Positive Thoughts and Actions: Outcomes, Mediators, and Moderators of a School-Based Depression Prevention Program

Introduction: Three recent meta-analyses show depression prevention programs are efficacious at post-intervention (Cohen's d range from .16 to .30), but effect sizes diminish over time. The impact of such programs on preventing depression diagnosis is mixed. Researchers recognize the potential for moderator and mediator analyses to inform theory refinement and intervention development. This project evaluates a school-based depression prevention program, Positive Thoughts and Actions (PTA) by examining its long-term impact on depressive symptoms, depression diagnosis, moderating factors and mediating mechanisms.

Methods: A randomized clinical trial was conducted with 120 adolescents (73 girls and 47 boys; age 12-14 years; 66.7% White, 9.2% Hispanic, 19.2% Asian or Pacific Islander; 6.7% Black, and 8.3% Multiracial). Youths were drawn from a school-based population with elevated depressive symptoms, and were randomized to the PTA intervention or an Individual Support Program (ISP), an active control group. Measures of depressive symptom and diagnosis, parent-child communication, health behavior, and school adjustment were obtained from parents and/or youth at baseline, post-intervention, and 6- and 12-month follow-up. Moderators tested included gender, baseline depressive symptoms, and ethnic minority status. We tested parent-child communication, health behavior, and school adjustment as mediators using MacKinnon et al.’s (1995, 2002) approach.

Results: Results from multilevel models indicated that PTA youth showed a reduction in depression symptoms post-intervention relative to the ISP group that persisted until 12-month follow-up, d = .36, d =.24, and d = .21 at post-intervention, 6-month follow-up, and 12 month follow-up, respectively. Survival analysis indicated that onset of new depressive episodes did not differ based on condition (21% ISP; 17% PTA). Gender and baseline symptoms did not moderate PTA impact. At post-intervention, all students showed a similar level of reduction in their depressive symptoms regardless of ethnicity. At 12-month follow-up, however, intervention gains were lost for ethnic minority participants but were maintained for non-minority students. Although PTA improved health behavior and attitudes toward school, there was no evidence that these variables mediated the impact of PTA on depressive symptoms. 

Conclusions: Results of this study highlight the potential of school-based depression prevention programs to have sustained impact. More work is needed to reduce rates of depression diagnosis, maintain improvement for ethnic minority youth, and identify mediating mechanisms.