Predicting and Monitoring Symptoms in Patients Diagnosed With Depression Using Smartphone Data: Observational Study
2024

Using Smartphone Data to Monitor Depression Symptoms

Sample size: 164 publication 10 minutes Evidence: moderate

Author Information

Author(s): Taiane de Azevedo Cardoso, Ruoyu Zhang, John Anjum Susan, Ali AL-Asadi, Johanna Kallio, Arsi Ikäheimonen, Luong Nguyen, Ilya Baryshnikov, Richard Darst, Roope Heikkilä, Joel Holmen, Annasofia Martikkala, Kirsi Riihimäki, Outi Saleva, Erkki Isometsä, Talayeh Aledavood

Primary Institution: Aalto University

Hypothesis

Can smartphone behavioral data accurately classify and monitor depression symptoms in patients diagnosed with depression?

Conclusion

Smartphone behavioral markers can help detect and monitor changes in depression symptoms, especially when combined with self-reported data.

Supporting Evidence

  • Smartphone data can classify depression presence with 82% accuracy.
  • 32 behavioral markers were identified as associated with changes in depressive states.
  • Participants provided data for up to 1 year, yielding 26 biweekly data points.

Takeaway

This study shows that data from smartphones can help doctors understand how depressed people are feeling and if their feelings are changing over time.

Methodology

A prospective cohort study collecting smartphone behavioral data for up to 1 year from 164 participants, including healthy controls and patients with various depressive disorders.

Potential Biases

The study may not fully account for external factors affecting behavior patterns and mood states.

Limitations

Participant dropouts and missing data increased over time, limiting the ability to capture symptom fluctuations.

Participant Demographics

Participants included 164 individuals: 31 healthy controls, 85 with major depressive disorder, 27 with comorbid borderline personality disorder, and 21 with major depressive episodes with bipolar disorder.

Statistical Information

Confidence Interval

95% CI 80%-84%

Statistical Significance

p<0.001

Digital Object Identifier (DOI)

10.2196/56874

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication