Using machine learning to develop a novel COVID-19 Vulnerability Index (C19VI)

Applying Machine Learning Techniques to Implementation Science

Machine learning (ML) approaches could expand the usefulness and application of implementation science methods in clinical medicine and public health settings. The aim of this viewpoint is to introduce a roadmap for applying ML techniques to address implementation science questions, such as predicting what will work best, for whom, under what circumstances, and with what predicted level of support, and what and when adaptation or deimplementation are needed. We describe how ML approaches could be used and discuss challenges that implementation scientists and methodologists will need to consider when using ML throughout the stages of implementation.

A modeling pipeline to relate municipal wastewater surveillance and regional public health data

As COVID-19 becomes endemic, public health departments benefit from improved passive indicators, which are independent of voluntary testing data, to estimate the prevalence of COVID-19 in local communities. Quantification of SARS-CoV-2 RNA from wastewater has the potential to be a powerful passive indicator. However, connecting measured SARS-CoV-2 RNA to community prevalence is challenging due to the high noise typical of environmental samples. We have developed a generalized pipeline using in- and out-of-sample model selection to test the ability of different correction models to reduce the variance in wastewater measurements and applied it to data collected from treatment plants in the Chicago area. We built and compared a set of multi-linear regression models, which incorporate pepper mild mottle virus (PMMoV) as a population biomarker, Bovine coronavirus (BCoV) as a recovery control, and wastewater system flow rate into a corrected estimate for SARS-CoV-2 RNA concentration. For our data, models with BCoV performed better than those with PMMoV, but the pipeline should be used to reevaluate any new data set as the sources of variance may change across locations, lab methods, and disease states. Using our best-fit model, we investigated the utility of RNA measurements in wastewater as a leading indicator of COVID-19 trends. We did this in a rolling manner for corrected wastewater data and for other prevalence indicators and statistically compared the temporal relationship between new increases in the wastewater data and those in other prevalence indicators. We found that wastewater trends often lead other COVID-19 indicators in predicting new surges.

Vulnerability assessment of heat waves within a risk framework using artificial intelligence

Current efforts to adapt to climate change are not sufficient to reduce projected impacts. Vulnerability assessments are essential to allocate resources where they are needed most. However, current assessments that use principal component analysis suffer from multiple shortcomings and are hard to translate into concrete actions. To address these issues, this article proposes a novel data-driven vulnerability assessment within a risk framework. The framework is based on the definitions from the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, but some definitions, such as sensitivity and adaptive capacity, are clarified. Heat waves that occurred between 2001 and 2018 in Quebec (Canada) are used to validate the framework. The studied impact is the daily mortality rates per cooling degree-days (CDD) region. A vulnerability map is produced to identify the distributions of summer mortality rates in aggregate dissemination areas within each CDD region. Socioeconomic and environmental variables are used to calculate impact and vulnerability. We compared abilities of AutoGluon (an AutoML framework), Gaussian process, and deep Gaussian process to model the impact and vulnerability. We offer advice on how to avoid common pitfalls with artificial intelligence and machine-learning algorithms. Gaussian process is a promising approach for supporting the proposed framework. SHAP values provide an explanation for the model results and are consistent with current knowledge of vulnerability. Recommendations are made to implement the proposed framework quantitatively or qualitatively.

Examining associations between social vulnerability indices and COVID-19 incidence and mortality with spatial-temporal Bayesian modeling

This study compares two social vulnerability indices, the U.S. CDC SVI and SoVI (the Social Vulnerability Index developed at the Hazards Vulnerability & Resilience Institute at the University of South Carolina), on their ability to predict the risk of COVID-19 cases and deaths. We utilize COVID-19 cases and deaths data for the state of Indiana from the Regenstrief Institute in Indianapolis, Indiana, from March 1, 2020, to March 31, 2021. We then aggregate the COVID-19 data to the census tract level, obtain the input variables, domains (components), and composite measures of both CDC SVI and SoVI data to create a Bayesian spatial-temporal ecological regression model. We compare the resulting spatial-temporal patterns and relative risk (RR) of SARS-CoV-2 infection (COVID-19 cases) and associated death. Results show there are discernable spatial-temporal patterns for SARS-CoV-2 infections and deaths with the largest contiguous hotspot for SARS-CoV-2 infections found in the southwest of the Indianapolis metropolitan area. We also observed one large contiguous hotspot for deaths that stretches across Indiana from the Cincinnati area in the southeast to just east and north of Terre Haute (southeast to west central). The spatial-temporal Bayesian model shows that a 1-percentile increase in CDC SVI was significantly (p ≤ 0.05) associated with an increased risk of SARS-CoV-2 infection by 6 % (RR = 1.06, 95 %CI = 1.04 -1.08). Whereas a 1-percentile increase in SoVI was significantly predicted to increase the risk of COVID-19 death by 45 % (RR = 1.45, 95 %CI =1.38 - 1.53). Domain-specific variables related to socioeconomic status, age, and race/ethnicity were shown to increase the risk of SARS-CoV-2 infections and deaths. There were notable differences in the relative risk estimates for SARS-CoV-2 infections and deaths when each of the two indices were incorporated in the model. Observed differences between the two social vulnerability indices and infection and death are likely due to alternative methodologies of formation and differences in input variables. The findings add to the growing literature on the relationship between social vulnerability and COVID-19 and further the development of COVID-19-specific vulnerability indices by illustrating the utility of local spatial-temporal analysis.

An early warning precision public health approach for assessing COVID-19 vulnerability in the UK: the Moore-Hill Vulnerability Index (MHVI)

BACKGROUND

Most COVID-19 vulnerability indices rely on measures that are biased by rates of exposure or are retrospective like mortality rates that offer little opportunity for intervention. The Moore-Hill Vulnerability Index (MHVI) is a precision public health early warning alternative to traditional infection fatality rates that presents avenues for mortality prevention.

METHODS

We produced an infection-severity vulnerability index by calculating the proportion of all recorded positive cases that were severe and attended by ambulances at small area scale for the East Midlands of the UK between May 2020 and April 2022. We produced maps identifying regions with high and low vulnerability, investigated the accuracy of the index over shorter and longer time periods, and explored the utility of the MHVI compared to other common proxy measures and indices. Analysis included exploring the correlation between our novel index and the Index of Multiple Deprivation (IMD).

RESULTS

The MHVI captures geospatial dynamics that single metrics alone often overlook, including the compound health challenges associated with disadvantaged and declining coastal towns inhabited by communities with post-industrial health legacies. A moderate negative correlation between MHVI and IMD reflects spatial analysis which suggests that high vulnerability occurs in affluent rural as well as deprived coastal and urban communities. Further, the MHVI estimates of severity rates are comparable to infection fatality rates for COVID-19.

CONCLUSIONS

The MHVI identifies regions with known high rates of poor health outcomes prior to the pandemic that case rates or mortality rates alone fail to identify. Pre-hospital early warning measures could be utilised to prevent mortality during a novel pandemic.

Natural and socio-environmental factors contribute to the transmissibility of COVID-19: evidence from an improved SEIR model

COVID-19 has ravaged Brazil, and its spread showed spatial heterogeneity. Changes in the environment have been implicated as potential factors involved in COVID-19 transmission. However, considerable research efforts have not elucidated the risk of environmental factors on COVID-19 transmission from the perspective of infectious disease dynamics. The aim of this study is to model the influence of the environment on COVID-19 transmission and to analyze how the socio-ecological factors affecting the probability of virus transmission in 10 states dramatically shifted during the early stages of the epidemic in Brazil. First, this study used a Pearson correlation to analyze the interconnection between COVID-19 morbidity and socio-ecological factors and identified factors with significant correlations as the dominant factors affecting COVID-19 transmission. Then, the time-lag effect of dominant factors on the morbidity of COVID-19 was investigated by constructing a distributed lag nonlinear model and standard two-stage meta-analytic model, and the results were considered in the improved SEIR model. Lastly, a machine learning method was introduced to explore the nonlinear relationship between the environmental propagation probability and socio-ecological factors. By analyzing the impact of environmental factors on virus transmission, it can be found that population mobility directly caused by human activities had a greater impact on virus transmission than temperature and humidity. The heterogeneity of meteorological factors can be accounted for by the diverse climate patterns in Brazil. The improved SEIR model was adopted to explore the interconnection of COVID-19 transmission and the environment, which revealed a new strategy to probe the causal links between them.

Social Vulnerability, COVID-19, Racial Violence, and Depressive Symptoms: a Cross-sectional Study in the Southern United States

BACKGROUND

In March 2020, the novel 2019 coronavirus disease (COVID-19) was declared a pandemic. In May 2020, George Floyd was murdered, catalyzing a national racial reckoning. In the Southern United States, these events occurred in the context of a history of racism and high rates of poverty and discrimination, especially among racially and ethnically minoritized populations.

OBJECTIVES

In this study, we examine social vulnerabilities, the perceived impacts of COVID-19 and the national racial reckoning, and how these are associated with depression symptoms in the South.

METHODS

Data were collected from 961 adults between June and November 2020 as part of an online survey study on family well-being during COVID-19. The sample was majority female (87.2%) and consisted of 661 White participants, 143 Black participants, and 157 other racial and ethnic minoritized participants. Existing social vulnerability, perceived impact of COVID-19 and racial violence and protests on families, and depressive symptoms were assessed. Hierarchical regression analysis was used to predict variance in depressive symptoms.

RESULTS

Half of the sample (52%) reported a negative impact of COVID-19, and 66% reported a negative impact of national racial violence/protests. Depressive symptoms were common with 49.8% meeting the cutoff for significant depressive symptoms; Black participants had lower levels of depressive symptoms. Results from the hierarchical regression analysis indicate social vulnerabilities and the perceived negative impact of COVID-19 and racial violence/protests each contribute to variance in depressive symptoms. Race-specific sensitivity analysis clarified distinct patterns in predictors of depressive symptoms.

CONCLUSION

People in the South report being negatively impacted by the confluence of the COVID-19 pandemic and the emergence of racial violence/protests in 2020, though patterns differ by racial group. These events, on top of pre-existing social vulnerabilities, help explain depressive symptoms in the South during 2020.

Preparing for the next outbreak: A review of indices measuring outbreak preparedness, vulnerability, and resilience

The COVID-19 pandemic has highlighted the need for relevant metrics describing the resources and community attributes that affect the impact of communicable disease outbreaks. Such tools can help inform policy, assess change, and identify gaps to potentially reduce the negative outcomes of future outbreaks. The present review was designed to identify available indices to assess communicable disease outbreak preparedness, vulnerability, or resilience, including articles describing an index or scale developed to address disasters or emergencies which could be applied to addressing a future outbreak. This review assesses the landscape of indices available, with a particular focus on tools assessing local-level attributes. This systematic review yielded 59 unique indices applicable to assessing communicable disease outbreaks through the lens of preparedness, vulnerability, or resilience. However, despite the large number of tools identified, only 3 of these indices assessed factors at the local level and were generalizable to different types of outbreaks. Given the influence of local resources and community attributes on a wide range of communicable disease outcomes, there is a need for local-level tools that can be applied broadly to various types of outbreaks. Such tools should assess both current and long-term changes in outbreak preparedness with the intent to identify gaps, inform local-level decision makers, public policy, and future response to current and novel outbreaks.

Wastewater-based epidemiology predicts COVID-19-induced weekly new hospital admissions in over 150 USA counties

Although the coronavirus disease (COVID-19) emergency status is easing, the COVID-19 pandemic continues to affect healthcare systems globally. It is crucial to have a reliable and population-wide prediction tool for estimating COVID-19-induced hospital admissions. We evaluated the feasibility of using wastewater-based epidemiology (WBE) to predict COVID-19-induced weekly new hospitalizations in 159 counties across 45 states in the United States of America (USA), covering a population of nearly 100 million. Using county-level weekly wastewater surveillance data (over 20 months), WBE-based models were established through the random forest algorithm. WBE-based models accurately predicted the county-level weekly new admissions, allowing a preparation window of 1-4 weeks. In real applications, periodically updated WBE-based models showed good accuracy and transferability, with mean absolute error within 4-6 patients/100k population for upcoming weekly new hospitalization numbers. Our study demonstrated the potential of using WBE as an effective method to provide early warnings for healthcare systems.

Multidimensional Machine Learning Model to Calculate a COVID-19 Vulnerability Index

In Colombia, the first case of COVID-19 was confirmed on 6 March 2020. On 13 March 2023, Colombia registered 6,360,780 confirmed positive cases of COVID-19, representing 12.18% of the total population. The National Administrative Department of Statistics (DANE) in Colombia published in 2020 a COVID-19 vulnerability index, which estimates the vulnerability (per city block) of being infected with COVID-19. Unfortunately, DANE did not consider multiple factors that could increase the risk of COVID-19 (in addition to demographic and health), such as environmental and mobility data (found in the related literature). The proposed multidimensional index considers variables of different types (unemployment rate, gross domestic product, citizens' mobility, vaccination data, and climatological and spatial information) in which the incidence of COVID-19 is calculated and compared with the incidence of the COVID-19 vulnerability index provided by DANE. The collection, data preparation, modeling, and evaluation phases of the Cross-Industry Standard Process for Data Mining methodology (CRISP-DM) were considered for constructing the index. The multidimensional index was evaluated using multiple machine learning models to calculate the incidence of COVID-19 cases in the main cities of Colombia. The results showed that the best-performing model to predict the incidence of COVID-19 in Colombia is the Extra Trees Regressor algorithm, obtaining an R-squared of 0.829. This work is the first step toward a multidimensional analysis of COVID-19 risk factors, which has the potential to support decision making in public health programs. The results are also relevant for calculating vulnerability indexes for other viral diseases, such as dengue.

Development of a COVID-19 Vulnerability Index (CVI) for the Counties and Residents of New Jersey, USA

The coronavirus disease 2019, or COVID-19, has impacted countless aspects of everyday life since it was declared a global pandemic by the World Health Organization in March of 2020. From societal to economic impacts, COVID-19 and its variants will leave a lasting impact on our society and the world. During the height of the pandemic, it became increasingly evident that indices, such as the Center for Disease Control's (CDC) Social Vulnerability Index (SVI), were instrumental in predicting vulnerabilities within a community. The CDC's SVI provides important estimates on which communities will be more susceptible to 'hazard events' by compiling a variety of data from the U.S. Census and the American Community Survey. The CDC's SVI does not directly consider the susceptibility of a community to a global pandemic, such as the COVID-19 pandemic, due to the four themes and 15 factors that contribute to the index. Thus, the objective of this research is to develop a COVID-19 Vulnerability Index, or CVI, to evaluate a community's susceptibility to future pandemics. With 15 factors considered for CDC's SVI, 26 other factors were also considered for the development of the CVI that covered themes such as socioeconomic status, environmental factors, healthcare capacity, epidemiological factors, and disability. All factors were equally weighted to calculate the CVI based on New Jersey. The CVI was validated by comparing index results to real-world COVID-19 data from New Jersey's 21 counties and CDC's SVI. The results present a stronger positive linear relationship between the CVI and the New Jersey COVID-19 mortality/population and infection/population than there is with the SVI. The results of this study indicate that Essex County has the highest CVI, and Hunterdon County has the lowest CVI. This is due to factors such as disparity in wealth, population density, minority status, and housing conditions, as well as other factors that were used to compose the CVI. The implications of this research will provide a critical tool for decision makers to utilize in allocating resources should another global pandemic occur. This CVI, developed through this research, can be used at the county, state, and global levels to help measure the vulnerability to future pandemics.

Incorporating resilience when assessing pandemic risk in the Arctic: a case study of Alaska

The discourse on vulnerability to COVID-19 or any other pandemic is about the susceptibility to the effects of disease outbreaks. Over time, vulnerability has been assessed through various indices calculated using a confluence of societal factors. However, categorising Arctic communities, without considering their socioeconomic, cultural and demographic uniqueness, into the high and low continuum of vulnerability using universal indicators will undoubtedly result in the underestimation of the communities' capacity to withstand and recover from pandemic exposure. By recognising vulnerability and resilience as two separate but interrelated dimensions, this study reviews the Arctic communities' ability to cope with pandemic risks. In particular, we have developed a pandemic vulnerability-resilience framework for Alaska to examine the potential community-level risks of COVID-19 or future pandemics. Based on the combined assessment of the vulnerability and resilience indices, we found that not all highly vulnerable census areas and boroughs had experienced COVID-19 epidemiological outcomes with similar severity. The more resilient a census area or borough is, the lower the cumulative death per 100 000 and case fatality ratio in that area. The insight that pandemic risks are the result of the interaction between vulnerability and resilience could help public officials and concerned parties to accurately identify the populations and communities at most risk or with the greatest need, which, in turn, helps in the efficient allocation of resources and services before, during and after a pandemic. A resilience-vulnerability-focused approach described in this paper can be applied to assess the potential effect of COVID-19 and similar future health crises in remote regions or regions with large Indigenous populations in other parts of the world.

Spatial relationship between COVID-19 and previous infectious disease outbreaks: Evidence from South Korea

As the novel coronavirus disease (COVID-19) has been rapidly spreading across the world, scholars have started paying attention to risk factors that affect the occurrence of the infectious disease. While various urban characteristics have been shown to influence the outbreak, less is known about whether COVID-19 is more likely to be transmitted in areas with a greater number of incidents of previous infectious diseases. This study examines a spatial relationship between COVID-19 and previous infectious diseases from a spatial perspective. Using the confirmed cases of COVID-19 and other types of infectious diseases across South Korea, we identified spatial clusters through regression and spatial econometric models. We found that COVID-19-confirmed case rates tended to be clustered despite no similarity with the spatial patterns of previous infectious diseases. Existing infectious diseases from abroad were associated with the occurrence of COVID-19, while the effect diminished after controlling for the spatial effect. Our findings highlight the importance of regional-level infectious disease surveillance for the effective prevention and control of COVID-19.

Data driven contagion risk management in low-income countries using machine learning applications with COVID-19 in South Asia

In the absence of real-time surveillance data, it is difficult to derive an early warning system and potential outbreak locations with the existing epidemiological models, especially in resource-constrained countries. We proposed a contagion risk index (CR-Index)-based on publicly available national statistics-founded on communicable disease spreadability vectors. Utilizing the daily COVID-19 data (positive cases and deaths) from 2020 to 2022, we developed country-specific and sub-national CR-Index for South Asia (India, Pakistan, and Bangladesh) and identified potential infection hotspots-aiding policymakers with efficient mitigation planning. Across the study period, the week-by-week and fixed-effects regression estimates demonstrate a strong correlation between the proposed CR-Index and sub-national (district-level) COVID-19 statistics. We validated the CR-Index using machine learning methods by evaluating the out-of-sample predictive performance. Machine learning driven validation showed that the CR-Index can correctly predict districts with high incidents of COVID-19 cases and deaths more than 85% of the time. This proposed CR-Index is a simple, replicable, and easily interpretable tool that can help low-income countries prioritize resource mobilization to contain the disease spread and associated crisis management with global relevance and applicability. This index can also help to contain future pandemics (and epidemics) and manage their far-reaching adverse consequences.

A counterfactual analysis of opioid-involved deaths during the COVID-19 pandemic using a spatiotemporal random forest modeling approach

The global pandemic of SARS-CoV-2 (COVID-19) has been linked to adversely impacting individuals with opioid use disorder in the United States. This study focuses on analyzing opioid-involved mortality in the context of COVID-19 in the U.S. from a geospatial perspective. We investigated spatiotemporal patterns of opioid-involved deaths during 2020 and compared the spatiotemporal pattern of these deaths with patterns for the previous three years (2017-2019) to understand changes in the context of the COVID-19 pandemic. A counterfactual analysis framework together with a space-time random forest (STRF) model were used to estimate the increase in opioid-involved deaths related to the pandemic. To gain further insight into the relationship between opioid deaths and COVID-19-related factors, we built a space-time random forest model for the City of Chicago, that experienced a steep increase in opioid-related deaths during 2020. High ranking indicators identified by the model such as the number of positive COVID-19 cases adjusted by population and the change in stay-at-home dwell time during the pandemic were used to generate a vulnerability index for opioid overdoses during the COVID-19 pandemic in Chicago.

Community vulnerability to the COVID-19 pandemic: A narrative synthesis from an ecological perspective

Background

We aimed to conduct a narrative synthesis of components and indicators of community vulnerability to a pandemic and discuss their interrelationships from an ecological perspective.

Methods

We searched from PubMed, Embase, Web of Science, PsycINFO, and Scopus (updated to November 2021) for studies focusing on community vulnerability to a pandemic caused by novel respiratory viruses on a geographic unit basis . Studies that reported the associations of community vulnerability levels with at least one disease morbidity or mortality outcome were included.

Results

Forty-one studies were included. All were about the COVID-19 pandemic. Suitable temperature and humidity environments, advanced social and human development (including high population density and human mobility, connectivity, and occupations), and settings that intensified physical interactions are important indicators of vulnerability to viral exposure. However, the eventual pandemic health impacts are predominant in communities that faced environmental pollution, higher proportions of socioeconomically deprived people, health deprivation, higher proportions of poor-condition households, limited access to preventive health care and urban infrastructure, uneven social and human development, and racism. More stringent social distancing policies were associated with lower COVID-19 morbidity and mortality only in the early pandemic phases. Prolonged social distancing policies can disproportionately burden the socially disadvantaged and racially/ethnically marginalized groups.

Conclusions

Community vulnerability to a pandemic is foremost the vulnerability of the ecological systems shaped by complex interactions between the human and environmental systems.

Registration

PROSPERO (CRD42021266186).

Social vulnerability amplifies the disparate impact of mobility on COVID-19 transmissibility across the United States

Although human mobility is considered critical for the spread of the new coronavirus disease (COVID-19) both locally and globally, the extent to which such an association is impacted by social vulnerability remains unclear. Here, using multisource epidemiological and socioeconomic data of US counties, we develop a COVID-19 pandemic vulnerability index (CPVI) to quantify their levels of social vulnerability and examine how social vulnerability moderated the influence of mobility on disease transmissibility (represented by the effective reproduction number, R _t) during the US summer epidemic wave of 2020. We find that counties in the top CPVI quintile suffered almost double in regard to COVID-19 transmission (45.02% days with an R _t higher than 1) from mobility, particularly intracounty mobility, compared to counties in the lowest quintile (21.90%). In contrast, counties in the bottom CPVI quintile were only slightly affected by the level of mobility. As such, a 25% intracounty mobility change was associated with a 15.28% R _t change for counties in the top CPVI quintile, which is eight times the 1.81% R _t change for those in the lowest quintile. These findings suggest the need to account for the vulnerability of communities when making social distancing measures against mobility in the future.

A district-level vulnerability assessment of next COVID-19 variant (Omicron BA.2) in Uttarakhand using quantitative SWOT analysis

COVID-19 has had an impact on the entire humankind and has been proved to spread in deadly waves. As a result, preparedness and planning are required to better deal with the epidemic's upcoming waves. Effective planning, on the other hand, necessitates detailed vulnerability assessments at all levels, from the national to the state or regional. There are several issues at the regional level, and each region has its own features. As a result, each region needs its own COVID-19 vulnerability assessment. In terms of climate, terrain and demographics, the state of Uttarakhand differs significantly from the rest of India. As a result, a vulnerability assessment of the next COVID-19 variation (Omicron BA.2) is required for district-level planning to meet regional concerns. A total of 17 variables were chosen for this study, including demographic, socio-economic, infrastructure, epidemiological and tourism-related factors. AHP was used to compute their weights. After applying min-max normalisation to the data, a district-level quantitative SWOT is created to compare the performance of 13 Uttarakhand districts. A COVID-19 vulnerability index (normalised R i ) ranging between 0 and 1 was produced, and district-level vulnerabilities were mapped. Quantitative SWOT results depict that Dehradun is a best performing district followed by Haridwar, while Bageshwar, Rudra Prayag, Champawat and Pithoragarh are on the weaker side and the normalised Ri proves Dehradun, Nainital, Champawat, Bageshwar and Chamoli to be least vulnerable to COVID-19 (normalised R i ≤ 0.25) and Pithoragarh to be the most vulnerable district (normalised R i > 0.90). Pauri Garwal and Uttarkashi are moderately vulnerable (normalised R i 0.50 to 0.75).

Exploring spatial distribution of social vulnerability and its relationship with the Coronavirus disease 2019: the Capital region of South Korea

Background & objective

The ongoing coronavirus disease 2019 (COVID-19) pandemic continues to cause death and socioeconomic problems worldwide. This study examined the spatial distribution of social vulnerability to COVID-19 and its relationship with the number of confirmed COVID-19 cases in 2020, focusing on the Capital region of South Korea.

Methods

A traditional social vulnerability index (SVI), healthy SVI, and the difference of each SVI were constructed in 2015 and 2019. The traditional SVI was constructed across five domains: age, socioeconomic disadvantage, housing, income, and environment. The healthy SVI domains were: prevention, health-related habits, chronic disease, healthcare infrastructure, and mortality. The spatial distribution of the traditional SVI, healthy SVI, and confirmed cases of COVID-19 was explored using ArcGIS 10.5. Pearson correlation was used to identify the relationship between confirmed COVID-19 cases and the two SVIs and their changes between 2015 and 2019. Four multiple linear regression models were used to identify the impact of the changes of the two SVIs on the confirmed COVID-19 cases for the three episodes and total period with control of population using STATA/MP 16.1.

Results

Confirmed COVID-19 cases were concentrated in a specific area of the Capital region. The traditional SVI was more vulnerable in the outer regions of the Capital region, and some central, western, and eastern areas reflected an increase in vulnerability. Healthy SVI was more vulnerable in the northern part of the Capital region, and increase in vulnerability showed in some central areas above Seoul. By multiple regression with the population controlled, the difference of the traditional SVI between 2015 and 2019 showed a positive relationship with the confirmed COVID-19 cases in all models at a significance level of 0.05, and the 2019 integrated SVI showed a negative relationship with confirmed COVID-19 cases in all models.

Conclusions

The results of this study showed that the confirmed COVID-19 cases are associated with increased traditional SVI vulnerability between 2015 and 2019 and have a high positive relationship with the spread of COVID-19. Policy efforts are needed to reduce confirmed COVID-19 cases among the vulnerable in regions with relatively increased traditional SVI.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-022-14212-7.

Temporal trend of COVID-19 incidence and mortality rates and their relationship with socioeconomic indicators in the state of Piauí, Brazil: an ecological study, 2020-2021

OBJECTIVE

To analyze the temporal trend of COVID-19 incidence and mortality rates and their relationship with socioeconomic indicators.

METHODS

This was an ecological time series study of COVID-19 cases/deaths in municipalities in Piauí, Brazil, between March, 2020 and May, 2021. Prais-Winsten linear regression model and Spearman's correlation test were used.

RESULTS

There were 271,228 cases and 5,888 deaths in the period. There was a rising trend in COVID-19 incidence rate, while the mortality trend was stable. The spatio-temporal analyses showed higher incidence/mortality in the second and fifth quarters of the period. There was no statistically significant correlation between COVID-19 and the Social Vulnerability Index (IVS). Significant correlations between the Municipal Human Development Index (IDHM) and COVID-19 incidence (p-value < 0.001) and mortality rates (p-value < 0.001) were found.

CONCLUSION

There was a rising trend in COVID-19 incidence and stability in COVID-19 mortality. Correlation between the MHDI and these two indicators was moderate and weak, respectively, demanding public service management decisions aimed at improving the population's quality of life.

Assessment of community vulnerability during the COVID-19 pandemic: Hong Kong as a case study

The COVID-19 pandemic continues to threaten global public health. Reliable assessment of community vulnerability is therefore essential to fighting and mitigating the pandemic. This study presents a framework that considers the roles of internal and external factors, including the components of social vulnerability, exposure, and sensitivity, to comprehensively and accurately assess community vulnerability to the pandemic. With respect to internal factors, we summarized the inherent social characteristics of people groups using census data and explored the roles of both overall and four major thematic social vulnerabilities in shaping community infection by COVID-19. We then designed two external factors to characterize exposure and sensitivity and implemented an aggregation by multiplying them with the internal social vulnerability to achieve a comprehensive vulnerability assessment. The role of the estimated vulnerability in shaping community infection was evaluated by statistical and spatial analysis as well as by risk factor classification using defined rules. This case study of Hong Kong demonstrated the value of our framework in vulnerability assessment and revealed the role of vulnerability in shaping community infection by COVID-19.

A machine learning-driven spatio-temporal vulnerability appraisal based on socio-economic data for COVID-19 impact prevention in the U.S. counties

A mature and hybrid machine-learning model is verified by mature empirical analysis to measure county-level COVID-19 vulnerability and track the impact of the imposition of pandemic control policies in the U.S. A total of 30 county-level social, economic, and medical variables and a timeline of the imposed policies constitutes a COVID-19 database. A hybrid feature-selection model composed of four machine-learning algorithms is developed to emphasize the regional impact of community features on the case fatality rate (CFR). A COVID-19 vulnerability index (COVULin) is proposed to measure the county's vulnerability, the effects of model's parameters on mortality, and the efficiency of control policies. The results showed that the dense counties in which minority groups represent more than 45% of the population and those with poverty rates greater than 24% were the most vulnerable counties during the first and the last pandemic peaks, respectively. Highly-correlated CFR and COVULin scores indicated a close agreement between the model outcomes and COVID-19 impacts. Counties with higher poverty and uninsured rates were the most resistant to government intervention. It is anticipated that the proposed model can play an essential role in identifying vulnerable communities and help reduce damages during long-term alike disasters.

The Impact of US County-Level Factors on COVID-19 Morbidity and Mortality

The effect of socio-economic factors, ethnicity, and other factors, on the morbidity and mortality of COVID-19 at the sub-population-level, rather than at the individual level, and their temporal dynamics, is only partially understood. Fifty-three county-level features were collected between 4/2020 and 11/2020 from 3,071 US counties from publicly available data of various American government and news websites: ethnicity, socio-economic factors, educational attainment, mask usage, population density, age distribution, COVID-19 morbidity and mortality, presidential election results, and ICU beds. We trained machine learning models that predict COVID-19 mortality and morbidity using county-level features and then performed a SHAP value game theoretic importance analysis of the predictive features for each model. The classifiers produced an AUROC of 0.863 for morbidity prediction and an AUROC of 0.812 for mortality prediction. A SHAP value-based analysis indicated that poverty rate, obesity rate, mean commute time, and mask usage statistics significantly affected morbidity rates, while ethnicity, median income, poverty rate, and education levels heavily influenced mortality rates. Surprisingly, the correlation between several of these factors and COVID-19 morbidity and mortality gradually shifted and even reversed during the study period; our analysis suggests that this phenomenon was probably due to COVID-19 being initially associated with more urbanized areas and, then, from 9/2020, with less urbanized ones. Thus, socio-economic features such as ethnicity, education, and economic disparity are the major factors for predicting county-level COVID-19 mortality rates. Between counties, low variance factors (e.g., age) are not meaningful predictors. The inversion of some correlations over time can be explained by COVID-19 spreading from urban to rural areas.

County-Level Assessment of Vulnerability to COVID-19 in Alabama

The COVID-19 pandemic has posed an unprecedented challenge to public health across the world and has further exposed health disparities and the vulnerability of marginal groups. Since the pandemic has exhibited marked regional differences, it is necessary to better understand the levels of vulnerability to the disease at local levels and provide policymakers with additional tools that will allow them to develop finely targeted policies. In this study, we develop for the State of Alabama (USA) a composite vulnerability index at county level that can be used as a tool that will help in the management of the pandemic. Twenty-four indicators were assigned to the following three categories: exposure, sensitivity, and adaptive capacity. The resulting subindices were aggregated into a composite index that depicts the vulnerability to COVID-19. A multivariate analysis was used to assign factor loadings and weights to indicators, and the results were mapped using Geographic Information Systems. The vulnerability index captured health disparities very well. Many of the most vulnerable counties were found in the Alabama Black Belt region. A deconstruction of the overall index and subindices allowed the development of individual county profiles and the detection of local strengths and weaknesses. We expect the model developed in this study to be an efficient planning tool for decision-makers.

Covid-19 vaccination priorities defined on machine learning

OBJECTIVE

Defining priority vaccination groups is a critical factor to reduce mortality rates.

METHODS

We sought to identify priority population groups for covid-19 vaccination, based on in-hospital risk of death, by using Extreme Gradient Boosting Machine Learning (ML) algorithm. We performed a retrospective cohort study comprising 49,197 patients (18 years or older), with RT-PCR-confirmed for covid-19, who were hospitalized in any of the 336 Brazilian hospitals considered in this study, from March 19th, 2020, to March 22nd, 2021. Independent variables encompassed age, sex, and chronic health conditions grouped into 179 large categories. Primary outcome was hospital discharge or in-hospital death. Priority population groups for vaccination were formed based on the different levels of in-hospital risk of death due to covid-19, from the ML model developed by taking into consideration the independent variables. All analysis were carried out in Python programming language (version 3.7) and R programming language (version 4.05).

RESULTS

Patients' mean age was of 60.5 ± 16.8 years (mean ± SD), mean in-hospital mortality rate was 17.9%, and the mean number of comorbidities per patient was 1.97 ± 1.85 (mean ± SD). The predictive model of in-hospital death presented area under the Receiver Operating Characteristic Curve (AUC - ROC) equal to 0.80. The investigated population was grouped into eleven (11) different risk categories, based on the variables chosen by the ML model developed in this study.

CONCLUSIONS

The use of ML for defining population priorities groups for vaccination, based on risk of in-hospital death, can be easily applied by health system managers.

Suburban Road Networks to Explore COVID-19 Vulnerability and Severity

The Delta variant of COVID-19 has been found to be extremely difficult to contain worldwide. The complex dynamics of human mobility and the variable intensity of local outbreaks make measuring the factors of COVID-19 transmission a challenge. The inter-suburb road connection details provide a reliable proxy of the moving options for people between suburbs for a given region. By using such data from Greater Sydney, Australia, this study explored the impact of suburban road networks on two COVID-19-related outcomes measures. The first measure is COVID-19 vulnerability, which gives a low score to a more vulnerable suburb. A suburb is more vulnerable if it has the first COVID-19 case earlier and vice versa. The second measure is COVID-19 severity, which is proportionate to the number of COVID-19-positive cases for a suburb. To analyze the suburban road network, we considered four centrality measures (degree, closeness, betweenness and eigenvector) and core-periphery structure. We found that the degree centrality measure of the suburban road network was a strong and statistically significant predictor for both COVID-19 vulnerability and severity. Closeness centrality and eigenvector centrality were also statistically significant predictors for COVID-19 vulnerability and severity, respectively. The findings of this study could provide practical insights to stakeholders and policymakers to develop timely strategies and policies to prevent and contain any highly infectious pandemics, including the Delta variant of COVID-19.

Model-based ensembles: Lessons learned from retrospective analysis of COVID-19 infection forecasts across 10 countries

Mathematical models of different types and data intensities are highly used by researchers, epidemiologists, and national authorities to explore the inherently unpredictable progression of COVID-19, including the effects of different non-pharmaceutical interventions. Regardless of model complexity, forecasts of future COVID-19 infections, deaths and hospitalization are associated with large uncertainties, and critically depend on the quality of the training data, and in particular how well the recorded national or regional numbers of infections, deaths and recoveries reflect the the actual situation. In turn, this depends on, e.g., local test and abatement strategies, treatment capacities and available technologies. Other influencing factors including temperature and humidity, which are suggested by several authors to affect the spread of COVID-19 in some countries, are generally only considered by the most complex models and further serve to inflate the uncertainty. Here we use comparative and retrospective analyses to illuminate the aggregated effect of these systematic biases on ensemble-based model forecasts. We compare the actual progression of active infections across ten of the most affected countries in the world until late November 2020 with "re-forecasts" produced by two of the most commonly used model types: (i) a compartment-type, susceptible-infected-removed (SIR) model; and (ii) a statistical (Holt-Winters) time series model. We specifically examine the sensitivity of the model parameters, estimated systematically from different subsets of the data and thereby different time windows, to illustrate the associated implications for short- to medium-term forecasting and for probabilistic projections based on (single) model ensembles as inspired by, e.g., weather forecasting and climate research. Our findings portray considerable variations in forecasting skill in between the ten countries and demonstrate that individual model predictions are highly sensitive to parameter assumptions. Significant skill is generally only confirmed for short-term forecasts (up to a few weeks) with some variation across locations and periods.

Vulnerability interactive geographic viewer against COVID-19 at the block level in Colombia: Analytical tool based on machine learning techniques

To mitigate the effects of the coronavirus disease 2019 (COVID-19) pandemic, different countries have developed computational tools and dashboards that generate value for decision-making in public health. We aimed to build an interactive geographic viewer for vulnerability to COVID-19 at the block level in Colombia to identify the location of populations that, because of sociodemographic characteristics and health conditions, could have more complications from COVID-19 infections. The vulnerability levels of the different blocks of 1,102 municipal capitals were calculated. Additionally, the institutions that provide health services and hotels were georeferenced, and changes in people's mobility dynamics in large cities were identified.

Assessment of the Occurrence of Forest Fires in Pandemic Period by COVID-19 in Chile. Preliminary Backgrounds

The spatial and temporal behavior of the occurrence of forest fires in Chile was evaluated in the presence of COVID-19 and mobility restrictions. The fire period from 2015-2016 to 2020-2021 was considered and statistics on mobility restrictions were granted by the Government of Chile. The analysis was developed at different scales of geographic perception. At the national and regional levels, the global behavior of the occurrence was determined, and later at the communal level, the political territorial unit, to determine internal variations attributable to the mobility dynamics in the quarantine period. In the process, the meteorological background of the fire activity was also considered. The results indicate that it is possible to rule out a meteorological effect, based on the variation of the moisture content of fine fuel. There was also no statistical association between the humidity of the fuel and the variation in the occurrence of fires. It is concluded that the communes that presented the greatest mobility of people before the pandemic were those that obtained the greatest reduction in fires. The variation in mobility, the product of restriction measures, is a statistical predictor of the increase or decrease in fires.

Evolutionary warning system for COVID-19 severity: Colony predation algorithm enhanced extreme learning machine

Coronavirus Disease 2019 (COVID-19) was distributed globally at the end of December 2019 due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Early diagnosis and successful COVID-19 assessment are missing, clinical care is ineffective, and deaths are high. In this study, we investigate whether the level of biochemical indicators helps to discriminate and classify the severity of the COVID-19 using the machine learning method. This research creates an efficient intelligence method for the diagnosis of COVID-19 from the perspective of biochemical indexes. The framework is proposed by integrating an enhanced new stochastic called the colony predation algorithm (CPA) with a kernel extreme learning machine (KELM), abbreviated as ECPA-KELM. The core feature of the approach is the ECPA algorithm which incorporates the two main operators that have been abstained from the grey wolf optimizer and moth-flame optimizer to improve and restore the CPA research functions and are simultaneously used to optimize the parameters and to select features for KELM. The ECPA output is checked thoroughly using IEEE CEC2017 benchmark to verify the capacity of the proposed methodology. Finally, in the diagnosis of COVID-19 using biochemical indexes, the designed ECPA-KELM model and other competing KELM models based on other optimization are used. Checking statistical results will display improved predictive properties for all metrics and higher stability. ECPA-KELM can also be used to discriminate and classify the severity of the COVID-19 as a possible computer-aided method and provide effective early warning for the treatment and diagnosis of COVID-19.

Space-Time Cluster's Detection and Geographical Weighted Regression Analysis of COVID-19 Mortality on Texas Counties

As COVID-19 run rampant in high-density housing sites, it is important to use real-time data in tracking the virus mobility. Emerging cluster detection analysis is a precise way of blunting the spread of COVID-19 as quickly as possible and save lives. To track compliable mobility of COVID-19 on a spatial-temporal scale, this research appropriately analyzed the disparities between spatial-temporal clusters, expectation maximization clustering (EM), and hierarchical clustering (HC) analysis on Texas county-level. Then, based on the outcome of clustering analysis, the sensitive counties are Cottle, Stonewall, Bexar, Tarrant, Dallas, Harris, Jim hogg, and Real, corresponding to Southeast Texas analysis in Geographically Weighted Regression (GWR) modeling. The sensitive period took place in the last two quarters in 2020 and the first quarter in 2021. We explored PostSQL application to portray tracking Covid-19 trajectory. We captured 14 social, economic, and environmental impact's indices to perform principal component analysis (PCA) to reduce dimensionality and minimize multicollinearity. By using the PCA, we extracted five factors related to mortality of COVID-19, involved population and hospitalization, adult population, natural supply, economic condition, air quality or medical care. We established the GWR model to seek the sensitive factors. The result shows that adult population, economic condition, air quality, and medical care are the sensitive factors. Those factors also triggered high increase of COVID-19 mortality. This research provides geographical understanding and solution of controlling COVID-19, reference of implementing geographically targeted ways to track virus mobility, and satisfy for the need of emergency operations plan (EOP).

Forecasting of COVID-19 using deep layer Recurrent Neural Networks (RNNs) with Gated Recurrent Units (GRUs) and Long Short-Term Memory (LSTM) cells

In December 2019, first case of the COVID-19 was reported in Wuhan, Hubei province in China. Soon world health organization has declared contagious coronavirus disease (a.k.a. COVID-19) as a global pandemic in the month of March 2020. Over the span of eleven months, it has rapidly spread out all over the world with total confirmed cases of ~ 41.39 M and causing a total fatality of ~1.13 M. At present, the entire mankind is facing serious threat and it is believed that COVID-19 may have been around for quite some time. Therefore, it has become imperative to forecast the global impact of COVID-19 in the near future. The present work proposes state-of-art deep learning Recurrent Neural Networks (RNN) models to predict the country-wise cumulative confirmed cases, cumulative recovered cases and the cumulative fatalities. The Gated Recurrent Units (GRUs) and Long Short-Term Memory (LSTM) cells along with Recurrent Neural Networks (RNN) were developed to predict the future trends of the COVID-19. We have used publicly available data from John Hopkins University's COVID-19 database. In this work, we emphasize the importance of various factors such as age, preventive measures, and healthcare facilities, population density, etc. that play vital role in rapid spread of COVID-19 pandemic. Therefore, our forecasted results are very helpful for countries to better prepare themselves to control the pandemic.