Abstract: Trusted data space is an infrastructure carrier that connects multiple subjects based on consensus rules and realizes the interoperability and sharing of data resources, which builds an application ecology for the co-creation of data element value and is also a key support for the construction of a national integrated data market. In the field of health data, its application can effectively integrate fragmented information to reduce the cost of repeated medical testing and waste of resources, support the development of personalized precision medicine services, and empower scientific research innovation, public health decision-making, and national strategic layout. However, the current construction of trusted health data space in China is facing multiple institutional barriers, including ambiguous definition of rights and interests leading to obstruction of authorization, lack of object classification inhibiting utilization efficiency, conflicts in normative systems exacerbating compliance risks, and lack of regulatory mechanisms leading to weak implementation. One of the reasons for this is that there are structural dilemmas in health data regulation theory—lagging theoretical iteration, governance formalism, and disconnection from new technology. To this end, it is necessary to take the theory of data space justice and balance of property rights and interests as the dual-core guidance, clarify the boundaries of rights and interests by promoting the reform of separation of powers, build dynamic grading standards to optimize data classification, accelerate the integration of special legislation and normative frameworks, and establish a closed loop of responsibility for the implementation of the gatekeeper system.