A recent systematic review surveyed the literature on using digital tools for detecting depression, in which the authors highlighted significant threats to developing robust and transparent digital biomarkers of depression. These include methodological challenges and threats to reproducibility and identified leading factors involving missing or vague data reporting. While the authors recommended using an open science framework and sharing research datasets, the review did not assess the same.

The team will be conducting a review of articles within the passive sensing field, as well as assessing the access to their methods and datasets for secondary research purposes. This study will provide further exposure to the issues surrounding reproducibility and access to FAIR (Findable, Accessible, Interoperable, Reusable) data.

We hope this work will help inform the research community of the need for greater transparency in the digital mental health field for the development of valid, reliable, and generalizable digital biomarkers.

• Systematic review: Using the protocol from the recent systematic review, the team will replicate the study’s search process and collect information from articles that have used RMTs (remote monitoring technologies) in studies with individuals with depressive symptoms
• Data and analytical methods coding: Evaluate the presence of available metadata and analytical methods, as well as associated data policies/statements
• Data analysis: Assess the metadata, methods and overall accessibility of information, discuss concerns and ways forward for reproducibility of studies and the role of FAIR data and data sharing requirements in research

Digital health tools for the passive monitoring of depression: a systematic review of methods