Exploring User Needs for a Mobile Behavioral-Sensing Technology for Depression Management: Qualitative Study

Jingbo Meng, Syed Ali Hussain, David C. Mohr, Mary Czerwinski, Mi ZhangJournal of Medical Internet Research 2018 · DOI

TL;DR

This study explores the design and utility of "iSee," a mobile sensing platform aimed at improving depression management for college students. Researchers found that while both clinicians and students see value in objective behavioral data, they have differing preferences for how this data should be presented and used for feedback, highlighting a key design challenge for such technologies.

What Problem Does This Paper Address?

College students are experiencing unprecedented rates of depression, overwhelming university counseling centers. These centers face limitations in capacity and efficiency, often relying on students' self-reports, which can be inaccurate or incomplete due to forgetfulness, embarrassment, or the nature of depression itself. This gap hinders timely assessment and effective treatment. The paper proposes a mobile behavioral-sensing platform, "iSee," as a potential solution to enhance the efficiency and accessibility of mental health services by providing objective data and supporting self-management.

Methods

The study employed a qualitative approach, conducting semi-structured interviews with 9 clinicians and 12 students diagnosed with depression from a Midwestern university counseling center. Interviews, lasting 40-50 minutes, were audio-recorded and transcribed. The data underwent thematic analysis using an inductive approach, involving iterative coding, memo writing, and theme refinement. Clinician and student interviews were analyzed separately to identify distinct perspectives. Participants were presented with conceptual prototypes of the iSee platform to gather feedback on its usefulness, preferred behavioral targets, and feedback mechanisms.

Datasets

Not discussed in the paper. The study focused on qualitative data gathered through interviews, not on pre-existing datasets for training or evaluation.

Key Insights and Ideas

  • Bridging the Subjective-Objective Divide: Clinicians and students alike recognized that objective behavioral data from a mobile sensing platform could offer a crucial reality check against the often negative and distorted self-perceptions associated with depression.
  • Clinician's Need for Objective Evidence: Clinicians saw iSee as a tool to verify student adherence to assigned "homework" (e.g., physical activity goals) and to inform their decision-making regarding treatment plans.
  • Student's Desire for Self-Management Support: Students found value in the platform's potential to ease the burden of manual self-tracking and provide integrated behavioral targets that help them understand the links between their actions and their mood.
  • Divergent Views on Data Presentation: A significant finding is the difference in how clinicians and students prefer behavioral targets to be presented. Clinicians favored individual targets linked to personalized goals, while students sought integrated targets that illustrate correlations between various behaviors and their depression levels.
  • Feedback Preferences Clash: Similarly, while both groups welcomed positive reinforcement, clinicians suggested using negative data for reflection or discussion during sessions, whereas students expressed a strong preference to avoid negative feedback altogether, fearing it would exacerbate their feelings of helplessness.

Key Results

  • Clinician Perspective:
    • 7 clinicians believed behavioral targets would provide objective checks against subjective beliefs.
  • Clinicians saw the platform as a way to confirm student compliance with assigned homework.
  • 5 clinicians noted the platform could fill communication gaps during limited counseling sessions.
  • 4 clinicians suggested organizing behavioral targets around personalized goals.
  • 5 clinicians highlighted the value of positive reinforcement and automated feedback.
  • 3 clinicians proposed reflection reminders for negative data.
  • Student Perspective:
    • 9 students had previously used tracking apps but found them difficult to maintain due to effort.
  • 7 students desired integrated behavioral targets to understand correlations between behaviors and depression.
  • 4 students wanted to avoid negative feedback, preferring improvement-focused insights.
  • 6 students expressed a strong need for positive supportive messages.
  • Students raised concerns about the potential burden of reviewing data, with some preferring clinicians to take the lead.

Limitations and Future Work

  • Limitations:
    • The study sample was small and drawn from a single university, limiting generalizability.
  • The study did not include participants with severe depression, whose opinions might differ.
  • The research relied on conceptual prototypes, and actual user experience with a deployed platform may vary.
  • Future Work:
    • Further research is needed to explore how sensor data and behavioral targets can inform the design and delivery of micro-interventions.
  • Investigating how frequently and when to send encouraging messages is a key area for future study.
  • Connecting sensor data, behavioral targets, feedback, and interventions in a comprehensive system is a crucial next step.
  • Future work should also explore the potential for tailoring data review to individual student needs and preferences to enhance engagement.
  • The authors suggest future research could explore perspective-taking questions to foster a shared understanding between clinicians and students regarding data presentation and feedback.

Summary updated

Abstract

Although clinician and student participants shared views on the advantages of iSee in supporting university counseling, they had divergent opinions on the types of behavioral targets and feedback to be provided via iSee. This exploratory work gained initial insights into the design of a mobile sensing platform for depression management and informed a more conclusive research project for the future.
BibTeX
@article{meng2018exploring,
  author = {Jingbo Meng and Syed Ali Hussain and David C. Mohr and Mary Czerwinski and Mi Zhang},
  title = {{Exploring User Needs for a Mobile Behavioral-Sensing Technology for Depression Management: Qualitative Study}},
  journal = {Journal of Medical Internet Research},
  year = {2018},
  doi = {10.2196/10139},
}
Figure