JMIR Publications

ABSTRACT

Background:

The COVID-19 pandemic has intensified mental health issues globally, highlighting the urgent need for remote mental health monitoring. Digital phenotyping using smart devices has emerged as a promising approach, but it remains unclear which features are essential for predicting depression and anxiety.

Objective:

This systematic review aimed to identify the types of features collected through smartphones, Actiwatch devices, smartbands, and smartwatches, and to determine which features are essential for mental health monitoring.

Methods:

A systematic review was conducted following the PRISMA 2020 guidelines. Searches were performed across Web of Science, PubMed, and Scopus on February 5, 2025. Inclusion criteria comprised quantitative studies involving adults (≥19 years) using smart devices to predict depression or anxiety based on passive data collection. Studies focusing solely on smartphones or qualitative designs were excluded. Risk of bias was assessed using the Mixed Methods Appraisal Tool (MMAT) and Quality Criteria Checklist (QCC). The results were synthesized descriptively.

Results:

From 1,382 records, 22 studies were included. These studies were conducted in various countries and involved diverse populations, from clinical patients to community samples. The most frequently utilized features were derived from accelerometer (ACC), such as step counts and activity levels, followed by heart rate (HR) features. Sleep-related features were also important, especially in studies using smartbands and smartwatches. However, features such as peripheral capillary oxygen saturation (SpO₂) and blood volume pulse (BVP) were rarely used. Smartphone-derived features (e.g., call logs, phone usage) were underutilized in smartwatch studies, likely due to data access restrictions.

Conclusions:

ACC and HR-derived features are essential in digital phenotyping for mood disorder prediction. Sleep features should be emphasized more, particularly in Actiwatch-based studies. Improving data accessibility and establishing standard reporting guidelines are crucial for advancing this field. Even with the various strengths of this study, variability in feature definitions, differences in study designs, and a lack of standardized reporting hindered direct comparisons across studies, making a meta-analysis infeasible. Clinical Trial: This review was registered in the Open Science Framework.