id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-289542-u86ujtur	Razavian, Narges	A validated, real-time prediction model for favorable outcomes in hospitalized COVID-19 patients	2020-10-06		.txt	text/plain	8068	379	45	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 this article, we describe how a collaboration among data scientists, electronic health record (EHR) programmers (vendorand health system-based), clinical informaticians, frontline physicians, and clinical leadership led to the development, prospective validation, and implementation of a machine learning model for real-time prediction of favorable outcomes within a 96 h window among hospitalized COVID-19 patients. Our approach differs from prior work in that we: (1) predict favorable outcomes (as opposed to adverse outcomes), (2) use a large COVID-19 patient cohort admitted across our hospitals, (3) design a model that can easily be extended to other institutions, (4) prospectively validate performance, and (5) integrate our model in the EHR to provide a real-time clinical decision support tool.	./cache/cord-289542-u86ujtur.txt	./txt/cord-289542-u86ujtur.txt