NIH Requirements

Scientists dealing with billions of pieces of raw data will find it much easier to do their work after Jan. 25, 2023. That’s when the NIH will adopt a new Data Management and Sharing Policy. Researchers funded by the NIH will be required to submit a formal plan detailing how they will share their research data. The new policy, in part, follows principles first laid out in a paper published in 2016 in Scientific Data.

The Scientific Data paper argued for scientific data to be findable, accessible, interoperable, and reusable (FAIR). That is, they need to be easy to find in a common database, accessible by those with disabilities, able for others to use and exchange the information, and able to be used again. The creation of FAIR Principles provides an essential set of guidelines for stakeholders who wish to go FAIR.

The NIH’s new policy requires that researchers share all scientific data so that others can more easily use findings from previous studies in future studies. Furthermore, researchers must include a plan for sharing their data whenever they apply for NIH funding. The NIH will cover the costs of data collection, sharing, and storage.

Many leaders in the field of biomedical informatics see the NIH policy as a giant step toward Open Science, a concept that makes science much more accessible, inclusive, and available for everyone. Open Science seeks to eliminate the barriers to access of data and to provide scientific credibility by allowing everyone to verify study findings. Stanford experts are in favor of the Open Science concept, but they differ in how they approach and use it.

Open Science in Action

Purvesh Khatri, PhD, an associate professor of biomedical informatics research and an associate professor in the Institute for Immunity, Transplantation and Infection, has used Open Science to create new findings using existing data. Khatri and his lab were able to comb through many publicly available data sets to examine and analyze what factors could lead to new developments in tuberculosis (TB) patterns in gene expression.

Khatri’s research in TB led to the development of TB testing using only finger-stick blood samples. The Khatri lab found that three gene signatures could be used to diagnose TB. Using publicly available data that represented various real-world patient populations, the lab concluded that three genes could indicate whether someone has TB. These findings led the lab to move 

past the discovery stage and produce a clinical result in five years.

For Khatri, the process of searching through different publicly available data remains a challenging endeavor. He says, “We know that these data sets exist out in the public domain, but we have to go out and look for them. Looking for this information is exhausting and labor-intensive. Since there is no centralized method of labeling data, every single piece of data could have different labels for many various things. Researchers who are reading these papers may not be able to find the information they need to get the information. There has to be a solution to this problem.”

“I was not thinking about the people who would read our data. We were creating data sets and maybe sharing them, but not really describing them in such a way that they could be found.” Creating and standardizing metadata and creating methods and forms to list them ensures that researchers can properly file and store data for researchers who wish to look up studies.

Metadata is a type of data that describes information inside something like a written document or file, which informs users of the content. Just as an abstract gives basic information about a scientific journal article, metadata offers the same for data. Consider a researcher who locates a book in a library on a subject of interest. Metadata would inform the researcher of what the book holds and where to find that information. The book metadata would include the title, subtitle, publication date and ISBN, keywords and key phrases, book description, and author bio. This information would be listed in the metadata, and the book’s information would be cataloged, thereby easily found in a library, similar to a data repository.

He plans to work with Borghi this fall to survey School of Medicine investigators and learn how databases are being managed to determine training and support needs. SPORR is also planning a school-wide symposium on Jan. 23, 2023 — the day the NIH requirements go into effect — to educate faculty and provide on-site consultation for investigators. It is also presenting a three-week mini-course on the topic for trainees during Stanford’s winter 2023 term.

Goodman stresses that the SPORR team is looking to support research groups that need assistance and to identify those that use best practices, recognizing them through prizes and dissemination. As Goodman says, “Our responsibility as an initiative and as a school is to make rigorous data handling easy; each researcher shouldn’t have to invent their own wheel. We want to help Stanford investigators learn from each other about what works best here.”

CEDAR is trying to streamline the authoring of this metadata, which will enable researchers to eliminate deviations when describing different data in the same subject. As mentioned above, the process of combing through different publicly available data is a continuing challenge for scientists like Khatri. The lack of standard terms within the metadata makes it hard for investigators to find what they’re looking for.

CEDAR’s primary contribution is to ensure that new metadata adheres to whatever standards the research community has settled upon. Once it’s simplified and centralized, researchers will have an easier time finding the information they need to continue with their research. As Musen explains, “The good news — or the bad news, depending on your perspective — is that scientists didn’t start putting their data in online repositories until the 1980s.” There has been a lot of data since the 1980s, and not all of it adheres to proper metadata search terms. However, there is hope.

The National Science Foundation has issued a grant to a small business called Metadata Game Changers to integrate CEDAR into Dryad, the general-purpose data repository. The expectation is that people who upload data sets into Dryad will be able to use CEDAR to create appropriate metadata. For Borghi, who is working with School of Medicine faculty, integrating CEDAR into Dryad will enable faculty to share data using Dryad if there is no other appropriate repository for them to use — just in time for the new NIH policy.

CEDAR is a starting point for researchers and scientists to correct the mistakes of past studies and to modernize how information from those studies is shared. If all things come together, using Open Science and FAIR principles will ensure that future studies based on previous landmark science papers will be able to reproduce results with few to no problems. What a difference that would have made for Amgen back in 2012!

When Swati DiDonato, MD, MBA, accompanies medical trainees on their morning rounds at Stanford Hospital to review admitted patients’ statuses and plans, she finds herself asking the same question again and again: Will that test (or medicine, or therapy) really improve the patient’s overall health and well-being? Is it necessary?

Over recent decades, health care costs in the United States have risen dramatically. Some of this increase is due to administrative costs and the complexity of the healthcare system, and some of it is due to an aging population and increases in chronic illnesses. But a part of the increase can also be attributed to how medicine is practiced — which tests or treatments doctors use for their patients.

DiDonato, a clinical assistant professor of hospital medicine, thinks clinicians can benefit from dedicated training in “value-based care.” Value-based care embraces the idea of prioritizing tests, procedures, and treatments that are likely to improve the quality of care that the patient receives, rather than turning to many tests or interventions right off the bat. It emphasizes the use of drugs and devices most probable to increase a patient’s quality of life or life span.

A New Curriculum

Teaching medical trainees how to think like a businessperson is one way to encourage systemic change in healthcare — over the coming years, it can lead to more junior faculty like DiDonato, who are inspired and equipped to make a difference. But getting more senior physicians to think and act in new ways is paramount as well, DiDonato says. That’s why she began working with Horomanski on the new CME course, with encouragement from General Medicine Section Chief Jeffrey Chi, MD, clinical associate professor of hospital medicine, and Hospital Medicine Division Chief Neera Ahuja, MD, clinical professor of medicine.

Just like her morning rounds lectures, the course that DiDonato developed aims to make doctors pause and think twice before they order a test or drug.

“We provide examples of what it looks like to use evidence-based guidelines and clinical pathways to guide the ordering of different tests and procedures,” says DiDonato.

The final lesson in the course focuses on how to talk to patients about value-based care. When a patient requests a particular test or drug, for instance, clinicians can struggle to communicate with the patient about why it’s not needed.

Initially, the Stanford Value-Based Care Academy is rolling out the virtual course for faculty members in the Department of Medicine. But DiDonato and Horomanski plan to adapt it for other departments, as well as medical students, residents, and fellows at Stanford. They’re also developing an in-person version of the course, in which staff of a given division can participate together, brainstorming how they can apply value-based care in their own day-to-day practices.

DiDonato hopes that eventually her efforts will inspire similar courses in other health systems.

“Public payers and insurance companies are all making pushes toward value-based models, but the education of physicians on value-based care has lagged behind,” she says. “Our hope is that we can meet the moment with this training to better serve our patients and our community.”

Many veterans in rural locations face the challenges of shortages of physicians and staff. Although veterans can get care in the community if necessary, that can sometimes lead to disjointed care, duplication of effort, and missed diagnoses, says William Fearon, MD, professor in the Division of Cardiovascular Medicine and director of interventional cardiology at the Stanford Department of Medicine, chief of cardiology at the VA. “Ideally, we’d like to keep the veterans within the system for all these reasons. So that’s how the idea of the Clinical Resource Hub, or CRH, came about.”

The Veterans Affairs Palo Alto Health Care System has a long history of providing veterans residing in Palo Alto and neighboring areas with top-notch health care. Starting in 2021, as part of a new program, the VA has offered this high-quality care to other VAs in the region, as it has become the Clinical Resource Hub in Cardiology for all of the VA Sierra Pacific Network, or VISN 21. VISN 21 is one of 18 Veterans Integrated Service Networks (VISNs) in the Veterans Health Administration (VHA). VISN 21 serves veterans in Northern and Central California, Nevada, Hawaii, the Philippines, and U.S. Territories in the Pacific Basin. The hub-and-spoke healthcare service model allows for veterans in rural areas of the VISN to access specialty care at the hub, which is the Palo Alto VA.

This means that veterans outside the Palo Alto area who need more complex cardiology care than they can find at their local VA can access the Palo Alto VA via video visits from either their home or their local clinic. If they need to have a procedure done, the VA arranges transportation and all the preop testing as well as the procedure itself so that the patient can come to Palo Alto for their care in a “one-stop shopping” trip, if at all possible.

“The VA is really focused on providing timely and quality care to veterans,” says Fearon. “And there have been studies that show that veterans have better outcomes when they get their care at the VA compared to in the community, particularly in cardiovascular medicine. So there’s a lot of incentive, from the standpoint of quality care, to keep veterans in the VA system.” Plus, he says, most veterans say they prefer to get their care at the VA.

One Patient’s Journey

One of the first patients to benefit from the CRH program was Marvin Rogers, an 85-year-old Army veteran who lives in Reno, Nevada. He had been experiencing shortness of breath and dizziness and had lost 45 pounds over the past year. His VA cardiologist in Reno diagnosed him with major problems in both his aortic and mitral valves.

“The CRH Cardiology team in Palo Alto performed a clinical assessment for Rogers via videoconference, along with his Reno cardiologists,” says Neha Mantri, MD, assistant professor in the Division of Cardiovascular Medicine and a specialist in structural imaging and noninvasive cardiology at Stanford Hospital and the VA, who was on the team that cared for Rogers. After determining which additional cardiac imaging and potential procedures he required, the team transferred him to the Palo Alto VA. Once the physicians on the team knew what Rogers needed, structural heart nurse coordinators Donna Lynch and Cheryl Christianson arranged his flights, transportation to and from the airport, and accommodations at the VA’s on-campus housing in Palo Alto.

“Within one week of his arrival, we were able to conduct all pre-procedural imaging and successfully complete two very specialized minimally invasive cardiac procedures — a TAVR and a MitraClip,” says Mantri. Rogers’ trip combined his clinic visits, scans, and valve procedures. After spending just three nights in the hospital after his procedures, he was home in Reno and back to his regular routine with his new aortic valve and repaired mitral valve. “Previously, the whole thing could have taken months, with the patient first coming out for an evaluation, then having to come out for one procedure and go back home, and so on,” says Fearon. “This allowed us to deliver that care really efficiently.”

Guson Kang, MD, assistant professor of cardiovascular medicine and a specialist in structural and interventional cardiology at Stanford Hospital and the VA, says that the CRH has provided the resources and staffing to refine their workflow into “a well-oiled machine. It really works well for us and for the veterans. We’ve had people come from Guam, Hawaii, Reno, Albuquerque, everywhere. Most veterans don’t mind traveling, especially when they know that the VA will take care of them and make sure everything is coordinated properly.”

The beauty of the CRH is that it allows a very streamlined method of delivering highly personalized care, says Fearon. “The veterans are getting really cutting-edge medical care, the kind that is only available at high-end academic facilities. And it’s a concierge-type care, where we have staff that can help them get their needs met.” Electronic records allow the VA to communicate seamlessly with the patient’s local VA for follow-up care as needed.

While in most cases patients fly to Palo Alto for specialty procedures and testing, such as structural imaging and electrophysiology, some may need their physicians to come to them at their local clinics. At other times, the team in Palo Alto might consult remotely with the local VA. Everything is individualized to best meet the patient’s needs.

department is able to offer general cardiology, electrophysiology, and structural/interventional cardiology, with heart failure care to launch soon.

As the CRH expands, the VA hopes to build a network of providing care for more diverse patient populations, says Chen. They have recently launched a Women’s Health Cardiology Clinic at the Palo Alto VA and will be offering service to all spoke sites within VISN 21 through CRH. Further networking across VISNs may help bring the numbers of patients needed for the VA to offer the expertise to places that might not otherwise have access to this kind of care. Video visits, electronic health records, and other supportive telehealth technologies facilitate providing this kind of care despite geographic barriers.

Back in Reno, Marvin Rogers recovered well. He is breathing easier and walking well without dizziness. “He has done very well,” says Mantri. “It has been incredibly gratifying to be able to provide specialty cardiac care to our veterans both near and far within our VISN.”