COVID-19 Electronic Healthcare Data Initiative

PCORnet®, the National Patient-Centered Clinical Research Network, has been working actively to build infrastructure to support COVID-19 surveillance across its sites. Since early in the COVID-19 pandemic, PCORnet has implemented critical changes to its data infrastructure to allow for capture of information on patients diagnosed with respiratory illnesses or tested for SARS-CoV-2. This data is refreshed frequently, often weekly, for a subset of patients receiving COVID-19-related health care and is stored in the PCORnet Common Data Model. In early April 2020, PCORnet began querying the COVID-19 Cohort, providing detailed characterizations of demographics, comorbidities, and treatments of patients diagnosed and testing positive for COVID-19as well as patients diagnosed with viral pneumonia and influenza. By July 21, 2020, 49 sites responded to queries, and PCORnet had information on nearly 90,000adult and 10,000 pediatric patients diagnosed with COVID-19 or another coronavirus and a similar number of adults and children who tested positive for SARS-CoV-2. Information is also available on more than 200,000 patients with influenza and nearly 1 million who tested negative for SARS-CoV-2.

Starting in fall of 2020, PHII and the CDC has provided funding for COVID-19 surveillance work in PCORnet, including the ongoing updates/refreshes to the COVID-19 Cohort, biweekly queries to generate aggregate reports, and surveillance-related research of patient-level data. Funding will support sites to maintain their data infrastructure for 11 months and will support the data and analytic teams for the same period. Examples of high priority topics identified by CDC and PCORnet include:

  • Characteristics of adults and children infected with COVID-19 and other infections, by geography and other designated subgroups;
  • Trends in COVID-19 infections, especially unusual complications and other rare outcomes; long-term sequelae of infection; and use of COVID-19 treatments over time;
  • Predictive modeling of factors associated with disease severity stratified by demographic (e.g., age, payor type) and clinical subgroups (e.g., race and ethnicity, diabetes and obesity, and use of certain medications such as immunosuppressive agents); and
  • Trends in hospitalizations for other diseases (e.g., myocardial infarction, appendicitis, diabetic ketoacidosis) during the COVID-19 pandemic, allowing for an assessment of the potential effect of delays in case.