Revealing the complexity of users’ intention to adopt healthcare chatbots: A mixed-method analysis of antecedent condition configurations

Healthcare chatbots provide a professional, immediate, and low-cost tool for advising people on health information. Unlike previous studies on chatbot adoption that focus on the one-way net effect of components, this study reveals the complex antecedent configurations behind online healthcare chatbot adoption through an asymmetric approach. This study attempts to explain the causal configuration of user adoption in terms of three dimensions: functional, social, and user motivation features. A sequential mixed-method approach was chosen and a two-stage study, fuzzy set comparison analysis (fsQCA), and semi-structured interviews were conducted to deepen users' understanding, perceptions, and attitudes toward online healthcare chatbots, expressing more fine-grained insights into variable relationships. Five configurations were found to explain the high willingness of users to adopt healthcare chatbots. Perceived social presence was a core condition for each configuration. However, the social features of chatbots can only lead to high levels of trust and satisfaction when combined with functionality. The findings of this paper's mixed-method design and complexity study contribute to the adoption and theoretical literature on intelligent health services and chatbots. It guides tailoring chatbot functionality to individual user needs and provides practical guidance for the development of chatbots for health services through diverse combinations of machine chatbot features and user-motivated features.