JMIR Publications

ABSTRACT

Background:

Digital health interventions based on self-management strategies aim to empower users’ self-reliance by utilizing self-monitoring, self-assessment and sensor-based output. The existing variety of digital devices utilizes a wide range of data sources and sensors to collect and monitor users’ output while little comparative data on parameter reliability and utility is available.

Objective:

This review aims to address the existing methodological and knowledge gap in understanding the efficient common parameters used among digital health interventions for depression that allow precise monitoring and prediction of the course of depression across different modes of digital intervention delivery.

Methods:

Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews digital databases including PubMed, Embase, Cochrane Library and Web of Science Core Collection were scoped for literature ranging from 2021 to 2024. A five-stage framework by Arksey and O’Malley (2005) was implemented to ensure systematic scoping of the literature. The quality of the retrieved studies was assessed using the Downs and Black Instrument and the Mixed Methods Appraisal Tool.

Results:

The overall five interdependent categories were defined including 1) Physical activity and Location, 2) Behavioural patterns, 3) Physiological data, 4) Sleep, and 5) Sociability and Self-reported assessments to best describe common assessment parameters across the literature. Eleven common clinical measures and self-report assessments were distinguished across defined categories as assessment combined with digital phenotyping methodology.

Conclusions:

Synthesis of result sections of the included studies indicated that predicting depressive symptoms by combining clinical assessment and digital phenotyping is a promising approach for further improvement of digital interventions. The overall strongest associations were found in combined approaches using parameters across categories combining sensor data and self-report assessment.