Author: Matthew Michael Vagnoni
Primary Advisor: Amy Franklin, Ph.D
Committee Member: Muhammad Walji, Ph.D
Masters thesis, The University of Texas School of Biomedical Informatics at Houston.
Background and objective.
Knowledge searching requirements have increased with recent health care reform measures, resulting in increased burdens and medical errors. Current approaches to finding and navigating clinical knowledge (such as terminologies) may not be as effective as more psychologically inspired approaches. One such approach is faceted browsing with the aim of ensuring accurate and rapid location of relevant information. Owing to a lack of comprehensive analysis of the existing literature, this review was undertaken to analyze psychological justifications for faceted browsing being an improvement over typical keyword and clustering searching strategies found in clinical user interfaces (e.g. coding systems and electronic medical records).
Materials and methods.
A literature search was performed using keywords in PubMed and Google Scholar to identify research and review studies with information related to either psychological underpinnings or the implementation of keyword search or faceted browsing functionalities.
Results and discussion.
Various information retrieval functionalities have been implemented in the clinical setting, but all appear to use the same recall orientated methods of keyword or single hierarchy/clustering search. Meanwhile, many medical errors are blamed on the design and functionality found inside of clinical computer systems. Recognition-based search and navigation functionalities, such as faceted browsing, could reduce these errors. Faceted browsing and its psychological underpinnings has been studied generally but has been little studied as an application to reduce medical errors in the clinical domain. Faceted browsing when studied has generally been viewed as more effective and subjectively higher rated than leading alternatives. There are challenges with faceted-browsing, but the clinical space is uniquely positioned to overcome them. Specifically, the clinical domain is endowed with a wealth of rich metadata and subject-specific concepts that have been curated for many decades.
Although certain faceted browsing studies have shown a benefit in user performance, there is no sufficient research in this area to be conclusive. Furthermore, the clinical domain and studies to tease out the psychological underpinnings are lacking. Further studies analyzing the effect of faceted browsing functionalities on clinical and medical errors are required.