Virus World

Purpose: To develop a reliable tool that predicts which patients are most likely to be COVID-19 positive and which ones have an increased risk of hospitalization. Methods: From February 2020 to April 2021, trained nurses recorded age, gender, and symptoms in an outpatient COVID-19 testing center. All positive patients were followed up by phone for 14 days or until symptom-free. We calculated the symptoms odds ratio for positive results and hospitalization and proposed a random forest machine-learning model to predict positive testing. Results: A total of 8,998 patients over 16 years old underwent COVID-19 RT-PCR, with 1,914 (21.3%) positives. Fifty patients needed hospitalization (2.6% of positives), and three died (0.15%). Most common symptoms were: cough, headache, sore throat, coryza, fever, myalgia (57%, 51%, 44%, 36%, 35%, 27%, respectively). Cough, fever, and myalgia predicted positive COVID-19 test, while others behaved as protective factors. The best predictors of positivity were fever plus anosmia/ageusia (OR=6.31), and cough plus anosmia/ageusia (OR=5.82), both p<0.0001. Our random forest model had an ROC-AUC of 0.72 (specificity=0.70, sensitivity=0.61, PPV=0.38, NPV=0.86). Having steady fever during the first days of infection and persistent dyspnea increased the risk of hospitalization (OR=6.66, p<0.0001 and OR=3.13, p=0.003, respectively), while anosmia-ageusia (OR=0.36, p=0.009) and coryza (OR=0.31, p=0.014) were protective. Conclusion: Present study and algorithm may help identify patients at higher risk of having SARS-COV-2 (online calculator http://wdchealth.covid-map.com/shiny/calculator/), and also disease severity and hospitalization based on symptoms presence, pattern, and duration, which can help physicians and health care providers.

Available as Preprint in medRxiv (August 10, 2021):

https://doi.org/10.1101/2021.08.09.21261729 

Researchers at the top-ranked hospital in Boston compared search interest in loss of taste and appetite, and diarrhea with the reported incidence of Covid-19 in 15 U.S. states from January 20 to April 20. Using Alphabet Inc.s Google Trends online tool, they found the volume of searches correlated most strongly with cases in New York, New Jersey, California, Massachusetts and Illinois - states with high disease burden - three to four weeks later. Internet searches on gastrointestinal symptoms predicted a rise in Covid-19 cases weeks later, researchers at Massachusetts General Hospital found, demonstrating a novel early warning system for hot spots of the pandemic disease. 

The research, published in the journal Clinical Gastroenterology and Hepatology, showed that the same approach used to monitor pandemic influenza trends more than a decade ago could be deployed for the coronavirus, the hospital said in a report this month. Patients with Covid-19 often report gastrointestinal symptoms, such as abdominal pain and diarrhea, sparking interest in conducting the study. “Our data underscore the importance of GI symptoms as a potential harbinger of Covid-19 infection and suggests that Google Trends may be a valuable tool for prediction of pandemics with GI manifestations,” Kyle Staller, a gastroenterologist and the director of Mass General’s gastrointestinal motility laboratory, and colleagues wrote in the study. Scientists are also testing for traces of the coronavirus in wastewater to identify places where Covid-19 is spreading.

Predictive models of immune protection from COVID-19 are urgently needed to identify correlates of protection to assist in the future deployment of vaccines. To address this, we analyzed the relationship between in vitro neutralization levels and the observed protection from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection using data from seven current vaccines and from convalescent cohorts. We estimated the neutralization level for 50% protection against detectable SARS-CoV-2 infection to be 20.2% of the mean convalescent level (95% confidence interval (CI) = 14.4–28.4%). The estimated neutralization level required for 50% protection from severe infection was significantly lower (3% of the mean convalescent level; 95% CI = 0.7–13%, P = 0.0004). Modeling of the decay of the neutralization titer over the first 250 d after immunization predicts that a significant loss in protection from SARS-CoV-2 infection will occur, although protection from severe disease should be largely retained. Neutralization titers against some SARS-CoV-2 variants of concern are reduced compared with the vaccine strain, and our model predicts the relationship between neutralization and efficacy against viral variants. Here, we show that neutralization level is highly predictive of immune protection, and provide an evidence-based model of SARS-CoV-2 immune protection that will assist in developing vaccine strategies to control the future trajectory of the pandemic. Estimates of the levels of neutralizing antibodies necessary for protection against symptomatic SARS-CoV-2 or severe COVID-19 are a fraction of the mean level in convalescent serum and will be useful in guiding vaccine rollouts.

Published in Nature Medicine (May 17, 2021):

https://www.nature.com/articles/s41591-021-01377-8 

The next Ebola outbreak could be predicted using a new UCL-developed model that tracks how changes to ecosystems and human societies combine to affect the spread of the deadly infectious disease. The model could help policymakers to decide where to target vaccine deployment, or develop healthcare infrastructure, to reduce the risk of zoonotic disease outbreaks—illnesses that spread between animals and humans.

Analysis using a mathematical model, published today in Nature Communications, shows that several countries in Africa, including Nigeria, could be at risk of Ebola outbreaks both presently, and in the future, despite having experienced no known cases to date. First author of the study, Dr. David Redding (UCL Genetics, Evolution & Environment), said: "It is vital that we understand the complexities causing animal-borne diseases to spill-over into humans, to accurately predict outbreaks and help save lives. "In our models, we've included more information about the animals that carry Ebola and, by doing so, we can better account for how changes in climate, land-use or human societies can affect human health." Designed by a UCL-led team of researchers, the model captures the impact of climate, land use and human population factors on the risk of Ebola and predicts the known set of previous outbreaks with a high degree of accuracy, even in the absence of case data. The results show that Ebola outbreaks, resulting from spill-over events, are 1.6 times more likely in scenarios with increased warming and slower socioeconomic development.

More than two thirds of all infectious diseases originate in animals, including Ebola, Lassa fever and West Nile virus. These diseases contribute to the global health and economic burden that disproportionately affects poor communities. The latest Ebola epidemic has claimed more than 2,100 lives since August 2018 and while there are signs it is in retreat, the risk of spread is still high according to a recent report by the UN....

"Ebola risk appears to worsen in future versions of our planet that have higher climate change and worse cooperation between societies. Working together to improve healthcare resources, which can contain dangerous diseases such as Ebola, appears to strongly reduce future risk, and this offers an important option for preventing future disease cases. We hope our model will help policy makers address this challenge."

Published in Nat. Communications on October 15, 2019:

https://doi.org/10.1038/s41467-019-12499-6