How Can Data Sharing Evolve Population Health Management?

Beginning the session, "The Future of Population Health Management—Data Sharing Is the Change Agent," Melanie Matthews, CEO of Physicians of Southwest Washington, highlighted a goal announced by CMS and the Learning Action Network last year to accelerate the percentage of US health care payments tied to quality and value in each market through the adoption of shared accountability alternative payment models (APMs).

While the coronavirus disease 2019 (COVID-19) pandemic has caused CMS to extend guidelines for many APMs, Matthews sought to reiterate the growing transition into data sharing. “I think it's really important for all of us involved in value-based care and attending the National Association of ACOs 2020 Virtual Spring Conference to recognize that the ability to understand how to use data is really important for this work to work well.”

Anna Routh, product manager for the Beneficiary Claims Data Application Programming Interface (API) at CMS, discussed why CMS and HHS began the implementation of application programming interfaces (APIs) in 2018 as these data interfaces aim to “reduce cost, improve quality by ensuring that information collected and the delivery of payment for health care is readily available to anyone entitled to it in a usable format.”

“A patient’s information is often trapped in siloed health systems, which prevent patients and providers from accessing complete health information aggregated in one usable health record,” said Routh.

So far, CMS has generated APIs for Medicare beneficiaries, accountable care organizations (ACOs), Part D plan sponsors, and providers. Notably, all of this data uses the same server, called the Beneficiary FHIR Data Server, so that all sectors involved in the care of a patient, ranging from the ACOs to the beneficiaries, are on the same page.

This pursuit for integrated health systems was further expanded by Anna Taylor, MS, program director at MultiCare Connected Care, as she noted that “it's not just clinical data that tells you the entire patient story.” Her organization is seeking to generate interoperability through partnerships that would enhance the availability of data and information from disparate health systems and community-based organizations that would be otherwise unaccounted for.

Taylor described 3 foundational data elements that would promote interoperability, which include member, provider, and claims data. By expanding the sources in which they derive information and data, their population health ecosystem, as Taylor described, would better reflect communities that may not align with national data who would then be provided with enhanced health care coverage.

So what technology is necessary for interoperability?

Taylor’s colleague, Ann Goldman, executive director of analytics and digital strategy at MultiCare Health System, spoke on the how-to when it comes to optimizing interoperability. As Goldman noted, this is no simple feat as it requires creating a platform that can handle and analyze these additional sources of data. Essentially, it is an investment in population health.

Taylor expanded that there are barriers to adopting these services, which include provider alignment, interpreting data from health plans and electronic health records that may not be in their system, and the monetary investment itself.

When it comes to convincing those within health care organizations to adopt these services, Goldman said to keep the quadruple aim in mind, which signifies better health, better experience, lower cost, and joy of work.

Lastly, how would these services look when finalized? Jen Perloff, director of research at the Institute for Accountable Care (IAC), discussed the current format and future of her organization’s custom analyses, called benchmarking comparison and performance analyses.

IAC’s data warehouse includes information from Medicare beneficiaries, comprehensive cost and quality measure sets, and ACO survey data. Additionally, it is now testing attribution models to profile physician groups to create synthetic ACOs.

Through a web-based tool that IAC plans to release next week, all of this data will be made available to the public. “So, do I have a trade off around the nuclear imaging versus the [magnetic resonance imaging]? You can begin to kind of use these reports to hopefully look for opportunities to steer the ship.”