, and Young Adult Well-Being: A Replication and Extension Study

Introduction: An earlier study examining whether delayed substance initiation during adolescence, resulting from universal preventive interventions delivered in middle school, contributed to increased school engagement in early high school, along with academic success in 12th grade. We extended this study to include indirect effects on young adult well-being at age 27. This extension was suggested by a current theoretical model of influences on school engagement in association with academic success and well-being. Intervention effects were predicted for (a) ninth grade adolescent substance initiation (ASI); (b) ninth grade school engagement (SE–assessing cognitive, affective, and behavioral aspects of engagement); (c) academic success (self- and parent-reported grades); and (d) age 27 well-being (WB).

Method: Data were collected from parents and youth from 36 Midwestern schools that were randomly assigned to experimental (N=223 and control conditions (N=218). Participants in the present study received the school-based LifeSkills Training program and also were offered the family-focused Strengthening Families Program: For Parents and Youth 10-14 (LST+SFP). Data were collected via written questionnaires from the 7^th through the 12^th grades and through telephone interviews and written questionnaires at age 27. The preliminary mediational SEM model was tested with Mplus 7.31 using the MLR estimator and FIML. Control variables included baseline measures of ASI, SE, Grades, intact biological families, and gender. To test for risk-related moderation of intervention effects, adolescents were classified as either higher- or lower-risk, based on baseline levels of gateway substance initiation (approximately 20% were classified as higher-risk). Analyses examined risk moderation with two-way interactions (Intervention Condition X Risk; CXR). Contrast coding (1 and -1) for the risk (R) and the intervention (C) variables created orthogonal terms to facilitate interpretation of main and interaction effects.

Results: The measurement model fit the data well and all factor loadings were significant. The hypothesized mediational model also fit the data well: c² (N=441; df =171) = 317.03; p < .001; CFI =.95; RMSEA=.044; and SRMR=.053. All structural paths were significant and significant indirect effects were found for C, R, and CXR to: WB through ASI and SE; Grades through ASI and SE; and SE through ASI.

Conclusions: This study supports the potential for universal preventive interventions delivered in early adolescence to reduce substance initiation and increase school engagement with subsequent positive impacts on academic success in high school and positive well-being in young adulthood, especially for those at higher risk.