Home Blogs & News

Unlocking the Value of SDEs: A Cost-Effective Approach to Sustainable Health Data Management

As health research increasingly depends on vast amounts of data, organisations face mounting pressure to manage this data efficiently, securely, and sustainably. The UK’s health data landscape, as highlighted in Professor Cathie Sudlow’s recent independent review, “Uniting the UK’s Health Data: A Huge Opportunity for Society” underscores the critical challenges and opportunities within this space. Among the most significant barriers identified in the report are the underfunding and undervaluation of data management and sustainability, alongside inefficiencies in accessing and curating health data.

These challenges present an urgent call for innovation—and Secure Data Environments (SDEs) are emerging as a leading solution. By addressing the inefficiencies of traditional data management models, SDEs offer a cost-effective and scalable approach that unlocks the potential of health data while significantly reducing the total cost of ownership (TCO). This blog explores how SDEs deliver cost savings compared to traditional Full-Time Equivalent (FTE) staffing and DIY research cloud or High-Performance Computing (HPC) systems, highlighting the critical insights from Sudlow’s report and their implications for researchers, funders, and policymakers.

The Cost Challenges in Health Data Management

Professor Sudlow’s report makes a compelling case for greater investment in health data infrastructure, pointing out that research projects and clinical activities often fail to adequately budget for the critical components of data management and sustainability. This oversight leads to delays, inefficiencies, and missed opportunities for meaningful research. For example:

Underfunded Data Custodians: Organisations responsible for curating and providing access to health data often lack the resources to do so efficiently. This leads to prolonged access times and bottlenecks that hinder progress.

Complex, Inefficient Systems: The UK’s fragmented data ecosystem—comprising general practices, hospitals, and other institutions—is costly to navigate and manage, with duplicative efforts driving up operational costs.

Short-Term Focus: Many initiatives prioritise immediate needs over long-term sustainability, resulting in repeated investments to address the same systemic challenges.

Increased Risk to Sensitive Health Data: A lack of sufficient funding often results in less than ideal approaches to data security and management. This can leave sensitive health data vulnerable to breaches, unauthorised access, and compliance failures, exposing organisations to significant financial, reputational, and legal risks.

Adding to these challenges is the need for sustainable funding mechanisms that recognise data management as a core component of research. A model that mirrors open access publishing fees could be established, where data management costs are treated as a per-project or per-participant expense, explicitly included in research budgets. Such costs could ensure the secure storage, curation, and governance of data, much like approved costs for animal or cellular research are routinely funded. This approach would enable a proactive and equitable allocation of resources for SDEs across all projects.

Why SDEs Are the Answer

Secure Data Environments (SDEs) provide a transformative approach to addressing these challenges. These platforms enable researchers to access and analyse de-identified health data in a secure, controlled environment, ensuring compliance with data privacy standards while fostering innovation. More importantly, they offer significant cost advantages over traditional methods.

1. Reducing Reliance on FTEs

Traditional data management systems often depend on large teams of data stewards, analysts, and IT staff to manually curate, clean, and govern data. This labour-intensive approach is not only slow but also expensive. By contrast, SDEs streamline these workflows through automation and built-in governance mechanisms, significantly reducing the need for manual intervention.

2. Avoiding Overinvestment in HPC Systems

High-Performance Computing (HPC) systems, while powerful, require substantial upfront capital investment and ongoing maintenance costs irrespective of whether you are building and maintaining your own HPC or acting as a user of public or private facilities. HPC are traditionally designed for non-sensitive research workloads, such as astronomy simulations or materials chemistry modelling, where strict data governance and privacy regulations are not required.

For sensitive health data, HPC environments often fall short in key areas:

Shared Compute Risks: HPC systems typically operate on shared infrastructure, increasing the risk of unauthorised access or cross-project data leakage.

Compliance Gaps: HPC centres may lack the rigorous controls needed to comply with frameworks like GDPR, NHS DSPT, or ISO 27001, which are mandatory for health data.

Limited Auditability: Many HPC environments do not offer detailed audit trails to track who accessed sensitive data, when, and for what purpose—a key requirement for health research.

These limitations can expose health data to security and compliance risks that outweigh the benefits of raw compute power.

For comparison, Secure Data Environments (SDEs) are designed specifically for sensitive data, with built-in capabilities for:

Granular Access Control and real-time audit logs.

Encryption during storage, transfer, and analysis.

Federated Analysis, allowing data to remain securely at its source while being analysed across multiple sites.

SDEs, by addressing these challenges, provide a flexible, scalable, and compliant alternative to traditional HPC systems while ensuring the highest standards of data security and governance.

High-Performance Computing (HPC) systems, while powerful, require substantial upfront capital investment and ongoing maintenance costs. Many research projects overestimate their computational needs, resulting in underutilised HPC resources. SDEs offer a more flexible, on-demand model, allowing researchers to scale computing power as needed, avoiding unnecessary expenditure.

3. Lower Total Cost of Ownership (TCO)

When considering the TCO of data management solutions, SDEs consistently outperform traditional approaches. Factors contributing to their cost-effectiveness include:

Integrated Security: SDEs come with built-in compliance frameworks, reducing the need for costly, standalone security audits and certifications.

Efficiency Gains: Faster data access and streamlined workflows translate into quicker project timelines, reducing labour and overhead costs.

Interoperability: By enabling seamless data integration and linkage, SDEs eliminate duplicative efforts and associated costs.

A case study within the Sudlow report emphasises the time and cost savings achieved when researchers could securely access linked datasets in real time—an efficiency that is far from standard practice in traditional systems. We’ve previously blogged on this issue in detail (The Economics of Building and Running a Trusted Research Environment; Building Versus Buying a TRE).

The Role of Federation in Enabling Research

In addition to their cost-effectiveness, SDEs are powerful enablers of federated research. Federation allows data stored in different SDEs to remain in place while still being securely accessed and analysed across multiple sites. This approach eliminates the need for data duplication, enhances collaboration, and significantly reduces data transfer costs.

Aridhia has taken a leadership role in advancing federation by open-sourcing our federated node technology. This initiative reflects our commitment to open science and empowering health and research communities to conduct secure, collaborative analytics on a global scale. By enabling seamless connections between SDEs, we are helping researchers unlock new insights while maintaining the highest standards of data privacy and security.

The Cost Effectiveness of SDEs: A Hypothetical Example

Consider a mid-sized research project involving 50-100 TB of data distributed across 3-5 sites. Analysis is performed by a 5-10 person research team over 12 months.

Traditional Model

Assumptions: Data stewards, analysts, and IT staff are needed to curate, clean, and govern the datasets manually for ~12 months of effort. Based on typical UK salaries, fully loaded costs per FTE (including benefits and overheads) range from £60,000+ per year.
High upfront capital investment to build or upgrade in-house HPC infrastructure, estimated at £400,000 or more. Operational expenses, including power, cooling, IT support, and hardware upgrades, typically cost an additional £50,000 to £100,000 annually.

Total FTE costs: £180,000 (3 FTEs x £60,000).

HPC setup and maintenance: £400,000 (setup) + £100,000 (annual maintenance).

Total: ~£680,000.

SDE Model:

Assumptions: The SDE automates significant portions of the data curation and governance process. This reduces the need for dedicated data stewards, cutting FTE costs by up to 50% (e.g., requiring 1-1.5 FTEs instead of 2-3), and streamlined workflows allow projects to progress faster, reducing overall labour requirements.

SDEs, leveraging Azure, provide compute resources on-demand. Researchers only pay for the resources they use, avoiding upfront infrastructure investments. By enabling data to remain in place while being accessed securely, SDEs reduce the need for expensive data transfer and storage duplication.

• Reduced FTE costs: £90,000 (1.5 FTEs x £60,000).
• On-demand compute: ~£100,000 (scalable costs based on workload).
• Federated capabilities: Minimal centralisation costs.
• Total: ~£190,000.

By adopting an SDE:

Data access and governance are automated, reducing FTE costs by up to 50%.

Computing resources are provisioned on demand, avoiding upfront capital expenses and delivering savings of 30-50% on infrastructure costs.

Federated capabilities eliminate the need for centralised data movement, further reducing operational overheads.

Overall, the research project could realise cost savings of 40-60%, freeing resources for additional studies or broader data exploration.

As noted earlier, see our earlier blog posts for a more detailed cost comparison (The Economics of Building and Running a Trusted Research Environment; Building Versus Buying a TRE).

Strategic Recommendations: A Call to Action

To fully unlock the potential of health data and realise the cost efficiencies of SDEs, we need coordinated action from stakeholders across the ecosystem. Here are three key recommendations:

1. Fund Data Management and Sustainability Proactively

Funders must recognise the critical role of data management in research success and allocate dedicated budgets for infrastructure. This aligns with Sudlow’s call for long-term investment strategies rather than short-term fixes.

2. Develop Sustainable Funding Models for Data Access and Management

SDE providers can help lead the way by developing transparent, tiered pricing models that reflect the true costs of data management.

However, sustainable funding for data access and management also requires advocacy from researchers and policymakers. A broader approach would incorporate per-project or per-participant funding models, where these costs are explicitly recognised and budgeted in research funding applications. This mirrors the approach already taken for open access publishing fees or approved unit costs in other research areas, such as animal and cellular studies. Early adoption incentives could drive widespread uptake while ensuring equitable support for underfunded research teams.

3. Advocate for SDE Accreditation and Standards

A UK-wide accreditation system, as recommended in the Sudlow report, will establish trust and ensure consistency across SDE platforms. Aridhia, as a leader in this space, can play a pivotal role in shaping these standards, and we’ve blogged previously on how our DRE delivers and exceeds the Five Safes and SATRE standards.

Looking Ahead: The Role of SDE Providers

The challenges outlined in the Sudlow report present a unique opportunity for SDE providers like Aridhia to demonstrate leadership. By addressing cost barriers and showcasing the value of SDEs through case studies and transparent cost analyses, we can build a compelling case for their widespread adoption. We note these challenges aren’t just restricted the UK, and relevant worldwide across the research landscape.

Federation between SDEs further enhances the value proposition by enabling collaborative research without compromising data security or privacy. Aridhia’s open-source federated node technology exemplifies our commitment to innovation and open science, providing researchers with the tools they need to tackle global health challenges.

Addendum: The Critical Role of Data Security and Governance in SDEs

One of the most valuable advantages of SDEs is their robust approach to data security and governance, which is crucial when working with sensitive health data. SDEs provide built-in compliance mechanisms that ensure adherence to stringent privacy regulations, such as GDPR, while offering a secure environment for managing and analysing de-identified data.

By contrast, other platforms or DIY solutions often lack these integrated safeguards, leaving organisations exposed to significant risks, including data breaches, unauthorised access, and non-compliance with legal requirements. The financial and reputational damages from such incidents can be catastrophic, particularly in healthcare, where trust is paramount.

SDEs also enhance auditability and transparency, enabling organisations to track data usage, access, and sharing. This level of oversight not only mitigates risks but also fosters greater collaboration and trust among stakeholders.

In an era where data breaches are increasingly common, investing in SDEs is not merely a cost consideration but a strategic necessity for any organisation handling sensitive health data.