Metrics of Mobility: Assessing the Impact of COVID-19 on Travel Behavior

The COVID-19 pandemic disrupted typical travel behavior worldwide. In the United States (U.S.), government entities took action to limit its spread through public health messaging to encourage reduced mobility and thus reduce the spread of the virus. Within statewide responses to COVID-19, however, there were different responses locally. Likely some of these variations were a result of individual attitudes toward the government and health messaging, but there is also likely a portion of the effects that were because of the character of the communities. In this research, we summarize county-level characteristics that are known to affect travel behavior for 404 counties in the U.S., and we investigate correlates of mobility between April and September (2020). We do this through application of three metrics that are derived via changepoint analysis—initial post-disruption mobility index, changepoint on restoration of a ‘‘new normal,'' and recovered mobility index. We find that variables for employment sectors are significantly correlated and had large effects on mobility during the pandemic. The state dummy variables are significant, suggesting that counties within the same state behaved more similarly to one another than to counties in different states. Our findings indicate that few travel characteristics that typically correlate with travel behavior are related to pandemic mobility, and that the number of COVID-19 cases may not be correlated with mobility outcomes. © National Academy of Sciences: Transportation Research Board 2021.

JUE Insight: Is Hospital Quality Predictive of Pandemic Deaths? Evidence from US Counties.

In the large literature on the spatial-level correlates of COVID-19, the association between quality of hospital care and outcomes has received little attention to date. To examine whether county-level mortality is correlated with measures of hospital performance, we assess daily cumulative deaths and pre-crisis measures of hospital quality, accounting for state fixed-effects and potential confounders. As a measure of quality, we use the pre-pandemic adjusted five-year penalty rates for excess 30-day readmissions following pneumonia admissions for the hospitals accessible to county residents based on ambulance travel patterns. Our adjustment corrects for socio-economic status and down-weighs observations based on small samples. We find that a one-standard-deviation increase in the quality of local hospitals is associated with a 2% lower death rate (relative to the mean of 20 deaths per 10,000 people) one and a half years after the first recorded death.

COVID-19 Epidemic Forecast using Machine Learning Regression Techniques

The COVID-19 Coronavirus (SARS-CoV-2), has caused destruction all around the world, since December 2019. It is still managing to grow at an unprecedented scale. It was declared as a health emergency for the entire globe by the World Health Organization (WHO) in January 2022. The virus continues to impact the lives of millions of people. An early detection system warning about the repercussions of the virus at a county level can be favorable for the residents as well and aid the government to enforce appropriate safety measures. This research aims at modeling such a warning system which predicts the positivity rate of COVID-19 for a geographical location. The proposed solution uses supervised machine learning techniques such as Random Forest, Linear Regression, Naive Bayes, and Gradient Boosting Regression. The prediction is made based on the analysis of the past data in each time frame with temporal input such as the population of the area, number of tests conducted, number of positive tests, reported cases in that area among others. The Gradient Boosting algorithm outperforms all the other algorithms used in this research. Machine learning based recommendation system for COVID-19 spread can help the public and government to take necessary precautions for suppressing its effect. The proposed modeling approach provides a reliable tool to predict COVID-19 transmission with an accuracy of 99.4%. © 2022 IEEE.

COVID-19 Epidemic Forecast using Machine Learning Regression Techniques

The COVID-19 Coronavirus (SARS-CoV-2), has caused destruction all around the world, since December 2019. It is still managing to grow at an unprecedented scale. It was declared as a health emergency for the entire globe by the World Health Organization (WHO) in January 2022. The virus continues to impact the lives of millions of people. An early detection system warning about the repercussions of the virus at a county level can be favorable for the residents as well and aid the government to enforce appropriate safety measures. This research aims at modeling such a warning system which predicts the positivity rate of COVID-19 for a geographical location. The proposed solution uses supervised machine learning techniques such as Random Forest, Linear Regression, Naive Bayes, and Gradient Boosting Regression. The prediction is made based on the analysis of the past data in each time frame with temporal input such as the population of the area, number of tests conducted, number of positive tests, reported cases in that area among others. The Gradient Boosting algorithm outperforms all the other algorithms used in this research. Machine learning based recommendation system for COVID-19 spread can help the public and government to take necessary precautions for suppressing its effect. The proposed modeling approach provides a reliable tool to predict COVID-19 transmission with an accuracy of 99.4%. © 2022 IEEE.

Predicting COVID-19 county-level case number trend by combining demographic characteristics and social distancing policies.

Objective: Predicting daily trends in the Coronavirus Disease 2019 (COVID-19) case number is important to support individual decisions in taking preventative measures. This study aims to use COVID-19 case number history, demographic characteristics, and social distancing policies both independently/interdependently to predict the daily trend in the rise or fall of county-level cases. Materials and Methods: We extracted 2093 features (5 from the US COVID-19 case number history, 1824 from the demographic characteristics independently/interdependently, and 264 from the social distancing policies independently/interdependently) for 3142 US counties. Using the top selected 200 features, we built 4 machine learning models: Logistic Regression, Naïve Bayes, Multi-Layer Perceptron, and Random Forest, along with 4 Ensemble methods: Average, Product, Minimum, and Maximum, and compared their performances. Results: The Ensemble Average method had the highest area-under the receiver operator characteristic curve (AUC) of 0.692. The top ranked features were all interdependent features. Conclusion: The findings of this study suggest the predictive power of diverse features, especially when combined, in predicting county-level trends of COVID-19 cases and can be helpful to individuals in making their daily decisions. Our results may guide future studies to consider more features interdependently from conventionally distinct data sources in county-level predictive models. Our code is available at: https://doi.org/10.5281/zenodo.6332944.

Comorbidities, sociodemographic factors, and determinants of health on COVID-19 fatalities in the United States.

Previous studies have evaluated comorbidities and sociodemographic factors individually or by type but not comprehensively. This study aims to analyze the influence of a wide variety of factors in a single study to better understand the big picture of their effects on case-fatalities. This cross-sectional study used county-level comorbidities, social determinants of health such as income and race, measures of preventive healthcare, age, education level, average household size, population density, and political voting patterns were all evaluated on a national and regional basis. Analysis was performed through Generalized Additive Models and adjusted by the COVID-19 Community Vulnerability Index (CCVI). Effect estimates of COVID-19 fatality rates for risk factors such as comorbidities, sociodemographic factors and determinant of health. Factors associated with reducing COVID-19 fatality rates were mostly sociodemographic factors such as age, education and income, and preventive health measures. Obesity, minimal leisurely activity, binge drinking, and higher rates of individuals taking high blood pressure medication were associated with increased case fatality rate in a county. Political leaning influenced case case-fatality rates. Regional trends showed contrasting effects where larger household size was protective in the Midwest, yet harmful in Northeast. Notably, higher rates of respiratory comorbidities such as asthma and chronic obstructive pulmonary disease (COPD) diagnosis were associated with reduced case-fatality rates in the Northeast. Increased rates of chronic kidney disease (CKD) within counties were often the strongest predictor of increased case-fatality rates for several regions. Our findings highlight the importance of considering the full context when evaluating contributing factors to case-fatality rates. The spectrum of factors identified in this study must be analyzed in the context of one another and not in isolation.

Coronavirus statistics, google mobility trend, and Census demographics in US counties

The Coronavirus had been viral and heavily influenced daily lives for more than two years. Two popular procedures to combat the pandemic were general vaccine implantation and limited social contact. We evaluated the effectiveness of these procedures by exploring the trends of vaccination rate, public mobility, and case/death rate at the county level in the mainland United States overtime during 2021. In addition, we investigated how demographic variables such as education and income hindered or promoted vaccination and social contact at the county level. We tested the associations between vaccination rate, public mobility, demographics, and case/death rate using Spearman Correlation tests, Student t-tests, and linear regression. Our findings provide domestic statistical feedback to the anti-epidemic policies and help public health officials make informed decisions in the future. © 2022 ACM.

Examining post-pandemic behaviors influencing human mobility trends

COVID-19 unleashed a global pandemic that has resulted in human, economic, and social crises of unprecedented scale. While the efficacy of mobility restrictions in curbing contagion has been scientifically and empirically acknowledged, a deeper understanding of the human behavioral trends driving the mixed adoption of mobility restrictions will aid future policymaking. In this paper, we employ associative rule-mining and regression to pinpoint socioeconomic and demographic factors influencing the evolving mobility trends. We compare and contrast short-distance and long-distance trips by analyzing Chicago county-level and US state-level mobility. Our study yields rules that explain the changing propensity in trip length and the collective effect of population density, economic standing, COVID testing, and the number of infected cases on mobility decisions. Through regression and correlation analysis, we show the influence of ethnic and demographic factors and perception of infection on short and long-distance trips. We find that the new mobility rules correspond to reduced long-And short-distance trip frequencies. We graphically demonstrate a marked decline in the proportion of long county-level trips but a minor change in the distribution of state-level trips. Our correlation study highlights it is hard to characterize the effect of perception of infection spread on mobility decisions. We conclude the paper with a discussion on the overlap between the analysis in the existing literature on both during-And post-lockdown mobility trends and our findings. © 2022 ACM.

An Analysis of Changes in County-Level Travel Behavior Considering COVID-19-Related Travel Restrictions, Immunization Patterns, and Political Leanings

The COVID-19 pandemic affected the world in extraordinary ways. Various measures were taken by state and local governments, including the introduction of stay-at-home orders and closures of nonessential businesses, as well as recommendations related to social distancing and the wearing of face coverings. The rollout of testing and vaccination programs were also key actions aimed at limiting the spread of COVID-19. Collectively, these restrictions have resulted in marked changes in travel behavior and patterns throughout the world. In this study, mixed-effects linear regression models are estimated to assess the impacts of COVID-19-related travel restrictions and vaccination rates on daily travel across the United States from January 1, 2020, through August 15, 2021. The results show that daily trips per person were reduced by 15% and 31% in March and April of 2020, respectively, prior to considering the impacts of any government-imposed restrictions. This suggests that government and media coverage of the pandemic played an important role in reducing travel levels. When accounting for the introduction of interventions, ranging from travel advisories to mandatory stay-at-home orders, travel was reduced by an additional 2%-9%. Interestingly, the reductions were less pronounced in areas that strongly supported the Republican candidate in the 2020 presidential election, raising important concerns as to the role of politics and trust in government. Along these same lines, as the duration of mandatory stay-at-home orders increased, travel tended to revert toward prepandemic levels, which may be attributed to quarantine fatigue. Travel levels were also higher among areas with lower median income, as well as those counties that exhibited greater variability in income, illustrating the inequitable impacts of the pandemic on these areas, which tend to include larger proportions of workers in essential industries. The results also show that trip-making increased with vaccination rates, particularly during the early stages of large-scale vaccination programs. Collectively, these insights are important in informing future strategies to mitigate the adverse impacts associated with future outbreaks of new COVID-19 strains and variants. © 2022 American Society of Civil Engineers.

US County-level Risk Factors Associated with COVID-19 Exacerbation During the Vaccination Era

The COVID-19 pandemic has caused unprecedented challenges to public health and disruption to everyday life. The news in 2020 was dominated by the worldwide spread of COVID-19, overwhelming healthcare providers and drastically changing people's lives. In 2021, the release of vaccines from multiple pharmaceutical companies changed the focus to ending the pandemic through mass inoculation. Nevertheless, the vaccine acceptance rate differs significantly across US counties, ranging from 99% to 0.1%. Our study investigates the principal risk factors in predicting COVID-19 infection and mortality rates at the county level during the early vaccination era. We are particularly interested in the role of vaccination in curbing the exacerbation of COVID-19. To this end, we first compare the efficacy of six established machine learning algorithms to predict county-level infection and mortality rates. Next, we perform risk factor analysis by identifying common principal predictors revealed by the models. Our experimental results suggest that vaccination plays an essential role in limiting COVID-19 infection and mortality. Furthermore, socioeconomic factors (e.g., severe housing problems and median household income) are more predictive of county-level mortality rate than intuitive features such as availability of healthcare resources (e.g., total numbers of hospitals/ICU beds/MDs). Our findings could provide additional insights to assist in COVID-19 resource allocation and priority setting. © 2022 IEEE.

An In-Depth Analysis Investigating the Ethnical Disparities of COVID-19 Cases and Death Rates in Georgia and Mississippi based on Socioeconomic Factors

This research aims to quantify racial disparities associated with COVID-19 cases and deaths in Georgia and Mississippi. It investigates ethnic disparities at the county level, based on socioeconomic factors. The factors used include the county population, median income, percentage of the county population per ethnic group, and county presidential election party major. In addition, COVID-19 cases and death rates by ethnicity were provided. The combined data was used for K-means clustering analysis and Analysis of Variances, to investigate the differences due to ethnicity per county and the differences due to aggregated cases and death rates per county. The results showed a significant difference in the ethnic group's COVID-19 cases and deaths as well as the socioeconomic factors that might have affected these rates. Specifically, counties with the Republican party as the presidential political party majority had significantly more cases and deaths for American Indian and Alaskan Native (AIAN), Black, and White ethnic groups in Mississippi and Georgia. There was no significant statistical difference between the Asian and Latinx groups. This research concluded that there is a significant difference in the COVID-19 deaths and cases based on the ethnic groups due to socioeconomic factors and the political party majority of the counties. In addition, counties with significant cases and death rates consist of large proportions of people of color than their population representation percentage based on the 2020 Census. © 2022 IISE Annual Conference and Expo 2022. All rights reserved.

Quantifying the Impact of Air Travel on Growth of COVID-19 Pandemic in the United States

This paper develops models to quantify the dynamics of the impact of air travel on the spread of the COVID-19 pandemic, using a wide range of datasets covering the period from March to December 2020. With the help of flight operation data, we first develop a novel approach to estimate the county-level daily air passenger traffic, which combines passenger load factor estimates and information about the air traffic distribution. Cross-sectional models using aggregated county-level variables are estimated. While this study focuses on air travel variables, we also control for potential spatial autocorrelation and other relevant covariates, including vehicle miles traveled (VMT), road network connectivity, demographic characteristics, and climate. The model results indicate that air travel has a strong and positive impact on the initial pandemic growth rate for both case-based and fatality-based aggregate models. © ATM 2021. All rights reserved.

An Analysis of the Impacts of Social Media on COVID-19 Vaccine Hesitancy in the U.S

The COVID-19 pandemic remains a public health emergency, although effective vaccines are available. Unfortunately, vaccine hesitancy (VH) has mitigated the effectiveness of vaccination campaigns. Preferred approaches to estimate VH are surveys, polls, and questionnaires;however, these methods are limited in scope because they capture VH at a single point in time and mostly in highly populated (urban) areas. A dearth of published research points to a knowledge gap that limits the ability to explain changes in VH over time at the county level. The proposed research uses open-access databases and data-driven approaches to fill this knowledge gap. We present a systematic, data-driven framework that analyzes sentiments on social media and news coverage on COVID-19 to estimate VH changes over time at the county level in the U.S. Understanding how sentiments and topics contribute to VH can help public health officials develop tailored strategies to increase vaccine uptake for different counties of the U.S., especially when data is limited. Our study indicates that the use of social media and online news coverage led to more reliable predictions of VH estimates for urban than rural areas. Future research will consider other factors which impact VH, such as social/demographic factors, weather conditions, vaccine mandates, etc. © 2022 IISE Annual Conference and Expo 2022. All rights reserved.

Using Neural Networks to Model the Spread of COVID-19

The spread of the novel coronavirus across the world in 2020 exposed the tenuous nature of hospital capacity and medical resource supply lines. Being able to anticipate surge events days before they hit an area would allow healthcare workers to pivot and prepare, critically expanding capacity and adjusting to resource loads. This work aims to enable advanced healthcare planning by providing adaptive forecasts into short range COVID-19 outbreaks and surge events. Here, we present a novel method to predict the spread of COVID-19 by using creative neural network architectures, especially convolutional and LSTM layers. Our goal was to create a generalizable method or model to predict disease spread on a county-level granularity. Importantly, we found that by using an adaptive neural network model with a frequent refresh rate, we were able to outperform simple feed forward estimation methods to predict county level new case counts on a daily basis. We also show the capabilities of neural network architectures by comparing performance on different sizes of training data and geographic inputs. Our results indicate that neural networks are well suited to dynamically modeling the spread of COVID-19 on a county-level basis, but that cultural and/or geographic differences in regions prevent the portability of fully-trained models. © 2022 IEEE.

Time Series Analysis of COVID-19 Cases in Humboldt County

The time series of COVID-19 daily cases in the U.S is analyzed by utilizing the county-level temporal data, from January 22, 2020 to October 18, 2021. Autocorrelation and partial autocorrelation show that time series of daily cases in Humboldt county has a 7-day seasonal pattern. Visualization and augmented Dickey-Fuller test show that time series of daily cases in Humboldt county is non-stationary. The seven-order difference reveals that the time series is stationary. There is a moderate positive correlation between daily cases and fully vaccination rate. Clustering analysis describes 33 counties have similar daily case pattern with Humboldt County by standard deviation of 0.003. This analysis can be used for future time-series forecasting and planning. © 2021 IEEE.

Exploring the Relationship between the Spatial Distribution of Different Age Populations and Points of Interest (POI) in China

Population spatialization data is crucial to conducting scientific studies of coupled human–environment systems. Although significant progress has been made in population spatialization, the spatialization of different age populations is still weak. POI data with rich information have great potential to simulate the spatial distribution of different age populations, but the relationship between spatial distributions of POI and different age populations is still unclear, and whether it can be used as an auxiliary variable for the different age population spatialization remains to be explored. Therefore, this study collected and sorted out the number of different age populations and POIs in 2846 county-level administrative units of the Chinese mainland in 2010, divided the research data by region and city size, and explored the relationship between the different age populations and POIs. We found that there is a complex relationship between POI and different age populations. Firstly, there are positive, moderate-to-strong linear correlations between POI and population indicators. Secondly, POI has a different explanatory power for different age populations, and it has a higher explanatory power for the young and middle-aged population than the child and old population. Thirdly, the explanatory power of POI to different age populations is positively correlated with the urban economic development level. Finally, a small number of a certain kinds of POIs can be used to effectively simulate the spatial distributions of different age populations, which can improve the efficiency of obtaining spatialization data of different age populations and greatly save on costs. The study can provide data support for the precise spatialization of different age populations and inspire the spatialization of the other population attributes by POI in the future.

Association between county-level risk groups and COVID-19 outcomes in the United States: a socioecological study.

BACKGROUND: Geographic heterogeneity in COVID-19 outcomes in the United States is well-documented and has been linked with factors at the county level, including sociodemographic and health factors. Whether an integrated measure of place-based risk can classify counties at high risk for COVID-19 outcomes is not known. METHODS: We conducted an ecological nationwide analysis of 2,701 US counties from 1/21/20 to 2/17/21. County-level characteristics across multiple domains, including demographic, socioeconomic, healthcare access, physical environment, and health factor prevalence were harmonized and linked from a variety of sources. We performed latent class analysis to identify distinct groups of counties based on multiple sociodemographic, health, and environmental domains and examined the association with COVID-19 cases and deaths per 100,000 population. RESULTS: Analysis of 25.9 million COVID-19 cases and 481,238 COVID-19 deaths revealed large between-county differences with widespread geographic dispersion, with the gap in cumulative cases and death rates between counties in the 90th and 10th percentile of 6,581 and 291 per 100,000, respectively. Counties from rural areas tended to cluster together compared with urban areas and were further stratified by social determinants of health factors that reflected high and low social vulnerability. Highest rates of cumulative COVID-19 cases (9,557 [2,520]) and deaths (210 [97]) per 100,000 occurred in the cluster comprised of rural disadvantaged counties. CONCLUSIONS: County-level COVID-19 cases and deaths had substantial disparities with heterogeneous geographic spread across the US. The approach to county-level risk characterization used in this study has the potential to provide novel insights into communicable disease patterns and disparities at the local level.

Time Warping Clustering for the Forecast and Analysis of COVID-19

This paper presents an effective algorithm for the clustering of confirmed COVID-19 cases at the county-level in the United States. Dynamic time warping and Euclidean distance are examined as the k-means clustering distance metrics. Dynamic time warping can compare time series varying in speed, as counties often experience similar outbreak trends without the timelines matching up exactly. The effect of data preprocessing on clustering was systematically studied. Further analyses demonstrate the immediate value of our clusters for both retrospective interpretation of the pandemic and as informative inputs for case prediction models. We visualize the time progression of COVID-19 from April 5, 2020 to August 23, 2020. We proposed a Monte-Carlo dropout feedforward neural network with the ability to forecast four weeks into the future. Predictions evaluated from July 24, 2020 to August 20, 2020 demonstrate the better empirical performance of the model when trained on the clusters, in comparison with the model trained on individual counties and the model trained on counties clustered by state. © 2020 IEEE.

Deep diffusion-based forecasting of COVID-19 by incorporating network-level mobility information

Modeling the spatiotemporal nature of the spread of infectious diseases can provide useful intuition in understanding the time-varying aspect of the disease spread and the underlying complex spatial dependency observed in people's mobility patterns. Besides, the county level multiple related time series information can be leveraged to make a forecast on an individual time series. Adding to this challenge is the fact that real-time data often deviates from the unimodal Gaussian distribution assumption and may show some complex mixed patterns. Motivated by this, we develop a deep learning-based time-series model for probabilistic forecasting called Auto-regressive Mixed Density Dynamic Diffusion Network (ARM3Dnet), which considers both people's mobility and disease spread as a diffusion process on a dynamic directed graph. The Gaussian Mixture Model layer is implemented to consider the multimodal nature of the realtime data while learning from multiple related time series. We show that our model, when trained with the best combination of dynamic covariate features and mixture components, can outperform both traditional statistical and deep learning models in forecasting the number of Covid-19 deaths and cases at the county level in the United States. © 2021 ACM.

The Impacts of Globalization on Inequality in the Post-COVID-19 World: Evidence From China.

Empirical studies suggest that globalization (FDI and international trade) has been greatly affected by the COVID-19 and related anti-pandemic measures imposed by governments worldwide. This paper investigates the impact of globalization on intra-provincial income inequality in China and the data is based on the county level. The findings reveal that FDI is negatively associated with intra-provincial inequality, intra-provincial inequality increases as the primary industry sector (agriculture) declines. The result also finds that the increase in inequality stems not from the development in the tertiary or secondary industry sectors per se, but the unevenness in the distribution of these sectors.