The HCPF draws on adaptive governance, adaptive policymaking, digital health strategy, and learning health systems but reorients these ideas toward national digital health strategic planning [22-25]. Its originality lies in explaining how long-term direction and short-cycle adaptation can be connected through a temporally layered planning architecture. The framework also highlights how top-down mobilization, bottom-up data feedback, and regional disparities together affect the quality and representativeness of policy recalibration. It is therefore best understood not as a new universal governance theory but as a digital health–specific lens for studying the shift from static planning toward adaptive governance in a large, heterogeneous health system.

The data-driven healthy China pathway can be conceptualized through the HCPF, which integrates long-range strategic direction with shorter-cycle policy, emergency, and technical adjustments. Unlike linear planning models, the HCPF operates as a temporally layered governance architecture that supports vertical alignment while allowing for adaptation to technological and epidemiological uncertainty. The architecture contains 5 interdependent tiers organized into strategic anchors and dynamic engines (Figure 1). Table 1 traces how these tiers emerged across China’s 3 planning phases and summarizes their implications for the framework.

At the apex, the long-term strategic vision (10- to 15-year horizon) defines enduring goals. Blueprints such as Healthy China 2030, formulated by the Central Committee and State Council, set broad priorities such as population health equity, whereas later directives elevated data, including health data, as a factor of production [1,26]. This layer sets strategic direction rather than concrete implementation pathways. These visions are translated into programmatic commitments through midterm planning (5-year horizon), exemplified by the 14th Five-Year Plan for National Health Informatization [18]. Through joint central agency issuances, this layer identifies priority domains, such as regional big data centers and telemedicine expansion, and sets measurable milestones [8]. Annual rolling adjustments and periodic midterm evaluations then moderate the rigidity of 5-year planning. This bridging layer translates midterm goals into actionable annual tasks, including interoperability targets and pilots for diagnosis-related group (DRG) and diagnosis-intervention packet (DIP) payment reform [27], allowing priorities to be resequenced in response to budget constraints, implementation feedback, and emerging technologies.

The framework’s adaptive capacity resides in its lower tiers. The agile emergency response layer operates on day-to-week cycles and enables rapid cross-agency mobilization during public health crises. During the COVID-19 pandemic, for example, health code risk classification rules were iteratively updated in response to changing case patterns and mobility needs [28]. The base layer is continuous technical iteration: foundational standards and tools are refined without waiting for the next formal planning cycle, often beginning as regional pilots before national scaling. Examples include the gradual tightening of EMR grading criteria and the staged rollout of DRG and DIP grouping standards. These standards function as payment and management tools that classify clinical cases and support standardized reimbursement and hospital performance management.

Functionally, the HCPF operates through dual-directional flows. Top-down mobilization promulgates mandates and allocates resources across administrative levels. Bottom-up data intelligence, derived from surveillance data, performance assessments, and pilot evaluations, informs strategic recalibration. In practice, this feedback loop operates through a tiered data governance pathway. National authorities define policy priorities, technical standards, data elements, security requirements, and performance indicators. Provincial platforms translate these requirements into regional health information architectures. Municipal and county systems aggregate data from hospitals, primary care institutions, public health agencies, and payment systems. Local institutions then generate, validate, and upload clinical, administrative, and public health data. This feedback loop connects top-level design with frontline implementation and allows the national strategy to adjust based on empirical signals [29,30]. However, the quality of this loop depends on local digital maturity. In digitally mature regions, data governance can be supported by integrated regional platforms, interoperability interfaces, hospital performance dashboards, and pilot-based feedback mechanisms. In less digitally mature regions, implementation may depend on minimum dataset requirements, provincial cloud-based services, targeted fiscal support, technical assistance from higher-level hospitals, and staged adoption of electronic health records and interoperability standards. The result is a hybrid planning model in which long-term and 5-year plans provide stability, annual and technical iterations enable incremental adjustment, and emergency mechanisms permit rapid reconfiguration. Its core contribution is temporal integration: aligning decisions made across decadal, 5-year, annual, and subannual time scales. This alignment helps explain how China has pursued long-term digital health goals while adapting to technological change and public health shocks.

China uses a whole-of-government approach. The National Health Commission serves as the lead health authority, whereas agencies responsible for cybersecurity, data regulation, and infrastructure planning, including the Cyberspace Administration of China and the National Development and Reform Commission, contribute to coordination. The National Data Administration further reflects the strategic elevation of data governance and data resource development [31]. This cross-agency structure can reduce fragmentation by aligning standards, infrastructure investment, data governance, and hospital performance requirements across different levels of government while linking these efforts to broader national digital economy strategies.

China has concentrated fiscal and regulatory support on designated pilot regions rather than pursuing uniform national rollouts [32]. Central authorities use fiscal transfers and “new infrastructure” (xin jijian) funding to support national and provincial health care big data centers, digital health pilot cities, and demonstration zones [33]. These sites receive targeted funding, policy support, and technical assistance for interoperability infrastructure, cloud platforms, and analytics capabilities [7]. They function as experimentation fields for EMR adoption, regional interoperability, DRG and DIP payment models, and data governance and implementation mechanisms [34]. This pilot-based strategy allows digital maturity models and governance arrangements to be tested in more digitally mature settings before wider diffusion.

The Chinese model combines public governance, private technological capability, and academic validation. Public authorities retain responsibility for data governance and safety regulation while drawing on the technical capabilities and consumer infrastructure of platform companies [35,36]. Academic medical centers and research universities provide clinical datasets and testing environments for refining AI algorithms and medical standards before wider scaling [37]. Companies such as Alibaba Health and Tencent Medipedia, together with medical AI developers, collaborate with hospitals, academic medical centers, and research institutes to develop telemedicine platforms, AI-assisted diagnostic tools, and health information portals [38,39]. Integrated services in WeChat, Alipay, and similar platforms provide a consumer-facing digital front door for appointment scheduling, payment, telehealth, and health information [40]. This configuration blends state steering with market experimentation while reducing the need for the health system to build all consumer-grade infrastructure independently.

A major driver is the integration of digital metrics into the National Tertiary Public Hospital Performance Appraisal (colloquially known as the Guokao), which assesses China’s tertiary public hospitals. Indicators such as EMR application-level grading, interoperability maturity, and smart hospital ratings are weighted components of performance assessment [41]. Because Guokao results can influence accreditation, resource allocation, and leadership appraisal, they function as a significant policy lever for digital transformation. By tying EMR maturity and data quality to hospital performance, the state incentivizes hospital executives to invest in digital infrastructure, data governance, and clinical adoption, turning digitalization from a technical support function into an institutional priority linked to competitiveness and leadership accountability [42].

China’s large-scale deployment of health codes (jiankang ma), itinerary codes, and nucleic acid testing registries during the COVID-19 pandemic response constituted one of the most extensive real-time collections of personal health and mobility data in recent history. Although these systems contributed substantially to pandemic control, they also exposed systemic governance challenges regarding data privacy and data protection [43]. The Personal Information Protection Law (PIPL; 2021) treats medical and health information as sensitive personal information and requires specific purpose, sufficient necessity, strict protective measures, and separate consent for its processing (PIPL Articles 28-29) [44]. The Data Security Law (2021) further establishes a broader framework for data classification, risk management, and data security obligations [45]. Despite these legal frameworks, local implementations in some jurisdictions appear to have exceeded statutory or proportionality limits. Many municipal platforms collect granular data beyond epidemiological necessity, whereas the algorithmic rules governing code assignment and reversion remain opaque [16,46]. Moreover, prior analyses have raised concerns about function creep, in which health code data could be repurposed for objectives such as social stability maintenance or public security rather than disease control alone [47]. Rebuilding legitimacy, therefore, requires more than reliance on existing framework legislation. Targeted reforms could include (1) dedicated regulations on personal information processing during public health emergencies, including clear retention limits such as mandatory deletion of identifiable data after the emergency period absent a separate lawful basis [46]; (2) an independent, cross-ministerial health data oversight mechanism with authority to audit algorithms and data flows; and (3) lower barriers to individual redress, including clarification of the public interest litigation threshold associated with PIPL Article 70 [48], so that affected individuals can more readily challenge erroneous code assignments or data misuse.

A major vulnerability of the pathway is the persistent data divide between digitally mature eastern or urban pilot zones and western, rural, or grassroots settings [9-11]. The risk is not only unequal infrastructure but also feedback bias: if planning data come mainly from high-resource regions, national strategies and AI models may underrepresent rural health needs. Usable feedback depends on data completeness, coding consistency, workflow integration, local informatics capacity, and sustained data quality work. In high-resource settings, these functions are supported by mature hospital information systems, regional platforms, dashboards, and automated quality control mechanisms. In low-resource settings, data may be incomplete, delayed, manually entered, or difficult to integrate across vendor-specific systems, weakening their influence on planning and policy recalibration.

To avoid a 2-speed digital transformation, governance must move from passive standardization to active redistribution [12,13]. Priority measures could include (1) earmarked central fiscal transfers for last-mile connectivity and basic clinical decision support in county-level facilities; (2) provincial cloud-native architectures that provide grassroots institutions with standardized software-as-a-service applications, that is, centrally hosted software delivered through the cloud rather than installed and maintained locally without heavy local maintenance costs; (3) simplified reporting requirements combined with stronger data quality auditing so that grassroots institutions are not overwhelmed by redundant data entry mandates; (4) formal frontline feedback channels before national scaling of new standards; and (5) “east-west” informatics twinning programs linking digitally mature hospitals with lower-resource institutions. These measures would help ensure that national data infrastructure reflects the epidemiological reality of the entire population, not only the urban core [12,14].

The HCPF’s performance-linked incentives mobilize investment but can also reinforce “heavy construction, light application” (zhong jianshe, qing yingyong) [49-51]. Under Guokao pressure, hospitals may prioritize EMR functional level 5 or 6 certification to improve rankings [52]. This can encourage documentation templates or structured data fields designed to satisfy scoring algorithms rather than improve care coordination. The result may be technically compliant but clinically burdensome systems that increase redundant data entry and reduce usability [53,54]. Sustainable incentive design should therefore shift from static technical specifications, such as interface counts, toward dynamic clinical value, such as effects on clinician workload, care coordination, and patient outcomes.

Reforms could include (1) outcome-based metrics, such as reductions in readmission rates tracked via integrated EMR–public health platforms assigned substantial weight to counterbalance process-oriented indicators, with penalties for unverifiable data; (2) hybrid evaluations combining automated data logic verification with unannounced on-site audits (feixing jiancha) to detect gaming behaviors and ensure consistency between reported data and actual clinical workflows [41]; and (3) incentives to mobilize private investment, such as tax rebates for enterprise-hospital collaborations, to supplement government subsidies and reduce overreliance on direct fiscal levers. Such adjustments, consistent with the value-based evolution of the US Meaningful Use program [55], would better align the HCPF with equitable, outcome-driven digital transformation.

A critical challenge within China’s data-driven pathway is the structural asymmetry between the central government’s rapid policy iteration and the limited absorptive capacity of frontline institutions. The convergence of high-frequency technical mandates from multiple ministries, including the National Health Commission, Cyberspace Administration of China, and Ministry of Industry and Information Technology, has created a fragmented compliance landscape in which grassroots providers must navigate overlapping interface standards and redundant reporting requirements [56]. This administrative burden can contribute to reform fatigue and digital formalism, leading local implementers to prioritize technical compliance and data upload quotas over genuine workflow re-engineering [57-59]. Although top-down performance incentives have accelerated infrastructure deployment [41,42], they may also decouple data generation from clinical utility, particularly in resource-constrained subcounty settings [59-61]. Resolving this asymmetry requires shifting from rapid mobilization to sustainable integration, specifically by introducing stability windows for technical standards and granting frontline institutions greater discretion to adapt noncore functions to local needs.