JMIR Publications

ABSTRACT

Background:

Mobile devices offer an emerging opportunity for research participants to contribute person-generated health data. There is little guidance, however, on how to best report findings from studies leveraging those data. Thus, there is a need to characterize current reporting practices so as to better understand the potential implications for producing reproducible findings.

Objective:

The primary objective of this scoping review was to characterize the reporting practices of publications on research that collect person-generated health data using mobile devices.

Methods:

The primary objective of this scoping review was to characterize the reporting practices of publications on research that collect person-generated health data using mobile devices.

Results:

Out of 3,602 publications screened, 100 were included in this review. We observed a rapid increase in all publications from 2016 to 2021 with the largest contribution from American (United States) authors with one exception - review article publications. Among all original research publications, very few used crowdsourcing platforms (7%, 3/45). In addition, among the original research publications that reported device ownership, the majority (75%, 21/28) reported the use of participant-owned devices for data collection (i.e., a Bring-Your-Own-Device or BYOD strategy). A significant deficiency in reporting completeness was observed for the Data and Ethics dimensions (five reporting factors missing in over half of the research publications), and reporting completeness worsened over time for data ownership and participant’s access to data after contribution.

Conclusions:

Our work depicts the reporting practices in publications that describe research involving person-generated health data collected using mobile devices. We found that very few articles reported the use of crowdsourcing platforms for data collection. The growing popularity of BYOD strategies to collect data highlights an opportunity for improved mechanisms to leverage crowdsourcing platforms to transfer data from device owners to researchers. In addition, given the substantial reporting deficiencies that exist, we found a need for consensus on best practices for reporting research with person-generated health data collected using mobile devices. Drawing from the five reporting dimensions characterized in this scoping review, we share our recommendations and justifications on nine items that require improved reporting, with the goal to improve data representativeness and quality and empower participants.