The National COVID Cohort Collaborative: Clinical Characterization and Early Severity Prediction

Abstract

BackgroundThe majority of U.S. reports of COVID-19 clinical characteristics, disease course, and treatments are from single health systems or focused on one domain. Here we report the creation of the National COVID Cohort Collaborative (N3C), a centralized, harmonized, high-granularity electronic health record repository that is the largest, most representative U.S. cohort of COVID-19 cases and controls to date. This multi-center dataset supports robust evidence-based development of predictive and diagnostic tools and informs critical care and policy.Methods and FindingsIn a retrospective cohort study of 1,926,526 patients from 34 medical centers nationwide, we stratified patients using a World Health Organization COVID-19 severity scale and demographics; we then evaluated differences between groups over time using multivariable logistic regression. We established vital signs and laboratory values among COVID-19 patients with different severities, providing the foundation for predictive analytics. The cohort included 174,568 adults with severe acute respiratory syndrome associated with SARS-CoV-2 (PCR >99% or antigen <1%) as well as 1,133,848 adult patients that served as lab-negative controls. Among 32,472 hospitalized patients, mortality was 11.6% overall and decreased from 16.4% in March/April 2020 to 8.6% in September/October 2020 (p = 0.002 monthly trend). In a multivariable logistic regression model, age, male sex, liver disease, dementia, African-American and Asian race, and obesity were independently associated with higher clinical severity. To demonstrate the utility of the N3C cohort for analytics, we used machine learning (ML) to predict clinical severity and risk factors over time. Using 64 inputs available on the first hospital day, we predicted a severe clinical course (death, discharge to hospice, invasive ventilation, or extracorporeal membrane oxygenation) using random forest and XGBoost models (AUROC 0.86 and 0.87 respectively) that were stable over time. The most powerful predictors in these models are patient age and widely available vital sign and laboratory values. The established expected trajectories for many vital signs and laboratory values among patients with different clinical severities validates observations from smaller studies, and provides comprehensive insight into COVID-19 characterization in U.S. patients.ConclusionsThis is the first description of an ongoing longitudinal observational study of patients seen in diverse clinical settings and geographical regions and is the largest COVID-19 cohort in the United States. Such data are the foundation for ML models that can be the basis for generalizable clinical decision support tools. The N3C Data Enclave is unique in providing transparent, reproducible, easily shared, versioned, and fully auditable data and analytic provenance for national-scale patient-level EHR data. The N3C is built for intensive ML analyses by academic, industry, and citizen scientists internationally. Many observational correlations can inform trial designs and care guidelines for this new disease.

A Machine-learning Parsimonious Multivariable Predictive Model of Mortality Risk in Patients With Covid-19

“…Other predictive models have been published previously, many of which report age, hematologic measures, Creactive protein and spO2 as the main variables explaining the predictive model 7,8 . Our results con rm and extend those of other large cohort studies [7][8][9][10][11][12][13] demonstrating the predictive value of renal function and, in particular, of blood urea nitrogen for mortality 14,16 . In addition, we share 4 of 9 variables from a machine-learning-based study with the largest included population 14 .…”

“…Unfortunately, treatment options are currently scarce, and as hospital resources are shrinking, systems to target respiratory support and other hospital resources to the highest-risk population, such as the ICU, is a priority. Several predictive models of adverse clinical outcomes in people with COVID-19 [7][8][9][10][11][12][13] as well as a systematic review 14 have been published. Having a clinical algorithm to predict patients who can bene t most from available resources is a valuable aid for decision making and capacity allocation.…”

Association Between Glucagon-Like Peptide 1 Receptor Agonist and Sodium–Glucose Cotransporter 2 Inhibitor Use and COVID-19 Outcomes

“…We analyzed data from a cohort study using COVID-19 data from health care systems across the U.S. contributing to the National COVID Cohort Collaborative (N3C) (19). The N3C cohort includes individuals with any encounter after 1 January 2020 and one or a combination of more than one of a set of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) laboratory tests, predefined based on diagnostic codes as defined by the N3C phenotype definition team (20,21).…”

NSAID use and clinical outcomes in COVID-19 patients: A 38-center retrospective cohort study

“…The dependent variable was COVID-19 severity, an ordered factor with levels "mild", "mild ED", "moderate", "severe", and tis rts hIt up er, e as nd "mortality/hospice". 21 We assessed the relationship between COVID-19 severity and treatment with each medication under consideration as a part of multiple OLR with age, race, ethnicity, gender, smoking status, Charlson Comorbidity Index, and BMI as additional predictors. For treatment with the medication we recorded the t value, the corresponding p value, and the odds ratio along with its 95% confidence interval.…”

“…A single index encounter was defined for each laboratory-confirmed positive patient as described previously. 21 Survival time was defined with respect to this encounter. For each patient, the latest visit and date of death (if observed) was recorded to measure right censoring (the last date for which death outcome can be ruled out) and survival, respectively.…”