The C3I COVID-19 Project

Description

This cohort study will obtain electronic health record (EHR) de-identified data from 21 health systems affiliated with the Cancer Center Cessation Initiative (C3I) network or health systems with large numbers of COVID-19 patients to explore whether smoking status, cancer history, and other risk factors among patients diagnosed with COVID-19 are associated with mortality and/or COVID-19 disease severity/complications. The Cancer Center Cessation Initiative (C3I) is a project launched by the US National Cancer Institute (NCI) to improve the rate at which NCI-designated Cancer Centers provide evidence-based smoking cessation to patients diagnosed with and treated for cancer. The C3I is coordinated at the University of Wisconsin-Center for Tobacco Research and Intervention (UW-CTRI) and the University of Wisconsin Carbone Cancer Center (UWCCC). Twenty-one health systems across the U.S. will provide EHR de-identified data to the UW-CTRI coordinating center on all COVID-19 patients identified during the period from February 1, 2020, through December 31, 2020.

Current EHR-based data elements collected will include:

Evidence of COVID-19: ICD-10-CM diagnosis of COVID-19, COVID-19 PCR lab test, and/or COVID-19 antibody lab test

Healthcare system encounter type: inpatient, outpatient, emergency department (ED), urgent care, or other

SES/Demographics variables: insurance status, education, housing status, sex, age, race/ethnicity, height, weight, body mass index

Comorbid diseases: chronic asthma, chronic COPD, chronic bronchiectasis, diabetes mellitus, cardiovascular disease, chronic renal disease, on dialysis, immunocompromised [due to SLE lupus, rheumatoid arthritis, organ transplant, HIV, Crohns], pregnant, cancer (lymphomas, leukemias, lung/respiratory, rectal, breast, prostate, pancreas), hypertension, depression, anxiety, alcohol abuse, pro- clotting disorders, and anti-clotting disorders

Tobacco use variables: smoking status (current, former, never), passive smoke exposure for never smoker, years since quitting (for former smokers), packs smoked per day, years of smoking, pack years, smokeless tobacco user, and marijuana use

Radiographic tests: chest CTs/MRIs, head CTs/MRIs, abdominal CTs/MRIs, lower extremities ultrasound, and cardiac echo

Signs and symptoms: temperature, pulse, systolic blood pressure, diastolic blood pressure, oxygen saturation, septic shock, pneumonia, chills, muscle aches/myalgia, rhinorrhea, sore throat, chronic cough, shortness of breath, nausea or vomiting, headache, abdominal pain, diarrhea, dizziness, impaired consciousness, acute cerebrovascular event, ataxia, seizure, taste impairment, smell impairment, vision impairment, nerve pain, and skeletal muscular pain

COVID-19 treatment variables: ICU admission, required supplemental oxygen, intubated for ventilator use, noninvasive positive pressure, and number of days hospitalized COVID-19 medications: chloroquine, hydroxychloroquine, tocilizumab, remdesivir, dexamethasone, convalescent plasma, and ascorbic acid

Classes of other medications used: nicotine replacement therapies (NRTs), varenicline medications; blood thinners, steroids, angiotensin converting enzyme (ACE) inhibitors, angiotensin-receptor blockers (ARBs), short-acting adrenergic bronchodilators, long-acting adrenergic bronchodilators, anticholinergic bronchodilators, bronchodilators combos, inhaled corticosteroids (includes combo medications), and inhaled corticosteroid bronchodilators

Lab tests: CBC, hematology (e.g., neutrophils, lymphocytes, hemoglobin, hematocrit), chemistry (e.g., potassium, BUN, creatinine, glucose), liver function tests, coagulation (INR, D-dimer), inflammatory markers (e.g., ESR, IL-6, C-reactive protein), troponin, Hemoglobin A1C, and tests for various infections

NOTE: As the COVID-19 pandemic evolves and additional relevant EHR variables are identified (e.g., convalescent plasma), they will be added to the list of variables collected.

For the initial paper(s) to be prepared based on these data, the main analytic methods will include GUIDE classification and regression tree models. However, whole sample methods will also be used as complementary analytic methods, which will vary with regard to outcome type: i.e., logistic regression for binary outcomes and Cox proportional hazard analyses for time-to-event outcomes. Initial analyses will focus on hospitalized COVID-19 patients. Later waves of analyses may use different analytic approaches and address different questions. These will be described in subsequent CT.gov filings.

Participating healthcare systems:

• Duke University (Duke Health)
• Hackensack Meridian Health
• Mayo Clinic
• Memorial Sloan Kettering Cancer Center
• University of Michigan (Michigan Medicine)
• Mount Sinai Health System
• Northwestern University
• New York University (NYU Langone Health)
• University of California Davis (University of California Davis Comprehensive Cancer Center / UC Davis Health)
• University of California San Francisco
• University of North Carolina at Chapel Hill (UNC Health)
• University of Chicago
• University of Illinois at Chicago
• University of Kansas (University of Kansas Medical Center)
• University of Maryland
• University of Utah (University of Utah Health)
• University of Wisconsin (UW Health)
• Vanderbilt University (Vanderbilt University Medical Center)
• Virginia Commonwealth University (VCU Health System/Massey Cancer Center)
• Washington University St Louis
• Yale University (Yale New Haven Hospital)

Details