The Future of Long COVID Care Lies in Data Interoperability
The Challenge:
Since late 2019, the novel coronavirus SARS-CoV-2 has disrupted the lives of millions and continues to burden healthcare systems worldwide. A significant portion of patients diagnosed with COVID-19 experience long COVID symptoms, a condition with lingering symptoms that severely affects quality of life.
The global ramifications of SARS-CoV-2 are unprecedented, affecting nearly every aspect of society. Researchers have been studying the infection's wide-ranging outcomes; however, the diversity in study methods and underlying assumptions has led to challenges in comparing data across studies. To improve our understanding of the long-term effects, often referred to as (1) Long COVID, (2) COVID-19 syndrome (PACS), and (3) post-acute sequelae of SARS-CoV-2 infection (PASC), standardization is essential. By identifying common data elements and formal definitions, researchers can more accurately capture and analyze variations in long-term outcomes. This consistency is critical to generating better insights that can inform interventions and better prepare health systems globally for the ongoing impacts of long COVID.
Our Impact:
Standardizing data elements for documenting and sharing information on long COVID is essential for understanding and addressing its lasting impacts. The EMI team, in collaboration with a Technical Expert Panel (TEP), led a consensus-driven process to identify key data elements for managing long COVID care. Through a two-part, iterative process, the TEP created a structured framework to improve data quality and data collection consistency. This initiative enhances patient care and provides a more transparent, unified picture of long COVID’s widespread effects on health systems. Details of this approach are published in the journal article “Establishing data elements and exchange standards to support long COVID healthcare and research". Key highlights of this work include:
1. Identifying and standardizing these data elements using nationally recognized terminologies to support the capture and electronic exchange.
Assessment Scales – LOINC®
Health concerns – ICD-10-CM (Diagnoses), SNOMED-CT® (Diagnoses, Symptoms, Observations)
Clinical, Imaging and Laboratory tests – LOINC®
Goals - LOINC®
Interventions – HCPCS, CPT® and SNOMED-CT®
2. Dissemination of these standards through the National Library of Medicine (NLM) Value Set Authority Center (VSAC)
Multiple Chronic Conditions (MCC) Questionnaire Response Value Sets (link)
MCC Chronic Conditions and Health Concern Value Sets (link)
MCC Symptoms Value Sets (link)
MCC Simple Observations Value Sets (link)
MCC Observation SDOH Assessment Value Sets (link)
MCC Clinical Tests Value Sets (link)
MCC Laboratory Results Value Sets (link)
MCC Diagnostic Report and Note Imaging Value Sets (link)
MCC Goal Value Sets (link)
MCC Interventions (Procedure and Service Request) Value Sets (link)
Outcomes:
Standardized data for long COVID provides a crucial foundation for interoperable data collection, enhancing care delivery and patient-centered outcomes research. By establishing these standards, we can more accurately identify patients for research studies, support quality improvement initiatives, and streamline clinical data for epidemiologic analysis. Standardization also enables efficient organization of clinical information in health applications and facilitates research on interventions addressing clinical and social needs. Together, these efforts accelerate the evidence needed to develop effective long COVID interventions and address the needs of caregivers supporting patients with multiple chronic conditions.
Figure 1: LONG COVID STANDARDIZED CARE COORDINATION
Real World Application:
The long COVID value sets developed during the MCC eCare project are now being utilized in an Assistant Secretary for Technology Policy (ASTP) project to enhance patient-centered outcomes research (PCOR) for COVID-19 using real-world data from health information exchanges (HIE). This project explores the probability of predicting adverse health outcomes, such as myocardial infarction, in COVID-19 patients with underlying co-morbidities. By leveraging a privacy-preserving split learning model, the project enables clinicians, without sharing raw data, to create personalized care plans for COVID-19 patients at higher risk of adverse health outcomes. The project integrates the long COVID value sets with other value sets identified in the MCC eCare Plan FHIR Implementation Guide and applies HL7 Clinical Quality Language (CQL)—an industry-standardized language designed to express clinical logic—to classify and summarize health data for thousands of patients in the research cohort population. The insights generated from this analysis serve as inputs to train and refine the split learning model. This innovative approach for PCOR highlights the transformative potential of standardized value sets in improving care for high-risk populations.
To learn how EMI Advisors can support your digital health strategy, contact hello@emiadvisors.net.