Experiences and Perceived Influence of the Artificial Intelligence–Based Health Education Accurately Linking System (AI-HEALS) on Health Behaviors Among Patients With Type 2 Diabetes: Qualitative Study
Journal of Medical Internet Research · 2026
DOI: 10.2196/84605 ↗Kurz gefasst
Randomisierte kontrollierte Studie in der Zeitschrift Journal of Medical Internet Research, 2026. Verfasst von Jing Wang und 6 weiteren Autor:innen. Beteiligte Einrichtung: School of Public Health, Peking University, 38 Xueyuan Road, Haidian District, Beijing, 100191, China, 86 13691212050. Hauptthemen laut MEDLINE: Diabetes Mellitus, Type 2, Artificial Intelligence, Health Behavior, Health Education. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.
Themen: Künstliche Intelligenz in der Pflege, Diabetes mellitus
Abstract (Originalsprache)
Background
The management of type 2 diabetes requires sustained self-management across diet, physical activity, medication adherence, and blood glucose monitoring; however, maintaining these behaviors in daily life remains difficult for many patients. Artificial intelligence–enabled and mobile health interventions have shown promise in supporting diabetes education and self-management, but evidence on how patients actually experience and use such systems in real-world primary care remains limited.
Objective
This study aimed to explore patients’ experiences with the Artificial Intelligence–Based Health Education Accurately Linking System (AI-HEALS) and its perceived influence on self-management behaviors among patients with type 2 diabetes.
Methods
This explanatory qualitative study was nested within the intervention arm of a cluster randomized controlled trial of AI-HEALS. Purposive maximum-variation sampling was used to recruit participants who varied by sex, age, diabetes duration, hemoglobin A 1c level, digital literacy, and level of platform use. Of the 25 patients approached from the intervention arm, 17 agreed to participate. Semistructured interviews were conducted 3 months after the intervention (August to December 2023) with participants recruited from 45 communities in the Daxing and Shunyi districts of Beijing, China. The qualitative component was designed to explain perceived mechanisms of behavior change, contextual facilitators and barriers, and implementation-related experiences not captured by trial outcomes alone. Interview transcripts were analyzed thematically in NVivo 12 by 2 independent researchers using consensus coding, an audit trail, and member checking.
Results
Participants’ experiences with AI-HEALS were reflected in four themes: (1) catalyzing health awareness and concern, (2) empowering self-management practices, (3) navigating usability and engagement, and (4) enhancing psychological adaptation. Participants perceived that AI-HEALS made diabetes more visible in daily life through repeated reminders and accessible educational content; supported practical self-management decisions related to diet, physical activity, medication adherence, and glucose monitoring; and, in some cases, improved confidence while reducing uncertainty and diabetes-related stress. The findings also suggested meaningful variation in how the intervention was experienced, particularly in relation to prior illness experience, routine stability, social support, and digital confidence. Many participants mainly engaged with lower-burden features such as pushed articles and reminders, whereas more interactive functions, such as the chatbot, were used less often.
Conclusions
Patients generally perceived AI-HEALS as a useful source of ongoing education, behavioral prompting, and everyday support for diabetes self-management. The findings suggest that artificial intelligence–enabled education may work not only by increasing knowledge but also by reinforcing awareness, translating guidance into feasible daily action, and supporting psychological adaptation. At the same time, the intervention’s perceived usefulness depended on whether its content was understandable, low burden, and compatible with users’ routines. These results support further refinement of literacy-sensitive, layered digital interventions and their integration into routine community-based primary care.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Public Health, Peking University, 38 Xueyuan Road, Haidian District, Beijing, 100191, China, 86 13691212050.
- Department of Nursing, the Fourth Affiliated Hospital of School of Medicine, International School of Medicine, International Institutes of Medicine, Zhejiang University, Yiwu, China.
- Jitang College, North China University of Science and Technology, Tangshan, China.
- Center for Clinical Epidemiology Research, Affiliated Hospital of Inner Mongolia Medical University, Hohhot, China.
- Capital Medical University, Beijing, China.
- School of Traditional Chinese Medicine, Capital Medical University, Beijing, China.
Zitieren
APA
Wang, J., Wu, Y., Jiang, Y., Bao, R., Gu, X., Kong, B., & Sun, X. (2026). Experiences and Perceived Influence of the Artificial Intelligence–Based Health Education Accurately Linking System (AI-HEALS) on Health Behaviors Among Patients With Type 2 Diabetes: Qualitative Study. Journal of Medical Internet Research. https://doi.org/10.2196/84605
Vancouver
Wang J, Wu Y, Jiang Y, Bao R, Gu X, Kong B, et al. Experiences and Perceived Influence of the Artificial Intelligence–Based Health Education Accurately Linking System (AI-HEALS) on Health Behaviors Among Patients With Type 2 Diabetes: Qualitative Study. Journal of Medical Internet Research. 2026. doi:10.2196/84605
BibTeX
@article{Wang2026,
author = {Wang, Jing and Wu, Yibo and Jiang, Yang and Bao, Rantong and Gu, Xinbao and Kong, Bingyang and Sun, Xinying},
title = {Experiences and Perceived Influence of the Artificial Intelligence–Based Health Education Accurately Linking System (AI-HEALS) on Health Behaviors Among Patients With Type 2 Diabetes: Qualitative Study},
journal = {Journal of Medical Internet Research},
year = {2026},
doi = {10.2196/84605}
}
EUVIMED