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

Abstract

The use of digital tools to measure physiological and behavioural variables of potential relevance to mental health is a growing field sitting at the intersection between computer science, engineering, and clinical science. We summarised the literature on remote measuring technologies, mapping methodological challenges and threats to reproducibility, and identified leading digital signals for depression. Medical and computer science databases were searched between January 2007 and November 2019. Published studies linking depression and objective behavioural data obtained from smartphone and wearable device sensors in adults with unipolar depression and healthy subjects were included. A descriptive approach was taken to synthesise study methodologies. We included 51 studies and found threats to reproducibility and transparency arising from failure to provide comprehensive descriptions of recruitment strategies, sample information, feature construction and the determination and handling of missing data. The literature is characterised by small sample sizes, short follow-up duration and great variability in the quality of reporting, limiting the interpretability of pooled results. Bivariate analyses show consistency in statistically significant associations between depression and digital features from sleep, physical activity, location, and phone use data. Machine learning models found the predictive value of aggregated features. Given the pitfalls in the combined literature, these results should be taken purely as a starting point for hypothesis generation. Since this research is ultimately aimed at informing clinical practice, we recommend improvements in reporting standards including consideration of generalisability and reproducibility, such as wider diversity of samples, thorough reporting methodology and the reporting of potential bias in studies with numerous features.

Using digital health tools for the Remote Assessment of Treatment Prognosis in Depression (RAPID): a study protocol for a feasibility study

“…In the interest of open and reproducible science, we will follow basic transparency recommendations,13 47–49 including the reporting of basic demographic and clinical data, attrition and participation rates, missing data, evidence of the validity or reliability of the sensors and devices used. For each behavioural feature, a full definition and description of feature construction will be provided, with links to GitHub repositories and source code, where available.…”

“…Given the relative recency of the field, and with an eye towards clinical implementation, studies on RMTs and depression have largely comprised proof-of-concept, feasibility and acceptability studies. While studies show RMTs to be generally feasible and acceptable,12 the data predominantly come from small, non-clinical or student samples with a median follow-up time of 2 weeks 13…”

Acceptability of Personal Sensing Among People with Alcohol Use Disorder: Observational Study

“…It has also often used very small sample sizes and/or short monitoring periods [15,17,21,22]. Recent reviews of this emerging literature have highlighted gaps in reporting on participant exclusions, attrition, and compliance that are necessary to assess selection biases and feasibility of these more novel personal sensing methods [36][37][38].…”

Increasing the Value of Digital Phenotyping Through Reducing Missingness: A Retrospective Analysis

“…Except for a slight difference in accelerometer between Black and White participants, the researchers did not find significant differences in missingness across race, gender, or education background 8 , but they did report a difference in GPS data coverage between Android and iOS users. Another large study with 623 participants using the RADAR platform examining data collected from November 2017-June 2020 found that only 110 participants had >50% data across all data types collected 10 . In addition, the researchers found no link between baseline depression and data availability in their sample, which suggests digital phenotyping methods are applicable regardless of illness severity.…”