Modeling older adults' continuance intention toward mobile health apps: a dual-path SEM-ANN approach

With the rapid growth of the aging population, older adults in China face significant challenges in health management, and their continuance intention to use mobile health applications remains lower than that of younger users.

Based on survey data from older adults, this study employed a hybrid approach combining structural equation modeling (SEM) and artificial neural networks (ANN) to examine both facilitating and hindering factors.

The results reveal that satisfaction (β = 0.42, p < 0.001) and perceived usefulness (β = 0.31, p < 0.01) exert significant positive effects on continuance intention, while resistance (β = -0.28, p < 0.05) has a significant adverse effect. The integrated model explains 56.6% of the variance in continuance intention. ANN analysis further shows that satisfaction is the most critical predictor (normalized importance = 100%), followed by confirmation (37.7%), perceived usefulness (21.2%), complexity barriers (12.2%), resistance (11.6%), and privacy concerns (11.0%).

This study confirms the suitability of integrating ECM and IRT to explain older adults' continuance intention toward mobile health apps. It highlights the multifactorial nature of their continuance behavior and provides theoretical and practical insights for enhancing their continued use of mobile health technologies.