Risk Prediction of Pneumonia in Maintenance Hemodialysis Patients with COVID-19: A Single-center Retrospective Study in China (preprint)

Objective:  The infection rate and mortality of COVID-19 in hemodialysis patients are extremely high. In this study, we analyzed the risk prediction of pneumonia in Chinese’s hemodialysis patients with COVID-19.  Method: We conducted a retrospective analysis of maintenance hemodialysis patients with COVID-19 admitted to Peking University Third Hospital from December 1, 2022 to January 31, 2023. We collected demographic data, underlying diseases, dialysis treatment data, and laboratory test results of these patients; logistic regression and ROC curve were used to analyze the risk factors of pneumonia.  Results:  A total of 209 hemodialysis patients with COVID-19 were enrolled in this study, of whom 80(38.3%) had pneumonia and 129(61.7%) normal. Multivariate logistic analysis revealed that older age (OR=1.030, 95%CI 1.002~1.059, p=0.036), lower hemoglobin (OR=0.968, 95%CI 0.942~0.995, p=0.019), lower albumin (OR=0.834, 95%CI 0.738-0.943, p=0.004) were risk factors for pneumonia. Patients with age >65 years old, hemoglobin <115g/L or albumin <36.8g/L had a higher risk of pneumonia, and we combined the three indexes to predict the risk of pneumonia (P= elogit(P)    / 1+  elogit(P) , logit(P)=9.593 +0.031×Age (years old) -0.038×Hemoglobin (g/L) -0.210×Albumin (g/L)), drew the ROC curve with the risk P value (AUC=0.756, 95% CI 0.687~0.825, p<0.001), when P=0.575 was selected as the cut-off value, the sensitivity and specificity of the three indexes combined to predict pneumonia were 44.0% and 94.5% respectively.  Conclusion: Older age, anemia and hypoalbuminemia were risk factors for pneumonia in MHD patients. We could reduce the incidence of pneumonia and improve the prognosis of MHD patients by correcting anemia and hypoalbuminemia.

Mobility during the COVID-19 Pandemic and Retail Investor Attention (preprint)

Does the impact to mobility due to state-level lockdown during the COVID-19 pandemic affect retail investors’ attention in equity markets? Using Google’s mobility, the lockdown dates, and companies’ Wikipedia page views data, we show that stay-at-home duration increases retail attention. The effects are more pronounced for firms with non-lottery-like stocks, high institutional ownership, and those located in states with higher proportions of young population and populations that did not work from home prior to the pandemic. We further find increased trading activity in volume and volatility associated with stay-at-home duration.

COVID-19 Pandemic: How Important is Face-to-Face Interaction for Information Dissemination? (preprint)

We examine the influence of mobility on the stock market reaction around earnings announcements during the COVID-19 pandemic. The pandemic has led to abrupt implementation of stay-at-home mandatory restrictions and social distancing requirements all around the world, resulting in the collapse of face-to-face interactions on a large scale. Using Google’s Community Mobility Reports to obtain daily, county-level mobility of U.S. residents, we find that firms located in counties with higher mobility experience stronger immediate price responses during earnings announcements and weaker post-announcement drift, suggesting that COVID-19 restrictions dampen price discovery in financial markets. We also show that investor attention is a potential mechanism by which mobility affects the market reaction. Together, our study finds empirical support for the importance of face-to-face interactions in facilitating stock market information.