The Potential for Digital Phenotyping in Understanding Mindfulness App Engagement Patterns: A Pilot Study

Area of Science:

• Digital mental health
• Behavioral science
• Human-computer interaction

Background:

• Low engagement is a key challenge for digital mental health interventions.
• Mindfulness apps are increasingly popular, necessitating research into engagement drivers.
• Digital phenotyping offers a novel method to study real-time user behavior.

Purpose of the Study:

• To investigate factors influencing engagement with mindfulness-based mental health apps.
• To explore the relationship between passive digital data (sleep, screen time) and app engagement.
• To examine demographic associations with app engagement and analyze high-engagement case studies.

Main Methods:

• Secondary analysis of data from two four-week studies involving college students (n=169).
• Utilized digital phenotyping to collect passive data (sleep duration, home time, screen duration).
• Measured app engagement through activity completion and frequency of use; performed cluster analysis.

Main Results:

• Demographics (gender, race/ethnicity, age) did not significantly predict mindfulness app engagement.
• Sleep duration and screen duration were significant predictors of weekly app engagement metrics.
• A significant interaction was found between screen duration, activity completion, and notification receipt.

Conclusions:

• Predicting mindfulness app engagement is complex, but digital phenotyping shows promise.
• Passive data analysis can reveal patterns related to engagement at individual and population levels.
• Further investigation into digital phenotyping signals may enhance the impact of widely available mindfulness apps.