Back to Blog List

Could data on cell phone use help identify depression symptoms?


Researchers are exploring how someone uses a smartphone can be used to identify and track behaviors, such as mood, fatigue, social connectedness and physical isolation. The data on phone behavior can be collected passively, as one goes about the day using a phone, requiring no action on the part of the user.


Many mobile phone apps are available to address mental health concerns. APA has developed a framework to help mental health professionals and individuals evaluate mental health apps. While most of these require user interaction, new research looks at use of data collected without user interaction.

A new study from researchers at Boston University, Bentley University, Harvard Medical School, MIT, VA Boston Healthcare System, and Cogito Corporation (an app developer) sought to test the use of a smartphone app to identify underlying mood and anxiety disorder symptoms. The researchers set out to test whether a smartphone could be used to collect, store, and analyze objective behavior data related to symptoms for depression and PTSD.

Learn More About Depression

Learn about depression, including symptoms, risk factors and treatment options.

Learn More

The ability to identify behavioral patterns can help doctors and individuals to recognize and treat symptoms early. Typically, monitoring symptoms and behavior relies on self-reporting through questionnaires or interviews. However, as the researchers note, self-reporting may not always be accurate.

The study involved 73 participants who had at least one symptom of depression or PTSD. Each participant had an initial visit, 12 weeks of gathering digital trace data through a cell phone app, and a follow-up visit. In addition to the passively collected data through a cell phone, each participant recorded an audio diary entry in the app at least once a week, talking about how they were feeling or how their days were going. The study focused on three specific depression symptoms (depressed mood most of the day, diminished interest or pleasure in all or most activities and fatigue or loss of energy) and one PTSD symptom (avoids activities, places, people).

Various types of digital trace data were collected:

  • Social – the time and de-identified descriptors of outgoing and incoming phone calls and text messages
  • Location – where the phone was physically located
  • Device interaction – record of when the device is actively being used
  • Voice cues – drawn from weekly audio diaries. Measurements relating to speaking, rate, stress pattern, loudness variations, pausing, intonation and voice quality were extracted and aggregated.
  • Activity – physical handling of the phone (via accelerometer and gyroscope)

The app makes it possible to condense millions of data points into a few important pieces of information. An important consideration for use of this type of technology is how acceptable and usable people find it. Participants were generally comfortable with sharing their data. Almost all (96 percent) completed the weekly audio diaries.

This approach is “not designed to replace clinical decision-making or diagnose a particular mental health condition, but rather to assess a number of behavioral indicators underlying mood and anxiety disorders,” the researchers note. This type of passively collected data could complement traditional methods that rely on self-reporting, such as questionnaires and diagnostic interviews.

While the research is in the early stages, it does show some potential for using data gathered passively via smartphone to help identify symptoms leading to better treatment of depression or other disorders. This continuous monitoring could not only provide information for clinicians or researchers, but could also help individuals to recognize change in behavior and warning signs. That knowledge can help them to better manage their own health.


  • Place, S, et al. Behavioral Indicators on a mobile Sensing Platform Predict Clinically Validated Psychiatric Symptoms of Mood and Anxiety Disorders. Journal of Medical Internet Research. 2017; 19(3):1-9.
  • Saebj, S. et al. Mobile Phone Sensor Correlates of Depressive Symptom Severity in Daily-Life Behavior: An Exploratory Study. Journal of Medical Internet Research. 2015; 17(7):e175.
  • Ben-Zeev, D. et al. Next-generation psychiatric assessment: Using smartphone sensors to monitor behavior and mental health. Psychiatric Rehabilitation Journal. 215; 38(3):218-226




Comments (0) Add a Comment


Add a comment

Enter the text shown in this image:*(Input is case sensitive)
* - Only comments approved by post author will be displayed here.