TY - JOUR
T1 - A validated, real-time prediction model for favorable outcomes in hospitalized COVID-19 patients
AU - Razavian, Narges
AU - Major, Vincent J.
AU - Sudarshan, Mukund
AU - Burk-Rafel, Jesse
AU - Stella, Peter
AU - Randhawa, Hardev
AU - Bilaloglu, Seda
AU - Chen, Ji
AU - Nguy, Vuthy
AU - Wang, Walter
AU - Zhang, Hao
AU - Reinstein, Ilan
AU - Kudlowitz, David
AU - Zenger, Cameron
AU - Cao, Meng
AU - Zhang, Ruina
AU - Dogra, Siddhant
AU - Harish, Keerthi B.
AU - Bosworth, Brian
AU - Francois, Fritz
AU - Horwitz, Leora I.
AU - Ranganath, Rajesh
AU - Austrian, Jonathan
AU - Aphinyanaphongs, Yindalon
N1 - Funding Information:
Implementation into the EHR could not have been possible without expertise from a variety of teams within Epic Systems. In particular, we would like to acknowledge: Adrienne Alimasa, Garry Bowlin, Erin Ello, Nick Krueger, Sean McGunigal, Joe McNitt, Ben Noffke, George Redgrave, and Owen Sizemore. Narges Razavian was partially supported by Leon Lowenstein Foundation Grant. Mukund Sudarshan and Rajesh Ranganath were partly supported by NIH R01HL148248. Mukund Sudarshan was partially supported by a PhRMA Foundation Predoctoral Fellowship. Yin Aphinyana-phongs was partially supported by NIH 3UL1TR001445-05 and National Science Foundation award #1928614.
Publisher Copyright:
© 2020, The Author(s).
PY - 2020/12/1
Y1 - 2020/12/1
N2 - The COVID-19 pandemic has challenged front-line clinical decision-making, leading to numerous published prognostic tools. However, few models have been prospectively validated and none report implementation in practice. Here, we use 3345 retrospective and 474 prospective hospitalizations to develop and validate a parsimonious model to identify patients with favorable outcomes within 96 h of a prediction, based on real-time lab values, vital signs, and oxygen support variables. In retrospective and prospective validation, the model achieves high average precision (88.6% 95% CI: [88.4–88.7] and 90.8% [90.8–90.8]) and discrimination (95.1% [95.1–95.2] and 86.8% [86.8–86.9]) respectively. We implemented and integrated the model into the EHR, achieving a positive predictive value of 93.3% with 41% sensitivity. Preliminary results suggest clinicians are adopting these scores into their clinical workflows.
AB - The COVID-19 pandemic has challenged front-line clinical decision-making, leading to numerous published prognostic tools. However, few models have been prospectively validated and none report implementation in practice. Here, we use 3345 retrospective and 474 prospective hospitalizations to develop and validate a parsimonious model to identify patients with favorable outcomes within 96 h of a prediction, based on real-time lab values, vital signs, and oxygen support variables. In retrospective and prospective validation, the model achieves high average precision (88.6% 95% CI: [88.4–88.7] and 90.8% [90.8–90.8]) and discrimination (95.1% [95.1–95.2] and 86.8% [86.8–86.9]) respectively. We implemented and integrated the model into the EHR, achieving a positive predictive value of 93.3% with 41% sensitivity. Preliminary results suggest clinicians are adopting these scores into their clinical workflows.
UR - http://www.scopus.com/inward/record.url?scp=85092029417&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092029417&partnerID=8YFLogxK
U2 - 10.1038/s41746-020-00343-x
DO - 10.1038/s41746-020-00343-x
M3 - Article
AN - SCOPUS:85092029417
SN - 2398-6352
VL - 3
JO - npj Digital Medicine
JF - npj Digital Medicine
IS - 1
M1 - 130
ER -