JMIR Publications

ABSTRACT

Diabetes has become a significant global chronic non-communicable disease with continuously rising prevalence, imposing a heavy long-term burden on public health systems. Its management is inherently long-term and complex, relying on continuous lifestyle adjustments and treatment adherence, including blood glucose monitoring, medication, dietary control, and exercise. However, the traditional medical model, which depends mainly on regular outpatient follow-ups, struggles to provide continuous intervention in patients’ daily behaviors, limiting long-term management effectiveness. Diabetes management therefore depends heavily on patients’ daily self-management, yet studies show widespread issues of insufficient compliance and difficulty sustaining behavioral changes, leading to unsatisfactory blood sugar control. With the increasing popularity of smartphones and wearable devices, mobile health (mHealth) has emerged as an important tool for chronic disease management by offering continuous and individualized support. In diabetes care, mHealth interventions enable real-time monitoring and promote self-management improvement through data feedback, reminders, and remote support, while also breaking through time and space limitations to improve healthcare accessibility. Although previous systematic reviews and meta-analyses indicate that mHealth interventions typically reduce HbA1c by approximately 0.3% to 0.5%, significant differences in effectiveness across studies suggest that outcomes may be influenced by functional design and implementation methods. Crucially, most existing reviews classify interventions by technology type rather than by functional modules, making it difficult to reveal the specific mechanisms of action of different functions. Furthermore, the prevailing “holistic assessment” approach limits in-depth understanding of intervention mechanisms and hinders optimization of intervention design. This review therefore systematically examines mHealth diabetes management interventions from a functional module perspective to address these gaps. This review aims to systematically evaluate the progress of mHealth interventions in diabetes management through the lens of functional modules. Specifically, it seeks to: (1) construct a classification framework for mHealth diabetes management interventions based on five core functional modules—self-monitoring and data collection, feedback and reminders, provider-patient interaction, health education and behavioral support, and personalized intervention; (2) analyze the influence of different functional modules on glycemic control, self-management behaviors, and treatment adherence; (3) explore the characteristics and advantages of integrated multi-functional interventions compared to single-module approaches; and (4) provide a systematic theoretical framework and practical reference for optimizing mHealth intervention design in diabetes care. This review draws on evidence from randomized controlled trials and systematic reviews to systematically evaluate mHealth interventions in diabetes management. The study employs a functional module-based analytical approach, categorizing mHealth interventions into five core modules: self-monitoring and data collection, feedback and reminders, provider-patient interaction, health education and behavioral support, and personalized intervention. Through this lens, the review examines the mechanisms linking functional modules to behavioral change and clinical outcomes, and compares the stability and efficacy of single-module versus integrated multi-functional interventions. The findings indicate that mHealth interventions generally yield a small-to-moderate reduction in HbA1c levels, alongside consistent improvements in self-management behaviors and treatment adherence. The underlying mechanism operates through a "functional module – behavioral change – clinical outcome" pathway, rather than through the isolated impact of individual modules. Further analysis reveals that integrated multi-functional interventions offer superior stability and efficacy compared to single-module approaches, attributable to cross-module synergy and continuous behavioral reinforcement mechanisms. Despite the demonstrated benefits of mHealth interventions in diabetes management, significant challenges remain, including declining user engagement, inconsistent data quality, limited integration with traditional clinical workflows, and concerns regarding health equity and data security. Future research should prioritize mechanism-based intervention design, advance personalized and intelligent solutions, and facilitate the deep integration of mHealth within conventional healthcare systems to enhance the sustainability and clinical utility of these interventions. This review provides a systematic theoretical framework and practical reference for optimizing mHealth intervention design in diabetes care.