Clinical Characterization and Prediction of Clinical Severity of SARS-CoV-2 Infection Among US Adults Using Data From the US National COVID Cohort Collaborative

作者：Tellen D. Bennett, Richard A. Moffitt, Janos Hajagos, Benjamin Amor, Adit Anand, Mark M. Bissell, Katie R. Bradwell, Carolyn Bremer, James Brian Byrd, Alina Denham, Peter E. DeWitt, Davera Gabriel, Brian T. Garibaldi, Andrew T. Girvin, Justin Guinney, Elaine Hill, Stephanie Hong, Hunter Jimenez, Ramakanth Kavuluru, Kristin Kostka, Harold P. Lehmann, Eli Levitt, Sandeep K. Mallipattu, Amin Manna, Julie A. McMurry, Michele Morris, John Muschelli, Andrew J. Neumann, Matvey B. Palchuk, Emily Pfaff, Zhenglong Qian, Nabeel Qureshi, Seth Russell, Heidi Spratt, Anita Walden, Andrew E. Williams, Jacob T. Wooldridge, Yun Jae Yoo, Xiaohan Tanner Zhang, Richard L. Zhu, Christopher P. Austin, Joel Saltz, Kenneth Gersing, Melissa Haendel, Christopher G. Chute, Joel Gagnier, Siqing Hu, Kanchan Lota, Sarah E. Maidlow, David A. Hanauer, Kevin J. Weatherwax, Nikhila Gandrakota, Rishikesan Kamaleswaran, Greg S. Martin, Jingjing Qian, Jason E. Farley, Patricia A. Francis, Dazhi Jiao, Hadi Kharrazi, Justin Reese, Mariam Deacy, Usman Ullah Sheikh, Jake Y. Chen, Michael Quinn Patton, T. Bennett Ramsey, Jasvinder A. Singh, James J. Cimino, Jing Su, William G. Adams, Timothy Q. Duong, John B. Buse, Jessica Y. Islam, Jihad S. Obeid, Stephane Meystre, Steve Patterson, Misha Zemmel, Ron Grider, A. Pérez Martínez, Carlos Antônio do Nascimento Santos, Julian Solway, Ryan G. Chiu, Gerald B. Brown, Jia-Feng Cui, Sharon X. Liang, Kamil Khanipov, Jeremy Harper, Peter J. Embí, David Eichmann, Boyd M. Knosp, William B. Hillegass, Chunlei Wu, James R. Aaron, Darren W. Henderson, Muhammad Gul, Tamela Harper, Daniel R. Harris, Jeffery Talbert, Neil Bahroos, Steven M. Dubinett, Jomol Mathew, Gabriel McMahan, Hongfang Liu, Claudia F. Lucchinetti, David L Schwartz, Ralph L. Sacco, Peyman Taghioff, Diane M. Harper, Denise B. Angst, Andrew Marek, Carlos E. Figueroa Castro, Bruce R. Blazar, Steve Johnson, Melissa Basford, Laura Jones, Gordon R. Bernard, Rosalind Wright, Joseph Finkelstein, Thomas R. Campion, Christopher E. Mason, Xiaobo Fuld, Alfred Anzalone, James C. McClay, Shyam Visweswaran, Connor Cook, Alexandra Dest, David H. Ellison, Rose Relevo, Andréa M Volz, Chengda Zhang, Martha M. Tenzer, David Bowers, Francis X. Farrell, Qiuyuan Qin, Martin S. Zand, Jeanne Holden‐Wiltse, Ramkiran Gouripeddi, Julio C. Facelli, Robert A. Clark, Benjamin J. Becerra, Yao Yan, Jimmy Phuong, Yooree Chae, Rena C. Patel, Christine Suver, Elizabeth Zampino, Ahmad Said, Philip Payne, Randeep S. Jawa, Peter L. Elkin, Farrukh M. Koraishy, George Golovko, Vignesh Subbian, Daniel Weisdorf, Lawrence I. Sinoway, Hiroki Morizono, Keith A. Crandall, Ali Rahnavard, Nawar Shara, Alysha J. Taxter, Brian Ostasiewski, Qianqian Song, Uma Maheswara Reddy Vangala, Katherine Ruiz De Luzuriaga, Rasha Khatib, John P. Kirwan, James von Oehsen, Jason H. Moore, Ankit Sakhuja, Joni L. Rutter · 发表于：JAMA Network Open · 年份：2021 · DOI：10.1001/jamanetworkopen.2021.16901 · 被引用次数：250 · 研究领域：COVID-19 Clinical Research Studies、Machine Learning in Healthcare、COVID-19 diagnosis using AI

Importance: The National COVID Cohort Collaborative (N3C) is a centralized, harmonized, high-granularity electronic health record repository that is the largest, most representative COVID-19 cohort to date. This multicenter data set can support robust evidence-based development of predictive and diagnostic tools and inform clinical care and policy. Objectives: To evaluate COVID-19 severity and risk factors over time and assess the use of machine learning to predict clinical severity. Design, Setting, and Participants: In a retrospective cohort study of 1 926 526 US adults with SARS-CoV-2 infection (polymerase chain reaction >99% or antigen <1%) and adult patients without SARS-CoV-2 infection who served as controls from 34 medical centers nationwide between January 1, 2020, and December 7, 2020, patients were stratified using a World Health Organization COVID-19 severity scale and demographic characteristics. Differences between groups over time were evaluated using multivariable logistic regression. Random forest and XGBoost models were used to predict severe clinical course (death, discharge to hospice, invasive ventilatory support, or extracorporeal membrane oxygenation). Main Outcomes and Measures: Patient demographic characteristics and COVID-19 severity using the World Health Organization COVID-19 severity scale and differences between groups over time using multivariable logistic regression. Results: The cohort included 174 568 adults who tested positive for SARS-CoV-2 ...