Influence of School-Based and Family-Focused Preventive Interventions Delivered in 7th Grade on Young Adult Relationships

Introduction: Family of origin characteristics are known to influence early adult romantic relationships. However, few studies have focused on how preventive interventions delivered to youth and their caregivers in middle school associate with young adult romantic relationship factors such as overall relationship success and positive problem solving. We tested a model based on the DEARR model (Development of Early Adult Romantic Relationships) linking parent-report of post-test parenting behaviors (PB), when participating youth were in 7^th grade, to early adult reports (age 25) of relationship success (RS) and positive problem-solving (PS) with the partner, through the mediator parent-report of low-levels of recurring conflict (LRC) with the youth (8^th grade); intervention-control differences were assessed.

Methods: Parents and their youth, enrolled in 36 Midwestern schools, were randomly assigned to two experimental and one control condition. Parents and youth who participated in both the school-based LifeSkills Training and the family-focused Strengthening Families: For Parents and Youth 10-14 (N=155) were compared to a limited contact control condition (N=146). Data were collected via written questionnaires through the 12^th grade and by telephone interviews and written questionnaires at age 25 (study participants had to be in a dating, romantic, cohabiting or married relationship). Multiple group comparisons (LST+SFP 10-14 vs Control) of the hybrid SEM models with latent (PB, RS, and PS) and manifest variables (LRC) were conducted with MPLUS 7.0 using the MLR estimator and FIML. Control variables included pre-test measures of PB and LRC, sex and relationship status at age 25.

Results: The final model fit the data adequately: Χ² (N=301; df =196) = 248.23; p = .007; CFI & TLI =.97; RMSEA=.042. The two age 25 relationship dependent latent variables, RS and PS, were significantly associated with the mediator, target youth 8^th grade report of LRC, and LRC was associated with PB. However, multiple group comparisons (LST+SFP vs Control) revealed that only for LST+SFP were all structural paths, PB to LRC (β=.37***), LRC to PS (β=.22**) and RS (β=.19**) significant, as were both indirect effects: PB to LRC to PS =.10* and PB to LRC to RS =.07*

Conclusions: Results are consistent with hypothesized predictions from the DEARR Model regarding pathways of influence. Additionally, approximately 13 years after delivery, the influence of a family-focused and a school-based preventive intervention delivered in 7^th grade indirectly affected participants’ relationship problem-solving behaviors through change in levels of recurring conflict between parents and youth in 8^th grade.