Skip to main content
ABCT 2023 Annual Convention Main banner
Final Program


Program Flipbook
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Ticketed Session
Ticketed Session
Recorded Session
Recorded Session
Back
ViewAttendees 11

Symposia

Technology

Session: Using Wearable Sensors and Smartphone Apps to Personalize Treatment: Promising Advances and Open Questions for Clinical Application

1 - (SYM 133) Passively Sensed Smartphone Data Predicts Depression and Generalized Anxiety Symptom Severity

Sunday, November 19, 2023
10:30 AM – 11:30 AM PST
Location: Chiiwack (402), Level 4

In order to optimize digital interventions, there is a need to better understand passively sensed predictors of mental health that may signal opportunities to deploy personalized content with minimal user burden. While studies show links between smartphone data and affective symptoms, we lack clarity on the specificity (e.g., to depression vs. anxiety), temporal scale, and person-specific (vs. group-level) nature of these associations. We conducted a large-scale (n=946) smartphone-based passive sensing study to identify within- and between-person digital markers of depression and anxiety symptoms over time. Participants (74.5% female; M age = 40.9) downloaded the LifeSense Passive Data Kit, which facilitated continuous passive data collection (e.g., GPS, app and device use, communication) across 16 weeks. Hierarchical linear regression models tested the within- and between-person associations of 2-week windows of passively sensed data with depression (PHQ-8) or generalized anxiety (GAD-7) severity. We used a shifting window to understand the time scale at which sensed features relate to mental health symptoms, predicting symptoms 2-3 weeks in the future (distal prediction), 1-2 weeks in the future (medial prediction), and 0-1 week in the future (proximal association). Spending more time at home relative to one's average was an early signal of PHQ-8 severity (distal β = .272, p = .007) and continued to relate to PHQ-8 at medial (β = .220, p = .025) and proximal (β = .253, p = .013) windows. In contrast, circadian movement was proximally related to (β = -.683, p = .027) but did not predict (distal β = -.534, p = .090; medial β = -.430, p = .170) PHQ-8. Distinct communication features were related to PHQ-8 and GAD-7. Findings have implications for digital health intervention development. Certain features (e.g., circadian movement) may represent correlates but not true prospective indicators of affective symptoms. Conversely, other features such as home duration may be early signals of intra-individual symptom change, indicating the potential utility of prophylactic intervention (e.g., behavioral activation) in response to person-specific increases in these signals.