Assessing Sustained Effects of Communities That Care on Youth Protective Factors in Grade 10

Introduction: Communities That Care (CTC) is a universal prevention strategy that uses a science-based approach to promote healthy youth development. Using data from a panel of youth followed since grade 5 in a community-randomized trial of CTC in 24 communities across 7 US states, we previously found significantly higher levels of overall protection among youth in CTC communities compared to those in control communities when they were in grade 8. The present study examined the sustained effect of CTC on youth protective factors, one year after study-provided funding, training, and technical assistance ended. Although CTC is not expected to differentially affect certain subgroups, some universal prevention programs have shown effects for specific subgroups. In this study, we examined whether CTC had a differential effect by gender and risk status at baseline.

Methods:The trial surveyed annually from Grade 5 through Grade 10 a longitudinal panel of 4,407 youths. To account for the nesting of the data, hierarchical linear models were estimated to compare Grade 10 mean levels of 15 protective factors in community, school, family, peer, and individual domains in CTC and control communities. Interaction analyses tested for moderation of the CTC effect on protective factors by gender and baseline risk. We defined high risk as individuals who initiated substance use or delinquency by grade 5 and individuals who had high levels of cumulative risk at baseline. Global test statistics (GTS) were calculated to examine effects on protection overall and by domain in each subgroup.

Results: Findings suggested that the effect of CTC on increasing youth protective factors was sustained beyond the implementation phase for males but not for females. There was little evidence to suggest that CTC had a differential effect by risk groups. Males in CTC communities compared to males in control communities reported significantly higher levels of protective factors across all domains (GTS t = 3.053, p = 0.006) as well as in the family domain specifically (GTS t = 2.356, p = 0.027). Within specific domains, males in CTC communities reported significantly higher levels of community opportunities (p = 0.038), family attachment (p = 0.050), and substance use refusal skills (p = 0.031) compared to the males in control communities. Gender interactions were statistically significant only for refusal skills (p = 0.011). Limited power to detect meaningful interactions in this study among 12 community pairs may explain the lack of statistically significant gender interactions. 

Conclusions: Finding differential effects of CTC on protective factors for males and females highlights the potential need for contextual adaption of this universal community-based prevention program.