How federated data networks can empower clinical research

Securely and efficiently sharing resources among healthcare institutions can provide the fuel to advance clinical trials and improve patient care.

In clinical research, the pursuit of reducing costs and increasing the predictability of outcomes has led to the emergence of an innovative solution – federated data networks. These networks provide a pivotal foundation for optimizing clinical trial operations, offering a more streamlined and efficient approach to managing the complexities of health data.

By enabling the secure and efficient sharing of resources across numerous healthcare institutions and national borders, federated data networks (FDNs) stand at the forefront of revolutionizing clinical research, making it more productive and patient centered.

Maximizing the value of AI

The integration of FDNs with artificial intelligence (AI) marks a significant leap forward in maximizing the effectiveness of clinical research. Artificial intelligence is a fixture in the spotlight, thanks to growing public interest in generative AI solutions such as ChatGPT. While AI has historically played an important role in clinical research by simplifying health data for more informed decision-making, its influence across the industry continues to evolve.

AI in clinical research is only as valuable as the data it leverages. For pharmaceutical companies and research institutions, health data can be incredibly difficult and sensitive to manage because of various national privacy laws. This has unfortunately resulted in the siloing of data across our healthcare ecosystem, which can hinder the speed of progress in clinical research.

In recent years, federated data networks — which enable several distinct locations to safely and efficiently share resources via a central framework — have emerged to mitigate many of these traditional barriers to data sharing. They do this by facilitating access to sensitive health data across healthcare institutions and national borders.

AI's capability to process and analyze vast amounts of data is well-recognized. However, its potential is fully unleashed only when it has access to comprehensive and diverse datasets. FDNs enable AI to tap into a rich repository of shared health data, including electronic health records (EHR) and electronic medical records (EMR), offering a groundbreaking solution by facilitating a shared repository of this crucial data.

While these networks present significant opportunities for optimizing clinical trial operations, perceived barriers often can stand in the way of their implementation. To move past these challenges, companies must first understand the vast benefits they provide.

The need for federated networks 

As the field of healthcare continues to rapidly expand, so does its potential to become incredibly fragmented — especially in relation to data. The silos this creates can be seen across each step of clinical development and decision-making and can have a tremendous impact on not only healthcare organizations but patients as well.

Many factors can be attributed to this issue. From lack of data standards, governance and harmonization, unclear and insufficient ownership, permissions and sharing of data to privacy concerns that span national borders, clinicians, researchers and patients alike are often left in the dark regarding data. Ultimately, this breakdown of trust around such critical issues prevents our progress.

The ability of federated networks to address current barriers related to data sharing is demonstrated in a recent study published in the Journal of Medical Internet Research. The findings of the report demonstrated the benefits of federated data networks in optimizing clinical trial operations. By sharing site performance information across multiple clinical research organizations, researchers were able to identify enrollment patterns, site performance metrics and factors affecting trial success, enabling them to make data-driven decisions to improve trial efficiency and enhance patient recruitment.

Additionally, across the industry, organizations such as the Clinical Research Data Sharing Alliance (CRDSA), comprised of biopharma companies, nonprofit data sharing platforms, academic institutions, patient advocacy groups and service and technology partners are beginning to lead the charge to drive this change and get life-saving medicines to patients sooner and maximize the value of clinical data to make clinical research more productive.

In parallel, the European Union is adopting a transformative approach toward the utilization of patient data through the inception of the European Health Data Space (EHDS). This initiative is designed to treat patient data as a fundamental resource for clinical research, establishing a federated health data network to advance healthcare delivery, research, innovation and policymaking across the EU.

Challenges of federated learning

Even with the benefits federal data networks present, there are several perceived challenges regarding their adoption and implementation.

Trust is a significant challenge in implementing federated data networks, as companies may be reluctant to share data that they see as their competitive advantage. A recent survey conducted by The Institute for Health Technology Transformation shows that trust remains a significant challenge in data sharing among healthcare organizations, with 78 percent of respondents expressing concerns about sharing sensitive data, including competitive information, with other organizations.

Reasons for this can include privacy, sensitive data and re-identification concerns. For example, the compilation of birth date, disease state and place of living could collectively identify a participant — demonstrating that convincing companies to share data in federated networks, where they perceive it as a competitive advantage, is a challenge that needs to be addressed. Overcoming this challenge is crucial for successful collaboration.

While federated data networks mitigate some privacy and security concerns, there are still considerations to be addressed, such as ensuring data sharing is done responsibly and addressing potential biases in datasets when combining data from different sources. A study published in Open Computer Science highlighted the importance of addressing privacy and security concerns when implementing federated data networks. The study emphasized the need for strict data governance and privacy protocols to ensure responsible data sharing as well as the importance of anonymization techniques to protect patient privacy while allowing for collaborative learning.

Another challenge encompasses the complexities surrounding the shared digital platform architecture, which facilitates cloud-to-cloud sharing. This challenge is particularly pivotal for CROs because of the necessity of ensuring that security and privacy standards are meticulously upheld within the federated environment. The question then arises: Could there be another significant hurdle for CROs in terms of implementation, adoption or potential data limitations?

Opportunities for optimization

The field of clinical research will continue to evolve, but with the power of federal data networks, there’s a significant opportunity to harness and enhance approaches to enable even greater progress. Across the industry, these networks hold the potential to greatly optimize clinical trial operations by sharing site performance information, identifying patient populations, and analyzing demographic and socioeconomic data.

To harness the full potential of the consortium model, which prioritizes patient welfare above all else, organizations must commit wholeheartedly to understanding and adopting this innovative approach. A patient-first model is not merely a guiding principle; it's the cornerstone of a revolutionary strategy that recognizes the immense value offered directly to patients. By adopting a holistic perspective on patient care, we enable the acceleration of the drug development process through the strategic use of federated data.

This approach is transformative, leveraging collective insights to expedite the delivery of groundbreaking medications. As a result, it’s imperative for organizations to invest time, effort and focus into mastering this new technology. Those that do will position themselves at the forefront of medical innovation, playing a pivotal role in bringing life-altering treatments to the patients who stand to benefit the most. This is not just an opportunity — it’s a call to action for those committed to making a significant impact on patient care and the future of medicine.

Amy Kissam-Sands is executive vice president of operational excellence and delivery for Parexel.

More for you

Loading data for hdm_tax_topic #better-outcomes...