Ong Hui Shan et al., Evaluation of a Chatbot to Support Telecarers in Patients’ Diabetes Management: An Embedded Mixed Methods Study. Public Health & Occupational Medicine Conference (PHOM) 2025.
Abstract:
BACKGROUND
Chatbots offer potential for better chronic disease management by enhancing patient–provider interactions. A clinical protocol-guided chatbot was developed to support nurses engaging patients with diabetes in a telehealth programme. By gathering information between calls, the chatbot enabled telecarers to tailor conversations to patients’ needs. This study explored the feasibility and acceptability of integrating the chatbot into the programme.
METHODS
We employed a qualitatively oriented embedded mixed methods design [QUAL(quan)], guided by the Theoretical Framework of Acceptability and a WHO Digital Health framework. Fourteen participants (7 patients with diabetes and 7 Telecarers) completed semi-structured interviews and a survey post-programme. Chatbot usage metrics were also collected. Interview data were analysed thematically, and survey data were analysed using descriptive statistics.
RESULTS
Participants viewed the chatbot positively, citing its ease of use, trustworthiness, and usefulness as an educational self-management aid. Telecarers found the chatbot reports easy to review and helpful for delivering more tailored calls. However, concerns were raised about its ability to address user-specific queries and fit into existing workflows. Issues included scheduling disruptions and uncertainty about its added value for telecarers. While patients noted improved communication, many still preferred human interaction. Survey findings echoed these themes. Participants also recommended more tailored chatbot educational content, user-friendly features, and onboarding support.
CONCLUSION
The chatbot showed promising feasibility and acceptability. While users appreciated its ease of use and educational value, limitations in content adaptation and workflow integration posed barriers. Addressing these concerns is key to integrating it into the programme.
License type:
Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)
Funding Info:
This research / project is supported by the Agency for Science, Technology and Research (A*STAR), Singapore under its Industry Alignment Pre-Positioning Fund (Grant No. H19/01/a0/023 – Diabetes Clinic of the Future). This research is approved by the SingHealth Centralized Institutional Review Board (Protocol No. 2019/2414) and the A*STAR Institutional Review Board (Protocol No. 2019-079) with a waiver of informed consent, and fully complied with all relevant ethical regulations. All data were anonymized before analysis. - Industry Alignment Pre-Positioning Fund (Grant No. H19/01/a0/023 – Diabetes Clinic of the Future)
Grant Reference no. : H19/01/a0/023