A Systematic Review on Healthcare Artificial Intelligent Conversational Agents for Chronic Conditions

Area of Science:

• Health Informatics
• Artificial Intelligence in Healthcare
• Human-Computer Interaction

Background:

• Conversational agents (CAs) are increasingly explored for managing chronic health conditions.
• Understanding the types, technologies, and user acceptance of these agents is crucial for their effective implementation.

Purpose of the Study:

• To review different types of conversational agents used in healthcare for chronic conditions.
• To examine their communication technology, evaluation measures, and AI methods.
• To assess user acceptance and identify gaps in evidence regarding efficacy.

Main Methods:

• A systematic literature search was conducted across six major databases (PubMed, EMBASE, PsycINFO, CINAHL, Web of Science, ACM Digital Library) in February 2021.
• Inclusion criteria focused on studies involving consumers, caregivers, or healthcare professionals, chronic diseases, conversational agents, and human user testing.
• Twenty-six studies met the criteria from an initial retrieval of 1087 articles.

Main Results:

• The review identified 26 conversational agents, including chatbots (16), embodied conversational agents (7), and others.
• Overall user acceptance of CAs for chronic condition self-management is promising, with over half of studies reporting user satisfaction, helpfulness, and ease of use.
• Users often express comfort interacting with CAs.

Conclusions:

• Conversational agents demonstrate potential for supporting self-management of chronic conditions.
• Despite positive user feedback, there is a lack of reliable, comparable evidence on the efficacy of AI-enabled CAs due to insufficient reporting of technical details.
• Further research with standardized reporting is needed to establish the clinical effectiveness of these technologies.