Development of a model by LASSO to predict hospital length of stay (LOS) in patients with the SARS-Cov-2 omicron variant. Issue 1 (31st December 2023)

• Record Type: Journal Article
• Authors: Zhang, Jianxiang; Li, Lei; Hu, Xiaobo; Cui, Guangying; Sun, Ranran; Zhang, Donghua; Li, Juan; Li, Yonghong; Shen, Shen; He, Ping; Yu, Zujiang; Ren, Zhigang
• Is Part Of: Virulence. Volume 14:Issue 1(2023)
• Publication Date: 2023-12-31
• Subjects: Length of stay (LOS) -- novel simple model -- omicron variant -- clinical characteristics -- variables; Virulence (Microbiology) -- Periodicals; Bacterial diseases -- Periodicals; Molecular microbiology -- Periodicals; 579.05
• DOI: 10.1080/21505594.2023.2196177
• Languages: English
• ISSNs: 2150-5608

Abstract

ABSTRACT: The length of stay (LOS) in hospital varied considerably in different patients with COVID-19 caused by SARS-CoV-2 Omicron variant. The study aimed to explore the clinical characteristics of Omicron patients, identify prognostic factors, and develop a prognostic model to predict the LOS of Omicron patients. This was a single center retrospective study in a secondary medical institution in China. A total of 384 Omicron patients in China were enrolled. According to the analyzed data, we employed LASSO to select the primitive predictors. The predictive model was constructed by fitting a linear regression model using the predictors selected by LASSO. Bootstrap validation was used to test performance and eventually we obtained the actual model. Among these patients, 222 (57.8%) were female, the median age of patients was 18 years and 349 (90.9%) completed two doses of vaccination. Patients on admission diagnosed as mild were 363 (94.5%). Five variables were selected by LASSO and a linear model, and those with P < 0.05 were integrated into the analysis. It shows that if Omicron patients receive immunotherapy or heparin, the LOS increases by 36% or 16.1%. If Omicron patients developed rhinorrhea or occur familial cluster, the LOS increased by 10.4% or 12.3%, respectively. Moreover, if Omicron patients' APTT increased by one unit, the LOS increased by 0.38%. Five variables were identified, including immunotherapy, heparin, familial cluster, rhinorrhea, and APTT. A simple model was developed and evaluated to predict the LOS of Omicron patients. The formula is as follows: Predictive LOS = exp(1*2.66263 + 0.30778* Immunotherapy + 0.1158*Familiar cluster + 0.1496*Heparin + 0.0989*Rhinorrhea + 0.0036*APTT ).