JMIR Publications

ABSTRACT

Background:

Various behavioral sensing research studies have found depressive symptoms are associated with human-smartphone interaction behaviors, including lack of diversity of unique physical locations, entropy of time spent in each location, sleep disruption, session duration, and typing speed. These behavioral measures are often tested against a total score of depressive symptoms and the recommended practice to disaggregate within- and between-person effects in longitudinal data is often neglected.

Objective:

We aimed to understand depression as a multi-dimensional process and explore the association between specific dimensions and behavioral measures computed from passively sensed human-smartphone interactions. We also aimed to highlight the non-ergodicity in psychological processes and the importance of disaggregating the within- and between-person effects in the analysis.

Methods:

Data used in this paper were collected by Mindstrong Health, a telehealth provider that focuses on individuals with Serious Mental Illness (SMI). Depressive symptoms were measured by DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure – Adult every 60 days for a year. Participants’ interactions with their smartphones were passively recorded and five behavioral measures were developed and expected to associate with depressive symptoms according to either theoretical proposition or previous empirical evidence. Multilevel modeling was used to explore the longitudinal relations between severity of depressive symptom and these behavioral measures. Furthermore, within- and between-person effects were disaggregated to accommodate the non-ergodicity commonly found in psychological processes.

Results:

This study included 982 records of DSM Level 1 depressive symptom measurements and corresponding human-smartphone interaction data from 142 participants (age from 29 to 77 years old with a mean of 55.1 and SD of 10.8; 96 of whom are female). Loss of interest in pleasurable activities was associated with app count (γ_10=-0.14, P=0.010, within-person effect). Depressed mood was associated with typing time interval (γ_05=0.88, P=0.047, within-person effect), and session duration (γ_05=-0.37, P=0.028, between-person effect).

Conclusions:

This study contributes new evidence of associations between human-smartphone interaction behaviors and the severity of depressive symptoms from a dimensional perspective, and it also highlights the importance of considering the non-ergodicity of psychological processes and analyzing the within- and between-person effects respectively.