The open sharing operation mechanism of health data in the digital healthcare era: A study based on grounded theory and interpretative structural modeling method

Abstract

Objective

In the evolving digital healthcare landscape, the strategic exploitation of open health data has become increasingly pivotal. The open sharing of health data is essential due to its irreplaceable role in digital health care. It is made feasible by the development of emerging technologies and the support of national policies. However, in practice, health data systems continue to face challenges such as data fragmentation, uneven data quality, poor platform connectivity, and a data regulatory vacuum, all of which hinder the effective integration of healthcare resources. This study proposes a health data governance framework to align stakeholder incentives, clarify operational responsibilities, and facilitate the compliant sharing and reuse of health data.

Methods

Innovatively employing grounded theory, this study meticulously examines the critical elements that influence health data sharing and provides a comprehensive explanation of its operational mechanisms through the Interpretive Structural Modeling – Cross-Impact Matrix Multiplication Applied to Classification model, delves into the operational mechanisms from three perspectives: the interest-driven and coordinated actions of multiple stakeholders, the open flow and secure sharing of the data ontology, and the internal support and external guarantees provided by the environmental context.

Results

This study proposes a structured data governance framework, establishing a health data sharing system built upon four core components: top-level guidance, collaborative governance, technological empowerment, and rights protection. By standardizing data quality assessment, enhancing platform interoperability, unifying data exchange protocols, and addressing critical regulatory gaps, the framework offers a systematic solution to the key challenges associated with open health data sharing.

Conclusion

This study advocates an open health data-sharing framework characterized by “top-level guidance, collaborative governance, technological empowerment, and robust rights protection.” Such a framework will optimize the economic and social value of open health data, address existing challenges, and harness future opportunities to create a more inclusive and efficient healthcare system.

Keywords

Health data digital healthcare open data sharing operation mechanism

Introduction

The convergence of national strategies such as “Digital China” and “Healthy China” has propelled electronic health data management to the forefront of healthcare agendas, driving the expansion of digital infrastructure and data-driven governance. With the shift in national policy focus from “health informatization” to “healthcare informatization,” the strategic importance of electronic health records (EHRs) and other forms of health data has been substantially magnified. Health data open sharing is considered not only a technical possibility enabled by advancements in big data, blockchain, and artificial intelligence but also a critical mechanism to support evidence-based decision-making, clinical innovation, and equitable healthcare delivery. However, despite institutional and technological progress, substantial barriers remain. Fragmentation of health data resources, poor interoperability, inconsistencies in data standards, and insufficient privacy protection mechanisms hinder the realization of integrated, high-quality data-sharing systems. Particularly, the challenges of cross-border data flows and the lack of a robust governance framework for data access and reuse raise both technical and ethical concerns. Recent studies emphasize that while patients and healthcare professionals recognize the significant value of health data sharing for research advancement and clinical innovation, there are persistent concerns regarding privacy risks, potential data misuse, and the need for robust ethical governance frameworks. Courbier et al. highlighted that rare disease patients strongly support health data sharing to promote scientific progress but simultaneously demand controlled, secure, and transparent mechanisms to protect their personal information.¹ Similarly, Goldstein demonstrated that physicians, while generally endorsing the secondary use of EHRs for research purposes, advocate for stringent privacy protections and robust informed consent processes to safeguard patient rights.²

In global discourse, public willingness to share health data is increasingly influenced by transparency, trust, and perceived benefit. Meanwhile, researchers continue to face challenges related to unclear regulations and inadequate infrastructure.³ Although various data-sharing models exist—from centralized open data platforms to privacy-preserving frameworks using federated learning—the operational mechanisms underpinning successful open health data sharing, particularly in the Chinese context, remain underexplored.⁴ Moreover, while conceptual models of digital healthcare transformation have been proposed, most lack empirical grounding in stakeholder dynamics and fail to reflect the complexity of sociotechnical conditions shaping health data governance.⁵ In this regard, there is a notable gap in systematic studies that integrate both stakeholder insights and systems analysis to construct a functional and scalable data-sharing framework.

This study addresses these gaps by combining grounded theory with Interpretive Structural Modeling (ISM) and Cross-Impact Matrix Multiplication Applied to Classification (MICMAC) analysis to investigate the multilevel operational logic of open health data sharing in China. Through a qualitatively driven, structure-integrated approach involving in-depth interviews and system modeling, we identify 23 key factors and their interrelationships that influence data-sharing practices. The goal is to establish a robust and context-sensitive framework guided by the principles of top-level design, collaborative governance, technological empowerment, and rights protection. By mapping the practical challenges and theoretical dimensions of health data openness, this study responds to the need for a more comprehensive and mechanism-focused analysis, contributing to both academic understanding and policy innovation in digital health governance. By integrating relevant theories from data science and adopting a qualitative research approach grounded in empirical evidence, this research proposes a structured operational framework, providing theoretical extensions and practical references to enhance the development of health data open sharing governance, offering critical insights into how health data can be shared effectively, securely, and equitably in the digital healthcare era.

Background

The strategic importance and challenges of public data sharing

Public data, as a vital strategic resource accessible to all members of society, play a fundamental role in the construction of Digital China. The strategic importance of public data sharing has been increasingly recognized both nationally and internationally, serving as a key driver for societal advancement, healthcare innovation, and public trust.^6,7 Initiatives such as Digital China and Healthy China highlight the government's commitment to building robust data infrastructures that enable wide-reaching benefits. Effective data sharing involves multiple aspects, including data quality, accessibility, privacy protection, and legal frameworks. Data quality is particularly critical, as the accuracy and completeness of shared data directly influence its usability and the reliability of research outputs. Interdisciplinary research has increasingly focused on data management within the information ecology, emphasizing the need for efficient integration of diverse data sources to support informed decision-making.⁸ Researchers have conducted studies on the construction of government data open ecosystems.⁹ At the same time, it was found that public and patient participation plays a key role in constructing these ecosystems. Transparent communication and education can enhance public trust and willingness to participate in data sharing, which is crucial for establishing a healthy data-sharing ecosystem.¹⁰

As big data and artificial intelligence technologies advance, there has been a substantial increase in the volume of public data, along with notable improvements in the speed, scope, and depth of data development and utilization.^11,12 The construction of platforms dedicated to public data openness has gradually become an important component of the sharing operation mechanism. However, despite these advancements, substantial challenges remain. Fragmentation of data resources, poor interoperability across information systems, inconsistencies in data standards, and insufficient privacy protection frameworks continues to hinder the realization of integrated, high-quality data-sharing systems.^13,14

Interoperability, in particular, remains problematic. Although frameworks such as HL7 Fast Healthcare Interoperability Resources (FHIR) have improved cross-platform data exchange, widespread implementation challenges persist globally.^14,15 Moreover, public skepticism regarding data ownership, transparency, and potential misuse has intensified by high-profile data breaches and regulatory gaps.^13,16

Security concerns have become critical points affecting the advancement of public data sharing. The demand for stronger data quality controls, enhanced privacy protections, and the standardization of data practices remains a pressing challenge.¹⁷ Technical solutions such as advanced data security technologies¹⁸ and ongoing improvements in regulatory policies19 are increasingly viewed as essential measures to protect citizens’ privacy while enabling responsible data sharing at scale.

Understanding the characteristics and value of health data

Health data, also called EHRs, have become a crucial component of the digital healthcare ecosystem. It offers substantial value for clinical decision-making, healthcare management, biomedical research, and public health interventions. As early as 2003, Maisie et al. developed a web-based personal health record (PHR) referral system for documenting orthopedic patient referrals.²⁰ In 2010, Koufi et al. created a chronic disease management system based on personal health data.²¹ Electronic health records are the core carriers of health data, encompassing comprehensive patient information such as medical histories, diagnostic results, treatment regimens, imaging reports, and medication records. With its inherent medical and research value, health data have gradually become a research hotspot in constructing medical and health information systems at home and abroad. Developed countries have generally studied EHRs earlier than China and have achieved specific research results, mainly in practice, technology, and open utilization of EHRs. Willis demonstrated the application of EHRs in public health infectious disease surveillance, emphasizing the advantages of EHRs data in disease symptoms, laboratory results, and medical treatment, and pointing out its important role in public health response and monitoring.²² Jones et al. explored the sustainable role of EHRs in training and guiding mental health rehabilitation in community-managed mental health organizations in Australia.²³ Schopf et al. investigated the support EHRs systems provide to doctors in performing basic clinical tasks in three Norwegian hospitals.²⁴ Duan et al. proposed an efficient and privacy-preserving distributed algorithm for protecting EHRs through extensive simulation studies and experiments on the University of Pennsylvania Health System.²⁵ The rapid development of information technologies, particularly artificial intelligence, big data analytics, and blockchain, has further amplified the potential value of health data. Secondary use of health data enables large-scale epidemiological studies, predictive modeling for disease outbreaks, healthcare cost control, and policy evaluation.² However, the fragmented storage of health data across diverse institutions and the lack of standardized interoperability frameworks significantly impede its effective utilization.²⁶ Privacy protection emerges as a paramount concern in health data sharing. The high sensitivity of health data demands stringent technical safeguards and necessitates the establishment of robust ethical and legal frameworks to ensure patient autonomy and prevent data misuse. Recent technological innovations, including blockchain-based secure sharing models²⁷ and privacy-enhancing techniques such as homomorphic encryption,²⁶ offer promising avenues for enabling responsible data sharing while preserving individual privacy.

Moreover, decentralized learning models like federated learning have been proposed as effective strategies to enable collaborative research across institutions without directly transferring sensitive patient data.²⁸ These models address both the technical barriers to interoperability and the ethical concerns regarding data sovereignty and consent.

Nevertheless, while technical solutions advance, ensuring data quality and consistency remains challenging. Health data are often incomplete, inconsistent, and context-dependent, undermining their usability in clinical and research settings. Furthermore, variations in data governance policies across jurisdictions complicate cross-institutional and cross-border data-sharing initiatives.^1,29 Overall, health data represent a high-value, high-risk resource whose optimal use requires integrating technological innovation, governance reforms, and stakeholder collaboration.

Addressing the challenges and gaps in health data open sharing

The drive to open health data for secondary uses has gained substantial momentum globally. Digital health platforms, national consortia, and international initiatives have been developed to facilitate health data sharing across institutions and borders. Nevertheless, realizing large-scale, sustainable, and ethically responsible health data sharing remains fraught with complex challenges.^30,31

Since 2016, China has accelerated the comprehensive integration of health and medical big data, with particular emphasis on open governance³² and regional sharing initiatives.^33,34 Theoretically, health data sharing refers to the behavior by which data controllers provide collected or processed data to third parties through transactions or exchanges.³⁵ In exploring open sharing mechanisms, scholars have analyzed the operational mechanisms empowered by emerging technologies such as blockchain from a technical perspective,³⁶ investigating the construction and operational mechanisms of data open sharing platforms.³⁷ Others have proposed multichannel approaches to promote the construction of public data open mechanisms,³⁸ integrating China's data industry development and data open policies.³⁹

Despite these advances, fragmentation remains a significant barrier. Health data are often siloed across disparate institutions, and technical interoperability is limited even with the advent of standards such as HL7 FHIR. Public trust in data-sharing initiatives remains fragile, shaped by concerns over data ownership, privacy risks, and potential misuse. Evidence-based healthcare relies on health data from diverse sources to inform decision-making across different domains. It is increasingly recognized that health data are highly sensitive and their subsequent reuse in research may raise privacy issues for individuals who provide the data. For instance, Zhang and Liu integrated social exchange theory and commitment-trust theory to analyze the moderating effects of trust on relational commitment and its antecedents from the perspective of privacy willingness, exploring factors that influence continuous sharing willingness.⁴⁰ Additionally, scholars have examined the present landscape, ongoing challenges, and future developments in data security governance, focusing on applying privacy protection measures in data openness.¹⁷ Subsequently, various scholars delve into maximizing the utilization of health data for scientific research while ensuring privacy protection,⁴¹ achieving deep integration and sharing of data.^42,43

Based on scholars’ research at home and abroad, it is evident that significant achievements have been made in the areas of public data open sharing, health data open sharing, factors affecting health data sharing, and open data governance. Especially after the outbreak of the COVID-19 pandemic, scholars have made substantial progress in theoretical research,^44–46 practical exploration,^47,48 and privacy protection concerning health data open sharing. Discussions encompassed data-sharing policies and standards,⁴⁹ technological applications,⁵⁰ management optimization,⁵¹ system platform construction,⁵² and open utilization frameworks.⁵³ Nonetheless, critical challenges persist in the practice of EHRs sharing. Issues such as low public participation and willingness at the data acquisition stage, lack of standardization, unclear responsibilities during the management phase, and severe security technology challenges in the open utilization phase have been widely reported. By reviewing the literature, it is found that existing research on health data open sharing has the following limitations: (1) In terms of research subjects, there are relatively few studies on health data open sharing, mainly focusing on the Open Government Data and Open Medical Data; (2) In terms of research content, there is a lack of systematic research, mostly staying at the level of practical and ethical challenges and macrolevel countermeasures in sharing, lacking exploration of the elements and operational mechanisms of health data open sharing, which is the core and key to effectively solving open governance issues. Therefore, in response to the current situation where existing research is not comprehensive and in-depth, this study aims to analyze the elements of health data open sharing in digital healthcare era, combined with relevant theories in data science, and on this basis, explore the overall framework of health data open sharing mechanisms to provide important extensions and references for enhancing the relevant theories of health data open sharing.

Methods

This study began with a review of the literature to establish a theoretical foundation and construct a methodological framework. During the research process, interviews were conducted with participants from diverse professional backgrounds to explore, from multiple perspectives, the factors influencing the open sharing of health data. Grounded theory was then employed to code and analyze the interview data, leading to the identification of key influencing factors. Reliability and validity tests were performed on the coding results to ensure the scientific rigor and credibility of the findings. Based on these results, the ISM method was applied to determine the hierarchical structure of the factors, and the MICMAC analysis was used to identify the driving and dependent relationships among them.

Context

The development of the interview outline in this study was informed by a systematic review of literature related to open data, health information sharing, and EHRs. Relevant academic and policy sources were retrieved from databases such as Web of Science, Elsevier, Emerald, Springer, IET Electronic Library, CNKI, and scholarly search engines such as Google and Baidu. The research team analyzed these materials to identify recurring themes, conceptual gaps, and practical challenges documented in prior studies. These findings provided the theoretical foundation for designing the semistructured interview outlines.

Building on this foundation, the study was conducted from February 2023 to May 2024, employing grounded theory and the ISM-MICMAC model to systematically identify, classify, and structure the factors influencing health data-sharing mechanisms in the digital healthcare era. The research team consisted of one professor and two students. The professor was a male with a PhD in management, and the female students were master's degree candidates. The interviews were primarily conducted by two master's degree candidates, and the raw data from the interviews were coded by three coders. The researchers had prior experience in open health data sharing and PHRs. The authors developed the interview outline used in this study to guide the semistructured interviews with different target groups, including healthcare institution professionals, government staff, professional technicians, and general users. These outlines were designed to explore participants’ experiences, perceptions, and attitudes towards health data open sharing in the context of digital health care. Our study ensured that participants were fully informed and participated voluntarily, with all data kept strictly confidential. The research collected only anonymous perspectives and suggestions regarding health data, without involving any personally identifiable information or sensitive content. Therefore, we employed a verbal informed consent process, and ethical approval was exempted.

Data collection and participant population

Qualitative research uses fewer cases, typically prioritizing detailed richness over larger sample sizes. Grounded theory imposes no strict limitations on sample size, though most related studies typically involve 10–30 samples. Moreover, grounded theory underscores the importance of diverse data sources and representative sample selection that are evenly distributed to reflect real societal conditions. In this study, interviewees were carefully chosen from among those involved in the acquisition, management, access, and utilization of health data. Accordingly, this study selected interview samples following theoretical saturation and targeted four distinct groups: healthcare professionals, government officials, technical experts, and general users. (The interview outlines are provided in Appendix.) To ensure high reliability and validity of the interview outlines, preinterviews were conducted with five participants before the formal interviews. These preliminary interviews aimed to ensure the clarity and logical coherence of the interview outlines, as well as the appropriateness of the interview duration. The process involved using preliminary semistructured interview outlines to interview volunteers. Based on feedback from the interviews and observations during the pilot study, revisions were made to the preliminary interview outlines, resulting in the refined semistructured interview outlines for the formal interviews. The content of the interview outline is tailored according to the interviewees’ professional background and personal experiences, setting different focal points accordingly.

Leveraging the convenience of new media tools in the digital era, interviewees were drawn from diverse regions and age groups. The participants included officials from health commissions and cybersecurity offices, healthcare workers, technical professionals, and other members of society, totaling 22 individuals. The detailed information is presented in Tables 1 and 2.

Table 1.

Demographic profile of professional participants.

Number	Gender	Age	Education level	Academic discipline	Department and position
1	Female	27	Master	Dermatology and Venereology	Resident Physician, Department of Dermatology, Xiangya Hospital
2	Female	30	Bachelor	Traditional Chinese Medicine	Former Gynecologist, Current Product Manager
3	Male	28	Master	Accountant	Administrative Clerk, Outpatient Office, Hefei BOE Hospital
4	Female	24	Bachelor	Health Economics	Pharmaceutical Marketing Department
5	Female	50	Bachelor	Pharmacy	Former Chief Physician, Current Pharmacist in Provincial Hospital
6	Female	35	Bachelor	Information Management	Administrative Staff of a University
7	Male	26	Bachelor	Linguistics and Literature	Government Office Secretary
8	Male	37	Bachelor	Computer Science and Technology	Information Center, The Second Norman Bethune Hospital of Jilin University
9	Male	30	Master	Law	Medical Insurance Department, The Second Norman Bethune Hospital of Jilin University
10	Female	51	Doctorate	Internal Medicine	Director, Department of Internal Medicine, The Second Norman Bethune Hospital of Jilin University
11	Male	32	Doctorate	Surgery	Department of Surgery, The Second Norman Bethune Hospital of Jilin University
12	Male	30	Bachelor	Information Management	Government Staff
13	Female	45	Master	Linguistics and Literature	Government Staff

Table 2.

Demographic profile of nonprofessional participants.

Participant	Gender	Age	Education level	Academic discipline	Department and position
1	Female	41	Master	Computer Science	Unemployed
2	Female	26	Bachelor	Human Resource Management	HR Specialist, Human Resources Department
3	Female	31	Master	Information Management	Information Designer
4	Female	24	Master's candidate	Business Administration	Student, School of Management and Economics
5	Male	28	Doctoral candidate	Educational Studies	Student, Faculty of Education
6	Female	30	Bachelor	Chinese Language and Literature	E-commerce Operations Specialist, E-commerce Department
7	Female	41	Bachelor	International Economics and Trade	Big Data Product Operations Manager
8	Female	36	Bachelor	Architecture	Project Lead, Design Department
9	Female	26	Master's candidate	Archival Science	Student, Information Management Department

The interviews were conducted using a combination of online and offline methods. Offline interviews were primarily conducted in Changchun, Jilin Province, while online interviews targeted nationwide interviewees, utilizing platforms such as WeChat and Tencent Meeting. Each interview lasted between 20 and 30 min. After each session, the audio recordings were promptly transcribed, resulting in approximately 140,000 words of interview data.

Specifically, the interviews can be divided into three stages: In the initial stage of sample collection, open sampling was employed to select interviewees of varying ages, job roles, regions, and work experience, aiming to obtain detailed descriptions of the elements of the open sharing mechanism for health record data. During this stage, it was observed that recent graduates and younger employees in the healthcare system or government agencies demonstrated relatively strong awareness and application abilities regarding data openness and sharing elements. In the mid-stage of data collection, following the principles of relational and variance sampling, younger professionals were targeted explicitly for interviews. Additionally, the number of male participants was increased to balance the gender ratio and minimize the influence of gender-based cognitive and behavioral differences on the research outcomes. In the final stage of data collection, interviewees who could refine or revise the theoretical model were selected based on the principle of discriminatory sampling. These included individuals less represented in earlier interviews, such as older professionals, those with longer work experience, and higher-ranking personnel from healthcare institutions and technical fields.

Thematic approach

In the methodological application and data analysis process, the grounded theory method was first used to summarize and identify the key factors of the health data open sharing mechanism. Subsequently, the ISM-MICMAC methodology was used to analyze the operational principles of these factors and the relationships among them. Building on this foundation, the OAIS model was incorporated to provide a systematic methodological perspective for health data governance. This integration led to the development of a logical framework for the open sharing mechanism of health data in the digital healthcare era, along with corresponding recommendations for its practical implementation. The specific analytical process is illustrated in Figure 1.

Figure 1.

Integrated methodological process for constructing the health data open sharing mechanism.

Grounded theory coding

Grounded theory, developed by sociologists Glaser and Strauss in 1967, is a qualitative research method that emphasizes deriving core concepts from raw data. Through continuous comparison, summarization, and analysis, this method explores the relationships among conceptual elements, thereby facilitating theory construction from the ground up. This approach is especially advantageous in research fields that lack established theoretical frameworks and require the development of new theories. Hudani argues that adopting grounded theory allows for a theoretical explanation of complex phenomena.⁵⁴ Constant data comparison, along with the three layers of analysis—open coding, axial coding, and selective coding—enables the identification of key influencing factors and clarifies deficiencies in collaboration and coordination processes between systems. This section utilizes the grounded theory method to systematically identify and extract the key elements of open health data sharing in the digital healthcare era. This foundational analysis sets the stage for subsequent examinations of mechanisms and the construction of related systems.

The interview data were analyzed using grounded theory methodology, using the qualitative analysis software NVivo to conduct a three-stage coding process. First, researchers deconstruct all original data and perform a sequence comparative analysis of the textual materials. The goal is to minimize personal biases and preconceptions from existing research, allowing for a fresh perspective in data recombination. By utilizing vocabulary from the original statements, researchers conceptualize and categorize the data to preserve the initial viewpoints embedded within the textual data. This process led to the identification of 70 initial concepts and 23 initial categories from the interview data. Building upon the categories identified through open coding, continuous exploration and analysis were undertaken to uncover implicit and explicit logical relationships among these categories. Through inductive and deductive methods, this analysis identified four primary categories and 10 secondary categories. Selective coding is employed to refine and integrate the principal categories derived from axial coding. This process synthesizes various theoretical components into a cohesive framework or storyline. Based on the results of open coding and axial coding analysis, this study identifies the core storyline as follows: “In the digital healthcare era, the open sharing mechanism of health data is composed of motivational elements, conditional elements, and operational elements.” Detailed coding results are shown in Table 3.

Table 3.

Results of the three-level coding in grounded theory.

Selective coding	Axial coding	Open coding
Motivational elements	B1 Individual Needs	A1 Disease Treatment
	B1 Individual Needs	A2 Health Management
	B2 Institutional Needs	A3 Medical Research
	B2 Institutional Needs	A4 Medical Efficiency
	B3 Social Needs	A5 Equalization of Medical Resources
	B3 Social Needs	A6 Intelligent Healthcare
Conditional elements	B4 National Support	A7 Legal Promotion
	B4 National Support	A8 Policy Orientation
	B5 Material Conditions	A9 Professional Talents
		A10 Basic Infrastructure
		A11 Financial Support
	B6 Technical Conditions	A12 Data Collection Technology
		A13 Data Management Technology
		A14 Data Preservation Technology
		A15 Data Utilization Technology
Operational elements	B7 Inter-agent Coordination	A16 Information Interconnectivity
	B7 Inter-agent Coordination	A17 User Benefits
	B8 Inter-object Coordination	A18 Data Standards
	B8 Inter-object Coordination	A21 Emergency Response
Assurance elements	B9 Legal Assurance	A20 Privacy Protection
	B9 Legal Assurance	A19 Data Security
	B10 Organizational Assurance	A22 Management System
	B10 Organizational Assurance	A23 Reward and Punishment System

Theoretical saturation testing involves adding interview samples and analyzing their characteristics based on the original interview data. Saturation is achieved when no new concepts or categories emerge while coding additional samples. In this study, a total of 22 individuals were interviewed. Nineteen interview transcripts were randomly selected from the sample for coding analysis and model construction, while the remaining three were reserved for theoretical saturation testing. Concepts were extracted from the reserved transcripts using the grounded theory's three-level coding process. The analysis showed that this study did not reveal any new significant concepts or categories that would modify the original core categories. Additionally, no new relationship structures emerged among the categories. Therefore, it can be concluded that the categories refined using grounded theory in this study have reached a state of theoretical saturation.

The Kappa coefficient is a statistical measure used to assess the level of agreement between two or more raters, enabling the quantification of the difference between observed concordance and chance-expected concordance, thereby reflecting the reliability of coding results. This study analyzed the two coding results regarding the open sharing of health data. The calculated Kappa coefficient was above 0.8, and the coding consistency scores exceeded 90%, indicating a high level of reliability in the coding.⁵⁵ Discrepancies in coding were resolved through team discussions and reexamination of the original transcripts to reach consensus.

Application of ISM-MICMAC

The ISM, proposed by Warfield, is primarily used to analyze elements within complex systems and their interrelationships.⁵⁶ The MICMAC builds on the principles of matrix multiplication to assess the degree of influence among system elements by constructing and analyzing a cross-impact matrix. The ISM-MICMAC method is well-suited for explaining and interpreting complex system issues. It can be effectively applied to analyze the elements involved in the open sharing of health data, thereby identifying the key influencing factors.

First, the open sharing of health data constitutes a complex system involving multiple interrelated factors. Compared with deep learning-based clustering or predictive analytics, the ISM-MICMAC method enables the construction of a structural model that captures the interactions among elements, clarifies causal relationships, and reveals the hierarchical structure and operational mechanisms of system components.⁵⁷ Second, the ISM-MICMAC method offers greater interpretability and transparency, making it particularly appropriate for studies on theory development and mechanism explanation. Finally, unlike data-driven approaches that require large volumes of high-quality training data, ISM-MICMAC relies on expert knowledge and experience, making it suitable in contexts where data are limited or historical samples are unavailable. This reliance also enhances the credibility and accuracy of research findings.⁵⁸

The preceding chapter identified 23 key factors influencing the open sharing of health data using grounded theory. However, these factors do not operate in isolation. There are intricate interactions between them. This chapter explores the complex interactions among these factors using the ISM-MICMAC method. Initially, an ISM was established to elucidate hierarchical relationships among the factors involved in health data open sharing. Subsequently, the MICMAC method is applied to create a “driver-dependency quadrant distribution analysis diagram.” This diagram facilitates the identification and representation of each factor's driving forces and dependency values. By analyzing the elements within each cluster, the study explores their interactive pathways and impact on the system's overall efficacy.

The adjacency matrix was constructed through expert scoring, with invited experts evaluating the interrelationships between elements. During the study, 22 valid scoring forms were collected from experts with relevant work experience and research backgrounds across various fields. A scoring form was independently designed for this purpose. The researchers computed the aggregated expert scores based on a weighted summation approach. According to the 23 influencing factors and their binary relationships listed in Table 1, the relationships were represented in the matrix using binary values “0” and “1.” Subsequently, by drawing a directed graph to illustrate the interactions among elements, the final 23 × 23 adjacency matrix was constructed. This process led to the creation of the adjacency matrix A, as shown in Figure 2.

Figure 2.

The adjacency matrix A.

Upon establishing the adjacency matrix, the reachable matrix M was derived through matrix transformation and calculations, as illustrated in Figure 3.

Figure 3.

The reachable matrix M.

To extract the correlations among the factors, the antecedent set Q(Aj), the reachable set R(Ai), and the standard set R(Ai)∩Q(Aj) were derived from the simplified reachable matrix M, as illustrated in Figure 4. Hierarchical division was then conducted by predefined rules.

Figure 4.

Heatmap of reachable and prior sets and intersection.

After five iterations, the final hierarchical decomposition of health data open sharing factors table was produced, as shown in Table 4.

Table 4.

Hierarchical decomposition of health data open sharing factors.

Hierarchy	Factors
Hierarchy 1(Top Hierarchy)	A16,A17
Hierarchy 2	A19,A20
Hierarchy 3	A12,A13,A14,A15,A21,A23
Hierarchy 4	A18,A22
Hierarchy 5	A10
Hierarchy 6	A9,A11
Hierarchy 7	A7,A8
Hierarchy 8(Bottom Hierarchy)	A1,A2,A3,A4,A5,A6

Finally, in line with the hierarchical structure theory of ISM, the factors were further categorized into three distinct levels: surface, middle, and root. Consequently, based on the reachable matrix and the hierarchical division results, the 23 factors were classified according to their respective levels, with the interactions between factors represented by directed lines. In this representation, the starting point of a directed line indicates the influencing factor, while the endpoint signifies the influenced factor. This framework was used to construct the interpretative structural model of health data open sharing factors, as shown in Figure 5.

Figure 5.

Hierarchical structure of health data open sharing factors.

Based on the reachable matrix M, each factor's driving force and dependency values are calculated using formulas (1) and (2). The results allow for constructing a “driving force-dependency” scatter plot, where the ordinate represents dependency, and the abscissa represents driving force. The scatter plot is then divided into four distinct groups, each representing different characteristics of the factors, as illustrated in Figure 6:

D_{i} = \sum_{j = 1}^{n} a_{i j}^{m}, (i = 1, 2, 3, \dots, n)

(1)

R_{i} = \sum_{i = 1}^{n} a_{i j}^{m}, (j = 1, 2, 3, \dots, n)

(2)

Figure 6.

Driver-dependency quadrant distribution of health data open sharing factors.

Results

Extraction of elements for health data open sharing

According to the grounded theory thorough analysis, it is determined that decisions regarding the open sharing of health data are made by data-sharing entities after a comprehensive evaluation of the needs for sharing, the perceived value of the sharing, and the perceptions of data security. These decisions are influenced by the technical, material, and environmental conditions surrounding data sharing and hinge on the availability of the data object. From the perspective of digital health care, this study identifies the motivational, conditional, operational, and assurance elements of health data open sharing, as illustrated in Figure 7.

Figure 7.

Health data open sharing elements model.

Based on the element diagram of health data open sharing presented in Figure 7, the key components of health data sharing can be categorized into four dimensions: shared motivation, shared conditions, shared operation, and shared security. Through the ISM-MICMAC methodology, the interrelationships and hierarchical structure among these elements can be further analyzed, identifying driving and dependent factors within the system and revealing the functional pathways of each element in the governance framework and their impact on the overall system. On this basis, the OAIS model is introduced to segment the health data life cycle into distinct stages, integrating the elements depicted in the diagram into the whole life cycle process of data—from ingestion, storage, management, preservation, transmission, and processing. Specifically, shared motivation reflects the original demand driving the data production and collection stage; shared conditions correspond to the data management and preservation stage, providing institutional, material, and technological support for open sharing; shared operation focuses on coordination mechanisms during the data circulation and usage stage; and shared security spans the entire data life cycle, ensuring risk control and regulatory compliance. The analysis of these elements contributes to the construction of a dynamically evolving and life cycle-controllable open data-sharing mechanism, thereby advancing high-quality governance and sustainable utilization of health data.

Factor analysis and mechanistic elucidation in health data open sharing

This chapter builds upon the factor model of health data open sharing by using the ISM-MICMAC method to conduct an in-depth analysis of the connections among these factors. The insights gained aimed to provide a reference for the development of mechanisms for health data open sharing.

ISM model construction and results analysis

According to this hierarchical structure diagram derived from the ISM model, it is evident that:

Surface-level factors represent the most direct influence on the health data open sharing, functioning as the highest-level factors. Notably, A19 (Data Security) is a prerequisite for enabling A20 (Privacy Protection), A16 (Information Interconnectivity), and A17 (User Benefits), serving as the cornerstone of health data governance and legal safeguards. Under the condition of A19 (Data Security), A20 (Privacy Protection) plays a significant role in safeguarding the privacy of relevant subjects, collectively fostering the establishment of A16 (Information Interconnectivity). This combination facilitates the realization of substantial social and economic benefits, manifesting as improved products and services, thus fulfilling A17 (User Benefits).

Middle-level factors, driven by root-level factors, indirectly influence the open sharing of health data through surface-level factors. Influenced by external environmental elements such as A7 (Legal Promotion) and A8 (Policy Orientation), the internal factors within the health data open sharing ecosystem are constantly improving. This level includes several internal operating factors such as A9 (Professional Talents), A10 (Basic Infrastructure), and A11 (Financial Support), all of which support and ensure the balanced development of health data open sharing. These factors, in turn, impact subsequent factors such as A22 (Management System), A18 (Data Standards), and A23 (Reward and Punishment System). Furthermore, A12 (Data Collection Technology), A13 (Data Management Technology), A14 (Data Preservation Technology), and A15 (Data Utilization Technology) are pivotal in the data governance processes of health data open sharing operations. A21 (Emergency Response) simultaneously safeguards the physical and information security of health data resources, enhancing their utility.

At the root level, families and individuals represented by A1 (Disease Treatment) and A2 (Health Management), medical institutions represented by A3 (Medical Research) and A4 (Medical Efficiency), along with government departments represented by A5 (Equalization of Medical Resources) and A6 (Intelligent Healthcare), collectively form the multisubjects of health data open sharing. Their interests and demands for the benefits of health data open sharing are continuously considered and fulfilled in practice, ultimately fostering an open environment primarily characterized by A7 (Legal Promotion) and A8 (Policy Orientation). Together, these elements constitute the root factors of health data open sharing, exerting a wide-ranging impact on other factors, and directly or indirectly influencing middle-level or surface-level factors.

Cross-impact matrix multiplication applied to classification model construction and results analysis

Based on the dependency-driving force quadrant distribution in Figure 6, the following can be observed:

Quadrant I: Autonomous Factors

This quadrant includes four factors: A18 (Data Standards), A21 (Emergency Response), A10 (Basic Infrastructure), A22 (Management System), and A23 (Reward and Punishment System). The factors in this quadrant exhibit low driving force and low dependency values, suggesting that their ability to influence and be influenced by other factors is relatively limited. However, they do impact other factors and maintain relative stability.

Quadrant II: Dependent Factors

This quadrant includes 10 factors: A12 (Data Collection Technology), A13 (Data Management Technology), A14 (Data Preservation Technology), A15 (Data Utilization Technology), A16 (Information Interconnectivity), A17 (User Benefits), A19 (Data Security), and A20 (Privacy Protection). The factors in this quadrant have low driving force but high dependency values, indicating they are highly susceptible to other factors.

Quadrant III: Linkage Factors

This quadrant is characterized by factors with both strong dependency and driving force. Altering these factors can lead to changes in other factors. Thus, their relationships with other factors are relatively complex and unstable. No such factors are present among those affecting the open sharing of health data.

Quadrant IV: Independent Factors

This quadrant includes nine factors: A1 (Medical Needs), A2 (Health Needs), A3 (Medical Research), A4 (Medical Efficiency), A5 (Equalization of Medical Resources), A6 (Intelligent Healthcare), A7 (Legal Promotion), A8 (Policy Orientation), A11 (Financial Support), and A9 (Professional Talents). The factors in this quadrant exhibit high driving force and low dependency values, indicating that they are less dependent on other factors but have a significant driving force, which drives other factors within the system. Therefore, these factors must be addressed first to facilitate the resolution of other factors. Usually, these are the most critical factors in the system, playing a crucial role in the health data open sharing and exerting sustained, high-intensity influence on other factors. Effectively addressing the factors in this quadrant will significantly promote the resolution of other related factors.

Comprehensive analysis of ISM-MICMAC

The analysis of the Interpretative Structural Model (ISM) and Matrix Impacts MICMAC reveals a clear correspondence and complementation between the ISM's hierarchical structure divisions and the MICMAC method's clustering of factor clusters.

Firstly, the factors located in quadrant I of the MICMAC distribution map are entirely situated within the middle level of the ISM model. These factors include A21 (Emergency Response), A23 (Reward and Punishment System), A22 (Management System), A18 (Data Standards), and A10 (Basic Infrastructure). Positioned at the middle level, these five factors exhibit relative stability and support other factors within the model. Influenced directly by root-level factors and impacting surface-level elements, their role is crucial in developing and refining mechanisms for open sharing of health data.

Secondly, the factors in quadrant II of the MICMAC distribution map are primarily located at the surface level, with some positioned in the middle level of the ISM model. Factors such as A12 (Data Collection Technology), A13 (Data Management Technology), A14 (Data Preservation Technology), and A15 (Data Utilization Technology) are found in the upper portion of the middle level. These factors directly influence surface-level factors, including A16 (Information Interconnectivity), A20 (Privacy Protection), A17 (User Benefits), and A19 (Data Security). The factors in this quadrant exhibit low driving force but high dependency, necessitating support from other factors. Surface-level factors are relatively more straightforward to identify than middle and root-level ones. However, effectively addressing these surface-level factors necessitates tracing them back to their origins and focusing on enhancing the underlying factors to drive improvements.

Finally, the factors in quadrant IV of the MICMAC distribution map are mainly located at the root level of the ISM model. These factors include A1 (Disease Treatment), A2 (Health Management), A3 (Medical Research), A4 (Medical Efficiency), A5 (Equalization of Medical Resources), A6 (Intelligent Healthcare), A7 (Legal Promotion), and A8 (Policy Orientation). Other factors do not influence these root-level factors but exert influence on the middle level and demonstrate a strong driving force. They should be considered primary targets in suggestions and countermeasures. The remaining factors, A9 (Professional Talents) and A11 (Financial Support), are influenced by root-level factors and directly and significantly impact subsequent factors. These factors should be key points of intervention and should be emphasized in the construction of mechanisms for health data open sharing.

The previous analysis shows that the internal operational mechanism of health data open sharing primarily revolves around data value as the core driving force. It relies on the flow of information during the input and output stages. This mechanism is shaped by the internal motivation of stakeholders in their professional activities or driven by self-interest, the external pull of user demands, and the influence of other environmental factors. Therefore, the operational mechanism of health data open sharing in the digital healthcare era is specifically manifested through the coupling of three dimensions: “data entity,” “environmental object,” and “stakeholder subject,” all empowered by digital healthcare technology. This mechanism is achieved through the forward information flow of health data entity during the input stage and the feedback information flow during the output stage, facilitating data open governance and value release, as illustrated in Figure 8.

The interest-driven and coordinated actions of multiple subjects constitute the foundational forces for the efficient open sharing of health data. As a critical linkage in a systematic framework, health data open sharing requires robust cross-departmental collaboration during the data opening stage and extensive participation from commercial entities, social organizations, and the public during the data utilization stage. The various demands of the government, medical institutions, enterprises, and individual citizens constitute the root-level driving forces behind health data open sharing. To ensure efficient data sharing and utilization, these stakeholders’ interests must be meticulously considered and adequately addressed.

A harmonious and stable environment for data openness is essential for the sustainable development of health data-sharing practices and the effective functioning of data openness mechanisms. From an external perspective, establishing and improving policy, laws, and regulations are critically important, fundamentally guiding and regulating the long-term development of health data-sharing work. From an internal perspective, factors such as organization, personnel, technology, and funding are vital in ensuring and supporting the comprehensive and stable development of middle-level activities in health data sharing. These factors should be given focused attention in the construction of mechanisms.

The health data entity and its flow are crucial for sustaining the regular operation and development of the open sharing mechanism. As the nexus where various stakeholders produce, transmit, utilize, and consume data, health data are voluminous and possesses high-value attributes derived from data matching processes. Data generation, preservation, sharing, openness, acquisition, utilization, value addition, and regeneration constitute forward data flow. Meanwhile, feedback information flow is shaped by data application requirements, suggestions, and corrections for data applications, user acquisition, and demand statistical feedback. Through this continuous cycle, the health data entity establishes a bidirectional connection within the data flow, continually expanding the depth and breadth of health services. In the era of big data, diverse data governance and security management technologies are essential for enabling the flow and sharing of data. These technologies serve as critical support that bridges the root and surface levels. Thus, they should be given significant attention in the following construction of mechanisms.

Figure 8.

The operating mechanism for health data open sharing elements.

Discussion

A review of the factors and operational mechanisms of health data open sharing reveals that data open governance is a systematic and dynamic process involving continuous stages and multiple subjects. This chapter synthesizes the analytical findings from the ISM-MICMAC model and the top-level architecture of the OAIS model, which supports the long-term preservation and utilization of open data. By analyzing the entire life cycle of data—from creation to final disposal through records life cycle theory—the health data open sharing system can be divided into four primary components: top-level guidance, collaborative governance, technical empowerment, and rights and interests protection. These components collectively contribute to the construction of a robust governance framework. This framework is designed to facilitate the ecological governance of open data, effectively balancing and integrating the various phases and participants involved in the process. The architecture and functional interactions of this system are illustrated in Figure 9.

Figure 9.

Health data open sharing system.

This study integrates the OAIS model with the ISM-MICMAC methodology, incorporating health data life cycle management and structural dependency analysis into the framework for open data sharing. This approach not only enhances the control over the entire life cycle of health data—from collection, management, and preservation to sharing and disposal—but also enables the precise identification of hierarchical relationships and influence pathways among key factors, thereby strengthening the model's structural coherence and strategic orientation. Compared with the commonly used ISM-DEMATEL model, the method adopted in this study demonstrates greater systematicity and logical consistency in mechanism construction, offering a more precise depiction of the evolutionary pathways in the operation of health data-sharing systems.

Top-level guidance system

Institutionalized and standardized guidance and regulation are crucial prerequisites for the sustainable development of health data open sharing practices in the long run. National top-level guidance and promotion through legal norms, policy directives, and an open environment are essential and effective measures to achieve this goal. Furthermore, the scalability of health data openness and sharing must be considered. As digital healthcare advances, health data sharing is expected to expand toward larger-scale data ecosystems, such as national-level databases. Therefore, it is essential to consider multiple influencing factors, including data availability, processing efficiency, and security. This study proposes a systematic framework for the health data open sharing, encompassing four key dimensions: driving forces, enabling conditions, operational mechanisms, and safeguard measures. The framework demonstrates a high level of domain specificity and supports comprehensive, whole-life cycle management of health data, while emphasizing the protection of individual rights and multistakeholder collaborative governance in the context of medical data. However, compared with well-established open data frameworks such as the Open Data Initiative, this framework still exhibits limitations, including relatively underdeveloped legal mechanisms and limited alignment with international standards.⁵⁹ Therefore, future efforts should draw on best practices from related research—such as interdepartmental collaboration, standardized interface design, and legal integration—to enhance the regulatory robustness, interoperability, and sustainability of health data governance in local contexts.

Currently, there are no unified laws and regulations on the open governance of health data in China. Health information sharing practices are primarily guided and regulated through guidelines on “The Internet Plus Government Services Initiative” and interim measures. These documents lack the authority and effectiveness of legal statutes, leading to practical issues such as unclear data ownership and incomplete data quality assurance systems. Therefore, it is imperative to align with the direction in health care and enhance the top-level design for health data sharing. This alignment should be done by considering the current state of data-sharing development and standardizing the objectives, scope, participating subjects, and operational processes of data sharing. Establishing unified and clear data processing standards and stringent punishment systems for offences will ensure policy uniformity and coordination, prevent policy fragmentation, safeguard personal privacy, and resolve the impasse in data sharing.

In terms of technical standards, efforts should be made to enhance consensus in standard development and ensure uniformity in implementation. These efforts include establishing standard data identifiers, storage protocols, sharing and utilization standards, and unified technical infrastructures across stakeholders to promote data availability and trustworthiness. Additionally, establishing cross-regional electronic health data-sharing platforms and digital health packages can facilitate data exchange and interoperability at the regional level, enabling both citizens and healthcare professionals to securely and efficiently access and utilize health data.⁶⁰ Emphasis should also be placed on the development of electronic medical record systems, alongside the formulation of relevant interoperability standards, in order to improve the efficiency and transparency of data access and sharing for data controllers.⁶¹

Collaborative governance system

Collaborative governance, a model involving multiple participation, emphasizes cooperative relationships and the consistency of organizational missions among various entities. The system for the health data open sharing involves diverse subjects, including government, enterprises, social organizations, and the public. Therefore, in open sharing, implementing a standardized data governance structure, a diversified interest balancing system, and a supply-demand matching system can help all participants reach a consensus and make governance decisions, thereby accelerating the release of data value. Firstly, from the perspective of data subjects, it is crucial to address the unfair discrepancies in costs and benefits associated with data resource sharing. This can be achieved by accounting for differences in institutional functions and resource allocations, and by providing reasonable reimbursements and incentives to entities that contribute health data resources. Establishing a fair and reciprocal exchange system will motivate all parties to actively engage in health data sharing. Secondly, from the perspective of data objects, it is essential to develop and implement unified data formats and exchange standards, establish a dynamic system for updating data standards, and standardize quality control measures across all data collection, processing, preservation, and transmission stages. These measures will ensure seamless data integration and interoperability across different systems. Lastly, in the digital healthcare era, the open sharing of health data should also address issues in traditional public data-sharing practices. By satisfying the diverse needs of various user groups, the supply and demand of health data can be effectively matched. This approach will foster the development of digital health care and related industries, achieving a win–win situation in public health and industrial advancement.

Technical empowerment system

In the digital healthcare era, information and network technology have undergone significant advancements. The operational characteristics of medical and health institutions now include intelligent medical services and agile application deployment. Consequently, the health information generated by these institutions has become more complex, specialized, and voluminous, raising higher demands for the connectivity of open systems. A technical empowerment system characterized by platform leadership, concurrent monitoring, and multidimensional sharing is thus essential to support the development of smart health care and health management in this new era. Specifically, health data platforms play crucial roles in facilitating the open sharing of health data. Serving as bridges between data providers and users, these platforms are instrumental in establishing networks among diverse stakeholders, thereby contributing to realizing a supply-demand matching system. The construction quality and level of these platforms lay the foundation for the digital governance of national health. Moreover, health data platforms enable multidimensional sharing and multiscenario application of data, progressively building a health management system based on community health service centers, with medical and health organizations as the core, supplemented by social health management agencies, and regulated by relevant government departments. This integrated approach will comprehensively advance the “Healthy China” strategy, ensuring a cohesive and effective implementation of health initiatives nationwide.

As the scope of data openness expands, the types and formats of health data are becoming increasingly complex and diverse, involving a broader range of stakeholders and longer temporal spans. This trend necessitates greater attention to information security and privacy protection to effectively mitigate data leakage and misuse risks. Regarding risk management, alignment with national regulations such as the Data Security Law and the Personal Information Protection Law is essential. Additionally, emerging technologies—such as blockchain and security-aware data provenance graphs—can support the development of a multistakeholder health data governance mechanism involving users, institutions, and government bodies.⁶² The decentralized, transparent, and tamper-resistant features of blockchain ensure the authenticity and reliability of health information, offering viable solutions for the organization and oversight of health data. These technologies enhance the controllability, reliability, and traceability of data throughout the sharing process, improve transparency in data use, and safeguard the confidentiality and integrity of sensitive information. Ultimately, such an approach facilitates the construction of a secure and efficient health data-sharing mechanism that achieves dual compliance with both institutional and technological requirements.

Rights and interests protection system

Whether it is the EU's GDPR, the US's HIPAA, or China's PIPL, these regulations all emphasize the importance of data security protection from distinct perspectives. The GDPR emphasizes balancing privacy protection with data flow by prioritizing data subject rights and cross-border transfer rules.⁶³ HIPAA implements the principle of risk management prioritization through mandatory healthcare data security measures and breach notification mechanisms.⁶⁴ The PIPL strengthens the synergy between national security and individual rights by enforcing data classification and tiering, localized storage requirements, and outbound security assessments.⁶⁵

The sustainable development of open sharing and the free flow of health data necessitate data security protection. It is critical to prioritize risk management and rights protection throughout the data-sharing process, ensuring reasonable regulation from multiple perspectives, including legal regulation, regulators, and factor allocation, to protect user rights. Health data often contain sensitive personal information and demands rigorous privacy protection measures. A robust legal framework for privacy protection is essential for safeguarding individual rights, advancing information technology, and facilitating the cross-regional flow of data. Firstly, enhancing the legal and policy frameworks for privacy protection is imperative. These frameworks should clearly define the rights associated with health data privacy, outline methods for safeguarding these rights, and establish tort liabilities. Forward-thinking legislation should empower public data open governance and personal information protection within the medical and health information technology sectors, promoting effective governance through sound legal practices. However, declarative rights may have limited practical effect, and the absence of effective regulatory systems can result in unclear privacy protection responsibilities among relevant entities, potentially creating regulatory blind spots. Therefore, it is crucial to establish the foundation and focus of regulatory governance by delineating the boundaries of departmental powers and defining the scope and functions of regulation. Lastly, the market-oriented allocation of health data factors should adopt an active defense approach. While facilitating data circulation and sharing, it is essential to implement proactive measures to mitigate potential security risks and prevent privacy breaches. Data security must be intensely defended in spatial dimensions and comprehensively protected throughout the entire life cycle and value chain from temporal dimensions. To achieve this, innovative upgrades to health data processing methods are necessary, along with the establishment of a comprehensive security governance framework through the integrated application of multiple technical measures.

Conclusion

In summary, the open utilization of health data is a dynamic process that spans multiple stages and involves diverse subjects to deliver inclusive public value to society. Distinct from traditional data resources, health data application services have a broader scope for exploration, expansion, and value extraction, offering substantial research opportunities. This study employs the grounded theory to collect and analyze data related to the open sharing of health record information, extracting key elements and further elucidating the operational mechanisms through the ISM-MICMAC model. Based on the analysis of these mechanisms, the study constructs a framework for an open health record data-sharing mechanism from four dimensions: top-level guidance, collaborative governance, technological empowerment, and rights and interests protection. Targeted policy recommendations are proposed to support the development of an open data-sharing system. The aim is to better match health management needs, enable efficient and precise personalized health services, and optimize the allocation of health service resources. Moreover, this research offers theoretical and methodological support for addressing the practical challenges of promoting equitable and inclusive digital public services.

However, the scope of this study was constrained by time and resource limitations, which restricted the channels and volume of data collection. Consequently, the study is limited by a small sample size and geographically narrow sampling, which may affect the representativeness of the findings. There is a need for further research to enhance the selection and operational mechanisms of factors influencing the open sharing of health data. Future research should comprehensively consider the integrated, complex, and multifaceted nature of primary-level health record data governance and employ diverse methodological approaches, such as fuzzy-set Qualitative Comparative Analysis, to explore the interactive effects and causal pathways among the factors influencing open data sharing. Additionally, attention should be given to the latest research on EHRs among community residents, focusing on thoroughly identifying open application scenarios. It is essential to differentiate user needs and preferences to develop and provide platforms and functionalities that align with personalized requirements, thereby facilitating health information sharing between patients and providers and enhancing the protection of public health.

Supplemental Material

sj-docx-1-dhj-10.1177_20552076251353694 - Supplemental material for The open sharing operation mechanism of health data in the digital healthcare era: A study based on grounded theory and interpretative structural modeling method

Supplemental material, sj-docx-1-dhj-10.1177_20552076251353694 for The open sharing operation mechanism of health data in the digital healthcare era: A study based on grounded theory and interpretative structural modeling method by Xinping Huang, Siyuan Zhu, Zheng Lv, Qianwen Zhou and Tianqi Kou in DIGITAL HEALTH

Footnotes

Acknowledgements

The authors would like to express our sincere gratitude to Ms. Siyuan Zhu, whose original thesis work and contributions to interview design and data analysis formed the empirical foundation of this study. The authors also extend our heartfelt appreciation to all the experts and participants who generously contributed their time, insights, and professional perspectives through interviews and evaluations. Their valuable input was essential in constructing the factor system and completing the expert scoring that supported the analysis.

ORCID iDs

Xinping Huang

Zheng Lv

Qianwen Zhou

Tianqi Kou

Ethical considerations

Our study did not require ethical board approval because it did not directly involve humans or animals. These data do not contain any data which can identify individuals. Therefore, ethical approval is exempted.

Consent to participate

Verbal informed consent was obtained from all participants, and these data does not contain any data which can identify individual.

Author contributions

XH contributed conceptualization, methodology, writing and revising. SZ contributed data collection, data analysis, and writing. ZL contributed data collection, software, writing, and visualization. QZ contributed writing and revising the paper. TK contributed writing and revising the paper and visualization.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by The National Natural Science Foundation of China, “Research on Intelligent Evidence-Based Policymaking Information Awareness and Its Dynamic Deduction Driven by Knowledge Computing” (Grant No. 72374082).

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material for this article is available online.

References

Courbier

Dimond

Bros-Facer

. Share and protect our health data: an evidence based approach to rare disease patients’ perspectives on data sharing and data protection - quantitative survey and recommendations. Orphanet J Rare Dis 2019; 14: 1–5.

Goldstein

. A qualitative study of physicians’ views on the reuse of electronic health record data for secondary analysis. Qual Health Res 2024. Epub ahead of print. DOI: https://doi.org/10.1177/10497323241245644.

Biasiotto

Johansson

Alemu

, et al. Public preferences for digital health data sharing: discrete choice experiment study in 12 European countries. J Med Internet Res 2023; 25: e44410.

Wickramasinghe

Ulapane

. A solution for the health data sharing dilemma: data-less and identity-less model sharing through federated learning and digital twin-assisted clinical decision making. Electronics (Basel) 2025; 14: 682.

Konopik

Blunck

. Development of an evidence-based conceptual model of the health care sector under digital transformation: integrative review. J Med Internet Res 2023; 25: e41512. https://www.jmir.org/2023/1/e41512 .

Davies

Walker

Rubinstein

, et al. (eds). The state of open data: histories and horizons. Cape Town and Ottawa: African Minds and International Development Research Centre, 2019, 592 p.

Abdelrahman

. Open government data: development, practice, and challenges. London: IntechOpen, 2021, https://doi.org/10.5772/intechopen.100465 .

Kang

Zeng

. Analysis and suggestions on the research of information ecosystem in China. Libr Inf Serv 2020; 64: 113–124.

Zhang

Wang

. Research on the theory, elements and model of open government data ecology system. Inf Stud Theory Appl 2022; 45: 42–49.

10.

Sánchez

Hernández Clemente

García López

. Public and patients’ perspectives towards data and sample sharing for research: an overview of empirical findings. J Empir Res Hum Res Ethics 2023 Dec; 18: 319–345.

11.

Alexopoulos

Loukis

Charalabidis

, et al. An evaluation framework for traditional and advanced open public data e-infrastructures. In: Proceedings of the 13th European Conference on Electronic Government , Como, Italy, Jun 13-14 2013, pp.102–111.

12.

Kim

. An iOS application for hospital information using Korean public open data. Adv Sci Lett 2017; 23: 10218–10222.

13.

The Pew Charitable Trusts. Closing gaps in data sharing is critical for public health [Internet]. Philadelphia, PA: The Pew Charitable Trusts, 2024 Jun 20, [cited 2025 Apr 25]. Available from: https://www.pewtrusts.org/en/research-and-analysis/articles/2024/06/20/closing-gaps-in-data-sharing-is-critical-for-public-health.

14.

Saberi

Mcheick

Adda

. From data silos to health records without borders: a systematic survey on patient-centered data interoperability. Information 2025; 16: 106.

15.

HL7 International. Fast Healthcare Interoperability Resources (FHIR) overview [Internet]. Ann Arbor (MI): HL7 International; 2023 [cited 2025 May 13]. Available from: https://www.hl7.org/fhir/overview.html.

16.

Deloitte. Tech companies have a trust gap to overcome-especially with women. Wall Street J [Internet]. 2024 Apr 4 [cited 2025 Apr 25]. Available from: https://deloitte.wsj.com/sustainable-business/tech-companies-have-a-trust-gap-to-overcomeespecially-with-women-2e50a5cc.

17.

Sun

Zhang

Fang

. Data security governance in the era of big data: status, challenges, and prospects. Data Sci Manag 2021; 2: 41–44.

18.

Custers

Uršič

. Big data and data reuse: a taxonomy of data reuse for balancing big data benefits and personal data protection. Int Data Priv Law 2016; 6: 4–15.

19.

Xing

Deng

Qin

. Influencing factors of data availability in open sharing of scientific data. Doc Inf Knowl 2025; 42: 1–11.

20.

Maisie

Christopher

, et al. Personal health information management system and its application in referral management. IEEE Trans Inf Technol Biomed 2004; 8: 287–297.

21.

Koufi

Malamateniou

Vassilacopoulos

. An agent-based application of personal health record in homecare. In: Proceedings of the 3rd International Conference on Pervasive Technologies Related to Assistive Environments , 2010 Jun, pp.1–6. https://doi.org/10.1145/1839294.1839309.

22.

Willis

Cocoros

Randall

, et al. Electronic health record use in public health infectious disease surveillance, USA, 2018-2019. Curr Infect Dis Rep. 2019; 21: 1–9.

23.

Jones

Deane

Wolstencroft

, et al. File audit to assess sustained fidelity to a recovery and wellbeing oriented mental health service model: an Australian case study. Arch Public Health 2019; 77: 1–8.

24.

Schopf

Nedrebø

Hufthammer

, et al. How well is the electronic health record supporting the clinical tasks of hospital physicians? a survey of physicians at three Norwegian hospitals. BMC Health Serv Res 2019; 19: 1–9.

25.

Duan

Boland

Liu

, et al. Learning from electronic health records across multiple sites: a communication-efficient and privacy-preserving distributed algorithm. J Am Med Inform Assoc 2020; 27: 376–385.

26.

Costa

Amorim

Maia

, et al. S3PHER: secure and searchable system for patient-driven health data sharing. arXiv preprint arXiv:2404.11372. 2024. Available from: https://arxiv.org/abs/2404.11372.

27.

Al Amin

Tummala

Shah

, et al. Balancing patient privacy and health data security: the role of compliance in protected health information (PHI) sharing. arXiv preprint arXiv:2407.02766. 2024. Available from: https://arxiv.org/abs/2407.02766.

28.

Fang

Dziedzic

Zhang

, et al. Decentralized, collaborative, and privacy-preserving machine learning for multi-hospital data. arXiv preprint arXiv:2402.00205. 2024. Available from: https://arxiv.org/abs/2402.00205.

29.

Reuters. China's data protection rules prompt pause from major European research funders [Internet]. London: Reuters, 2025 Apr 25, [cited 2025 May 13]. Available from: https://www.reuters.com/sustainability/society-equity/chinas-data-protection-rules-prompt-pause-major-european-research-funders-2025-04-25/.

30.

Chen

Huang

, et al. A model for predicting factors affecting health information avoidance on WeChat. Digit Health 2025; 11: 20552076251314277.

31.

Dubovitskaya

Novotny

, et al. Applications of blockchain technology for data-sharing in oncology: results from a systematic literature review. Oncology 2020; 98: 403–411.

32.

Huang

Lai

. Barriers of open government data in China from the perspective of data life cycle. Inf Stud Theory Appl 2018; 41: 7–13.

33.

Zhou

. Study on the value cognition and application advancement strategy of electronic health records. J Med Inform 2018; 39: 2–8.

34.

Zheng

. Implementation strategies for electronic health records data sharing. Arch Manag 2021; 2: 92–93.

35.

Wang

. Data sharing and personal information protection. Mod Law Sci 2019; 41: 45–57.

36.

Cai

Liu

. Secure sharing solution for medical data based on Ethereum. Appl Res Comput 2022; 39: 24–30.

37.

Tong

Chen

Yang

, et al. Novel utilization of a paper-level classification system for the evaluation of journal impact: an update of the CAS journal ranking. Quant Sci Stud. 2023; 4: 1–16.

38.

Dawes

Cresswell

Pardo

. From “need to know” to “need to share”: tangled problems, information boundaries, and the building of public sector knowledge networks. Public Adm Rev 2009; 69: 392–402.

39.

Yang

Wang

, et al. Collaborative measurement of data opening policy in China's municipal government data management system: taking a regional central city as an example. PLoS One 2023; 18: e0289550.

40.

Zhang

Liu

. Understanding relationship commitment and continuous knowledge sharing in online health communities: a social exchange perspective. J Knowl Manag 2022; 26: 592–614.

41.

Tamuhla

Lulamba

Mutemaringa

, et al. Multiple modes of data sharing can facilitate secondary use of sensitive health data for research. BMJ Glob Health 2023; 8: e013092.

42.

Burnap

Anthi

Reinecke

, et al. Mapping automated cyber attack intelligence to context-based impact on system-level goals. J Cybersecur Priv 2024; 4: 340–356.

43.

Guo

Tian

. Medical data privacy protection scheme supporting controlled sharing. J Xidian Univ 2023; 51: 165–177.

44.

Ismail

Materwala

Karduck

, et al. Requirements of health data management systems for biomedical care and research: scoping review. J Med Internet Res 2020; 22: e17508.

45.

Schwalbe

Wahl

Song

, et al. Data sharing and global public health: defining what we mean by data. Front Digit Health 2020; 2: 612339.

46.

Tenopir

Rice

Allard

, et al. Data sharing, management, use, and reuse: practices and perceptions of scientists worldwide. PLoS One 2020; 15: e0229003.

47.

Dash

Shakyawar

Sharma

, et al. Big data in healthcare: management, analysis and future prospects. J Big Data 2019; 6: 1–25.

48.

Min

Xing

. Literature analysis-based construction of virtual communities in APP for electronic health records. Chin J Med Libr Inf Sci 2019; 28: 68–72.

49.

Xiang

Wang

. Comparison of personal healthcare data protection standards between China and foreign countries. Inf Stud Theory Appl 2022; 45: 188–193,187.

50.

Gao

Duan

Pan

, et al. Blockchain-enabled supervised secure data sharing and delegation scheme in Web3.0. J Cloud Comput 2024; 13: 21.

51.

Zhang

. Does open data boost journal impact: evidence from Chinese economics. Scientometrics 2021; 126: 3393–3419.

52.

Guo

Wang

. Business model of health data sharing platform based on data crowdsourcing in the context of big data. Manag Rev 2019; 31: 56–64.

53.

Chen

. Discussion on the cross-boundary integration and sharing service model of disease control archives for the construction of healthy China. Arch Manag 2023; 6: 118–119.

54.

Hudani

Labonté

Yaya

. Where's the disconnect? Exploring pathways to healthcare coordinated for youth experiencing homelessness in Toronto, Canada, using grounded theory methodology. Qual Health Res 2024; 34: 298–310.

55.

Venkatesh

Brown

Sullivan

. Guidelines for conducting mixed-methods research: an extension and illustration. J Assoc Inf Syst 2016; 17: 2.

56.

Warfield

. Development of interconnection matrices in structural modeling. IEEE Trans Syst Man Cybern 1974; SMC-4: 81–87.

57.

Liu

Chen

. Enablers for maritime supply chain resilience during pandemic: an integrated MCDM approach. Transp Res Part A Policy Pract 2023; 175: 103777.

58.

Chen

Jiang

Song

. Factors influencing enterprise digital capability: an ISM-MICMAC analysis based on the TOE framework. Sci Technol Prog Policy 2025; 42: 118–127.

59.

Attard

Orlandi

Scerri

, et al. A systematic review of open government data initiatives. Gov Inf Q 2015; 32: 399–418.

60.

Poppe

Sæbø

Braa

. WHO Digital health packages for disseminating data standards and data use practices. Int J Med Inform 2021; 149: 104422.

61.

Wang

Xia

. On the practical dilemma and perfecting improvement of data rights and interests protection in the utilization of health and medical data - take the European health data space as an example. Libr Inf. 2024; 44: 55–68.

62.

Jaigirdar

Rudolph

Bain

. Risk and compliance in IoT-health data propagation: a security-aware provenance based approach. In: 2021 IEEE International Conference on Digital Health (ICDH) , 2021 September 5; pp.27–37.

63.

European Union. Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data (General Data Protection Regulation) [Internet]. Luxembourg: Publications Office of the European Union; 2016 Apr 27 [cited 2025 May 13]. Available from: https://eur-lex.europa.eu/eli/reg/2016/679/oj.

64.

Health Insurance Portability and Accountability Act of 1996, Pub. L. No. 104-191, 110 Stat. 1936 (1996) [Internet]. Washington, D.C.: United States Congress; 1996 [cited 2025 May 13]. Available from: https://www.congress.gov/bill/104th-congress/house-bill/3103

65.

Standing Committee of the National People's Congress. Personal Information Protection Law of the People's Republic of China [Internet]. Beijing: National People's Congress, 2021, [cited 2025 May 13]. Available from: http://www.npc.gov.cn/npc/c30834/202108/a8c4e3672c74492a80b53a172bb753fe.shtml.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB