Conversation Analysis and Clinical Encounters 2024

Adam spoke at the 8th International Meeting on Conversation Analysis and Clinical Encounters (CA&CE 2024, 1-3 July 2024, Oxford). This is a conference very much organised around finding practical applications of Conversation Analysis for assisting in the work of the healthcare professional, the patient, family members and policy makers.

Adam introduced our work with Ufonia on developing Dora.

Abstract

Automating clinical encounters – developing healthcare voice assistants for routine calls to patients

Interactive technologies such as Alexa and Siri have over the past few years become established as everyday devices through which people manage their affairs or access information from the wider Web. Although most common is still the text-based chatbot, designed to manage communications between a company or organisation and a client, increasingly we find voice-based automated Conversational User Interfaces being used, including in the healthcare sector. The benefit for healthcare providers is that this frees up valuable time for clinical staff to attend to patients who need further care, rather than phone calls to those for whom no further interventions are needed. However, the technology must be implemented without negatively impacting those routine calls to patients, or the quality of the information gathered.

This presentation reports on work carried out by the authors in partnership with a healthcare start-up. Ufonia has developed a voice assistant, Dora, for carrying out routine telephone-based clinical conversations across a number of hospital trusts (see Brandt et al. 2023). The collaboration set out to investigate whether Conversation Analysis could provide the interface designers with tools and insights to further improve these automated phone calls. The project focused on how to draw on insights from conventional human-human clinical encounters to feed into the conversation design of Dora. This collaborative work, a partnership between Conversation Analysts, conversation designers and clinicians trialled and implemented changes to the automated system, with a view to optimising the outcome for the healthcare provider, while at the same time delivering a better experience for the patient.

The project demonstrates how CA can feed into the design of automated user interfaces, enabling a more natural user experience for patients by modelling the design on the normative patterns of equivalent clinical encounters.

Keywords: clinical calls; automation; voice-based conversation user interface