JMIR Publications

ABSTRACT

Background:

Meditation offers well-documented health benefits, yet engagement with meditation apps is low. The SuMMed stage model used for meditation habit formation offers a framework to better understand how users engage with meditation through these apps.

Objective:

The objective of this study was to map the SuMMed stage model to meditation app usage using digital trace data. In doing so, we can identify app users who are in different stages of behavioral change in order to obtain a more nuanced perspective of both meditation habit formation and meditation app usage.

Methods:

This study is based on digital trace data from April 2021 to August 2023. The data includes 268,124 event logs from 13,865 users of the top three most popular meditation apps (Calm, Headspace, and Insight Timer). Using previously validated behavioral thresholds for meditation behavioral habit formation, we classified users into stages of behavior change and subsequently tested for differences between groups.

Results:

96% of popular meditation app users are in the Intentional Phase of behavior change and are likely experimenting with meditation apps, while 3.9% have an established habit. Additionally, more than half of meditation app users in the Intentional Phase only used a meditation app one time. However, those who returned often did so after a long period of inactivity (mean [SD], 26.1 [57.8] days).

Conclusions:

Meditation app usage is predominantly characterized by trial, casual experimentation, and sporadic engagement, rather than clear habitual usage. Standard retention metrics, such as the share of users still active at day 7 or day 30, can miss this dynamic. Thus, mapping behavioral stages provides a more nuanced understanding of app usage. This research demonstrates a methodology for applying stage models to digital health data without reliance on self-reporting. It also further reinforces the important role of experimentation in meditation app usage and habit formation.