JMIR Publications

ABSTRACT

Background:

In the burgeoning area of clinical digital phenotyping research, there is a dearth of literature that details methodology, including the key challenges and dilemmas in developing and implementing a successful architecture for technological infrastructure, patient engagement, longitudinal study participation, and successful reporting and analysis of diverse passive and active digital data streams.

Objective:

This article provides a narrative rationale for our study design in the context of current evidence base and best practices with an emphasis on our initial lessons learned from the implementation challenges and successes of this digital phenotyping study.

Methods:

We describe the design and implementation approach for a digital phenotyping pilot feasibility study with an attention to synthesizing key literature and the reasoning for pragmatic adaptations in implemenating a multi-site study encompassing distinct geographic and population settings. This methodology was used to recruit patients as study paritcipants with a clinician validated diagnostic history of either unipolar depression, bipolar I disorder, bipolar II disorder, or health control in two geographically distinct healthcare systems for a longitudinal digital phenotyping study of mood disorders.

Results:

We describe the feasibility of a multi-site digital phenotyping pilot study for patients with mood disorders in terms of passively and actively collected phenotyping data quality and enrollmet of patients. Overall data quality (assessed as the amount of sensor data obtained vs expected) was high compare to related studies. Results were reported on relevant demographic features of study participants revealing recruitment properties of age (mean sub-group age ranged from 31 in the health control sub-group to 38 years old in the bipolar I disorder sub-group), sex (predominance of female participants which ranged from 64% female in the bipolar II disorder sub-group), and study subjects’ smartphone operating system (iOS versus Android) which ranged from 71% iOS in the bipolar II disorder subgroup to 91% iOS in the healthy control sub-group. We also describe implemention cnsideratons around digital phenotyping research for mood disorders and other psychiatric conditions.

Conclusions:

Digital phenotyopiung in affective disorders is feasible on both Android and Apple smartphones and resulting data quality using an open-soruce plafrom is higher than comparable studies. While the digital phentoyping data quality are independent of gender and race, the reported demographic features of study participants reveals important information on possible selection biases that may result from naturalistic research in this domain. We believe the methodology described will be readily reproducible and generalizable to other study settings and patient populations given our data on deploying at two unique sites.