What?

Real time data visualization for users of UKRDC. These visualizations will appear in the user interface for the UKRDC. The core calculations will be done by the dedicated dashboard-stats library. This will be usable as a standalone library as well as callable from the ukrdc-fastapi. The dashboard stats will be tailored appropriately to serve the visualization needs of different UKRDC stakeholders.

Who?

Renal data managers: The UKRDC contains uncleaned data sent directly from renal units. Visualization will aid in identification and correction of erroneous or missing data. Examples of this might include:

• Visualization of the age of patients in a unit to identify infeasibly old patients resulting from incorrect demographic data.

• A visualization of biomarkers data might help identify incorrect test results which do not fit the expected pattern.

Clinicians: Visualization will provide comparison to audit standards on a day by day basis. For example, displaying quality of care metrics such as vascular access on the first dialysis session would allow real time comparison to both NICE guidelines and the audit standard in the annual report.

Statisticians: Overlap in scope with annual report other stats activity at UKKA will provide the ability to cross check with processed and cleaned data.

Special Use Case User: Do we want allow for users at some point in the future to be able to specify calculators for stand alone/special interest use cases which use the UKRDC data?

How?

• Flexible - development will focus on delivering value to users quickly and using feedback to inform future design choices.

• Modular and configurable - the challenge of a broad user based will be addressed by a series of calculators. Each will produce a self contained set of stats and will be designed with a specific use case in mind. The priority given to each calculator and whether to pre-calculate on schedule - some calculators for some units will inevitably be too slow to run on demand - will depend on demand. Dashboards will be configurable for individual users depending on their permissions and needs.

• User friendly - users should be able to find what they need to see without having to set too many options. Design should be kept clean, simple and tailored to the end user. In general stats should be available with load times shorter the time taken to calculate them. This will be handled by scheduled calculation and caching with ukrdc-fastapi.

• Granular - data correction should be possible by drilling down to data in its full granularity. An example of this could be using click events to select an area of the data and open a set of records corresponding to it.

When?

As of January 2023 basic stats are available for the units with RDA feeds. By the summer an array of different metrics will be available covering a broad range of use cases. Medium term development will roll out interactivity such as the ability to open a cohort of patients based on a section of the visualization. In addition the ability to create and download reports will be added. Longer term convergence with other projects such as the current UKKA data portal and the provision of a similar service for the RADAR initiative is expected.

See also