From Fragmentation to Flow: Unlocking the Full Potential of Health Data

Fragmentation continues to limit the full potential of health data across systems. Progress depends on establishing interoperability, aligning regulations and investing in sustainable digital infrastructure. Strategic actions include enforcing open standards, financing ongoing digital operations, developing workforce skills and empowering patients as active data stewards. With quality data and structured access, healthcare can transition from silos to intelligent, connected care.

Key Points

1. Health data remains underused due to fragmentation and lack of interoperability.
2. Open standards and regulatory support are vital for connected digital health systems.
3. AI in healthcare needs quality data, clear governance and inclusive access models.
4. Patients must be empowered as active data stewards through better consent tools.
5. Sustainable change requires financing, training, smart regulation and coordination.

Introduction

Health systems today generate vast volumes of data; however, much of it remains underutilised. Fragmented IT landscapes, siloed infrastructures and multi-layered regulations continue to obstruct the path to a data-driven healthcare model. With demographic pressures, rising expectations and workforce shortages mounting, the key question is no longer whether to use health data, but how.

The insights in this article are based on a whitepaper authored by Jens Kögler, developed in collaboration with healthcare executives, digital health experts and policy leaders from the German-speaking healthcare ecosystem.A broad-based expert consultation identified the primary enablers of meaningful digital transformation. While technologies are available, the core issues lie in structural and cultural barriers that hinder implementation. Achieving intelligent use of health data requires more than digitisation; it calls for integrated strategies that span interoperability, governance, financing and patient participation.

Interoperability First: The Cornerstone of Digital Health

The lack of interoperability continues to be one of the most pressing barriers to digital health. Despite the existence of established standards such as HL7 FHIR, SNOMED CT and LOINC, healthcare providers often operate with incompatible systems, disconnected interfaces and historically grown data silos. This fragmentation hinders the seamless exchange of information, creating inefficiencies and missed opportunities across sectors.

The absence of binding implementation requirements allows proprietary systems to persist. Experts emphasise the need for a regulatory push towards open standards and for procurement policies to prioritise interoperable architectures. Interoperability is not merely a technical concern—it is foundational to coordinated care, efficient workflows and the sustainable integration of innovations such as artificial intelligence.

Strategic models such as federated data spaces, data lakes and Clinical Data Repositories (CDRs) can help enable secure and structured data sharing. These must be backed by clear responsibilities, standardised data formats and support mechanisms for medical and administrative personnel to boost digital competence. Without a cultural shift and a common understanding of shared responsibilities, true interoperability will remain elusive.

AI-Readiness Depends on Data Quality and Access

Artificial intelligence is widely seen as a game-changer for clinical decision-making, diagnostics and operational efficiency. However, its deployment depends critically on the availability of structured, high-quality, longitudinal datasets. In most healthcare systems, these remain scarce or are held in formats unsuitable for algorithmic training and analysis.

Regulatory frameworks such as the European AI Act currently classify most clinical AI systems as high-risk, imposing stringent documentation and approval requirements. While important for patient safety, this creates high barriers for smaller organisations and limits experimentation. There is growing support for differentiated risk assessment, sandbox environments for testing, and access structures based on neutral data trustees that preserve privacy while enabling innovation.

Without pragmatic governance models, including consent frameworks like Broad Consent and mechanisms for pseudonymised data access, AI will remain confined to isolated pilot projects. Standardisation of data quality is equally essential—poor input data invariably leads to flawed models. Collaborative partnerships between providers, researchers and developers can create a more inclusive and adaptive ecosystem for AI in healthcare.

Patients as Data Stewards, Not Just Subjects

Empowering patients to manage their own health data is another key enabler of transformation. Many individuals remain unaware of how their data is used—or why it matters. Simplified and transparent consent frameworks can strengthen trust while reducing bureaucratic hurdles. Broad Consent, allowing individuals to approve data use for predefined research or care purposes, is widely viewed as a practical compromise between autonomy and efficiency.

The availability of secure digital identities will be a crucial step forward. Current registration procedures for services such as electronic health records remain cumbersome and often exclude less digitally literate groups. Streamlined authentication protocols based on national electronic IDs could unlock everyday use while maintaining data protection and sovereignty.

The electronic patient record (ePA) in Germany offers an instructive case. Despite political support and regulatory momentum, it has struggled to gain broad adoption due to limited usability and unclear added value for users. The inclusion of patient-generated data—such as symptom tracking, wearables or therapy feedback—alongside incentives for clinical use, could improve uptake. However, integration must be seamless, structured and supported by compensation schemes to be sustainable.

System-Wide Enablers: What Needs to Happen Now

Health data cannot be unlocked through isolated projects or pilot schemes. Systemic change is required—grounded in structural investment, human capability and strategic oversight. Based on expert consultations, four strategic enablers were identified:

1. Sustainable financing: Digital infrastructure must be financed as a long-term operational component, not just through time-limited project funding. Recurrent costs for cloud services, platform maintenance and innovation should be integrated into routine financing mechanisms, including DRG-based systems.
2. Skills development: Medical, administrative and IT professionals alike need access to targeted training in digital tools and data governance. Interdisciplinary teams and updated curricula are essential. The discontinuation of qualifications such as medical informatics risks exacerbating the digital talent gap.
3. Governance and oversight: Responsibilities for digital transformation remain fragmented. National agencies or agile coordination bodies could help steer implementation, define clear priorities and ensure alignment across sectors. Institutions like Gematik should focus on standard-setting and enforcement, not product development.
4. Smart regulation: Legal frameworks should include sunset clauses or periodic impact assessments. If policies fail to deliver demonstrable improvements in care, access or efficiency, they must be revised. Data protection remains vital, but it must be reconciled with innovation through harmonised interpretation and practical application.

Conclusion

Unlocking health data is both an opportunity and a necessity. More than just technological infrastructure, it requires trust, collaboration and strategic intent. By building systems that are interoperable, inclusive and accountable, healthcare can move from data fragmentation to data flow. This shift underpins digital transformation, but also sustainability and resilience of future care delivery.

Conflict of Interests

None.