When healthcare data collaborates, science moves faster

Many of the most important questions in healthcare cannot be answered with data from a single organization. Whether it’s researchers studying rare diseases across multiple institutions, pharmaceutical companies relying on provider claims to generate real-world evidence or clinical trial sponsors identifying patient populations across diverse networks—the true value comes from these distinct organizations being able to securely collaborate on their datasets. When researchers united patient data from four academic health systems across California, Minnesota, Texas and Utah to evaluate COVID-19 monoclonal antibody treatments (opens in new tab), they demonstrated exactly why.

By pooling records into a single, centralized registry of over 167,000 patients, they achieved the massive statistical scale required to track incredibly rare adverse drug events and pinpoint exactly which high-risk, immunocompromised populations benefited most. Linking this massive clinical dataset with localized viral genomic sequencing allowed them to map, in real time, precisely when the therapies lost their effectiveness against mutating variants like Omicron.

This collaborative effort proved that the most complex puzzles in medicine cannot be solved in isolation; it requires a unified, cross-network approach to transform fragmented real-world data into definitive, life-saving evidence.

The problem is that each of those collaborations required extraordinary effort to execute (opens in new tab). For this single study, teams had to manually align disparate databases to a specialized data model, clear massive regulatory hurdles to satisfy strict HIPAA deidentification standards and secure individual institutional review board approvals across multiple states.

It was a highly synchronized operation (opens in new tab) funded by federal contracts and managed by a dedicated centralized coordinator—a long, cumbersome process that the vast majority of healthcare organizations simply do not have the time, budget or infrastructure to replicate at scale.

Introducing Databolt Connect

As Capital One Software launched and began scaling Databolt, our data protection platform, we saw a unique opportunity to help the healthcare and life sciences ecosystem solve its massive multi-institution data bottleneck. That is why, as an early-adopter partner for the launch of the new Databricks Marketplace (opens in new tab), we are introducing Databolt Connect.

Available as one of the first native applications on the new marketplace, Databolt Connect is a zero-trust data collaboration solution built specifically to turn a historically cumbersome process into secure, friction-free collaboration at scale.

Instead of requiring healthcare organizations to spend months manually engineering custom data pipelines or building bespoke cloud silos, Databolt Connect allows organizations to collaborate right where their data already lives.

Integrated natively within the Databricks ecosystem, healthcare organizations can securely prepare and link datasets without raw data ever leaving their environment. The entire linkage process stays within each organization's own controlled Databricks instance, supporting HIPAA compliance and eliminating the dependency on third-party identity resolution providers.

What this unlocks

With Databolt Connect, healthcare organizations can pursue collaborations that may have historically been too complex or too risky to attempt, such as:

• Streamlined research & cohort discovery: Finding the right patients for clinical trials or observational studies often requires scanning millions of records across multiple hospital networks. Databolt Connect is designed to help researchers securely run multi-party feasibility queries to better identify highly specific patient cohorts.

• Real-world evidence (RWE) generation: Bringing a drug to market or expanding its label often involves evaluating how treatments perform in actual clinical settings outside of controlled trials. By enabling secure collaboration between life sciences companies, insurance providers and hospital networks, Databolt Connect can help accelerate the timeline to satisfy post-approval commitments and support formulary decisions.

• 360-degree longitudinal patient views: A patient’s medical journey is rarely confined to a single healthcare system; it is fragmented across electronic health records (EHRs), pharmacy dispensing logs, claims databases and wearable devices. Databolt Connect can enable the goal of safely connecting these disparate data points, without exposing raw patient identities or moving data out of an organization's environment.

Recognized at Databricks Data + AI Summit

We’re proud that Databolt Connect was recognized this week at Databricks Data + AI Summit (opens in new tab) as an early-adopter of the new Databricks Marketplace, an open marketplace for Databricks customers to discover and use directly inside their own governed workspace. We are incredibly excited to support the healthcare and life sciences organizations already leveraging the Databricks platform to drive medical innovation. By combining Capital One Software’s deep focus on data protection with the scalability of Databricks, Databolt Connect is positioned to help remove the traditional, exhausting friction from multi-institution research.

Ultimately, our goal is to empower organizations to collaborate with confidence, knowing their data remains secure and under their full control. We look forward to seeing how teams utilize Databolt Connect to securely unite fragmented health data, break down historic silos and safely accelerate the insights that can improve patient outcomes.

To learn more about Databolt Connect or to explore how it can support your organization's data collaboration goals, visit the Databricks Marketplace (opens in new tab) or reach out to our team today.