The roles of meteorological variables, demographic factors, and policy response measures in the variation of COVID-19 daily cases: Evidence from different climate zones (preprint)

It is widely considered that weather conditions affect the spread of COVID-19, but to date, the collective influence of demographic factors and government policy response measures have hardly been considered. The objective of this study is to utilize a machine learning method to assess the corresponding roles of meteorological variables, demographic factors, and government response measures in daily new cases of COVID-19 among multiple climate zones at city/county level. The overall model showed good performance with a validated R2 of 0.86, as satisfactory as individual climate zone models. Population density ranked the most important factor, followed by meteorological variables and response measures. Ultraviolet radiation and temperature dominated among meteorological factors, but the association with daily new cases seemed to be inconsistent among different climate zones. Implementing stricter response measures could help effectively contain the spread of COVID-19, but did so with a lagged effect, and the typical lockdown measures might not be applicable to all climate conditions. This study preliminarily analyzed the roles of certain factors in the transmission of COVID-19, and provided practical evidence for developing an early health warning system of global pandemics by leveraging big data technology and multiple sourced data fusion.

Risk factors associated with fatal outcomes of novel coronavirus infection pneumonia (COVID-19): A systematic review and meta-analysis (preprint)

Background: The COVID-19 infection has caused 111652 deaths worldwide as of 13 April 2020. Risk factors for fatal outcomes of COVID-19 have varied across studies due to limited samples and have lacked effective qualitative and quantitative measurements. We performed a meta-analysis to evaluate risk factors for fatal outcomes of COVID-19.Methods: Data on demographic, clinic, laboratory findings and complications were extracted. Quantitative and qualitative synthesis was conducted for weighted-mean-difference (WMD) and odds-ratio (OR).Results: A total of 30 studies involving 5741 survivors and 1670 deaths were included. The death cases were significantly older than survivors (WMD=15.36, 95% CI: 12.90-17.82), male and smoking history showed higher risk to develop fatal outcome (OR=3.37, 95% CI: 2.27-5.01; OR=1.37, 95% CI: 1.02-1.83, respectively). The clinical symptoms including dyspnea (OR=4.63, 95% CI: 2.85-7.54), hemoptysis (OR=3.11, 95% CI: 1.26-7.56), malaise (OR=2.44, 95% CI: 1.49-3.97). comorbidities with coronary heart disease (OR=4.36, 95% CI: 1.91-9.97), COPD (OR=3.70, 95% CI: 2.03-6.73) and cardiovascular disease (OR=3.45, 95% CI: 2.54-4.70). Compared to survivors, many laboratory indexes increased in deaths group, including serum ferritin (WMD=741.47, 95% CI: 566.77-916.16), lactate dehydrogenase (WMD=226.86, 95% CI: 177.08-276.64) and myoglobin (WMD=102.58, 95% CI: 65.12-140.04), and the decreased indexes included PaO2/FiO2 (WMD=-71.61, 95% CI: -134.11 to -9.11), platelets (WMD=-41.09, 95% CI: -47.33 to -34.85) and PaO2 (WMD=-26.09, 95% CI: -38.9 to -13.29). Main complications contributed to the fatal outcome included sepsis (OR=184.61, 95% CI: 33.43-1019.42), shock (OR=133.76, 95% CI: 36.86-485.34) and respiratory failure (OR=47.37, 95% CI: 20.65-108.66). Conclusion: The main risk factors associated with fatal outcome of COVID-19 involved male, older age, smoking history, chronic medical conditions including coronary heart disease, COPD and cardiovascular disease, clinical symptoms including dyspnea, hemoptysis and malaise, the increased laboratory indexes including serum ferritin, lactate dehydrogenase and myoglobin, the decreased indexes including PaO2/FiO2, platelets and PaO2, main complications including sepsis, shock and respiratory failure. These factors could be considered in triaging patients and allocating medical resources when such medical resources are scarce, devising improved protocols for patient diagnosis and management, and developing new drugs and other therapies to treat COVID-19 patients.

Risk factors and clinical features of deterioration in COVID-19 patients in Zhejiang, China: a single- centre, retrospective study (preprint)

BackgroundSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection swept through Wuhan and spread across China and overseas beginning in December 2019. To identify predictors associated with disease progression, we evaluated clinical risk factors for exacerbation of SARS-CoV-2 infection.MethodsA retrospective analysis was used for PCR-confirmed COVID-19 (coronavirus disease 2019)-diagnosed hospitalized cases between January 19, 2020, and February 19, 2020, in Zhejiang, China. We systematically analysed the clinical characteristics of the patients and predictors of clinical deterioration.ResultsOne hundred patients with COVID-19, with a median age of 54 years, were included. Among them, 49 patients (49%) had severe and critical disease. Age ([36-58] vs [51-70], P=0.0001); sex (49% vs 77.6%, P=0.0031); Body Mass Index (BMI ) ([21.53-25.51] vs [23.28-27.01], P=0.0339); hypertension (17.6% vs 57.1%, P<0.0001); IL-6 ([6.42-30.46] vs [16.2-81.71], P=0.0001); IL-10 ([2.16-5.82] vs [4.35-9.63], P<0.0001); T lymphocyte count ([305- 1178] vs [167.5-440], P=0.0001); B lymphocyte count ([91-213] vs [54.5-163.5], P=0.0001); white blood cell count ([3.9-7.6] vs [5.5-13.6], P=0.0002); D2 dimer ([172-836] vs [408-953], P=0.005), PCT ([0.03-0.07] vs [0.04-0.15], P=0.0039); CRP ([3.8-27.9] vs [17.3-58.9], P<0.0001); AST ([16, 29] vs [18, 42], P=0.0484); artificial liver therapy (2% vs 16.3%, P=0.0148); and glucocorticoid therapy (64.7% vs 98%, P<0.0001) were associated with the severity of the disease. Age and weight were independent risk factors for disease severity.ConclusionDeterioration among COVID-19-infected patients occurred rapidly after hospital admission. In our cohort, we found that multiple factors were associated with the severity of COVID19. Early detection and monitoring of these indicators may reduce the progression of the disease. Removing these factors may halt the progression of the disease. In addition, Oxygen support, early treatment with low doses of glucocorticoids and liver therapy, when necessary, may help reduce mortality in critically ill patients.

Retrospective study of low-to-moderate dose glucocorticoids on viral clearance in patients with novel coronavirus pneumonia / 中华临床感染病杂志

Objective@#To study the effect of low-to-moderate dose glucocorticoid therapy on viral clearance time in patients with COVID-19.@*Methods@#A total of 72 patients diagnosed with COVID-19 from January 19 to February 17, 2020 at the First Affiliated Hospital, School of Medicine, Zhejiang University were recruited. All patients received oral abidol and/or combined lopinavir/ritonavir, darunavir antiviral, and symptomatic supportive care. Among them, 51 patients received methylprednisolone (0.75-1.50 mg·kg-1·d-1) (glucocorticoid treatment group), and 21 patients who did not use glucocorticoid were the control group. The time of stable virologic conversion insputumand the time of radiologic recovery in lungsince onset were compared between the two groups and among the normal patients.The Kruskal-Wallis test or Fisher exact test was used to compare the difference between groups.@*Results@#The median ages of the glucocorticoid group and the control group were 52 [interquartile range (IQR):45, 62] years and 46 (IQR: 32, 56)years, and the differences were significant (P<0.05). The clinical conditions at hospital admission were different between the two groups (P<0.01). There were 52.0% critical ill patients in the glucocorticoid treatment group, compared to that of 71.4% normal patients in the control group. The median times from the onset tostable virologic conversion to negative in the two groups were 15 (IQR:13,20) days and 14 (IQR:12,20) days (P>0.05), and the difference was no statistically significant. The median times from onset to radiologic recovery were 13 (IQR: 11,15) days and 13 (IQR:12,17) days in the two groups, and there was no difference (P>0.05). In ordinary patients, the median timesfrom the onset tostable virologic conversion insputum were no difference (P>0.05), with 13 (IQR:11,18) days in the glucocorticoid group and 13 (IQR:12,15) days in the control group; The median times from onset to radiologic recovery in lungwere also no difference (P>0.05), with 12 (IQR: 10,15)days in the glucocorticoid group and 13 (IQR: 12,17) days inthe control group.@*Conclusions@#Low-to-moderate glucocorticoid treatment has no effect on the time of virus clearance in patients with different clinical types of COVID-19. The glucocorticoid is not recommended since no effectiveness on accelerating the improvement of radiologic recovery in lung has been observed.