As information becomes increasingly digital and data operations become ubiquitous in the federal workforce, agencies are turning to data literacy to help inform workforce training programs, drive decision-making and improve overall health care delivery.
Data literacy is also gaining attention from lawmakers. The Data Science and Literacy Act of 2023, introduced to the House in February, is one way Congress aims to increase access to data science education, reduce course equity gaps for all students and help build the 21st century STEM workforce.
“Data touches everything we do. Data education is integral to bolstering our global competitiveness, unlocking good-paying jobs and fostering a well-informed society,” Rep. Haley Stevens said in a press release for the bill.
In the public sector, agencies like the departments of Defense and Veterans Affairs, along with the National Institutes of Health, are also spearheading new programs and goals to help meet this national priority.
Enterprise Intelligence and Data Solutions Program
Within the Military Health System, the Enterprise Intelligence and Data Solutions (EIDS) Program established in 2018 integrates all disparate data capabilities around military health care into one program office.
Five years later, EIDS has streamlined 23 warehouses of various sizes and capabilities into what is now the Military Health Systems Information Platform (MHS MIP).
The effort aims to connect and deliver all of the military health data that is not just limited to clinical data of the 9.6 million beneficiaries. It also includes all the data adjacent to it, such as financial data or materiel data.
"That doesn't mean we're taking over all those systems, but it's really about integrating the data ... the MIP becoming the nexus of all secondary data for military health care," Chris Nichols, EIDS program manager at the Defense Health Agency, told GovCIO Media & Research. "Without bringing it all into a single source, we can't really identify and work through all the insights that we can leverage from our information."
Nichols also said military health care is an integrated delivery network with a significant portion of health care administered by third parties, including behavioral health practices or hospital systems. This means a substantial amount of care for the 9.6 million beneficiaries happens outside the walls of the network's physical hospitals and clinics.
"We have to get all of that data back in and integrate all of that to really be able to manage the large population from a health care perspective and also then manage the readiness of our force," Nichols said. "That is all over 2 million-plus active duty soldiers, sailors [and] airmen, but then it's also the readiness of our medical staff of all types and varying capabilities and their readiness to deploy and support the warfighter mission."
Once all the disparate data sources are brought together, the next step is integrating that data in a usable, understandable and, most importantly, actionable way.
During the pandemic, Nichols' office reported information to the Centers for Disease Control and Prevention (CDC) and DOD about COVID-19 case rates while simultaneously partnering with the clinical community to change and modify behavior and manage the patient population.
"In a short amount of time, we were able to pull that data and make it more actionable and make reaction changes to how we managed our patient population," Nichols said.
Over the last seven months, the EIDS successfully tore down data silos and developed a deliberate data strategy around the work being done. Data literacy, Nichols said, is all about education and process changes.
Partnerships with analysts, clinicians and business development also help develop a data-literate agency.
"A lot of that is really having a consistent conceptual framework around these things, developing the right data strategy and bringing the right personnel to the table to then drive those decision points in a collective manner," Nichols said.
NIH Funds Data Literacy Research Programs
The National Institutes of Health (NIH) relies on multiple programs across its 27 institutes and centers to improve data literacy throughout all levels of its workforce. NIH has intramural research programs as well as extramural programs funding activities at training institutions around the country.
Everyone doesn’t have to be a data scientist, but if we’re going to work in a very data-intensive organization, it makes sense for people to at least be data literate, National Library of Medicine Data Science and Open Science Librarian Lisa Federer told GovCIO Media & Research.
NLM recently participated in a data science training activity where all staff completed a pretest to see where they were with what they knew. Federer said some were data scientists who were extremely savvy and for others this was brand-new.
“Understanding where they were starting out and what their goals were through this persona-based learning program that allowed us to pair people with the right training for the level they were starting at and the level they wanted to get to based on what it is that is most relevant for their work,” Federer said. “It was a successful program, it was really personalized and gave people an opportunity to start out wherever they were, and I think people really enjoyed that.”
NLM funds different programs based at universities focusing on data literacy training. Federer said NLM also offers fellowships and also relies on its network of regional medical libraries conducting outreach and work on data literacy with different audiences around the U.S.
“We also have a National Center for Data Services through the network and they focus on upskilling librarians in data literacy particularly as the new NIH data management and sharing policy has come into effect, which is a big change for some researchers who haven’t done this," Federer said. “Biomedical librarians have started to play a really important role in helping researchers to meet some of the requirements of the policy.”
One of the major challenges NIH identified with the data literacy program is finding the right fit for such a diverse and large audience. Not everyone is at the same starting line in terms of their existing baseline knowledge and not everyone has the same finish line.
“It’s really hard to come up with a one-size-fits-all program in the way you would for other types of training, so I think it's really important to have solutions that do allow for customization for whatever level you’re at," Federer said. “I also think it's been a challenge for people to find advanced courses. There seems to be a lot for the beginner level, but when you get to a point where you want to expand and learn more it can be harder for people to find resources even though there are a lot available if you know where to look.”
Federer said there are also collaborative efforts underway at the Department of Health and Human Services where agencies can share best practices and partner to provide data-driven value.
“Within HHS, a group was formed to focus on how we augment the workforce in terms of both data science and data literacy expertise among our existing staff, and also how do we enable ourselves to hire people to come in and fill in some of the gaps that we have,” Federer said.
VA’s ASPIRE Program
The Department of Veterans Affairs oversees the delivery of care to more than 9 million enrolled veterans. With modernizing electronic health records (EHR) and providing equitable services, data is the crux of VA’s health care system and effective health care delivery.
The agency launched several new programs to upskill and train its workforce so VA employees — including health care providers, technology developers and data scientists — can effectively manage and understand the massive amount of data it has.
The All Services Personnel and Institutional Readiness Engine, or APSIRE, is one program bolstering data literacy not only at VA, but also across government. ASPIRE is an AI-driven system that will automatically test and populate a learning pathway specific to individuals' skills and needs within the federal workforce. Housed within VA, the interagency effort aims to upskill the federal workforce to better manage, use and understand data.
"ASPIRE does a lot of things, but at the core, it's an education system. I would say it's an assessment education system,” Anthony Boese, interagency programs manager and ethics officer at VA’s National Artificial Intelligence Institute, told GovCIO Media & Research.
As VA continues to build out ASPIRE, the agency will conduct validation testing to mitigate bias in the learning pathways. Boese said the new training system could also help prepare for future public health emergencies.
“Increasing data literacy will be very helpful for the VA — as we did some of this for COVID-19, and I imagine we're going to have to do it again, unfortunately — for pandemic detection and preparedness, using a large fleet of data-literate people to help get extremely excellent models about when and where pandemics may or may not occur and what will happen when they do,” Boese added. “The VA level we really want to get to make sure that we have all the competence we need to both minimize risks and maximize use of technology for the veterans ... For VHA, it always comes back to veteran health care.”
Under VA’s data governance process, the agency stood up a people subcouncil in April 2022. During the Feb. 22 ASPIRE AI Tech Sprint Golden Envelope event, subcouncil Co-Chair and VA Chief Human Capital Officer Tracey Therit said the group is uniting expertise in technology and human capital management to invest in VA’s workforce.
VA’s subcouncil takes a four-pronged approach to strengthen its workforce: improve hiring, development, engagement and collaboration. Under the strategy, the agency focuses on two key projects. First, VA is reclassifying its data science positions under occupational series 1560.
“We really want to make sure that our workforce has that credential, understands that's their identity and how they define themselves in terms of the work that they do [and] the technical expertise that they bring to VA,” Therit said.
Second, VA is developing a data training plan to outline the skillsets the agency has as well as skills professionals should bring into the data and analytics workforce.
“We are well on our way to completing both of those milestones this year: reclassifying all our data science positions under the 1560 series, and we have a working group that is developing the plan and will test the plan for the pilot or subset of our workforce, and then gather some satisfaction and get some feedback and input on that plan,” Therit said.