Process and Effect Evaluation of a School-Based Prevention Program for Childhood Anxiety and Depression

Introduction: Anxiety is highly prevalent in children, and associated with anxiety and depressive disorders in later life. Prevention is therefore of utmost importance. In the Amsterdam, the Netherlands, we implemented FRIENDS for Life, a school-based indicated prevention program against anxiety and depression in children aged 9-12. In several other countries, including the USA, the program has been found effective. The study aimed to evaluate whether the program is also effective in the Netherlands, and test potential moderating effects of child characteristics and implementation quality.

Methods: The design of the study was a controlled trial. In the intervention group, children were screened for symptoms of anxiety and depression (T1), and children who met the inclusion criteria and whose parents consented participated in the program. Post-intervention (T2, after 3 months), and at 6 and 12 months (T3 and T4) after the program was finished follow-ups were performed. In control schools, no selection procedure took place: all children in class participated in the study at the four times of assessment. The program was delivered by trained prevention professionals.

Results: In total, at T1 screening data of 3636 children were collected, 331 of which participated in the program. The screening covered gender, age, socio-economic status, ethnicity, previous treatment for anxiety or depression, severity of symptoms of anxiety and depression as reported by the child, the teacher report, and all classmates. In the intervention group also data on treatment integrity using classroom based observations and prevention professionals’ reports tapping adherence to protocol, quality of delivery, participant responsiveness, and exposure. In addition, participants’ and parents’ satisfaction with the program were assessed.

Conclusions: This poster will present participants’ and parents’ evaluation of the program, and relate them to T1 child characteristics and parts of implementation integrity assessments. Satisfaction is expected to be positively related to implementation quality, and with favorable child characteristics. Because using a sufficiently large sample the current study aims to answer questions not addressed well in other studies on the implementation of FRIENDS for Life, we refrained from formulating more specific a priori hypotheses. Implications for practice from these analyses will be forwarded.