# The Impact of Mechanisms of Action on Adherence and Outcomes in Self-Guided Digital Mental Health Interventions: Protocol for a Randomized Controlled Trial With Mediation Analysis

## TL;DR

This study explores how persuasive design in digital mental health tools affects user adherence and outcomes in parent training programs for children's behavior issues.

## Contribution

The study introduces a novel approach to evaluating therapeutic persuasiveness in unguided digital interventions through a randomized controlled trial with mediation analysis.

## Key findings

- Examines the causal impact of therapeutic persuasiveness on program completion and outcomes in digital parent training.
- Aims to identify mediational pathways linking program adherence to changes in child behavior and parenting variables.
- Monitors treatment gains over time using multiple assessment points.

## Abstract

One of the main recognizable challenges in the digital mental health interventions field is that users adhere to these interventions in their unguided forms poorly. Studies have shown that a persuasive system design focused on encouraging users to make positive behavior changes in their lives can increase user engagement and a program’s efficacy. This design approach can be referred to as therapeutic persuasiveness (TP) and includes a call to action, monitoring, ongoing feedback, and program adaptation based on user state. The goal of this study is to examine the causal impact of TP on program completion and outcomes in unguided digital mental interventions. We aim to examine these questions in digital parent training programs (DPTs) aimed at treating children’s behavior problems.

This study aims to (1) examine the impact of TP quality on usage, reduction in child behavior problems, and improvement in parenting variables; (2) examine the maintenance of treatment gains over a follow-up period; and (3) examine mediational pathways, including whether adherence to the program (measured by module completion rates) mediates reported changes.

A randomized controlled trial will be conducted to compare 2 interventions that use the same evidence-based components of established DPTs, but that differ in terms of the quality of TP (standard: DPT-STD; enhanced TP: DPT-TP). We will recruit parents from 160 families with children aged 3-7 years with behavior problems who will be randomized into one of the 2 intervention arms. We will measure child behavior problems and related parenting variables at 5 time points: before (T1), during (T2 and T3), and after the intervention (T4 and T5). Program usage will be passively collected.

The study was funded in October 2023, and enrollment began in September 2024. As of the end of December 2024, 100 participants were enrolled in the study. Analyses are expected to be completed by September 2027.

Identifying conceptual scientific theories that draw a link between active ingredients embedded within a digital intervention function and their outcomes is crucial in advancing our understanding of what influences usage and outcomes.

ClinicalTrials.gov NCT06514326; https://clinicaltrials.gov/study/NCT06514326

DERR1-10.2196/71238