STIQI: an epistemic EHR interface (demo)
Designed with the clinician community using an epistemic approach to visualisation
Dave Pao
2 min read
November 2023
‘…a graphical summary should encourage doctors and nurses to reshape, perhaps re-invent, the medical record before computer programmers cast institutional convenience into silicon.'
- Powsner & Tufte (1994)
The Sexually Transmitted Infection Query Interface (STIQI) was originally built as a provocative prototype, or 'provotype' (Ruecker, 2015)—a research tool to guide the clinician community into articulating their intuitive sense of clinical usability.
The central research question was: 'How can a clinician community contribute clinical usability knowledge to the design of their own visual EHR interface?'
This later iteration, the STIQI development prototype, is shaped by that engagement. It is the first EHR interface that uses Cheng & Barone's (2007) Representational Epistemic (REEP) approach to visualise patient data. The REEP approach is unique in providing an explicit, granular blueprint for the design of epistemic visualisations by asking: 'What makes a problem complex?’
In other words, Cheng & Barone operationalise the epistemic visualisation process, such that clinical theory can be incorporated within the interface design specification, to any level of sophistication. They do this by describing how the entire conceptual structure of a knowledge domain can be analysed in terms of: (1) domain concepts; (2) their constituent conceptual dimensions; (3) the intra- and inter-relationships within and between them; and (4) precisely how these will be expressed visually.
Perhaps the major usability requirement voiced by the clinician community during this research was the essential requirement to input data whilst simultaneously reviewing data. For this reason, the EHR interface is envisaged as having two screens (see image):
Screen 1: Data of Subjective Origin (DoSO)—data that have been interpreted by a clinician (coding, narrative history, clinical opinion etc.)
Screen 2: Data of Objective Origin (DoOO)—data that have not been interpreted by a clinician (this is the STIQI prototype). These data are represented by five domain concepts: STI-related tests, Blood-borne infections and vaccines, Other tests, Contraception and Medication. A high-level overview (Timeline) sits alongside fully granular data views (Detail Panels).
A key benefit of this epistemic approach is that visual elements are determined not only by the inherent characteristics of the data but also by clinical theory. This means that data can be represented in a way that is clinically appropriate to the zoom level—one example is the use of aggregated 'sets' (e.g., six different gonorrhoea tests are visible in the Detail Panels but are aggregated into a single set on the Timeline).
The STIQI prototype (desktop only) can be accessed at https://stiqi.netlify.app/. On this demo version, DoOO data is only viewable from the 2022 perspective (i.e., do not navigate away from 2022).
Cheng, P. C.-H. and Barone, R. (2007). Representing complex problems: A representational epistemic approach. In: Jonassen, D. H. (Ed). Learning to Solve Complex Scientific Problems. Milton Park, UK: Routledge.
Powsner, S. M. and Tufte, E. R. (1994). Graphical summary of patient status. The Lancet, 344 (8919), pp.386–389.
Ruecker, S. (2015). A brief taxonomy of prototypes for the digital humanities. Scholarly and Research Communication, 6 (2).
