Behaviour Analytics of Users Completing Ecological Momentary Assessments in the Form of Mental Health Scales and Mood Logs on a Smartphone App

Abstract

Behavioural data analytics and user log analysis can be useful to gain insight into how users interact with technologies. In this study, data analytics were conducted on maternal mental health data generated from the Moment Health app to address the question: What is the temporal behaviour of users when completing ecological momentary assessments (EMA) on a mental health app, with EMAs in the form of full mental health scales versus EMAs in the form of mood logs? The Health Interaction Log Data Analytics (HILDA) pipeline was used to analyse 1,461 users of the app. More users completed single mood logs EMAs (n=6,993) compared to scaled EMAs (n=2,129). Distinct temporal patterns were identified, with more users willing to log moods at 9am and 12pm as opposed to completing a scale. The most common hours for users to complete scaled EMAs are between 8pm and 10pm. The least number of mood logs and scale completions take place on Saturday. Whilst happiness is the dominant mood during day times, anxiety and sadness peak during the night at 1am and 4am respectively. The data indicates that postnatal depression decreases over time for some users (r = -0.23, p-value < 0.01). The overall finding from this work are that users prefer simple EMA approaches and that the temporal behavior of users engaging with the two forms of EMA are distinctly different.