BACKGROUND: With the rapid digitalization of healthcare and an aging population, understanding the factors influencing older adults' sustained adoption of Internet medical services is critical. However, existing research often oversimplifies these factors by relying on linear models. This study integrates Partial Least Squares Structural Equation Modeling (PLS-SEM) and fuzzy-set Qualitative Comparative Analysis (fsQCA) to explore the complex pathways driving continued use. METHODS: A survey of 1,920 older adults (60-75 years) in China assessed satisfaction, e-health literacy, self-efficacy, social support, social influence, social participation, and willingness to use Internet medical services. PLS-SEM examined the relationships between variables, while fsQCA identified multiple configurations leading to sustained use. RESULTS: PLS-SEM identified satisfaction as the strongest predictor of sustained use (β = 0.281, p <
0.001), acting as both a direct determinant and a mediator for e-health literacy and social participation. Social influence (β = 0.189, p <
0.001) and social support (β = 0.172, p <
0.001) also contributed significantly. FsQCA revealed six distinct configurations, with satisfaction and e-health literacy as core conditions across most pathways. CONCLUSIONS: By integrating linear and configurational approaches, this study provides a nuanced understanding of older adults' digital healthcare behaviors. Enhancing satisfaction, digital literacy, and social engagement is key to fostering sustained adoption. Tailored interventions based on distinct configurations can maximize the effectiveness of digital health programs. IMPLICATIONS: This research bridges gaps in understanding complex behaviors and provides actionable insights for policymakers and healthcare providers, highlighting the critical role of digital literacy and social support.