Digital Health

Industry Perspectives: Public Health Data

Two doctors meeting with a third who is on a laptop.

For the final installment of our Industry Perspective Series, we will focus on public health data and the trust, or lack thereof, that is necessary for continued innovation in this space. We asked thought leaders in healthcare to provide perspectives on what a successful public health data infrastructure looks like, what gaps may already exist in the public health data industry, and what strategies they suggest for improving public trust.

Altarum Institute:

  • Jim Kamp, director of strategic partnerships, public health systems

  • Craig Newman, subject matter expert, public health interoperability

  • Laura Rappleye, senior director, public health interoperability 

What are some of the strategies to strengthen the trust in public health data?

The primary sources of public health data are health systems and labs, which vary in the data standards followed, the levels of standards adopted and the level of detail reported. Reducing discrepancies and variations can help strengthen trust in public health data.

One strategy to address this is to encourage a standard approach to the collection and presentation of data across jurisdictions by establishing and promoting inclusive communities of practice across public health data areas — vital statistics, immunization, chronic disease case reporting and others.

The Centers for Disease Control and Prevention’s (CDC) National Center for Health Statistics Vital Statistics Modernization Community of Practice is an example of an approach that is making an impact. Focused on modernizing the vital records system, this community of practice facilitates multiple forums per month open to public health officials in all jurisdictions, their partners and vendors. Another good example is HL7 International, which provides a forum for volunteers to participate in workgroups dedicated to establish and extend health data standards across a variety of healthcare use cases, including public health data.

Another strategy to strengthen trust is to improve data quality through use of a National Patient Identifier (NPI). Recent legislative changes now allow the U.S. Department of Health and Human Services to fund NPI efforts, which seek a better alternative to using a social security number when deduping data. By making it easier to share data bidirectionally, without introducing duplicate data between health systems and public health registries, the public health data would improve in quality and become more trustworthy.

What are some resilient/sustainable ways to build new public health data infrastructure? (What “regular” infrastructure can we use better for public health data?)

Building a more sustainable and resilient infrastructure starts with reviewing an approach that has not worked well. For example, a closed, proprietary, and completely localized public health systems that make it difficult to scale data sharing across jurisdictions of varying resource and capability.

A good way to address this challenge is to reinvigorate efforts to promote the coordinated use of open-source software. Noam Artz, president of HLN Consulting, has an excellent blog post discussing how using open-source software could change the dynamics of public health systems procurement for years to come. He argues that agencies should shift their approach to one that is challenge-based and value-focused, embracing modular systems, continuous development cycles and Agile methodologies.

The adoption of health data standards, such as those developed by the HL7 community, offers another way to overcome the challenges posed by closed, proprietary systems. Such standards take time to develop and implement across our complex network of public health systems, labs, health information exchanges, public health agencies, and health information technology (HIT) vendors, so it is unrealistic to expect rapid widespread adoption of promising new standards like the Fast Healthcare Interoperability Resource (FHIR). When prioritizing investment in public health infrastructure, focus should be on pathways not yet functioning well or at all, even if that means using older pre-FHIR standards. For example, immunization systems make efficient use of HL7 Version 2, so it does not make sense to focus modernization funding on upgrading this part of our public health infrastructure.

The focus, instead, should be on modernizing the areas that have the most impact, such as those identified by the Helios FHIR Accelerator for Public Health, a public-private project with initial funding from the Office of the National Coordinator for Health Information Technology and the CDC: It promotes the adoption of existing HL7 specifications suitable for public health that are scalable, adaptable, and sustainable.

Helios is encouraging investment in three areas: (1) make public health data available in bulk; (2) deliver aggregate data to public health systems; and (3) align and optimize public health data sharing. These investments should include incentives for HIT vendors to add support for standards in their products and funding for the resources needed to implement them.

Where are the gaps in public health data? What are they, from your point of view? Are we gathering the right information?

Gaps in public health data exist primarily because of historic underinvestment in community health, limited integration of data already present in public health systems, and a lack of policy incentives for, and investment in, public health data reporting and standards implementation.

Efforts like the recently formed Community Information Exchange (CIE) Task Force in Michigan are filling some of these gaps by bringing together voices that have been underinvested in the past, such as behavioral health, tribal health, nursing homes, and providers in communities with fewer financial resources. The work of the CIE Task Force is the next logical step in building on the Gravity Project’s work to develop social determinants of health data standards. Thanks to those foundational efforts, data gaps around food, transportation, and housing insecurity can be targeted for future inclusion in public health data collection efforts.

Gaps also stem from public health data silos. Data across vital records, immunization, and disease surveillance systems often contain disconnected information on the same individual. Without consistent linkage across these data sets, the picture painted by public health data is impressionistic, with some squinting required to understand what’s happening! The good news is that the silo problem has been identified and a more holistic approach is being taken as part of the CDC’s Data Modernization Initiative.

Finally, the recent Public Health Data Systems Task Force 2022 report to the Health Information Technology Advisory Committee does an excellent job outlining recommendations to address existing pain points in public health reporting. Among these recommendations are those that highlight the need to raise the reporting expectations for certified technology. Many existing and emerging reporting standards have not been fully embraced by certified technology, and therefore, have not been widely implemented. Examples include standards for newborn screening, birth-defect surveillance, and situational awareness. It is critical to formalize recognition of additional reporting requirements to promote broader adoption in the real world. These additional reporting requirements will, in turn, help fill gaps in public health data.



Femi Ladega, chief digital officer

What are some of the strategies to strengthen the trust in public health data?

The underlining core principle for any industry to engage the public simply is trust. Trust first formulates on a personal level and revolves around everything that affects the consumer in one way or another. The more my healthcare journey applies to or is designed around “me,” the more trust and information I provide back into the larger public health pool. The principal of “data donation” to improve outcomes for others is also a motivating factor. This cycle is very similar to how navigation systems are strengthened by crowd sourcing information from their larger user base. Each individual driver builds trust based on how safely and reliably the navigation directs them to their specific destination and how transparent health systems are with data use. With that earned trust, the consumer relies entirely on the system and in doing so provides valuable and relevant information back into the pool of data.

The best way to strengthen the trust in public health data is by making the information contextually relevant to its population of one and to generate information useful on a macro-population scale. Allowing citizens to participate in systemic improvement with their data or even become “citizen scientists” allows further participation for those motivated to get involved.

What are some resilient/sustainable ways to build new public health data infrastructure? (What “regular” infrastructure can we use better for public health data?)

With the urgency created by COVID-19, leveraging health information to fight life-threatening disease became priority number one. It became clear that not every institution was prepared to create an infrastructure that can empower me, the consumer, or the network around me. The wide spectrum of healthcare needs, coupled with social determinants that are economic, environmental, educational, professional, and personal, magnify the fact that current healthcare is more episodic than we would like it to be.

To establish a more proactive approach that benefits the larger population, it’s important to focus on whole-person care, which is unfortunately difficult to capture during an impersonal 10-to-15-minute doctor visit addressing the issue, and not the larger problem at hand. To make the right adjustments for a more holistic approach to health, the infrastructure must change and be able to bring together multiple data sets. Consumers must be able to contribute their own data but in a strategic way to allow us to personalize care and provide more precise treatment plans.  

Public health data infrastructure needs to interlace clinical information, social determinants, and environmental information across all consumer channels.

Understanding consumer behavior means capturing their attention where and when appropriate. In this way, the approach to population wellness is more proactive, created through awareness on the consumer’s terms. The right information technology infrastructure should stratify the population to see who prefers SMS texts, social media or phone calls. As an example, when seeing a tropical storm alert on their phone from their care team, a chronic obstructive pulmonary disease patient can take precautions and avert an unnecessary episode. We need a multi-channel personalized approach to maximize personal positive impact.

Where are the gaps in public health data? What are they, from your point of view? Are we gathering the right information?

Public health data is not organized with the ability to combine and stratify data effectively because there are so many variations of conditions which require just as many types of interventions. There are multiple registries of information, but they're isolated from one another. This is called a “data lake,” and is in fact limiting for public health data registries, even though many believe it is the solution to interoperability. The dictionary Britannica defines a lake as “a large body of slowly moving or standing water that is surrounded by land,” so, if we've not combined or distributed the meaningful data set in the appropriate way, it is isolated and stagnant.

A true and well-functioning interoperability engine would be a delta, a body of water that flows into a larger body of water. Having the ability to extrapolate information from a wider public source of data that flows back and forth is essential for building a constant source of updated information and is what we need to engineer for the future of individual consumers as well as populations.