Clemson University
Exploring Healthcare Chatbot Information Presentation: Applying Hierarchical Bayesian Regression and Inductive Thematic Analysis in a Mixed Methods Study
Abstract
dc:description.abstractHigh blood pressure, also known as hypertension, significantly increases the risk of heart disease and stroke, which are leading causes of death in the United States. While contributing to over 691,000 deaths in 2021 alone in the United States (U.S.), it also imposes immense economic burden on the healthcare system, costing approximately $131 billion annually. One way to address this issue is for increased self-care behaviors and medication adherence, both of which require sufficient health literacy. Despite the importance of health literacy, 90% of U.S. adults struggle with health-related subjects. Overcoming the issues associated with health literacy requires addressing the inherent challenges of presenting health information as it is often unfamiliar, complicated, and exceptionally technical. However, with the advent of large language models (LLMs) such as ChatGPT, there are new opportunities to deliver health information more effectively and accessibly in the form of chatbots and conversational agents (CAs). This thesis takes a mixed-methods approach to investigate the effects of varying information presentation styles on a healthcare chatbot’s effectiveness, trust, and usability to assist users in a health-related information seeking task. Controlling for the communication style (conversational or informative) and language style (technical or non-technical), participants engaged with a healthcare chatbot to learn about blood pressure and hypertension. They then engaged in a semi-structured interview detailing their experience and ideating on future health information presentation designs for chatbots and CAs. Hierarchical Bayesian regression models were created to provide inference for how varying information presentations affected the chatbot’s effectiveness, trust, and usability. Moreover, inductive thematic analysis was conducted to analyze the qualitative interviews. The findings revealed dynamic interactions between the various information presentation styles and the outcome measures. Moreover, the qualitative interviews provide evidence for diverse information presentations styles to enhance usability and ease cognitive load for health-related information. These results suggest the importance of tailoring health information presentation to users’ cognitive abilities and preferences to enhance learning of health-related information. As chatbots and other CAs become increasingly used to disseminate health-related subjects, ensuring effective communication and usability will become increasingly more prevalent.
Degree
thesis:*- Name thesis:degree_name
- Master of Science (MS)
- Level thesis:degree_level
- Thesis
- Discipline thesis:degree_discipline
- Industrial Engineering
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Koscelny, Samuel Nelson
- Contributors dc:contributor
-
- Dr. David Neyens
- Dr. Emily Tucker
- Dr. Emma Dixon
Subjects
dc:subject × 21- Human-Computer Interaction
- Bayesian Analysis
- Health Literacy
- Artificial Intelligence
- Health Informatics
- Statistics
- Applied Statistics
- Artificial Intelligence and Robotics
- Data Science
- Design of Experiments and Sample Surveys
- Experimental Analysis of Behavior
- Graphics and Human Computer Interfaces
- Health Communication
- Health Information Technology
- Human Factors Psychology
- Industrial Engineering
- Other Statistics and Probability
- Probability
- Public Health Education and Promotion
- Quantitative Psychology
- Statistical Models
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://open.clemson.edu/all_theses/4342