BACKGROUND
A large body of evidence highlights the important role of parental behavior in reducing the risk of, and increasing the protective factors for, adolescent internalizing disorders. The possible benefits offered by preventive parenting programs may be undermined by low rates of engagement by parents. Online interventions have the potential to mitigate the common barriers to participation and engagement in face-to-face preventive parenting programs. However, there is a surprising lack of studies that report on the relationship between intervention engagement and improvement in target outcomes.
OBJECTIVE
This study evaluated the predictive power of several measures of engagement in a Web-based intervention aimed at increasing parental protective factors and reducing risk factors for adolescent depression and anxiety. We aimed to ascertain which measures of program engagement best predict 3-month post-intervention and 12-month follow-up scores on preventive parenting, parental self-efficacy (PSE), and adolescent depressive and anxiety symptoms.
METHODS
Our sample comprised 176 parents who received the Partners in Parenting (‘PiP’) intervention and their adolescents. Engagement was measured through multiple measures derived from server logs on a web database. Potential predictors included total modules completed, percentage of intended modules completed, percentage of quiz questions answered correctly, and percentage of goals completed.
RESULTS
Hierarchical multiple regressions indicated that the total modules completed predicted parent rated preventive parenting at post-intervention, and a combination of the total modules completed, percentage of intended modules completed, and percentage of quiz questions answered predicted PSE scores at post-intervention. At 12-month follow up, a combination of engagement measures predicted preventive parenting scores, and the percentage of intended modules completed predicted PSE scores. The percentage of intended modules completed predicted parent report of adolescent depressive symptoms at 12-month follow-up. None of the engagement measures predicted adolescent report of preventive parenting or their own symptoms.
CONCLUSIONS
Our findings suggest that future programs will benefit from including multiple measures of engagement and reinforce the importance of examining the longer-term effects of engagement. Our study provides evidence for the benefits of including goal-setting exercises as a persuasive feature and ensuring quiz questions are challenging. The implications of these findings for future program development, as well as our understanding of improvement in behavior and symptoms through intervention, are discussed.