ABSTRACT
Background: The COVID-19 pandemic has been striking for three years and, despite the regular arise of new variants, populations are now widely immune and protected from severe symptoms. However, immunocompromised patients still have worse clinical outcomes, higher mortality and rarely develop effective immunity through vaccination or infection. Here, we studied the temporal distribution of infections, viral loads (VL) as well as the viral genetic diversity among an immunocompromised patient cohort, between January 2021 and September 2022. Method(s): Overall, 478 immunocompromised patients (solid organ transplant, HIV positive, cancer, autoimmune disease) and 234 controls (healthcare workers) from Pitie-Salpetriere and Bichat Claude-Bernard University hospitals (Paris, FRANCE) were diagnosed with SARS-CoV-2 infection by RT-qPCR. Whole genome sequencing was performed according to ARTIC protocol on Oxford Nanopore platform. All 712 full viral genomes were used to determine lineages and mapped to Wuhan-Hu-1 reference to produce a maximum likelihood phylogenetic tree (IQTree, 1000 bootstraps). Differences in temporal distributions of infections and VL were assessed using nonparametric statistical tests. Result(s): According to phylogenetic analysis, genomes from SARS-CoV- 2 infecting immunocompromised patients and those infecting healthy individuals are distributed in a similar way. No significant genetic differences can be observed between viral genomes from patients and controls within the different lineages. Temporal distribution of COVID-19 infections were also similar between immunocompromised patients and controls, with the exception of BA.2 variant for which controls were infected earlier (p< 0.001). VL were significantly lower in immunocompromised patients infected with Omicron variants (p=0.04). No differences in VL were observed for Alpha and Delta variants. Conclusion(s): At diagnosis, no intrinsic genetic divergence was observed in virus infecting immunocompromised patients compared to those circulating in the general population. Similarities in temporal distribution of infections between controls and patients suggest that these different groups become infected concomitantly. VL appeared to be lower for Omicron variants in immunocompromised patients. An earlier VL peak of Omicron and a testing of immunocompromised patients hospitalized once severe symptoms have appeared could indicate a delayed testing in these patients, once the replicative phase over. (Figure Presented).
ABSTRACT
This article seeks to systematically identify and model antecedents of entrepreneurial bootstrapping and bricolage to determine and interpret the relationships and hierarchy between them. Entrepreneurial bootstrapping and bricolage are key dynamic capabilities that help entrepreneurs access, accumulate and enhance resources to adapt to scarce business environments. The article employs a modified total interpretive structural modelling analysis to determine hierarchical inter-relationships between the antecedents and a Matrice d' Impacts Croises Multiplication Applique An Classment analysis to understand their driving and dependence powers. The results highlight that founder characteristics and human capital are placed at the lower levels, making them critical driving elements of the model along with environmental hostility and resource constraints. Entrepreneurial orientation, slack, external financial capital and entrepreneurial frugality are dependent variables, with social capital as a linkage variable. This study will guide entrepreneurs trying to implement resourcefulness behaviours to respond to the coronavirus disease-2019 crisis by prioritising driving antecedents to impact the dependent factors further. © 2023 Entrepreneurship Development Institute of India.
ABSTRACT
Globally, traffic accidents are of main concern because of more death rates and economic losses every year. Thus, road accident severity is the most important issue of concern, mainly in the undeveloped countries. Generally, traffic accidents result in severe human fatalities and large economic losses in real-world circumstances. Moreover, appropriate, precise prediction of traffic accidents has a high probability with regard to safeguarding public security as well as decreasing economic losses. Hence, the conventional accident prediction techniques are usually devised with statistical evaluations, which identify and evaluate the fundamental relationships among human variability, environmental aspects, traffic accidents and road geometry. However, the conventional approaches have major restrictions based on the assumptions regarding function kind and data distribution. In this paper, Aquila Anti-Coronavirus Optimization-based Deep Long Short-Term Memory (AACO-based Deep LSTM) is developed for road accident severity detection. Spearman's rank correlation coefficient and Deep Recurrent Neural Network (DRNN) are utilized for the feature fusion process. Data augmentation method is carried out to improve the detection performance. Deep LSTM detects the road accident and its severity, where Deep LSTM is trained by the designed AACO algorithm for better performance. The developed AACO-based Deep LSTM model outperformed other existing methods with the Mean Square Error (MSE), Root-Mean-Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) of 0.0145, 0.1204 and 0.075%, respectively. [ FROM AUTHOR] Copyright of International Journal of Pattern Recognition & Artificial Intelligence is the property of World Scientific Publishing Company and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full . (Copyright applies to all s.)
ABSTRACT
Stress is integral to biological survival. However, without an appropriate coping response, high stress levels and long-term stressful situations may lead to negative mental health outcomes. Since the COVID-19 pandemic, remote assessment of mental health has become imperative. The majority of past studies focused on detecting users' stress levels rather than coping responses using social media. Because of the diversity of human expression and because people do not usually express stress and the corresponding coping response simultaneously, it is challenging to extract users' tweets about their coping responses to stressful events from their daily tweets. Consequently, there are two goals being pursued in this study: to anchor users' stress statuses and to detect their stress responses based on the existing stressful conditions. In order to accomplish these goals, we propose a framework that consists of two phases: the construction of stress dataset and the extraction of coping responses. Since the stressed users' data are lacking, the first phase is to construct a stress dataset based on stress-related hashtags, personal pronouns, and emotion recognition. In addition, to ensure the collection of enough tweets to observe the coping responses of stressed users, we broadened the survey's scope by collecting all tweets from the same user. In the second phase, stress-coping tweets were extracted by utilizing bootstrapping-based patterns and semantic features. The bootstrapping method was used to enrich word patterns for text expression and the semantic feature to assess the meaning of sentences. The collected data included the tweets of the stressed users identified in Phase 1 and the various coping responses from Phase 2 can contribute to developing a tool for the remote assessment of mental health. The experimental results show that our two-phase method outperforms the baseline and can help improve the efficiency of extracting stress-coping tweets. © 2022 IEEE.
ABSTRACT
The COVID-19 virus has affected many people around the globe with several issues. Moreover, it causes a worldwide pandemic, and it makes more than one million deaths. Countries around the globe had to announce a complete lockdown when the corona virus causes the community to spread. In real-time, Polymerase Chain Reaction (RT-PCR) test is conducted to detect COVID-19, which is not effective and sensitive. Hence, this research presents the proposed Caviar-MFFO-assisted Deep LSTM scheme for COVID-19 detection. In this research, the COVID-19 cases data is utilized to process the COVID-19 detection. This method extracts the various technical indicators that improve the efficiency of COVID-19 detection. Moreover, the significant features fit for COVID-19 detection are selected using proposed mayfly with fruit fly optimization (MFFO). In addition, COVID-19 is detected by Deep Long Short Term Memory (Deep LSTM), and the Conditional Autoregressive Value at Risk MFFO (Caviar-MFFO) is modeled to train the weight of Deep LSTM. The experimental analysis reveals that the proposed Caviar-MFFO assisted Deep LSTM method provided efficient performance based on the Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), and achieved the recovered cases with the minimal values of 1.438 and 1.199, whereas the developed model achieved the death cases with the values of 4.582 and 2.140 for MSE and RMSE. In addition, 6.127 and 2.475 are achieved by the developed model based on infected cases.
ABSTRACT
The parameters, reliability, and hazard rate functions of the Unit-Lindley distribution based on adaptive Type-II progressive censored sample are estimated using both non-Bayesian and Bayesian inference methods in this study. The Newton–Raphson method is used to obtain the maximum likelihood and maximum product of spacing estimators of unknown values in point estimation. On the basis of observable Fisher information data, estimated confidence ranges for unknown parameters and reliability characteristics are created using the delta approach and the frequentist estimators' asymptotic normality approximation. To approximate confidence intervals, two bootstrap approaches are utilized. Using an independent gamma density prior, a Bayesian estimator for the squared-error loss is derived. The Metropolis–Hastings algorithm is proposed to approximate the Bayesian estimates and also to create the associated highest posterior density credible intervals. Extensive Monte Carlo simulation tests are carried out to evaluate the performance of the developed approaches. For selecting the optimum progressive censoring scheme, several optimality criteria are offered. A practical case based on COVID-19 data is used to demonstrate the applicability of the presented methodologies in real-life COVID-19 scenarios. © 2022 The Author(s)
ABSTRACT
INTRODUCTION: Although dexmedetomidine is a widely used sedative in the ICU, risk factors for bradycardia associated with dexmedetomidine use are not well characterized. Identifying factors that place a patient at increased risk for bradycardia with dexmedetomidine use may help guide interventions or limit complications associated with the medication's negative chronotropic effects. The aim of this analysis is to determine risk factors for development of bradycardia with dexmedetomidine use. METHOD(S): This single center, retrospective nested casecontrol included adult patients in cardiac and non-cardiac intensive care units with an intravenous dexmedetomidine duration of at least 1 hour. A univariate analysis was used to compare patients with and without bradycardia with dexmedetomidine use, and a predictive model was used to evaluate factors associated with bradycardia. Step-down backward variable selection was used based on Akaike's Information Criterion (AIC) and Bayesian Information Criteria (BIC) to identify the final model. The discriminatory power and absolute predictive ability of the final model was evaluated by the concordance index (c-index), which was internal validated by bootstrapping. Multiple imputation was performed before model selection to fill in missing values in pulse at initiation and Child Pugh Score before modeling. RESULT(S): Of the 1,838 patients receiving dexmedetomidine, 110 patients (6.0%) developed bradycardia within 72 hours of initiation. In patients that experienced bradycardia, 31 (28.1%) required an intervention. The initial full predictive model for bradycardia included age, sex, BMI, COVID19 positive test, hypothermia, pulse at initiation, ICU location (Cardiac vs non-cardiac), Child Pugh Score, use of fentanyl and propofol. Step-down backward variable selection identified 4 predictors in the final model, including COVID positive test, hypothermia, pulse at initiation, and ICU location. The final model achieved a good performance in discriminatory capability (c-index: 0.758, 95%CI 0.713-0.806) using the smallest number of predictors. CONCLUSION(S): Patients with COVID-19, hypothermia, non-cardiac ICU locations and lower pulse at initiation are at increased odds of developing bradycardia. Recognition of risk may be used to guide monitoring or alternative sedation strategies.
ABSTRACT
INTRODUCTION: Serum albumin (ALB) is inversely associated with COVID-19 severity through an unclear mechanism. We addressed this gap using machine learning to identify traits exhibiting explanatory variance (EV%) in mortality risk within 12h of admission versus near end of hospitalization in the context of hypoalbuminemia. METHOD(S): Data were extracted under IRB exemption from medical records of COVID-19 patients at least 18 years old in the ICU with at least two ALB measures from March 2020 through September 2021. ALB, COVID-19 inflammation and injury traits were characterized across hospitalization. Hypoalbuminemia present on admission (POA) was defined as ALB < 3.2 g/dL. Traits associated with mortality were controlled for age, sex, COVID-19 directed treatment and the four COVID surges. Bootstrap Forest (BF) evaluated EV% of traits in mortality. Continuous data were compared using KruskalWallis. Discrete data were compared with chi-squared test. RESULT(S): Among 878 patients, 631 (72%) vs. 247 (28%) POA respectively with ALB < 3.2g/dL vs. >3.2g/ dL. Median age 68(57,77) years distributed across 64% males with 75% Whites, 10% Blacks and 15% other races exhibiting hypertension (53%), coagulopathy (28%), chronic pulmonary disease (22%) and heart failure (22%). Excess comorbidity associated with hypoalbuminemia included obesity (48% vs. 38%, p=.004), anemias (42% vs. 28%, p<.0001), and diabetes (39% vs. 32%, p=.03). Respective hypoalbuminemia near end of hospitalization increased to 97% (p<.0001) and 84% (p<.0001) with hospital mortality of 51% vs. 31% (p=<.0001). Associated ALB declines were 0.5(0.1, 1.0) vs. 1.0 (0.6, 1.0) g/dL. BF modeling (RSquared=0.69) identified POA traits EV% including CRP (21%), AST/SGOT (11%), proBNP (9%), WBC (9%), and ferritin (9%) among others. BF modeling (RSquared=0.84) identified near end visit traits EV% including WBC (31%), CRP (13%), platelet (12%), and ANC (11%) among others. CONCLUSION(S): The EV% of CRP at presentation corroborates an inverse relationship with ALB suggesting acute phase signaling may evoke hypoalbuminemia. Specifically, increased endothelial permeability allowing ALB extravasation as evidenced by proBNP EV%. Secondary etiology may derive from inhibited albumin synthesis secondary to liver injury as suggested by AST/SGOT at presentation and visit end.
ABSTRACT
Background. In Washington State, COVID-19 cases in long-term care facilities (LTCF) have accounted for less than 3% of all cases, yet 30% of all COVID-19 deaths. Understanding transmission patterns and outbreak epidemiology informs outbreak management. From April to October 2021, two large LTCFs experienced COVID-19 outbreaks. Whole genome sequencing and phylogenetic analysis were leveraged to explore transmission patterns and complement outbreak epidemiology. Methods. Epidemiologic data was exported from the Washington Disease Reporting System. Sequences, retrieved from GISAID, were aligned to the Wuhan-1 reference genome using Nextalign version 1.11.0. Pairwise single nucleotide polymorphism (SNP) distance matrices were calculated using SNP-Dists version 0.8.2. Phylogenetic trees for each outbreak were generated using IQ-Tree multicore version 2.2.0-beta COVID-edition using the GTR+F+G4 nucleotide substitution model with 1000 bootstrap replicates. MicrobeTrace was used to visualize the phylogeny, SNP heatmap, and identify clusters among sequences. Results. Weekly, LTCF A tested 162 residents and 800 staff, and LTCF B tested 60 residents and 144 staff. Of all cases in LTCF A (n= 119), 23% (n =27) were residents and 77% (n = 92) were staff, compared to 78% (n =28) residents and 22% (n =7) staff among total LTCF B cases (n=36). In LTCF A, 34% (n=40) of the cases had highquality sequences available. Seven clusters of two or more genetically related sequences and thirteen genetically unrelated sequences were identified. Five of the clusters involved resident and staff cases, linked by unit. Two clusters and remaining unrelated sequences were among staff. In LTCF B, 40% (n=14) of the cases had high-quality sequences available. One cluster of genetically related sequences was identified, all from residents of two floors. The SNP differences between sequences from LTCF A ranged from 0 to 70, whereas SNP differences between LTCF B sequences ranged from 0 to 6. Conclusion. Phylogenetic analysis of the two outbreaks confirms differences in disease transmission patterns. Multiple independent introductions of SARS-CoV-2 were identified in LTCF A, compared to a single introduction in LTCF B. Genomic epidemiology is a valuable resource for outbreak investigation and management.
ABSTRACT
Background. Rapid COVID-19 tests can offer significant advantages and reduce health disparities. The LumiraDx SARS-CoV-2 platform can perform microfluidic fluorescence assays for the rapid detection of SARS-CoV-2 antigen (Ag) and antibodies (Ab). We evaluated both tests in a longitudinal cohort to evaluate performance during acute SARS-CoV-2 infection and recovery. Methods. We collected nasal samples from 71 unique participants at four clinic visits spanning 0-21 days since symptom onset (DSSO);blood samples were collected from the same participants over six visits spanning 0-87 DSSO. For Ag testing, 232 anterior nasal swabs were assayed by: 1) the LumiraDx Ag test, 2) a laboratory-based electrochemiluminescence immunoassay for N Ag, 3) RT-PCR (Hologic Panther Fusion), and 4) culture (growth in VeroE6AT cells). For Ab testing, 308 serum samples were assayed by: 1) the LumiraDx Ab test and 2) Roche Elecsys Anti-S SARS-CoV-2 total Ab test. Measures of concordance [positive predictive agreement (PPA), negative predictive agreement (NPA), and Cohen's Kappa (K)] were estimated for qualitative results of the LumiraDx tests versus corresponding lab reference tests. Confidence intervals were estimated via bootstrapping. Results. LumiraDx Ag results had strong agreement with lab N-Ag results (K > 0.80) across all samples. Between 0-5 days, agreement was perfect, except for one sample resulting positive by LumiraDx Ag and negative by lab Ag. Agreement with PCR results was moderate overall (K=0.60), though substantial (K > 0.6) for both 0-5 DSSO (PPA=0.96/NPA=0.80) and 6-10 DSSO (PPA=0.96/NPA=0.59). Agreement with culture results was moderate overall (K=0.46): substantial (K=0.6) between 0-5 DSSO (PPA=0.96/NPA=0.60) and fair (K=0.29) between 6-10 DSSO (PPA=1.0/NPA=0.32). LumiraDx Ab results showed almost perfect agreement with lab Ab results across all samples (K=0.88), with substantial agreement (K > 0.7) for samples collected 0-10 DSSO (PPA=0.93/NPA=0.89) and 11-28 DSSO (PPA=0.99/NPA=0.69). Longitudinal agreement of LumiraDx antigen test result and culture positivity, by PCR Ct value. Nasal samples grouped by participant (lines) and agreement of results between LumiraDx antigen test result and culture positivity (proxy for infectiousness). Conclusion. LumiraDx rapid tests perform well compared to more costly and time-consuming lab methods of Ag and Ab detection. The rapid Ag test may be helpful in identifying patients infectious between 0-5 DSSO, given the substantial concordance of the rapid Ag test and culture positivity.
ABSTRACT
The parameters, reliability, and hazard rate functions of the Unit-Lindley distribution based on adaptive Type-II progressive censored sample are estimated using both non-Bayesian and Bayesian inference methods in this study. The Newton-Raphson method is used to obtain the maximum likelihood and maximum product of spacing estimators of unknown values in point estimation. On the basis of observable Fisher information data, estimated confidence ranges for unknown parameters and reliability characteristics are created using the delta approach and the frequentist estimators’ asymptotic normality approximation. To approximate confidence intervals, two bootstrap approaches are utilized. Using an independent gamma density prior, a Bayesian estimator for the squared-error loss is derived. The Metropolis-Hastings algorithm is proposed to approximate the Bayesian estimates and also to create the associated highest posterior density credible intervals. Extensive Monte Carlo simulation tests are carried out to evaluate the performance of the developed approaches. For selecting the optimum progressive censoring scheme, several optimality criteria are offered. A practical case based on COVID-19 data is used to demonstrate the applicability of the presented methodologies in real-life COVID-19 scenarios.
ABSTRACT
Objectives: We explored in this study whether insomnia, viral anxiety, reassurance-seeking behavior, and preoccupation with COVID-19 are related among the general population. As well, we explored the possibility that insomnia may mediate the association between COVID-19 viral anxiety and preoccupation. Method(s): During November 9-15, 2021, 400 participants voluntarily completed this survey, and participants' age, sex, living location, and marital status were collected. Responses to questions about COVID- 19, were also gathered, and their symptoms were rated using the Obsession with COVID-19 scale (OCS), Coronavirus Reassurance- Seeking Behaviors Scale (CRBS), Fear of COVID-19 scale (FCV-19S), and Insomnia Severity Index (ISI). The mean and standard deviation of participants' demographic characteristics and rating scale scores are summarized. Two-tailed significance was determined by a p value of 0.05. Correlation analysis was performed using Pearson's correlation coefficient. We used linear regression to examine which variables can predict obsession with COVID-19. The bootstrap method with 2,000 resamples was implemented to determine whether insomnia mediates the influence of viral anxiety or reassurance seeking behavior on preoccupation with COVID-19. Result(s): A total of 400 participants were analyzed in this study. Preoccupation with COVID-19 was predicted by young age (beta = -0.08, p = 0.012), CRBS (beta = 0.52, p < 0.001), FCV-19S (beta = 0.30, p < 0.001), and ISI (beta = 0.07, p = 0.029) (adjusted R2 = 0.62, F = 163.6, p < 0.001). Mediation analysis showed that insomnia partially mediates the influence of reassurance seeking behavior and viral anxiety on preoccupation with COVID-19. Conclusion(s): Sleep disturbances can contribute to a vicious cycle of hypochondriacal preoccupation with COVID-19. In order to reduce an individual's viral anxiety, insomnia symptoms must be addressed.
ABSTRACT
Background: During the lockdown related to the recent COVID-19 pandemic, many oncological patients, in addition to the fear of the COVID-19, have also experienced a sense of hopelessness related to the difficult management of their medical and oncological care. However, several studies showed that the physician's empathic communication about medical care management may act as a buffer against this negative relationship. Thus, the study aimed to test a model in which the physician's empathic communication mediates the relationship between fear of COVID-19 and hopelessness related to aforementioned difficulties. Methods: A sample of 70 oncological outpatients [36 females (51.4%), aged from 40 to 81 years (mean = 65.94, SD = 11.13)] were recruited at the Department of Medical Oncology, Presidio Ospedaliero di Saronno, ASST Valle Olona. An ad hoc structured interview was developed to assess: (A) fear of COVID-19 (KR20: 0.86);(B) feelings of hopelessness related to the COVID- 19 (KR20: 0.80);and (C) the patient's perception of empathy shown by the physician in communicating care management during the pandemic (KR20: 0.71). The factor score of each dimension was used as the total score. A mediation analysis (10,000 bootstrap resampling) was performed. Results: The mediation model showed statistical significance: F = 31.852, p < 0.001;R2 = 0.487. The relationship between 'fear of COVID-19' and 'hopelessness' was partially mediated by physician's empathic communication (path c': β = 0.369, se = 0.096, 95%CI[0.177;0.561], p < 0.001) - revealing its buffering role;path a: β = -0.488, se = 0.073, 95%CI[-0.634;-0.342], p < 0.001;path b: β = -0.375, se = 0.124, 95%CI[-0.623;-0.128], p = 0.003. Conclusions: These results highlight the protective role (buffer) of the physician's empathic communication in the process that leads to hopelessness from the fear of COVID-19. Results suggest how physicians as well as psycho-oncologists should structure stronger interventions - based on the improvement of communication strategies - that may lead to an improvement of patients' quality of life.
ABSTRACT
Introduction: Molnupiravir, a prodrug of the antiviral Nhydroxycytidine (NHC), is one of the limited treatment options that has recently gained emergency use authorization for treating mild-tomoderate SARS-CoV-2 cases. While NHC is shown to follow linear pharmacokinetics with similar exposures in healthy and SARS-CoV-2 subjects, its pharmacokinetics has not been characterized in the Egyptian population. Research Question or Hypothesis: We aimed to develop a population pharmacokinetic model for NHC and evaluate through simulations the current molnupiravir dosage of 800 mg twice daily for five days in the Egyptian population. Study Design: An open label, single arm pharmacokinetic study. Methods: Twelve healthy volunteers received 800 mg molnupiravir oral dose. Model development using non-linear mixed effect modeling and internal validation using bootstrapping and visual predictive check were conducted in MonolixSuite. Simulation-based maximum concentration (Cmax) 'the safety metric' and area under the curve (AUC0-12h) 'the efficacy metric' were computed for 1000 virtual subjects. Geometric mean ratios (GMR) and 90% confidence intervals (CI) compared to previously reported values were calculated. Results: A total of 132 NHC plasma concentrations were analyzed. Six transit compartments for absorption and one-compartment with weight on apparent clearance (CL/F) and volume of distribution (Vd/F) for disposition best described NHC's pharmacokinetics. The pharmackokinetic parameters were estimated with good precision and the population estimates for mean transit time, first-order absorption rate constant, CL/F and Vd/F were 0.49 hours, 2.32 hour-1, 75.12 L/hour.70 kg and 118 L/70 kg, respectively. Geometric means of simulation-based Cmax and AUC0-12 were 3827 ng/mL (GMR = 1.05;90% CI= 0.96-1.15) and 9320 ng.hr/mL (GMR = 1.04;90% CI= 0.97-1.11), respectively. Conclusion: Population pharmacokinetic model was developed for NHC. Simulations showed that current molnupiravir dosage can achieve the therapeutic targets and dose adjustment may not be required for the Egyptian population. The developed model could be used in the future to refine molnupiravir's dosage once further therapeutic targets are identified.
ABSTRACT
AFRICAN DEVELOPMENT : Swift response models are vital tools for emergency assistance agencies. The COVID-19 pandemic revealed the lack of economic models for short-run policy relevant research to anticipate local impacts and design effective policy responses. The most direct effects of the pandemic and lockdown tended to be concentrated in urban areas;however, markets quickly transmitted impacts to rural areas as well as among poor and non-poor households. General equilibrium modeling is a tool of choice to capture indirect, spillover effects of exogenous shocks. This article describes an unusual micro general-equilibrium (GE) modeling approach that we developed to quickly simulate impacts of the pandemic and lockdowns on poor and non-poor rural and urban households across sub-Saharan Africa. Monte Carlo bootstrapping was used to construct four stylized regional GE models from 34 existing local economy-wide impact evaluation (LEWIE) models. Simulations revealed that the pandemic and policy responses to curtail its spread were likely to affect rural households at least as severely as urban households. Simulated income losses are greater in poor households in both urban and rural settings. These findings are relatively consistent across models spanning sub-Saharan Africa. Because COVID-19 impacts are so far-reaching, all types of economies experience downturns. Our research underlines the importance of modeling assumptions. We find total annualized impacts of around a 6-percent loss of GDP, smaller than estimates from single-country models that ignore price effects, such as SAM-multiplier models, but in line with The World Bank's baseline forecast of a 5.2% contraction in global GDP in 2020. The largest negative impacts are on poor rural households
ABSTRACT
Background and Aims: While the relationship between elevated liver enzymes and COVID- 19 related adverse events is well-established, a liver-dependent prognostic model that predicts the risk of death is helpful to accurately stratify admitted patients. In this study, we use a bootstrapping-enhanced method of regression modeling to predict COVID-19 related deaths in admitted patients. Method: This was a single-center, retrospective study. Univariate and multivariate Cox regression analyses were performed using 30-day mortality as the primary endpoint to establish associated hepatic risk factors. Regression-based prediction models were constructed using a series of modeling iterations with an escalating number of categorical terms. Model performance was evaluated using receiver operating characteristic (ROC) curves. Model accuracy was internally validated using bootstrapping-enhanced iterations. Results: 858 patients admitted to hospital with COVID-19 were included. 78 were deceased by 30 days (9.09%). Cox regression (greater than 20 variables) showed the following core variables to be significant: INR (aHR 1.26 95%CI 1.06-1.49), AST (aHR 1.00 95%CI 1.00- 1.00), age (aHR 1.05 95%CI 1.02-1.08), WBC (aHR 1.07 95%CI 1.03-1.11), lung cancer (aHR 3.38 95%CI 1.15-9.90), COPD (aHR 2.26 95%CI 1.21-4.22). Using these core variables and additional categorical terms, the following model iterations were constructed with their respective AUC;model 1 (core only): 0.82 95%CI 0.776-0.82, model 2 (core + demographics): 0.828 95%CI 0.785-0.828, model 3 (prior terms + additional biomarkers): 0.842 95%CI 0.799-0.842, model 4 (prior terms + comorbidities): 0.851 95%CI 0.809-0.851, model 5 (prior terms + life-sustaining therapies): 0.933 95%CI 0.91-0.933, model 6 (prior terms + COVID-19 medications): 0.934 95%CI 0.91-0.934. Model 1 demonstrated the following parameters at 0.91 TPR: 0.54 specificity, 0.17 PPV, 0.98 NPV. Bootstrapped iterations showed the following AUC for the respective models: model 1: 0.82 95%CI 0.765-0.882, model 2 0.828 95%CI 0.764-0.885, model 3 0.842 95%CI 0.779-0.883, model 4: 0.851 95%CI 0.808-0.914, model 5: 0.933 95%CI 0.901-0.957, model 6: 0.934 95%CI 0.901- 0.961. Conclusion: Model 1 displays high prediction performance (AUC >0.8) in both regression-based and bootstrapping-enhanced modeling iterations. Therefore, this model can be adopted for clinical use as a calculator to evaluate the risk of 30-day mortality in patients admitted with COVID-19. (Table Presented)
ABSTRACT
Background Although a higher body mass index (BMI) has been reported to be associated with severe COVID-19 pneumonia (severe disease), it is unclear if metabolic status plays a role. Being metabolically unhealthy (MU) is defined as having either hypertension, hyperlipidemia, type 2 diabetes mellitus/pre-diabetes, or non-alcoholic fatty liver disease. We aimed to derive a risk score to predict severe disease in patients with obesity or overweight according to metabolic status. Methods A retrospective study was performed for patients hospitalized with COVID-19 pneumonia between March 2020 and August 2021 at a single tertiary center. Patients were excluded if they were immunocompromised or had a BMI < 25.0. Wilcoxon rank sum test or Fisher's Exact test were performed. Univariate logistic regression was performed followed by multivariate logistic regression to derive a risk score to predict severe disease. Variables with the highest p-values were sequentially removed until removal led to less than a 1-point reduction (improvement) in the Akaike information criterion. Accuracy of the model was calculated using bootstrap resampling estimates of the area under the receiver operating characteristic curve (AUROCC). Results 334 of 450 patients hospitalized with COVID-19 pneumonia (74.2%) were MU. Older age, higher BMI, being a former smoker, and having been vaccinated for SARS-CoV-2 were associated with being MU. There was no difference in treatments for COVID-19 pneumonia according to metabolic status. Patients who were MU had a higher death rate (10.5% vs. 2.6%) and longer total length of stay (median 6 vs. 5 days). Figure 1. On univariate analysis, age at admission, male gender, Asian race, hypertension, and type 2 diabetes mellitus were significant predictors of severe disease, whereas being MU was not, p=0.27. On multivariate logistic regression, older age, male gender, and Asian race were associated with having severe disease. Not being vaccinated was associated with a doubled odds of severe disease (OR 2.24, 95% CI: 1.07, 4.59). Figure 2. The AUROCC of the final model was 0.66 (95% CI: 0.60 to 0.71). The risk score at the lowest quintile had a 33.1% to 65.5% predicted risk and a 58.7% observed risk of severe disease, whereas at the highest quintile there was an 85.7% to 97.7% predicted risk and an 89.7% observed risk of severe disease in our cohort. Conclusion In this retrospective study of hospitalized patients with COVID-19 pneumonia, being MU was not a predictor of severe disease, even though mortality rate and total length of stay were higher in this group despite having higher rates of vaccination. Older age at admission, male gender, Asian race, and being unvaccinated were associated with severe disease. Using this risk score may help to predict severe disease in hospitalized patients with obesity or overweight. External validation is recommended (Table Presented)(Table Presented)
ABSTRACT
Introduction: The SARS-CoV-2 pandemic highlighted the need for a way to predict progression to critical illness and ICU admission amongst infected patients. Previous liver disease is a known risk for progression to critical illness. Attempts to identify biomarkers for progression to critical illness suggest inflammatory markers and coagulation markers as useful. We used a machine learning approach to compare the admission liver panel and inflammatory biomarker assays in hospitalized COVID-19 patients with extant mild or severe hepatic disease who progressed to critical illness (ICU admission) versus those who were progression-free. Methods: We included data ed under IRB exemption from electronic medical records (EMR) for SARS-CoV-2 patients admitted to the hospital with chronic liver disease ICD-10-CM codes. Demographics, laboratory results and administrative data were archived and analyzed (SAS, Cary, NC). Generalized regression identified inflammatory and liver panel biomarkers assayed within 8h of hospital admission associated (p<.05) with progression to critical illness. Retained biomarkers underwent bootstrap forest analysis forming a receiver operating characteristic (ROC) that optimized area under ROC (AUROC) estimating model accuracy (precision). Continuous data summarized with median [IQR] were compared using Kruskal-Wallis Test. Discrete data summarized as counts or proportions were compared with chi-squared test. Two-tailed p<.05 was significant. Results: Out of the 4411 COVID-19 patients who were discharged between March 14, 2020 and September 30, 2021, 333 with a previous diagnosis chronic liver disease were included in this study. Demographics for this population are presented in Table 1. Statistical values for biomarkers and progression to critical illness are seen in table 1. Statistically significant markers are compared via explained variance and ROC curve in Figure 1. Although AST and D-dimer were statistically significant markers of progression to critical illness, when modelled as a predictive biomarker, they were not informative in the aggregated ensemble. Therefore, they were not included in the modeling analysis. Conclusion: Hypoalbuminemia, inflammatory markers, D-dimer, and AST were significantly associated with progression to critical illness. Indexing liver specific synthetic function (albumin) to CoV-2 evoked inflammatory markers improves explained variance for progression to critical illness. Alternative liver synthetic function biomarker (INR), ALT, and ALP were not a significant prognostic indicator for progression to severe illness. To our knowledge, this is debut of modeling hypoalbuminemia indexed with multiple routinely assayed inflammatory biomarkers for baseline risk assessment in COVID-19 patients with liver disease. (Table Presented) (Figure Presented)
ABSTRACT
Introduction: In 2020, poverty in the United States increased as the COVID-19 pandemic led to the loss of work and/or income. Recent research has also shown that stress caused by the pandemic has led to increased rates of poor sleep. While insomnia rates have increased nationwide, it is not yet known if those living in poverty experienced insomnia symptoms at disproportionate rates. This study examined the effect poverty has had on insomnia symptom severity, as well as whether perceived stress mediated this association. Methods: Survey data was collected from 3,775 U.S. adults (83.1% White, 78.6% female, age = 18 - 86 years old) during the initial months of the COVID-19 pandemic (April-June 2020). These data were used for a secondary analysis. Participants completed an online survey aimed to assess basic demographics, sleep, physical activity, social engagement, and overall stress levels. Poverty was defined using the poverty guidelines provided by the Department of Health and Human Services (i.e., based on self-reported income and family/household size). The Insomnia Severity Index (ISI) was used to assess insomnia symptoms. Perceived stress was assessed using the Perceived Stress Scale (PSS). Results: 316 participants (8.4%) met criteria to be considered living below the poverty threshold. Those below the poverty threshold had a mean ISI of 10.20 (95% CI: 9.54, 10.86), while those above the poverty threshold had a mean ISI of 8.33 (95% CI: 8.13, 8.53). Put differently, 26.6% of those below the poverty threshold met criteria for clinical insomnia (i.e., ISI > 14), whereas 15.9% of those above the poverty threshold met criteria for clinical insomnia. Finally, a mediation test (with bootstrapping) confirmed that the association between poverty and insomnia was partially mediated by perceived stress (indirect effect = 1.15, 95% CI: 0.76, 1.55). Conclusion: While poverty guidelines vary by state, these data generally support that there are notable disparities in sleep and insomnia based on family/household income, and that these differences are, in part, due to greater perceived stress. This may be due to increased stress related to loss of work or income. Future studies examining the impact of pandemic stress on insomnia should consider the role of socio-economic status.