Optimizing Diabetes Self-Care in Patients with Limited Health Literacy: A Comparative Study of SCT-Based Education with and without an AI-Designed Photo-Novel

Background: Type 2 diabetes mellitus (T2DM) requires continuous self-care to manage the condition and prevent complications. Many patients, particularly those with limited health literacy, face significant challenges in adhering to recommended self-care behaviors. Limited health literacy can hinder understanding of health information and following treatment guidelines, often leading to poorer health outcomes. Therefore, educational interventions tailored to improve both self-care behaviors and health literacy are crucial. This study evaluated and compared the effectiveness of a Social Cognitive Theory-based educational intervention, with and without health literacy strategies, in enhancing self-care behaviors in patients with T2DM and limited health literacy. Methods: This interventional study included 150 patients with T2DM and limited health literacy, randomly assigned to one of three groups: a Social Cognitive Theory-based group (TBG), a theory-based group with an AI-designed photo-novel (TBG + AIPN), and a control group. At baseline, all participants completed questionnaires assessing key variables. Both intervention groups received five training sessions grounded in Social Cognitive Theory. In the TBG + AIPN group, additional health literacy strategies—such as an AI-designed photo-novel and plain language materials—were incorporated into the training. Follow-up questionnaires were administered one and three months post-intervention. Data were analyzed using SPSS version 24. Results: Baseline demographic and clinical characteristics were comparable across the three groups, with no statistically significant differences observed. Longitudinal analyses indicated significant within-group improvements from baseline to follow-up in both intervention groups for all measured constructs, including diabetes knowledge, self-efficacy, outcome expectations, perceived social support, and self-care behaviors (all p < 0.001). Between-group comparisons revealed that the TBG + AIPN demonstrated significantly greater improvements than the TBG in self-care behaviors, diabetes knowledge, self-efficacy, outcome expectations, and self-regulation (all p < 0.05). However, no significant between-group difference was observed in perceived social support (p > 0.05), suggesting that while the literacy-sensitive enhancements improved most theoretical constructs, perceived social support may require distinct or additional targeted strategies. Conclusion: This study demonstrates that an intervention based on Bandura’s Social Cognitive Theory, when combined with health literacy strategies including an AI-designed photo-novel, is more effective in improving self-care behaviors among patients with T2DM and limited health literacy. Such a comprehensive approach not only targets psychological aspects of self-care but also integrates tailored communication methods, offering a holistic and adaptable solution for improving health outcomes in this vulnerable group. It is recommended that health systems adopt and implement these strategies to enhance self-care and reduce the burden of chronic disease.