Assessing the design of Conversational Agents for Asthma support: Protocol for an Observational Pilot Study (Preprint)

BACKGROUND A significant number of people have poorly controlled asthma – 60% according to some estimates. One likely reason is poor self-assessment of risk by those with asthma, an issue that could be addressed with a conversational agent that assesses and communicates risk accurately. Such a system could further improve outcomes by providing follow-up recommendations to address other common obstacles to asthma control such as poor inhaler technique and insufficient understanding of asthma triggers and management strategies. OBJECTIVE The aims of this study are to: 1) Determine the feasibility and usability of a text-based conversational agent (i.e. chatbot) that processes a patient’s text responses and short sample voice recordings to calculate an estimate of their risk for an asthma exacerbation and then offers follow-up information for improving asthma control; 2) Assess the level of engagement of different types of users, particularly those who do not control their asthma well; 3) Assess self-reported level of asthma control and symptoms and 4) explores associations between asthma control and engagement with the conversational agent. METHODS For this pilot study we will recruit 90 adults with asthma through NHS outpatient clinics across primary and secondary care. Participants will have access to the conversational agent through WhatsApp on their mobile phone. Participants will be sent scheduled and randomly timed messages to invite them to engage in a dialogue about their asthma management during the period of the study. After a data collection period (28 days), participants will respond to survey items related to the quality of the interaction. A pre and post questionnaire will measure asthma control and symptoms before and after the intervention. RESULTS This study was funded in March 2021 and started in January 2022. We developed a pilot conversational agent, which will be improved in iterations of work with nurses, starting in September 2022. This pilot evaluation will start recruitment in January 2023. The anticipated completion of the study is July 2023. CONCLUSIONS This pilot study will support a follow-up study to assess the impact of a conversational agent on asthma care.