Local Lexicon Extraction and Language Processing in Facilitating Language Awareness and Informing User-Centered Design in the Health Care Environment

The written word is a ubiquitous, critical, and time-sensitive design feature in the health care environment. Language critically influences understanding, decision making and subsequent actions. Words can influence usability, be confusing and cluttering, contribute to distraction and task disruption, and subsequently contribute to issues with clinician workload and patient safety. Design team members often hotly and at length deliberate over word selection during user centered design (UCD) efforts. Usability evaluation methods, when applied, can be valuable in facilitating a match between the system and the language spoken by users. In particular, usability evaluations can help designers identify confusing and ambiguous words and may suggest alternative choices. However, even with a hearty approach to design, word selection can be a bear of a task. This task can be time consuming, resource intensive, and frustrating for the design team, and suboptimal term selection may increase the risk of patient harm. The purpose of this paper is to encourage discussion on the use of natural language processing (NLP) approaches in facilitating local language awareness and informing UCD. The use of NLP in the health care domain is well documented when applied to narrative text for assisting in behind-the-scenes clinical decision support mechanisms and for assisting in risk analysis. Additionally, NLP may also be a valuable arrow in the designer’s quiver to inform conceptual design and the selection of user-facing terminology. This discussion attempts to expand upon documented uses of NLP in health care to its use in informing the selection of terminological design features with an emphasis on local design efforts in fostering safer patient care environments.