Skip to Main Content

New NIH-Wide Strategic Plan Emphasizes Data, Diversity

The plan looks to leverage data science and improve health for gender and ethnic minorities.

7m read
Written by:
Leadership and successful business ideas concept 3d rendering of crowd 3d low polygon people arrow shape form walk together on white floor color tone image
Photo Credit: whyframestudio/iStock

The National Institutes of Health’s new strategic plan highlights data and diversity as critical components to realizing the future mission of its institutes,

The strategy — which updates the first five-year NIH-wide strategic plan released in 2016 — emphasizes three objectives: advance biomedical and behavioral science, develop and maintain scientific research capacity, and promote scientific integrity and public accountability.

The updates maintain goals of advancing basic biomedical research and enhancing the research workforce, and also includes new crosscutting themes that will drive the new objectives forward. These include focuses on:

  • Improving minority health and reducing health disparities
  • Enhancing women’s health
  • Addressing public health challenges across the lifespan
  • Promoting collaborative science
  • Leveraging data science for biomedical discovery

NIH has made recent strides that embrace diversity and inclusion, such as NIH UNITE, as well as data and technology, such as in the 2018 NIH Science and Technology Research Infrastructure for Discovery, Experimentation, and Sustainability Initiative (STRIDES). However, the strategic plan marks overall longer-term commitment to strengthening these efforts across all of NIH’s activities.

NIH Director Dr. Francis Collins noted that reinforcing diversity and inclusion across NIH research and the scientific community will help address health inequities.

“NIH will further enhance the science of tomorrow by continuing its efforts to build a generation of researchers that better reflects the rich, creative diversity of our great nation,” Collins said in the strategic plan. “The increasingly complex scientific questions that our society will face in the future will require not only diversity of scientific disciplines, but also diversity of thought, experience and demographics.”

On the data maturity and resource front, the new strategic plan builds on the 2018 NIH Strategic Plan for Data Science, a roadmap of the agency’s data science activities. The 2018 plan also reinforces crosscutting collaboration across different agencies and groups of expertise, said NIH Associate Director for Data Science and Office of Data Science Strategy Director Susan Gregurick.

“Advances in biomedical data science require cross-disciplinary researchers to engage in data harmonization and sharing and have awareness of privacy and security, ethics and bias,” Gregurick told GovernmentCIO Media & Research in an emailed statement. “NIH is partnering with [National Science Foundation] to engage data and computer scientists and engineers with clinicians, research participants, ethicists, and the public in its plans to address future challenges and opportunities.”

Gregurick said that to help NIH meet its goals in leveraging data in biomedical research over the next 5 years, her office is expanding on preexisting initiatives it started and also launching new ones. This includes building on STRIDES, for instance, to allow NIH to explore the use of cloud environments to streamline NIH data resources by partnering with commercial cloud providers.

“By leveraging the STRIDES Initiative, NIH and NIH-funded institutions can begin to create a robust, interconnected ecosystem that breaks down silos related to generating, analyzing and sharing research data,” Gregurick said.

To improve interconnectivity and remove barriers to access data resources, NIH launched the Research Auth Service initiative to facilitate access to NIH’s open data assets and repositories in a “consistent and user-friendly manner,” Gregurick added. NIH has recently undergone the process of building off the effort in creating a federated data ecosystem across five large data resources, including BioData Catalyst, Kids First Data Commons, Cancer Research Data Commons, AnVIL and NCBI.

Another newer effort NIH is pushing is with new advancements in Fast Healthcare Interoperability Resources, or FHIR. Gregurick said this will couple advances in data science with advances in clinical and health care data to aid in patient and provider decision-making, as well as improve disease prevention and health promotion strategies across communities.

As NIH expands its data maturity — which includes committing to findable, accessible, interoperability and reusable (FAIR) standards — it aims to make it more readily available for artificial intelligence and machine-learning applications. Although NIH has launched a number of initiatives around this, such as Bridge2AI and the AIM-AHEAD program, Gregurick said that NIH has a vast and diverse repository of data that can lead to new opportunities with AI in the future.

“NIH has unique needs that can drive the development of novel approaches and application of existing tools in AI/ML,” Gregurick said. “From electronic health record data, omics data, imaging data, disease-specific data and beyond, NIH is poised to create and implement large and far-reaching applications with AI.”

The theme around data also overlaps with the strategic plan’s other crosscutting themes. Not only will data interoperability, availability and advancement encourage collaborative science, but also further representation of ethnic and gender minorities in data science and AI applications for research and biomedical science.

“Underrepresented communities have untapped potential to contribute new expertise, data, recruitment strategies and cutting-edge science to the data science and AI/ML fields,” Gregurick said. “To close the gaps in the field and to better engage underrepresented communities, NIH has launched the Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity (AIM-AHEAD) program.”

Amid these efforts, Gregurick said that in data science, AI and machine learning overall lack diversity in its researchers and data, which can lead to harmful biases. In that regard, she said that the crosscutting themes to improve health outcomes and studies of racial, ethnic and gender minority health can help data and data applications better address underrepresented communities.

Woman typing at computer

Stay in the know

Subscribe now to receive our curated newsletters

Related Content
Woman typing at computer

Stay in the Know

Subscribe now to receive our newsletters.