Early Prediction Model for Critical Illness of Hospitalized COVID-19 Patients Based on Machine Learning Techniques.

Motivation: Patients with novel coronavirus disease 2019 (COVID-19) worsen into critical illness suddenly is a matter of great concern. Early identification and effective triaging of patients with a high risk of developing critical illness COVID-19 upon admission can aid in improving patient care, increasing the cure rate, and mitigating the burden on the medical care system. This study proposed and extended classical least absolute shrinkage and selection operator (LASSO) logistic regression to objectively identify clinical determination and risk factors for the early identification of patients at high risk of progression to critical illness at the time of hospital admission. Methods: In this retrospective multicenter study, data of 1,929 patients with COVID-19 were assessed. The association between laboratory characteristics measured at admission and critical illness was screened with logistic regression. LASSO logistic regression was utilized to construct predictive models for estimating the risk that a patient with COVID-19 will develop a critical illness. Results: The development cohort consisted of 1,363 patients with COVID-19 with 133 (9.7%) patients developing the critical illness. Univariate logistic regression analysis revealed 28 variables were prognosis factors for critical illness COVID-19 (p < 0.05). Elevated CK-MB, neutrophils, PCT, α-HBDH, D-dimer, LDH, glucose, PT, APTT, RDW (SD and CV), fibrinogen, and AST were predictors for the early identification of patients at high risk of progression to critical illness. Lymphopenia, a low rate of basophils, eosinophils, thrombopenia, red blood cell, hematocrit, hemoglobin concentration, blood platelet count, and decreased levels of K, Na, albumin, albumin to globulin ratio, and uric acid were clinical determinations associated with the development of critical illness at the time of hospital admission. The risk score accurately predicted critical illness in the development cohort [area under the curve (AUC) = 0.83, 95% CI: 0.78-0.86], also in the external validation cohort (n = 566, AUC = 0.84). Conclusion: A risk prediction model based on laboratory findings of patients with COVID-19 was developed for the early identification of patients at high risk of progression to critical illness. This cohort study identified 28 indicators associated with critical illness of patients with COVID-19. The risk model might contribute to the treatment of critical illness disease as early as possible and allow for optimized use of medical resources.

Development and Validation of a Predictive Nomogram with Age and Laboratory Findings for Severe COVID-19 in Hunan Province, China.

Purpose: To identify more objectively predictive factors of severe outcome among patients hospitalized for coronavirus disease 2019 (COVID-19). Patients and Methods: A retrospective cohort of 479 hospitalized patients diagnosed with COVID-19 in Hunan Province was selected. The prognostic effects of factors such as age and laboratory indicators were analyzed using the Kaplan-Meier method and Cox proportional hazards model. A prognostic nomogram model was established to predict the progression of patients with COVID-19. Results: A total of 524 patients in Hunan province with COVID-19 from December 2019 to October 2020 were retrospectively recruited. Among them, 479 eligible patients were randomly assigned into the training cohort (n = 383) and validation cohort (n = 96), at a ratio of 8:2. Sixty-eight (17.8%) and 15 (15.6%) patients developed severe COVID-19 after admission in the training cohort and validation cohort, respectively. The differences in baseline characteristics were not statistically significant between the two cohorts with regard to age, sex, and comorbidities (P > 0.05). Multivariable analyses included age, C-reactive protein, fibrinogen, lactic dehydrogenase, neutrophil-to-lymphocyte ratio, urea, albumin-to-globulin ratio, and eosinophil count as predictive factors for patients with progression to severe COVID-19. A nomogram was constructed with sufficient discriminatory power (C index = 0.81), and proper consistency between the prediction and observation, with an area under the ROC curve of 0.81 and 0.86 in the training and validation cohort, respectively. Conclusion: We proposed a simple nomogram for early detection of patients with non-severe COVID-19 but at high risk of progression to severe COVID-19, which could help optimize clinical care and personalized decision-making therapies.

Empirical Research on the Relationship between Industry Working Capital Shortfall and Company Cash Holding in the Same Industry

Affected by the fluctuation of the market and economic environment during the epidemic period, the capital pressure of companies has increased sharply, which increases the possibility of risk transmission in the same industry and poses new challenges to the operation and financing of companies. From the perspective of preventive motivation of corporate cash holding, we creatively use the industry working capital shortfall as the explanatory variable to construct an Extended Cash Holding Model and a Cash Holding Value Regression Model. Taking the panel data of A-share listed companies in Shanghai and Shenzhen as samples, this paper uses the Classical Linear Regression Model and Fixed Effects Regression Model to study the relationship between industry working capital shortfall and cash holdings in the same industry, as well as the relationship between industry working capital shortfall and cash holding value. The empirical results show that industry working capital shortfall has an important impact on the cash holding level within the same industry, and the cash holding level is significantly and positively correlated with industry working capital shortfall. Moreover, this study also reveals that the industry working capital shortfall has a dual impact on cash holdings. Specifically, the higher the risk of industry working capital shortfall, the lower the value of company cash holdings. The conclusions of this paper not only extend the research on cash holding but also provide support and reference for companies to optimize the cash holding value.