VA, DOE join forces to harness power of big health data

Joint initiative between federal agencies will enable researchers to tap into database of genetic, clinical, lifestyle and military exposure information for medical research.


The Departments of Veterans Affairs and Energy are partnering on a big data initiative designed to help facilitate medical breakthroughs for veteran healthcare, with new studies in the areas of cardiovascular disease, prostate cancer and suicide prevention.

The VA-DOE Big Data Science Initiative will leverage the VA’s Million Veteran Program (MVP), a national voluntary research effort that includes the country’s largest database linking genetic, clinical, lifestyle and military exposure information.

For its part in the collaboration, DOE is contributing expertise and technologies in artificial intelligence, big data and high-performance computing to identify trends supporting the development of new treatments and preventive care.

“The main driving force was that our Million Veteran Program—the largest genomic cohort of its kind— was in need of a computing partner, and there really is none better than the Department of Energy,” says Michael Gaziano, MD, principal investigator with MVP.

Part of the new VA-DOE initiative is MVP-Computational Health Analytics for Medical Precision to Improve Outcomes Now (CHAMPION), under which the agencies will make health and genomic data more broadly available to researchers for specific studies.

One planned study will determine what sets of risk factors are the best predictors of certain forms of cardiovascular disease, with the goal of informing individualized therapy and treatments for patients based on their individual risk factors.

Another research area will focus on prostate cancer in an effort to find new ways to assess which cancer tumors are lethal—and therefore require treatment—versus those that are slow-growing and unlikely to cause any symptoms.

In addition, a third study seeks to build algorithms to generate highly tailored personalized risk scores for suicide, which could be used to help predict which patients are at the highest risk, and to develop prevention strategies.

According to Gaziano, MVP has enrolled more than 560,000 veterans so far. Ultimately, the goal is to recruit 1 million veterans to participate in MVP’s genomic database—which includes DNA specimens.

The VA-DOE initiative will access the VA’s electronic health record system, which includes EHRs for 24 million veterans who have used VA for healthcare during the past two decades. The Veterans Health Information Systems and Technology Architecture (VistA) is the agency’s legacy EHR, which Gaziano calls the “backbone” of veteran patient records.

Also See: VA leveraging health IT to address its 5 top priorities

“We have some of the best healthcare data resources on the planet and—as we utilize the data and add to it the genetic information—we’re creating an unprecedented big data problem. Not only is the data big—it’s extremely complex,” adds Gaziano. The VA-DOE initiative will be based within DOE’s National Laboratory system, which has one of the world’s preeminent supercomputing capabilities.

Among the reasons the VA partnered with DOE is the fact that the DOE’s National Laboratory system is in a secure environment that’s able to safeguard patient data, Gaziano contends. “We needed to do this in a way that protects our veterans, and security was paramount,” he says. “We wanted a big, robust computing environment that would allow investigators to log into that space so that we could keep an eye on what’s going on. But the data itself does not leave the environment.”

The VA-DOE initiative will also securely link MVP data with other sources of health information, including non-VA sources of data from the Department of Defense, Centers for Medicare and Medicaid Services, and the Center for Disease Control’s National Death Index.

“Those are datasets that we in the VA have used for research purposes over the years and—in some cases—for clinical purposes,” concludes Gaziano. “There are many other datasets within the VA, such as disease-specific registries like the cancer registry, which we use on a regular basis. We even have the capacity to use the images from various diagnostic tests like MRIs and CT scans.”

More for you

Loading data for hdm_tax_topic #better-outcomes...