Health Information Exchange

Health Data Collaboration During COVID-19 and the Long-Term Potential

Health data sharing during COVID-19

Amit Trivedi, MS

Joyce Sensmeier, MS, RN-BC, FHIMSS, FAAN

Our understanding of the impact of the COVID-19 pandemic to date evolves from data that are tracked from a local, regional, state and national perspective. However, there are many information gaps, in part because these data are captured separately by healthcare provider organizations, public health departments and other settings.

Healthcare organizations are challenged to cohesively and efficiently tap their individual data repositories to extract the necessary health data from their IT systems including admission/discharge/transfer, EHR, enterprise resource planning and supply chain systems. The lack of data sharing agreements and foundational infrastructure across organizations also create barriers for meeting today’s health information needs.

While data from individual health systems are being compiled and used in aggregate to inform evolving strategies, the pandemic requires an integrated approach for obtaining and applying data-driven insights to address specific needs such as case management, contact tracing, disease surveillance, or epidemiological modeling.

Share Best Practices

Convening multi-institutional, collaborative activities and sharing best practices and resources across entities is a catalyst for community-wide dialogue that will inform future strategies.

Washington University has developed several approaches in response to the COVID-19 pandemic by leveraging a variety of data types and sources. In one approach they have established a COVID-19 data commons which includes regional data for operational and research use such as institutional dashboards, registries, data sharing networks and ad-hoc data exchanges to inform real-time decision-making. They also work with various public health departments and healthcare provider organizations to identify shared common data elements and transactional standards, which can be used to harmonize and manage data sharing requests across multiple entities.

Clinical informatics resources for COVID-19 have been assembled by nurses and other clinicians handling health informatics challenges related to COVID-19. Among these resources are podcast interviews with nursing informatics leaders from across the nation who are sharing insights about their work in dealing with the pandemic. In one interview, experts from Vanderbilt University Medical Center described a standard operating procedure developed to reduce nursing documentation burden. The standard procedure, which was shared with other organizations nationwide as an exemplar, outlines the critical information that is necessary to capture during a crisis, focusing on documenting by exception, and therefore alleviating additional workload for nurses, freeing them to focus on timely and efficient patient care.

The University of California, San Diego Health has developed a 24-hour incident command center to address their evolving needs during the COVID-19 pandemic and provide real-time feedback on key interventions. EHR tools to support clinical care were designed and implemented including rapid screening processes, clinical decision support, reporting tools and patient-facing technologies. These COVID-19-specific tools include scripted triaging, electronic check-in, standard ordering and documentation templates, secure messaging, real-time data analytics, and telemedicine capabilities. The team highlights the importance of their multidisciplinary approach, and they are offering their tools to help other health systems facing the crisis.

...

Learn what organizations around the world are doing to test, triage and treat COVID-19 in the HIMSS COVID-19 Think Tank.

Develop Digital Public Health Data Architecture and Reporting Structure

Current public health data architecture and reporting capabilities are inadequate to address the current crisis. Public health information systems were a small part of the American Recovery and Reinvestment Act and Meaningful Use stimulus programs, and requirements were directed toward EHR and health IT system developers focused on exchanging limited sets of patient health information. While there were reporting requirements to public health agencies, there were no similar requirements or support for public health agencies or public health information registries to be able to receive the standardized information, or use that data.

Today, in the face of COVID-19, public health officials continue to struggle with limited resources, broken data systems, incomplete results and outdated formats such as faxes—that produce data that can’t be easily consumed or reused.

This fractured environment results in a resource intensive process that requires a great deal of human intervention. For example, researchers at the Mass General Brigham Boston hospital network extract and clean COVID-19 data for global research purposes—the data curation process takes up to eight hours for a single patient record. This manual process is a stark contrast when compared to the enthusiasm and promise of leveraging big data and analytics to help solve our larger public health challenges.

One of the biggest challenges is that preparation is paramount in order to react to an emergency, and the COVID-19 pandemic has exposed a lack of preparation and coordination to communicate a uniform national response. Public health officials are left struggling to discover in real-time what specific data points are most relevant and where they must be obtained from.

The need for a digital public health data architecture, deployable across local, regional, national and international domains is clear. A conceptual model has been outlined for such an architecture that would facilitate focused data sharing and analytics, leverage syntactic and semantic standards, as well as a foundational common data model and data elements, to create a network of networks.

The authors note that this infrastructure could also be linked with similar platforms across a state, or nationally, to create a scalable, reusable, sustainable infrastructure. Key elements of their approach include:

  • Repurposing existing technologies and platforms to enable rapid data sharing
  • Evolving data transmission activities from manual abstraction to automated queries and standards-based transmission via secure file transfer protocols, standard HL7® feeds or HL7 FHIR® approaches
  • Harmonizing, storing and retrieving data to enable analytic activities such as reporting, surveillance, benchmarking and on-demand analyses

Open Data Sources

Public health agencies focus on gathering and sharing data to control the spread of the disease. The Situational Awareness for Novel Epidemic Response (SANER) Project is a collaborative industry effort convened in response to the pandemic, leveraging existing standards to inform critical decisions regarding both the allocation of scarce resources and reducing reporting burden on the healthcare facilities at the front line of response. This open-source, data sharing effort streamlines and accelerates real-time transmission of de-identified health data among healthcare facilities, critical infrastructure and response authorities during public health emergencies and disasters.

The project leverages application programming interfaces (APIs) using FHIR, an international data sharing standard designed to scale collaboration across disparate infrastructures. FHIR contains discrete data elements, giving it an advantage over other data-sharing methods, by allowing for the specification and transmission of only the most essential pieces of information. Awareness of healthcare resource availability is important in any public health emergency, so it is likely this project will continue to provide value long after the COVID-19 pandemic has resolved.

Build for the Future

The COVID19 pandemic and response is being described as data-driven—however the public health informatics community has concerns about the accuracy and consistency of public health reporting data. In the United States, public health reporting requirements can vary state to state and regionally, which impacts the validity of nationally reported data.

For example, COVID-19 is categorized as an infectious disease, and positive results must be reported to the state public health agency. However, negative tests are not reported by all states—which leads to questionable statistics when reporting statistics nationally. Additionally, outdated infrastructure results in data being transmitted through less-than-ideal means—from fax, to spreadsheets, to manual counts. Without a uniform and consistent set of requirements and reporting mechanism, the country is forced to make decisions such as whether to reopen a region, or how much bed capacity a regional hospital system has based on an incomplete picture.

This makes the need for a robust public health reporting infrastructure vital not only to population health, but stresses the importance of investing in the accurate and efficient data collection, capture and reporting if those data are going to be used to drive economic recovery and further analysis as to where there are hot-spots and where it is safe to re-open the country.

Public health data should be considered critical infrastructure as it not only fuels the nation’s economic growth, but also is related to overall national security—as evidenced by the pandemic’s impact on the global economy, supply chains, workforce, vulnerable populations and more. The UN has commented on how associating national security with health policy requirements can “elevate the level of priority given to an issue and deliver results.”

While the healthcare ecosystem is awash in data, the accessibility and usability of those data for individual, organizational and public health purposes remains an ongoing challenge. The COVID-19 pandemic has created an opportunity to address such challenges, both short and long term, but it will require collaboration and cooperation across local, regional, national and international domains.

By building on the foundation of a digital public health data architecture, and applying data-driven insights and tools realized during the pandemic, we can make strides toward realizing the vision of a digital health ecosystem that optimally supports health information needs.

Machine Learning & AI for Healthcare

December 1–2, 2020 | Digital Summit

While healthcare is beyond the hype and already seeing the influence of machine learning and AI technologies within their workflows, the successful implementation that drives results depends on achieving analytics maturity and ensuring data quality and governance. Join HIMSS and take a holistic, workshop approach with a focus on implementation.

Be ready for what’s next