5 - (IOP 14) Interindividual Differences in Digital Phenotypes of Major Depressive Disorder Based on Smartphone and Wearable Sensor Data

Saturday, November 22, 2025

4:00 PM - 4:15 PM CST

Location: Imperial 9, Level 4

Keywords: Depression, Measurement, Technology / Mobile Health
Recommended Readings: Fried, E. I., & Nesse, R. M. (2015). Depression is not a consistent syndrome: An investigation of unique symptom patterns in the STAR*D study. Journal of Affective Disorders, 172, 96–102. https://doi.org/10.1016/j.jad.2014.10.010, Oetzmann, C., Cummins, N., Lamers, F., Matcham, F., Siddi, S., White, K. M., Haro, J. M., Vairavan, S., Penninx, B. W. J. H., Narayan, V. A., Hotopf, M., & Carr, E. (2025). Identifying depression subtypes and investigating their consistency and transitions in a 1-year cohort analysis. PLOS ONE, 20(1), e0314604., , ,

author(s)

Elizabeth W. Lampe, M.S. (she/her/hers)

Doctoral Candidate
Center for Technology & Behavioral Health, Dartmouth College
Lebanon, NH, United States

Co-Author(s)

Amanda C. Collins, Ph.D. (she/her/hers)

Post-doctoral Fellow
Center for Technology and Behavioral Health, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States; Department of Biomedical Data Science, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States
Hanover, NH, United States
AL

Ahhyun Lee, B.A.

Research Coordinator
Center for Technology and Behavioral Health, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States; Department of Biomedical Data Science, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States
Hanover, NH, United States
NE

Nicholas Enbar-Salo, None

Research Assistant
Center for Technology and Behavioral Health, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States; Department of Biomedical Data Science, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States
Hanover, NH, United States
TG

Tess Z. Griffin, Ph.D.

Research Coordinator
Center for Technology and Behavioral Health, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States
Hanover, NH, United States
AP

Arvind Pillai, M.S.

Graduate Student
Department of Computer Science, Dartmouth College, Hanover, NH, United States
Hanover, NH, United States
Michael V. Heinz, M.D.

Assistant Professor of Psychiatry
Center for Technology and Behavioral Health, Geisel School of Medicine, Dartmouth College, Lebanon, NH, United States; Department of Psychiatry, Geisel School of Medicine, Dartmouth College, Hanover, NH, United States
Hanover, NH, United States
DL

Damien Lekkas, PhD (he/him/his)

Research Assistant Professor
Northwestern University
Lebanon, NH, United States
MN

Matthew D. Nemesure, Ph.D.

Senior Data Scientist
Digital Data Design Institute, Harvard Business School, Boston, MA, United States
Boston, MA, United States
DM

Daniel M. Mackin, Ph.D.

Assistant Professor
Dartmouth-Hitchcock Medical Center and Dartmouth College
Lebanon, NH, United States
SN

Subigya Nepal, Ph.D.

Post-doctoral Fellow
Stanford Institute for Human-Centered-AI, Stanford University, Palo Alto, CA, United States
Stanford, CA, United States
AC

Andrew Campbell, PhD (he/him/his)

Professor
Dartmouth College
Lebanon, NH, United States
Nicholas C. Jacobson, Ph.D.

Associate Professor
Dartmouth College
Lebanon, NH, United States

Major depressive disorder (MDD) is characterized by high levels of heterogeneity in symptom presentation across individuals. While previous research has identified heterogeneous patterns of MDD subtypes, existing studies have largely relied on self-report assessment to create MDD symptom phenotypes which is subject to important biases and limits clinical utility. This is an especially important limitation given that the symptoms which seem to differentiate phenotypes most robustly involve the intersection between biology and behavior (e.g., insomnia, somatic symptoms). Passive sensors, which can continuously collect data across a range of biological and behavioral phenomena without requiring active participant input, offer three key advantages over traditional self-report methods for phenotyping MDD symptoms: 1) increased data availability and validity, 2) decreased patient burden that in turn offers high potential for scalability, and 3) improved understanding of variability over time. Few studies to our knowledge have leveraged objective data from smartphones and wearable devices to phenotype MDD symptoms.

Thus, we identified latent profiles based on digital biomarkers of depression collected from smartphones and Garmin smartwatches among 297 individuals with MDD (Mage = 40.07, SDage = 11.48, 83.84% female, 78.79% White, 11.78% Hispanic). Digital biomarkers included sleep duration and efficiency, daily step count, daily screen time, digital and in-person social engagement (i.e., outgoing calls and texts, phone detected conversation duration), and heart rate variability. An exploratory aim examined whether identified profiles were associated with MDD symptom severity and global assessment of social and occupational functioning assessed via SCID-5 interview.

A two-profile solution demonstrated best fit with the data: Profile 1 (85.7% of the sample) showed moderate and consistent scores across all digital biomarkers, while Profile 2 (14.3%) showed evidence of deficient sleep, anhedonia characterized by chronically low heart rate variability, and low social engagement across both digital and in-person contexts as compared to Profile 1. While the two profiles did not significantly differ on MDD symptom severity (est = 0.322, S.E. = 0.767, p = .444), Profile 2 demonstrated significantly lower social and occupational functioning compared to Profile 1 (est = -5.309, S.E. = 2.321, p = .023).

Taken together, our findings provide new insights into digital phenotyping for MDD. To our knowledge, this is the first study to show that solely using passive sensing can detect differences in depressive symptomatology. Passive sensor indices of sleep dysregulation, anhedonia, and low social engagement across digital and in-person contexts emerged as key differentiators, closely linked to social and occupational impairment. Future research should incorporate additional digital biomarkers to refine the identification of digital phenotypes of MDD and validate these profiles against other clinical severity metrics in larger, more diverse samples. Finally, future studies should evaluate the utility of digital phenotypes of MDD symptoms for predicting treatment-seeking behavior and treatment response.

Learning Objectives:

At the end of this session, the learner will be able to:

Discuss heterogeneity in depressive symptom phenotypes and the implications for passive sensing to differentiate these phenotypes.