Conversational Agents for Managing Chronic Diseases: A Systematic Review

Simple item page
AuthorSomayyeh Zakerabasalien
AuthorParastoo Amirien
AuthorJahanpour Alipouren
OrcidSomayyeh Zakerabasali [0000-0002-0125-5473]en
OrcidParastoo Amiri [0000-0002-5654-1987]en
OrcidJahanpour Alipour [0000-0002-8139-1140]en
Issued Date2025-11-30en
AbstractContext: Chronic medical conditions (CMCs) are the major causes of universal morbidity and mortality. Conversational agents (CAs) are promising solutions for managing these conditions. Objectives: This study aimed to examine the clinical and technical insights of CAs in managing CMCs. Data Sources: We conducted a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines, utilizing the Web of Science, PubMed, Scopus, IEEE databases, and the Google Scholar search engine to find articles published until June 17, 2025, with the keywords "Conversational agents" and "chronic diseases", along with their synonyms. Study Selection: The studies were assessed for quality using the Mixed Methods Appraisal Tool (MMAT) in its 2018 version. Data Extraction: We collected clinical and technical insights regarding CAs for the management of CMCs. Results: After reviewing 1035 articles, 22 ultimately met the inclusion criteria. The quality scores of the articles varied from 60 (moderate, n = 4, 18.2%) to 100 (high, n = 18, 81.8%) according to the MMAT. Among the studies, diabetes mellitus was the most commonly targeted condition, self-management was the most frequently applied approach, and usability was the most commonly measured outcome. Improvement of clinical outcomes (n = 11, 50%), usability (n = 12, 54.5%), and user satisfaction (n = 5, 22.7%) were the most frequent key findings of CAs. Most of the studies used the Android platform (n = 12, 54.5%), and English language (n = 14, 63.6%), and generating responses using artificial intelligence (AI; n = 14, 63.6%) was the most frequent technical insight. In most studies, the CAs took on the role of starting the conversation (n = 12, 54.5%), and most of the dialogues were text-based (n = 11, 50%). Various software (e.g., Python, RASA toolkit, Flutter, and React Native) were used to design CAs. The small sample size was the most frequent limitation in the reviewed studies (n = 6, 27.3%). Conclusions: This review highlights the potential of CAs to revolutionize chronic disease management by improving communication, enhancing patient engagement, and facilitating personalized care. However, it is important to acknowledge the significant heterogeneity between studies, which may affect the reliability of the findings. Despite these advancements, challenges such as privacy concerns, data security, and technological accessibility persist, necessitating the development of appropriate solutions. Future research and development should focus on addressing challenges and creating standard frameworks for maximizing the benefits of these technologies.en
DOIhttps://doi.org/10.5812/healthscope-163131en
KeywordArtificial Intelligenceen
KeywordChronic Diseaseen
KeywordSystematic Reviewen
KeywordSelf-managementen
PublisherBrieflandsen
TitleConversational Agents for Managing Chronic Diseases: A Systematic Reviewen
TypeSystematic Reviewen

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
healthscope-14-4-163131-publish-pdf.pdf
Size:
302.17 KB
Format:
Adobe Portable Document Format
Description:
Article/s PDF
Download

Collections

Health Scope