Internet adaptability on artificial intelligence (AI) healthcare platforms is a key factor influencing users' continued usage and the effectiveness of platform outcomes. It has emerged as a major challenge in the era of digital healthcare transformation. However, it remains unclear to what extent users' Internet adaptability and platform usage behaviors interact, predict each other, and sustain a dynamic pattern of co-evolution. Therefore, this study employed cross-lagged panel network (CLPN) analysis with a multi-wave longitudinal design to uncover the network structure and dynamic interaction mechanisms underlying the co-occurrence of users' network adaptability and usage behaviors on AI-driven healthcare platforms. The results show that (1) In the cross-sectional network, there was a relatively dispersed structure during the early stage. As user experience accumulated, the network became increasingly centralized around a few core pathways, with self-efficacy and disease prevention emerging as key nodes. (2) According to the CLPN analysis, network adaptability factors (such as information protection, learning ability, and self-control) significantly promoted later usage behavior on AI-driven healthcare platforms (particularly self-diagnosis and disease prevention), forming a causal chain from adaptation to usage. (3) There are gender differences in the predictive effects of various dimensions of Internet adaptability on platform usage behaviors. Female users tend to adopt a socially oriented and holistic approach to health information processing, whereas male users are more inclined towards a tool-oriented and functional usage pattern. Interpreting user behavior evolution in intelligent healthcare environments, this research provides theoretical insights for the personalized design and precision service of AI-driven healthcare platforms.
Zhao et al. (2026) studied this question.