Cluster-based quality analysis of mobile health applications based on health behavior theories

Objectives: This study evaluated Korean mobile health (mHealth) applications, analyzing the extent to which they incorporated elements of health behavior theory. We aimed to classify application types and compare their structural characteristics and quality using cluster analysis. Methods: We evaluated 238 mHealth applications from Apple App Store and Google Play Store. Applications were assessed using the Health Care Smartphone Application Evaluation Tool, which considers content, design, and security as the criteria for quality evaluation. Elements from the Health Belief Model (HBM) and Theory of Planned Behavior (TPB), including perceived benefits/attitudes, cues to action, self-efficacy/perceived behavioral control, and subjective norms, were mapped onto application features. K-means clustering was performed using binary-coded theory elements, followed by validation and quality comparison. Results: Three clusters were identified: information-focused (36.6%), engagement-empowered (56.3%), and social influence-driven (7.1%) applications. Information-focused applications emphasized benefits but lacked interactivity. Engagement-empowered applications comprised diverse theory components, and social influence-driven applications emphasized peer influence. Application quality scores significantly differed across clusters ( p<.001), with information-focused apps scoring highest. Conclusion: mHealth applications can be meaningfully classified based on behavioral theory integration, which is associated with quality. Integrating structured theoretical frameworks may enhance both effectiveness and credibility of mHealth interventions.