Real-world Effectiveness of BNT162b2 in Children and Adolescents in Preventing Infection and Severe Diseases with SARS-CoV-2 During the Delta and Omicron Periods: Findings from Trial Emulation Using an EHR-based Cohort (preprint)

BACKGROUND The current understanding of the long-term effectiveness of the BNT162b2 vaccine for a range of outcomes across diverse U.S. pediatric populations is limited. In this study, we assessed the effectiveness of BNT162b2 against various strains of the SARS-CoV-2 virus using data from a national collaboration of pediatric health systems (PEDSnet). METHODS We emulated three target trials to assess the real-world effectiveness of BNT162b2: adolescents aged 12 to 20 years during the Delta variant period (Target trial 1), children aged 5 to 11 years (Target trial 2) and adolescents aged 12 to 20 years during the Omicron variant period (Target trial 3). The outcomes included documented infection, COVID-19 illness severity, admission to an intensive care unit (ICU), and two cardiac-related outcomes, myocarditis and pericarditis. In the U.S., immunization records are often captured and stored across multiple disconnected sources, resulting in incomplete vaccination records in patients' electronic health records (EHR). We implemented a novel trial emulation pipeline accounting for possible misclassification bias in vaccine documentation in EHRs. The effectiveness of the BNT162b2 vaccine was estimated from the Poisson regression model with confounders balanced via propensity score stratification. RESULTS During the Delta period, the BNT162b2 vaccine demonstrated an overall effectiveness 98.4% (95% CI, 98.1 to 98.7) against documented infection among adolescents, with no significant waning after receipt of the first dose. During the Omicron period, the overall effectiveness was estimated to be 74.3% (95% CI, 72.2 to 76.2) in preventing documented infection among children, which was higher against moderate or severe COVID-19 (75.5%; 95% CI, 69.0 to 81.0) and ICU admission with COVID-19 (84.9%; 95% CI, 64.8 to 93.5). In the adolescent population, the overall effectiveness against documented Omicron infection was 85.5% (95% CI, 83.8 to 87.1), with effectiveness of 84.8% (95% CI, 77.3 to 89.9) against moderate or severe COVID-19, and 91.5% (95% CI, 69.5 to 97.6) against ICU admission with COVID-19. The effectiveness of the BNT162b2 vaccine against the Omicron variant declined after 4 months following the first dose and then stabilized with higher levels of uncertainty. Across all three cohorts, the risk of cardiac outcomes was approximately 65% to 85% lower in the vaccinated group than that of the unvaccinated group accounting for possible misclassification bias. CONCLUSIONS This study suggests BNT162b2 was effective among children and adolescents in Delta and Omicron periods for a range of COVID-19-related outcomes and is associated with a lower risk for cardiac complications. Waning effectiveness over time suggests that revaccination may be needed in the future.

Risk Factors and Predictive Modeling for Post-Acute Sequelae of SARS-CoV-2 Infection: Findings from EHR Cohorts of the RECOVER Initiative (preprint)

Background Patients who were SARS-CoV-2 infected could suffer from newly incidental conditions in their post-acute infection period. These conditions, denoted as the post-acute sequelae of SARS-CoV-2 infection (PASC), are highly heterogeneous and involve a diverse set of organ systems. Limited studies have investigated the predictability of these conditions and their associated risk factors. Method In this retrospective cohort study, we investigated two large-scale PCORnet clinical research networks, INSIGHT and OneFlorida+, including 11 million patients in the New York City area and 16.8 million patients from Florida, to develop machine learning prediction models for those who are at risk for newly incident PASC and to identify factors associated with newly incident PASC conditions. Adult patients aged  20 with SARS-CoV-2 infection and without recorded infection between March 1st, 2020, and November 30th, 2021, were used for identifying associated factors with incident PASC after removing background associations. The predictive models were developed on infected adults. Results We find several incident PASC, e.g., malnutrition, COPD, dementia, and acute kidney failure, were associated with severe acute SARS-CoV-2 infection, defined by hospitalization and ICU stay. Older age and extremes of weight were also associated with these incident conditions. These conditions were better predicted (C-index >0.8). Moderately predictable conditions included diabetes and thromboembolic disease (C-index 0.7-0.8). These were associated with a wider variety of baseline conditions. Less predictable conditions included fatigue, anxiety, sleep disorders, and depression (C-index around 0.6). Conclusions This observational study suggests that a set of likely risk factors for different PASC conditions were identifiable from EHRs, predictability of different PASC conditions was heterogeneous, and using machine learning-based predictive models might help in identifying patients who were at risk of developing incident PASC. 

Blue LED light promoting the growth, accumulation of high-value isoflavonoids and astragalosides, antioxidant response, and biosynthesis gene expression in Astragalus membranaceus (Fisch.) Bunge hairy root cultures (preprint)

The root of Astragalus membranaceus (Fisch.) Bunge is one of the most frequently used herbs in traditional Chinese medicine (TCM) formulae for fighting COVID-19 infections, due to the presence of isoflavonoids and astragalosides associated with antiviral and immune-enhancing activities. For the first time, the exposure of A. membranaceus hairy root cultures (AMHRCs) to different colors of LED lights i.e., red, green, blue, red/green/blue (1/1/1, RGB), and white, was conducted to promote the root growth and accumulation of isoflavonoids and astragalosides. LED light treatment regardless of colors was found beneficial for root growth, which might be a result of the formation of more root hairs upon light stimulation. Blue LED light was found most effective for enhancing phytochemical accumulation. Results showed that the productivity of root biomass in blue-light grown AMHRCs with an initial inoculum size of 0.6% for 55 days was 1.40-fold higher than that in dark (control), and yields of high-value isoflavonoids and astragalosides including calycosin, formononetin, astragaloside IV, and astragaloside I increased by 3.17-fold, 2.66-fold, 1.78-fold, and 1.52-fold relative to control, respectively. Moreover, the photooxidative stress together with transcriptional activation of biosynthesis genes might contribute to the enhanced accumulation of isoflavonoids and astragalosides in blue-light grown AMHRCs. Overall, this work offered a feasible approach for obtaining higher yields of root biomass and medicinally important compounds in AMHRCs via the simple supplementation of blue LED light, which made blue-light grown AMHRCs industrially attractive as plant factory in controlled growing systems.

Exploring the roles of fear and powerlessness in the relationship between perceived risk of the COVID-19 pandemic and information-avoidance behavior

The COVID-19 has seriously impacted various aspects of the society on a global scale. However, it is still unclear how perceived risk influences epidemic information-avoidance behavior which generally helps us understand public information avoidance. This study aimed to assess the relationship between the perceived epidemic risk and information-avoidance behavior and the mediating role of fear and powerlessness during the COVID-19 pandemic in China. A total of 557 Chinese respondents with COVID-19 treated in modular hospitals ranging from 16 to 72 years old were recruited and completed questionnaires in the face-to-face manner containing scales of the perceived epidemic risk of COVID-19, fear, powerlessness, and information-avoidance behavior. To test the conceptual model, we adopted structural equation modeling (SEM) with the perceived risk of the COVID-19 pandemic as a predictor, fear and powerlessness as mediating variables, and information-avoidance behavior as the outcome. The results indicated a significant and positive association between the perceived epidemic risk of COVID-19 and information-avoidance behavior. Powerlessness acted as the mediator between the perceived epidemic risk of COVID-19 and information-avoidance behavior. The perceived epidemic risk of COVID-19 influenced information-avoidance behavior through fear and powerlessness in turn. Findings from this study implied that public health managers should consider the mediating roles of negative emotions such as fear and powerlessness for coping with behaviors in public health emergencies, especially the information avoidance behaviors related to risk perception.

Efficacy and safety of granulocyte–macrophage colony-stimulating factor (GM-CSF) antibodies in COVID-19 patients: a meta-analysis

Objective This study aims to determine the efficacy and safety of granulocyte–macrophage colony-stimulating factor (GM-CSF) antibodies in COVID-19 patients. Methods We searched Cochrane Library, PubMed, Embase, and ClinicalTrials.gov databases until July 27, 2022. Both randomized control trials (RCTs) and cohort studies were included and analyzed separately. The outcomes included mortality, incidence of invasive mechanical ventilation (IMV), ventilation improvement rate (need oxygen therapy to without oxygen therapy), secondary infection, and adverse events (AEs). The odds ratio (OR) with a 95% confidence interval (CI) was calculated by a random-effects meta-analysis model. Results Five RCTs and 2 cohort studies with 1726 COVID-19 patients were recruited (n = 866 in the GM-CSF antibody group and n = 891 in the control group). GM-CSF antibodies treatment reduced the incidence of IMV, which was supported by two cohort studies (OR 0.16;95% CI 0.03, 0.74) and three RCTs (OR 0.62;95% CI 0.41, 0.94). GM-CSF antibodies resulted in slight but not significant reductions in mortality (based on two cohort studies and five RCTs) and ventilation improvement (based on one cohort study and two RCTs). The sensitive analysis further showed the results of mortality and ventilation improvement rate became statistically significant when one included study was removed. Besides, GM-CSF antibodies did not increase the risks of the second infection (based on one cohort study and five RCTs) and AEs (based on five RCTs). Conclusion GM-CSF antibody treatments may be an efficacious and well-tolerant way for the treatment of COVID-19. Further clinical evidence is still warranted. Supplementary Information The online version contains supplementary material available at 10.1007/s10787-022-01105-9.

A potential alternative systemic treatment option for epithelioid sarcoma (preprint)

Background: Epithelioid sarcoma is a rare soft tissue sarcoma characterized by SMARCB1/INI1 deficiency. Much attention has been paid to the selective EZH2 inhibitor tazemetostat, where other systemic treatments are generally ignored. To explore alternative treatment options, we studied the effects of irinotecan-based chemotherapy in a series of epithelioid sarcoma patients. Methods: We retrospectively reviewed data from patients with metastatic or unresectable epithelioid sarcoma at the Peking University People’s Hospital treated with irinotecan (50 mg/m2/d d1-5 Q3W) in combination with Anlotinib (12 mg Qd, 2 weeks on and 1 week off) from July 2015 to November 2021. Results: A total of 54 courses were administered. With a median follow up of 21.2 months (95% CI, 12.2, 68.1), the 5-year overall survival rate was 83.3%. Five of eight (62.5%) patients presented with unresectable localized lesions, including local tumor thrombosis and lymphatic metastasis. The other patients had unresectable pulmonary metastases. Six of eight (75%) patients had progressed following two lines of systemic therapy. The objective response rate reached 37.5% (three of eight patients) while stabilized disease was observed in 62.5% (five of eight) of patients. No patient had progressed at initial evaluation. At the last follow up, two patients were still using the combination and three patients had ceased the therapy due to toxicities such as diarrhea, nausea, and emesis. One patient changed to tazemetostat for maintenance and one patient stopped treatment due to coronavirus disease 2019 (COVID-19). Another patient stopped therapy as residual lesions had been radiated. Conclusions: The combination of irinotecan and Anlotinib as a salvage regimen may be considered another effective treatment option for refractory epithelioid sarcoma. Trial registration: This trial was approved in the Medical Ethics Committee of Peking University People’s Hospital on October 28, 2022 (No.: 2022PHD015-002). The trial was registered in Clinicaltrials.gov with identifier no. NCT05656222.

Identifying Contextual and Spatial Risk Factors for Post-Acute Sequelae of SARS-CoV-2 Infection: An EHR-based Cohort Study from the RECOVER Program (preprint)

Post-acute sequelae of SARS-CoV-2 infection (PASC) affects a wide range of organ systems among a large proportion of patients with SARS-CoV-2 infection. Although studies have identified a broad set of patient-level risk factors for PASC, little is known about the contextual and spatial risk factors for PASC. Using electronic health data of patients with COVID-19 from two large clinical research networks in New York City and Florida, we identified contextual and spatial risk factors from nearly 200 environmental characteristics for 23 PASC symptoms and conditions of eight organ systems. We conducted a two-phase environment-wide association study. In Phase 1, we ran a mixed effects logistic regression with 5-digit ZIP Code tabulation area (ZCTA5) random intercepts for each PASC outcome and each contextual and spatial factor, adjusting for a comprehensive set of patient-level confounders. In Phase 2, we ran a mixed effects logistic regression for each PASC outcome including all significant (false positive discovery adjusted p-value < 0.05) contextual and spatial characteristics identified from Phase I and adjusting for confounders. We identified air toxicants (e.g., methyl methacrylate), criteria air pollutants (e.g., sulfur dioxide), particulate matter (PM2.5) compositions (e.g., ammonium), neighborhood deprivation, and built environment (e.g., food access) that were associated with increased risk of PASC conditions related to nervous, respiratory, blood, circulatory, endocrine, and other organ systems. Specific contextual and spatial risk factors for each PASC condition and symptom were different across New York City area and Florida. Future research is warranted to extend the analyses to other regions and examine more granular contextual and spatial characteristics to inform public health efforts to help patients recover from SARS-CoV-2 infection.

Synergistic effect and kinetic analysis of catalytic co-pyrolysis of waste cotton swabs and non-woven masks

In the context of the COVID-19, the co-pyrolysis of waste cotton swabs and non-woven masks is a promising way of biomass utilization. On the one hand, it alleviates the energy crisis to some extent, and on the other hand, it is conducive to the proper disposal of medical waste. In this paper, the pyrolysis process and kinetic behavior of two materials under different conditions including mixing ratios, heating rates and catalyst additions were studied. The results show that the heating rate has little effect on the pyrolysis process. Mask has better pyrolysis performance than cotton swabs and could promote the decomposition of biomass as an auxiliary material. All the blends show a certain synergistic effect. The difference between the actual mass loss and the theoretical mass loss (ΔW) is positive and the activation energy is lower than that of any single component. The addition of catalyst can further promote the pyrolysis reaction, and the residual mass is greatly reduced. This paper aims to provide some suggestions for further exploration in related fields.

Factors associated with prolonged viral shedding in older patients infected with Omicron BA.2.2 (preprint)

Background To explore the risk factors associated with the viral shedding time in the elder Chinese patients infected with SARS-CoV-2 omicron.Methods Participants infected with SARS-CoV-2 omicron were enrolled in a retrospective study, and divided into two groups according to the shedding time of omicron [≥ 10 days, "late clearance group" and < 10 days, "early clearance group"].Results 180 patients were enrolled in the study (88 early, 92 late), with a median time of viral shedding was 10 days and a mean age of 77.02 years. When comparing patients between either group, prolonged SARS-CoV-2 omicron shedding was associated with old age (P = 0.007), unvaccinated (P = 0.001), delayed admission to hospital after illness onset (P = 0.001), D-dimer (P = 0.003) and methylprednisolone treatment (P = 0.048). In the multivariable analysis, vaccinated [OR], 0.319 [95% CI, 0.130–0.786], P = 0.013), paxlovid (OR, 0.259 [95% CI, 0.104–0.643], P = 0.004), and time from onset to admission (OR, 1.802 [95% CI, 1.391–2.355], P = 0.000) were significantly associated with viral clearance.Conclusions The older age, methylprednisolone therapy, and D-dimer were associated with prolonged duration of omicron viral shedding. The time from onset to hospitalization, unused paxlovid and unvaccinated were independent risk factors in patients infected with SARS-CoV-2 omicron.

Machine Learning for Identifying Data-Driven Subphenotypes of Incident Post-Acute SARS-CoV-2 Infection Conditions with Large Scale Electronic Health Records: Findings from the RECOVER Initiative (preprint)

The post-acute sequelae of SARS-CoV-2 infection (PASC) refers to a broad spectrum of symptoms and signs that are persistent, exacerbated, or newly incident in the post-acute SARS-CoV-2 infection period of COVID-19 patients. Most studies have examined these conditions individually without providing concluding evidence on co-occurring conditions. To answer this question, this study leveraged electronic health records (EHRs) from two large clinical research networks from the national Patient-Centered Clinical Research Network (PCORnet) and investigated patients’ newly incident diagnoses that appeared within 30 to 180 days after a documented SARS-CoV-2 infection. Through machine learning, we identified four reproducible subphenotypes of PASC dominated by blood and circulatory system, respiratory, musculoskeletal and nervous system, and digestive system problems, respectively. We also demonstrated that these subphenotypes were associated with distinct patterns of patient demographics, underlying conditions present prior to SARS-CoV-2 infection, acute infection phase severity, and use of new medications in the post-acute period. Our study provides novel insights into the heterogeneity of PASC and can inform stratified decision-making in the treatment of COVID-19 patients with PASC conditions.

Understanding Post-Acute Sequelae of SARS-CoV-2 Infection through Data-Driven Analysis with Longitudinal Electronic Health Records: Findings from the RECOVER Initiative (preprint)

Recent studies have investigated post-acute sequelae of SARS-CoV-2 infection (PASC) using real-world patient data such as electronic health records (EHR). Prior studies have typically been conducted on patient cohorts with small sample sizes 1 or specific patient populations 2,3 limiting generalizability. This study aims to characterize PASC using the EHR data warehouses from two large national patient-centered clinical research networks (PCORnet), INSIGHT and OneFlorida+, which include 11 million patients in New York City (NYC) and 16.8 million patients in Florida respectively. With a high-throughput causal inference pipeline using high-dimensional inverse propensity score adjustment, we identified a broad list of diagnoses and medications with significantly higher incidence 30-180 days after the laboratory-confirmed SARS-CoV-2 infection compared to non-infected patients. We found more PASC diagnoses and a higher risk of PASC in NYC than in Florida, which highlights the heterogeneity of PASC in different populations.

Synthetic Biology-Powered Biosensors Based on CRISPR/Cas Mediated Cascade Signal Amplification for Precise RNA Detection

Synthetic biology enabling technologies have been harnessed to create new diagnostic technologies. However, most strategies involve error-prone amplification steps and limitations of accuracy in RNA detection. Here, a cell-free synthetic biology-powered biosensing strategy, termed as SHARK (Synthetic Enzyme Shift RNA Signal Amplifier Related Cas13a Knockdown Reaction), could efficiently and accurately amplify RNA signal by leveraging the collateral cleavage of activated Cas13a to regulate cell-free enzyme synthesis. Based on cascade amplification and tailored enzyme output, SHARK behaves broad compatibility in different scenarios. The portable device based on SHARK was successfully used as SARS-CoV-2 biosensors with high sensitivity and selectivity, and the results were highly consistent with Ct values of qRT-PCR. In addition, when combined with machine learning, SHARK performs bio-computations and thus for cancer diagnosis and staging based on 64 clinical samples. SHARK shows characteristics of precise recognition, cascade amplification and tailored signal outputting comparisons with established assays, presenting significant potential in developing next-generation RNA detection technology.

The comprehensive performance analysis on a novel high-performance air-purification-sterilization type PV-Trombe wall

PV-Trombe wall is a superior system because it realizes space heating and electricity generation. However, the application on thermal energy is single. Besides, the indoor air pollutions caused by volatile organic compounds (VOCs) and bioaerosols formed by bacteria or virus (COVID-2019) have been more and more serious. Thus, a high-performance air-purification-sterilization type PV-Trombe wall that meets electricity, heating, formaldehyde and bacteria removal fully driven by solar energy is proposed. The thermal, electrical and purification performances with different PV coverages are analyzed based on the experimental and modelling works. Main results are: (1) With PV coverage of 0, the air thermal efficiency and formaldehyde single-pass ratio are 0.46 and 0.35, respectively. Five kinds of bacteria are fully thermal inactivated. (2) The PV coverage plays an important role on the electrical, thermal and purification performances. With optimal PV coverage, the thermal and electrical efficiencies are 0.328 and 0.136, respectively. The generated clean air by removing formaldehyde and bacteria are 85.1 and 157.1 m3, respectively. (3) The total air heat intake and output electricity in Xining, Hefei and Beijing are 1301.2, 376.2, 932.6 MJ, and 171.8, 57.4, 120.9 MJ, respectively. The saving fresh air thermal energy are 396.6, 87.5 and 198.3 MJ, respectively.

Development of a federated learning approach to predict acute kidney injury in adult hospitalized patients with COVID-19 in New York City (preprint)

Federated learning is a technique for training predictive models without sharing patient-level data, thus maintaining data security while allowing inter-institutional collaboration. We used federated learning to predict acute kidney injury within three and seven days of admission, using demographics, comorbidities, vital signs, and laboratory values, in 4029 adults hospitalized with COVID-19 at five sociodemographically diverse New York City hospitals, between March-October 2020. Prediction performance of federated models was generally higher than single-hospital models and was comparable to pooled-data models. In the first use-case in kidney disease, federated learning improved prediction of a common complication of COVID-19, while preserving data privacy.

Effects of TCM guidance physical exercise on prevention and treatment of novel Coronavirus pneumonia

Starting from prevention period, medical observation period, convalescence intervention guidance method of novel coronavirus pneumonia (NCP), this article analyzed the effects of TCM guidance physical exercise on the prevention and control of NCP, summarized the social significance and value of current promotion of TCM guidance physical exercise, and recommended that further research on TCM guidance physical exercise should be done so that it will not only contribute to the health of the entire population, but also give full play to the advantages of TCM in this outbreak of NCP.

Low-voltage driven Ag-Co3O4 textile device for multifunctional air cleaning

Human exposure risks to airborne pollutants, bacteria and viruses in confined spaces have attracted tremendous attention. It is a challenge to degrade these harmful materials over a single device by electrical method other than conventional thermal method. We fabricated an air cleaning device based on a conductive Ag-Co3O4 coating with Ag nanoparticles on a glass fibre cloth (GFC). The device possessed good flexibility and high permeability of the GFC. Powered by low-voltages (< 20 V), the device exhibited a 3-fold formaldehyde conversion in its conventional thermal counterpart, and energy savings of > 90% were achieved. The electrically treated device completely killed Escherichia coli and Staphylococcus aureus and SARS-CoV-2 pseudovirus within a few minutes at a low surface temperature of < 50 °C. The excellent efficiency of the devices was attributed to the confinement of electric power to the coating. The device can serve as a flexible filter for air cleaners or conditioners to ensure that human health is maintained amid the pandemic.

A meta-analysis on the risk factors adjusted association between cardiovascular disease and COVID-19 severity (preprint)

Background Cardiovascular disease (CVD), one of the most common comorbidities of coronavirus disease 2019 (COVID-19), has been suspected to be associated with adverse outcomes in COVID-19 patients, but their correlation remains controversial.Method This is a quantitative meta-analysis on the basis of adjusted effect estimates. PubMed, Web of Science, MedRxiv, Scopus, Elsevier ScienceDirect, Cochrane Library and EMBASE were searched comprehensively to obtain a complete data source up to January 7, 2021. Pooled effects (hazard ratio (HR), odds ratio (OR)) and the 95% confidence intervals (CIs) were estimated to evaluate the risk of the adverse outcomes in COVID-19 patients with CVD. Heterogeneity was assessed by Cochran’s Q-statistic, I²test, and meta-regression. In addition, we also provided the prediction interval, which was helpful for assessing whether the variation across studies was clinically significant. The robustness of the results was evaluated by sensitivity analysis. Publication bias was assessed by Begg’s test, Egger’s test, and trim-and-fill method.Result Our results revealed that COVID-19 patients with pre-existing CVD tended more to adverse outcomes on the basis of 203 eligible studies with 24,033,838 cases (pooled ORs = 1.41, 95% CIs: 1.32–1.51, prediction interval: 0.84–2.39; pooled HRs = 1.34, 95% CIs: 1.23–1.46, prediction interval: 0.82–2.21). Further subgroup analyses stratified by age, the proportion of male, study design, disease types, sample size, region and disease outcomes also showed that pre-existing CVD was significantly associated with adverse outcomes among COVID-19 patients.Conclusion Our findings demonstrated that pre-existing CVD was an independent risk factor associated with adverse outcomes among COVID-19 patients.

Synthetic Homogeneous Glycoforms of the SARS-CoV-2 Spike Receptor-Binding Domain Reveals Different Binding Profiles of Monoclonal Antibodies

SARS-CoV-2 attaches to its host receptor, angiotensin-converting enzyme 2 (ACE2), via the receptor-binding domain (RBD) of the spike protein. The RBD glycoprotein is a critical target for the development of neutralizing antibodies and vaccines against SARS-CoV-2. However, the high heterogeneity of RBD glycoforms may lead to an incomplete neutralization effect and impact the immunogenic integrity of RBD-based vaccines. Investigating the role of different carbohydrate domains is of paramount importance. Unfortunately, there is no viable method for preparing RBD glycoproteins with structurally defined glycans. Herein we describe a highly efficient and scalable strategy for the preparation of six glycosylated RBDs bearing defined structure glycoforms at T323, N331 and N343. A combination of modern oligosaccharide, peptide synthesis and recombinant protein engineering provides a robust route to deciphering carbohydrate structure?function relationships.

Federated Learning of Electronic Health Records Improves Mortality Prediction in Patients Hospitalized with COVID-19 (preprint)

Machine learning (ML) models require large datasets which may be siloed across different healthcare institutions. Using federated learning, a ML technique that avoids locally aggregating raw clinical data across multiple institutions, we predict mortality within seven days in hospitalized COVID-19 patients. Patient data was collected from Electronic Health Records (EHRs) from five hospitals within the Mount Sinai Health System (MSHS). Logistic Regression with L1 regularization (LASSO) and Multilayer Perceptron (MLP) models were trained using local data at each site, a pooled model with combined data from all five sites, and a federated model that only shared parameters with a central aggregator. Both the federated LASSO and federated MLP models performed better than their local model counterparts at four hospitals. The federated MLP model also outperformed the federated LASSO model at all hospitals. Federated learning shows promise in COVID-19 EHR data to develop robust predictive models without compromising patient privacy.