AUTHOR=Dai Zhenyu , Zeng Dong , Cui Dawei , Wang Dawei , Feng Yanling , Shi Yuhan , Zhao Liangping , Xu Jingjing , Guo Wenjuan , Yang Yuexiang , Zhao Xinguo , Li Duoduo , Zheng Ye , Wang Ao , Wu Minmin , Song Shu , Lu Hongzhou 

TITLE=Prediction of COVID-19 Patients at High Risk of Progression to Severe Disease

JOURNAL=Frontiers in Public Health

VOLUME=Volume 8 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2020.574915

DOI=10.3389/fpubh.2020.574915

ISSN=2296-2565

ABSTRACT=Background. We aimed to develop a novel scoring model for prediction of COVID-19 patients at high risk of severity.
Methods. We retrospectively studied 419 patients from fiver hospitals in Shanghai, Hubei and Jiangsu provinces from January 22 through March 30, 2020. Multivariate COX regression and orthogonal projections to latent structures (OPLS‐DA) were both used to identify the high-risk factors with severity for COVID-19 patients. The prediction model was developed based on all the four high-risk factors.
Results. The multivariable analysis showed the comorbidity (HR 3.17, 95% CI 1.96 - 5.11), ALB (HR 3.67, 95% CI 1.91 - 7.02), CRP (HR 3.16, 95% CI 1.68 - 5.96) and age >=60 (HR 2.31, 95% CI 1.43 - 3.73) were independent risk factors with severity for COVID-19 patients. OPLS‐DA identified that the top five influences parameters with severity for COVID-19 patients were CRP, ALB, age >=60, comorbidity, and LDH. Incorporating above four factors, the nomogram had a good concordance index of 0.86 (95% CI 0.83 - 0.89), and also had an optimal agreement between the predictive nomogram and actual observation with a slope of 0.95 (R2 = 0.89) in 7-day prediction and 0.96 (R2 = 0.92) in 14-day prediction after 1000 bootstrap sampling. The AUROC of COVID-19-AACC model was 0.85(95% CI: 0.81–0.90). According to the probabilities to severity, the model divided the cases into three groups: low risk, intermediate risk and high risk.
Conclusion. The COVID-19-AACC model can provide an effective method for clinicians to screen patients with severe progression tendency.