Development of dashboards to improve data quality and completeness for clinical teams’ use in RADAR-AD

RADAR-AD technical and clinical teams co-developed dashboards to support the work of clinicians when collecting data on participants’ behavior within the study

One of the major advantages of a mHealth or Remote Monitoring Study is the availability of continuous feeds of data, in real or near-real time. In research studies, one of the many uses of this type of data is to better understand the completeness of this and the quality of the data being collected. Oftentimes, it can prove difficult to track if a person wears their device, has some technical issue within the clinical study or if they filled in their questionnaire at a specific time. Therefore, the goal of the development of these dashboards is to assist clinical teams and researchers. For instance, within the RADAR-CNS (Remote Assessment of Disease and Relapse – Central Nervous System) project, which is RADAR-AD’s predecessor, data completeness was checked manually in the storage server and individually per participant. This required technical effort can make it hard to get a full overview of the data at any given point in time.

**This dashboard shows the data completeness from two passive data sources: Passive RMT app and Fitbit. Hovering over the graph shows the values. The passive RMT record count visualisation shows the number of records (battery level and relative location records) received from the application daily.

To address this, initial efforts to create dashboards for RADAR-CNS were carried out in 2019 with the use of Grafana – a dashboard software. This marked the start of adding dashboards within the RADAR-base platform in an effort to mitigate the effect of missing data and issues with tracking.

**The Fitbit visualisation shows the wear time. This is calculated from the heart rate data available per day.

In a similar fashion, the RADAR-AD consortium will incorporate the use of Grafana dashboards to support to work of clinicians. The team working on the dashboard development comprises members from the clinical teams from VUmc and technical teams from The Hyve and KCL. The clinical teams worked on forming the requirements for dashboards to be able to support the work of the clinicians. The technical teams created their design and architecture and decided on the way to develop the dashboards.

The first version of these dashboards was released in the first week of August this year.  Further RADAR-AD features are planned to be added soon to the initial cohort dashboards, such as the function to filter by study sites.