ABSTRACT
Background: Human-associated microbial communities have been linked to host immune response to respiratory viral infections. Prior investigations have observed shifts in the composition of the gut or respiratory microbiome in severe COVID-19. However, there has been no comprehensive metagenomic evaluation of the interaction between lower respiratory and gut microbiomes and host immune factors in COVID-19. Methods: From April 2020 to May 2021, we prospectively enrolled 153 hospitalized patients with mild (n=12), moderate (n=65), and severe (n=76) COVID-19 infection categorized using established clinical criteria. We longitudinally collected stool (n=270) for metagenomic profiling, and in a subset, we generated comprehensive host-microbiome-molecular profiles by collecting sputum metagenomes (n=87 participants with 212 samples) and blood cytokine levels (n=109 with 181 samples) weekly until hospital discharge. We performed omnibus testing of overall gut and respiratory community structure, species-level differential abundance testing using mixed effects modeling accounting for repeated sampling, hierarchical clustering of paired gut and respiratory metagenomic profiles, and multi-omic machine learning classification of disease severity. Results: Patients with severe COVID-19 tended to be older, were more frequently male, had higher rates of overweight/obesity, and a greater mean Charlson Comorbidity Index. Patients with severe COVID-19 infection had significantly decreased stool and respiratory microbiome a-diversity irrespective of antibiotic administration. COVID severity accounted for a small proportion of variance in stool (R2=2.4%, p=0.002) and sputum (R2=4.4%, p= 0.03) profiles. Hierarchical clustering of paired gut and respiratory samples from patients with severe COVID revealed the joint expansion of oral-typical taxa typically present during systemic inflammation (i.e., increases in Streptococcus and Peptostreptococus spp. in both gut and sputum). A pro-inflammatory milieu defined by a composite elevation of circulating plasma cytokines (e.g., IL-6, TNF-a, and IL-29 among others) were linked to broad microbial excursions in community structure for both stool and sputum as measured by Bray-Curtis distances. A random forest classifier incorporating either stool or sputum taxonomic features and accounting for age, sex, body mass index, and recent antibiotic use achieved excellent classification of biospecimens from patients with severe vs. non-severe COVID patients (AUROC > 0.80). Conclusions: Alterations of the gut and respiratory microbiome were associated with differences in host immune response and COVID-19 disease severity. Further studies are needed to identify the potential role of human-associated microbial communities as a biomarker for poor patient outcomes in COVID-19 who may warrant escalated levels of care.(Figure Presented) Fig. 1. (A) Using unsupervised feature selection (species abundance > 0.001) inclusive of taxa differentially abundant by non-parametric Wilcoxon rank-sum testing (nominal p-value < 0.05), (B) we performed random forest classification using a twice-repeated 5-fold crossvalidation scheme to predict COVID-19 disease severity from shotgun metagenomic stool profiles (C) yielding an AUROC of 0.91.
ABSTRACT
This study analyzes the dynamic connectedness (i.e., spillovers and spillbacks) of financial stress across advanced and emerging economies. As proxy for financial stress, we reconstruct the financial stress index (FSI) for 16 advanced economies and 15 emerging economies from January 1997 to August 2020. The constructed FSIs reflect combined stress level in banking sectors, equity markets, capital markets and exchange rate markets. Using frameworks proposed by Diebold and Yilmaz (Better to give than to receive: Predictive directional measurement of volatility spillovers. International Journal of Forecasting, 28(1), 57-66) and Barunik and Krehlik (Measuring the frequency dynamics of financial connectedness and systemic risk. Journal of Financial Econometrics, 16(2), 271-296), we find that there is strong connectedness of financial stress across economies. Moreover, the connectedness of the financial stress is stronger after the global financial crisis and during the COVID-19 pandemic. Although the spillover of shocks is strongest in the short-term horizon, the spillovers in the longer-term horizons are not trivial. Our results also show that the US is the largest shock transmitter as well as one of the largest shock receivers. Our results also suggest that shocks originating in advanced economies have strong effects on other economies, but shocks originating in emerging economies also play an increasing role. Global factors such as global economic policy uncertainty and geopolitical risks influence the magnitude of the spillover of financial stress.
ABSTRACT
Saliou (Eur J Epidemiol, 1994, 10 (4), 515-517) argued that pandemics are special kinds of crises and requires the public health sector to focus on: 1) reducing uncertainty, 2) rumor mitigation, and 3) ensuring the public reduces their risk of contracting the disease. With this as a backdrop, the central aim of this research is to better understand the connections between public information seeking, evaluation, and self-protective behaviors in the COVID-19 pandemic and focuses on a comparison between the Republic of Korea and Vietnam to provide insights into the influence of the individual, institutional, and information factors influencing people's experience with COVID-19. Thus, there are two major contributions of this study. First, it provides a cross-theory evaluation of the factors that contribute to information seeking, evaluation, and self-protective behaviors. Second, the study identifies potentially critical differences in information seeking, evaluation, and self-protective behaviors based on acute disease reproduction in countries with a successful pandemic suppression history. Findings suggest that in countries where there are high levels of trust and satisfaction even small changes in the infection rates lead to different information seeking and self-protective behaviors.
ABSTRACT
OBJECTIVES: Diagnostic work-up following any COVID-19 associated symptom will lead to extensive testing, potentially overwhelming laboratory capacity whilst primarily yielding negative results. We aimed to identify optimal symptom combinations to capture most cases using fewer tests with implications for COVID-19 vaccine developers across different resource settings and public health. METHODS: UK and US users of the COVID-19 Symptom Study app who reported new-onset symptoms and an RT-PCR test within seven days of symptom onset were included. Sensitivity, specificity, and number of RT-PCR tests needed to identify one case (test per case [TPC]) were calculated for different symptom combinations. A multi-objective evolutionary algorithm was applied to generate combinations with optimal trade-offs between sensitivity and specificity. FINDINGS: UK and US cohorts included 122,305 (1,202 positives) and 3,162 (79 positive) individuals. Within three days of symptom onset, the COVID-19 specific symptom combination (cough, dyspnoea, fever, anosmia/ageusia) identified 69% of cases requiring 47 TPC. The combination with highest sensitivity (fatigue, anosmia/ageusia, cough, diarrhoea, headache, sore throat) identified 96% cases requiring 96 TPC. INTERPRETATION: We confirmed the significance of COVID-19 specific symptoms for triggering RT-PCR and identified additional symptom combinations with optimal trade-offs between sensitivity and specificity that maximize case capture given different resource settings.
Subject(s)
COVID-19 , COVID-19 Vaccines , Fever , Humans , Prospective Studies , SARS-CoV-2ABSTRACT
Background: The COVID-19 pandemic and response underscore the urgent need for real-time population-leveldata, especially for vulnerable populations (e.g., cancer patients, racial and ethnic minorities). Smartphoneapplications (apps) facilitate the collection of self-reported data at scale, the results of which can then be rapidlyredeployed to inform the public health response. The COVID Symptom Study is an app that was launched March24, 2020, and is now used by nearly 4 million people in the U.S., U.K., and Sweden. Methods: COVID Symptom Study app users self-report health status (e.g., symptoms, COVID-19 testing, healthcare utilization), comorbidities, demographics, and key risk factors for infection on a daily basis. Multivariableadjusted logistic regression models were used to determine the association of cancer and race with COVID-19prevalence, adjusting for age, sex, comorbidities, and risk factors for infection, from app launch through May 25,2020. Results: Among 23,266 individuals with cancer and 1,784,293 without cancer, we documented 155 and 10,249 self-reports of COVID-19, respectively. Compared to individuals without cancer, those with cancer had an increased riskof COVID-19 (adjusted odds ratio (aOR): 1.60;95% confidence interval (CI): 1.36-1.88). The association wasstronger among older participants >65 compared to younger participants (Pinteraction<0.001) and among males(aOR: 1.71;95%CI: 1.36-2.15) compared to females (aOR: 1.43;95%CI: 1.14-1.79;Pinteraction=0.02).Chemotherapy/immunotherapy was associated with a 2-fold increased risk of COVID-19 (aOR: 2.22;95% CI: 1.68-2.94) and risk of COVID-related hospitalization (aOR:2.47;95% CI: 2.22-2.76). In a separate analysis, wedocumented 8,990 self-reported cases of positive COVID-19 testing among 2,304,472 non-Hispanic whiteparticipants (93.6% of cohort);93 among 19,498 Hispanic participants;204 among 19,498 Black participants;608among 64,429 Asian participants;and 352 among 65,046 mixed race/other racial minorities. Compared with non-Hispanic white participants, the ORs for reporting a positive COVID-19 test for racial minorities ranged from 1.44(mixed race/other races) to 2.59 (Black). After accounting for risk factors for infection, comorbidities, andsociodemographic characteristics, the aORs were 1.37 (95% CI 1.09-1.72) for Hispanic participants, 1.42 (95% CI1.23-1.64) for Black participants, 1.44 (95% CI 1.33-1.57) for Asian participants, and 1.18 (95% CI 1.06-1.32) formixed race/other minorities. Conclusion: Our results demonstrate an increase in COVID-19 risk among ethnic minorities and individuals withcancer, particularly those on treatment with chemotherapy/immunotherapy. The association with minorities was notcompletely explained by other known risk factors for COVID-19 or sociodemographic characteristics. These findingshighlight the utility of app-based syndromic surveillance for quantifying the impact of the COVID-19 pandemic on at-risk populations.
ABSTRACT
OBJECTIVES: Diagnostic work-up following any COVID-19 associated symptom will lead to extensive testing, potentially overwhelming laboratory capacity whilst primarily yielding negative results. We aimed to identify optimal symptom combinations to capture most cases using fewer tests with implications for COVID-19 vaccine developers across different resource settings and public health. METHODS: UK and US users of the COVID-19 Symptom Study app who reported new-onset symptoms and an RT-PCR test within seven days of symptom onset were included. Sensitivity, specificity, and number of RT-PCR tests needed to identify one case (test per case [TPC]) were calculated for different symptom combinations. A multi-objective evolutionary algorithm was applied to generate combinations with optimal trade-offs between sensitivity and specificity. FINDINGS: UK and US cohorts included 122,305 (1,202 positives) and 3,162 (79 positive) individuals. Within three days of symptom onset, the COVID-19 specific symptom combination (cough, dyspnoea, fever, anosmia/ageusia) identified 69% of cases requiring 47 TPC. The combination with highest sensitivity (fatigue, anosmia/ageusia, cough, diarrhoea, headache, sore throat) identified 96% cases requiring 96 TPC. INTERPRETATION: We confirmed the significance of COVID-19 specific symptoms for triggering RT-PCR and identified additional symptom combinations with optimal trade-offs between sensitivity and specificity that maximize case capture given different resource settings.