# Designing a Visual Analytics Tool to Support Data Analysis Tasks of Digital Mental Health Interventions: Case Study

## TL;DR

This paper introduces a visual analytics tool to help researchers analyze data from digital mental health interventions, improving understanding and effectiveness of these interventions.

## Contribution

A novel analysis task model and a visual analytics tool prototype for digital health intervention data analysis.

## Key findings

- The analysis task model integrates user characteristics, engagement, and intervention effectiveness.
- The visual analytics tool simplifies data analysis tasks and supports multidisciplinary communication.
- User study participants found the tool helpful for identifying users needing care and analyzing intervention effectiveness.

## Abstract

Digital health interventions (DHIs) are widely used to manage users’ health in everyday life through digital devices. The use of DHIs generates various data, such as records of intervention use and the status of target symptoms, providing researchers with data-driven insights for improving these interventions even after deployment. Although DHI researchers have investigated these data, existing analysis practices have been fragmented, limiting a comprehensive understanding of the data.

We proposed an analysis task model to help DHI researchers analyze observational data from a holistic perspective. This model was then used to prototype an interactive visual analytics tool. We aimed to evaluate the suitability of the model for DHI data analysis and explore task support using a visual analytics tool.

We constructed a data analysis task model using 3 key components (ie, user grouping criteria) for DHI data analysis: user characteristics, user engagement with DHIs, and the effectiveness of DHIs on target symptoms based on comparisons before and after the intervention. On the basis of this model, we designed Maum Health Analytics, a medium-fidelity prototype of an interactive visual analytics tool. Each feature of the prototype was mapped one-to-one to the analysis task described in the model. To investigate whether the proposed model adequately reflects real-world DHI analysis needs, we conducted a preliminary user study with 5 groups of researchers (N=15). Participants explored the tool through scenario-based analysis tasks using in-the-wild data collected from a mobile DHI service targeting depressive symptoms. Following the session, we conducted interviews to assess the appropriateness of the defined tasks and the usability and practical utility of the visual analytics tool.

DHI researchers responded positively to both the analysis task model and the visual analytics tool. In the interviews, participants noted that the tool supported the identification of users who needed additional care, informed content recommendations, and helped analyze intervention effectiveness in relation to user characteristics and engagement levels. They also appreciated the tool’s role in simplifying analytic tasks and supporting communication across multidisciplinary teams. Additional suggestions included improvements for continuity across tasks and more detailed engagement metrics.

We proposed an analysis task model and designed an interactive visual analytics tool to support DHI researchers. Our user study showed that the model allows a holistic investigation of DHI data by integrating key analysis components and that the prototype tool simplifies analytic tasks and enhances communication among researchers. As DHIs grow, the proposed model and tool can effectively meet the data analysis requirements of researchers and improve efficiency.