Statistical Areas Research Articles

Detecting Elevated Air Pollution Levels by Monitoring Web Search Queries: Algorithm Development and Validation

Real-time air pollution monitoring is a valuable tool for public health and environmental surveillance. In recent years, there has been a dramatic increase in air pollution forecasting and monitoring research using artificial neural networks. Most prior work relied on modeling pollutant concentrations collected from ground-based monitors and meteorological data for long-term forecasting of outdoor ozone (O3), oxides of nitrogen, and fine particulate matter (PM2.5). Given that traditional, highly sophisticated air quality monitors are expensive and not universally available, these models cannot adequately serve those not living near pollutant monitoring sites. Furthermore, because prior models were built based on physical measurement data collected from sensors, they may not be suitable for predicting the public health effects of pollution exposure. This study aimed to develop and validate models to nowcast the observed pollution levels using web search data, which are publicly available in near real time from major search engines. We developed novel machine learning-based models using both traditional supervised classification methods and state-of-the-art deep learning methods to detect elevated air pollution levels at the US city level by using generally available meteorological data and aggregate web-based search volume data derived from Google Trends. We validated the performance of these methods by predicting 3 critical air pollutants (O3, nitrogen dioxide, and PM2.5) across 10 major US metropolitan statistical areas in 2017 and 2018. We also explore different variations of the long short-term memory model and propose a novel search term dictionary learner-long short-term memory model to learn sequential patterns across multiple search terms for prediction. The top-performing model was a deep neural sequence model long short-term memory, using meteorological and web search data, and reached an accuracy of 0.82 (F1-score 0.51) for O3, 0.74 (F1-score 0.41) for nitrogen dioxide, and 0.85 (F1-score 0.27) for PM2.5, when used for detecting elevated pollution levels. Compared with using only meteorological data, the proposed method achieved superior accuracy by incorporating web search data. The results show that incorporating web search data with meteorological data improves the nowcasting performance for all 3 pollutants and suggest promising novel applications for tracking global physical phenomena using web search data.

Changes in subjective well-being predict changes in U.S. presidential, Senate, & House of Representatives election outcomes

Free and fair elections enable the nation’s citizens to elect candidates whom they believe best represent their interests. When deciding who to vote for, individuals may consider a host of factors that ultimately improve their subjective well-being. Using data from the Gallup Sharecare Well-being Index (N = 3,208,924), we examined whether changes in subjective well-being predicted U.S. presidential, Senate, and House of Representatives election outcomes from 2010 to 2020. We tested this effect at county (n = 1,652–3,061), metropolitan statistical area (n = 191–363), state (n = 50), and district (n = 389–427) levels. Pre-registered multilevel models supported the notion that regions with growing discontent tended to have larger increases in non-incumbent vote shares. Establishing a link between subjective well-being and electoral outcomes has the potential to realign policymakers’ priorities with what truly matters to their constituents, thereby facilitating the promotion of population well-being.

Exploring young adults' e-cigarette use behavior during COVID-19.

Changes in daily life related to COVID-19 have impacted e-cigarette use, particularly in young adults. This cross-sectional mixed-methods study explored young adults' perceptions regarding how COVID-19 influenced their e-cigarette use. We analyzed Fall 2020 survey data from 726 past 6-month e-cigarette users (mean age=24.15 years, 51.1% female, 35.5% sexual minority, 4.4% Black, 10.2% Asian, 12.1% Hispanic) and Spring 2021 semi-structured interview data among a subset of 40 participants (mean age=26.30 years, 35.0% female, 45.0% sexual minority, 5.0% Black, 22.5% Asian, 12.5% Hispanic). Participants were drawn from 6 metropolitan statistical areas with varied tobacco and cannabis legislative contexts. Among survey participants, 44.4% also smoked cigarettes, 54.0% other tobacco products, and 60.1% used cannabis. They reported various changes in their daily lives, including changes in the nature and/or status of employment (e.g. 15.3% were laid off, 72.8% experienced household income loss). Regarding changes in e-cigarette use since COVID-19, 22.6% tried to cut down and 16.0% tried to quit. Interview participants commonly indicated that they increased their use due to stress, boredom, changes in accessibility, and/or changes to daily environment that made e-cigarette use more feasible. Results highlight the importance of promoting opportunities for young adults to build relationships to decrease stress, foster a sense of belonging, and increase quality of life (e.g. increasing the accessibility to mental health and social support services, intentionally engaging young adults in pandemic-appropriate community-building and extracurricular activities). This research may help to inform future e-cigarette cessation interventions that consider the unique challenges of societal stressors, such as pandemics.

2059. Estimated Background HIV Diagnosis Rates among People Who Could Benefit from Pre-Exposure Prophylaxis (PrEP) at the Metropolitan Statistical Area (MSA) level in the United States, 2012-2019

Abstract Background Identifying areas of high background HIV diagnosis rates (bHIV) is critical for future PrEP trials using counterfactual study designs to estimate efficacy and to facilitate effective implementation of HIV prevention interventions. In this study, we used a multiple data source model to estimate bHIV between 2012 and 2019 in people who could benefit from PrEP (PWBP) but are not taking PrEP to identify with greater precision areas of high HIV transmission at the Metropolitan Statistical Area (MSA) level. We compared bHIV in PWBP in PURPOSE2 (NCT04925752) trial sites to those in the DISCOVER trial (NCT02842086) to demonstrate how this approach can inform site selection in future PrEP clinical trials and support delivery of PrEP interventions. Methods Number of new HIV diagnoses, number of individuals with a PrEP indication, and person-time on PrEP or after HIV diagnosis (obtained from published surveillance data) and pharmacy claims data (IQVIA PharMetrics Plus) were used to calculate estimated bHIV in a multivariate Poisson regression model for PWBP in 108 MSAs and across states between 2012 and 2019. Results Overall, bHIV among PWBP decreased from 4.42 per 100 person-years (PY) (95% CI 4.38 – 4.47) in 2012 to 3.14 per 100 PY (95% CI 3.10 – 3.17) in 2019. However, bHIV among PWBP in MSAs with PURPOSE2 trial sites were consistently higher (from 4.69 per 100 PY [95% CI 4.56 – 4.81] in 2012 to 3.58 per 100 PY [95% CI 3.49 – 3.67] in 2019) compared to areas with DISCOVER-only sites (from 3.77 per 100 PY [95% CI 3.53 – 4.01] in 2012 to 2.83 per 100 PY [95% CI 2.63 – 3.03] in 2019; Figure 1A). Within states, we identified specific localities with high bHIV among PWBP, such as in Illinois where state level HIV diagnosis rate was 2.65 per 100 PY [95% CI 2.50 – 2.80] but as high as 6.04 per 100 PY [95% CI 5.67 - 6.40] in the MSA with a PURPOSE2 site (Figure 1B). Figure:Estimated background HIV diagnosis rates (bHIV) in people who could benefit from PrEP by MSAs selected by DISCOVER/PURPOSE2 trials (A) from 2012 to 2019 and (B) in 2019 in the US. Conclusion We found considerable variation in bHIV at the MSA level among PWBP who are not taking PrEP, including a concentration of MSAs in the Southeastern US with high rates of new HIV diagnosis. Our methodology for leveraging existing surveillance data to identify areas of high bHIV with greater precision may provide a valuable approach to support future counterfactual clinical trial designs and prioritize resources for delivery of PrEP services. Disclosures Melanie de Boer, PhD, Gilead Sciences: Employee|Gilead Sciences: Stocks/Bonds Jason Yuan, n/a, Gilead Sciences: Employee|Gilead Sciences: Stocks/Bonds Juan Yang, PhD, Gilead Sciences: Employee|Gilead Sciences: Stocks/Bonds Lillian Brown, MD, PhD, Gilead Sciences: Stocks/Bonds Carlo Hojilla, PhD, Gilead Sciences: Employee|Gilead Sciences: Stocks/Bonds Christoph C. Carter, MD, PhD, Gilead Sciences: Employee|Gilead Sciences: Employee|Gilead Sciences: Stocks/Bonds|Gilead Sciences: Stocks/Bonds Moupali Das, MD, Gilead Sciences: Employee|Gilead Sciences: Employee|Gilead Sciences: Stocks/Bonds|Gilead Sciences: Stocks/Bonds Li Tao, MD, PhD, Gilead Sciences: Employee|Gilead Sciences: Employee|Gilead Sciences: Stocks/Bonds|Gilead Sciences: Stocks/Bonds.

568. Using Machine Learning to Predict Place-Based Risks for Staphylococcus aureus Infections in Children

Abstract Background Community-onset Staphylococcus aureus (CO-S. aureus) pediatric infections, methicillin-resistant S. aureus (MRSA) and methicillin-susceptible S. aureus (MSSA) continue to contribute to the burden of infections seen in the ambulatory setting in the US. Individual risk factors have been identified, but place-based factors and specific geographic locality have not been well-studied. The purpose of this study is to predict place-based factors that contribute to the spread of CO-S. aureus in a major urban area using maximum entropy (MaxEnt), a machine learning technique. Methods Electronic medical records from two pediatric hospitals (2002 to 2016) were retrospectively reviewed. Inclusion criteria: a confirmed S. aureus infection within 48 hours of hospital admission (CO-S. aureus), < 19 years old, and a geo-referenced address within Atlanta’s metropolitan statistical area (MSA). Fourteen place-based factors, at the US Census block group level, were included in the MaxEnt models: < 18 years old, Caucasian, African American, ethnicity, poverty, education attainment, crowding, daycare, kindergarten enrollment, distance to K-12 school, distance to a children’s hospital, distance to a daycare center, and population density. A total of four models (CO-MRSA early, CO-MSSA early, CO-MRSA later, and CO-MSSA later) were run using the MaxEnt software. For each model, 75% and 25% of data was randomly assigned to training and testing groups, respectively. Models were assessed by jack-knife tests. Results 16,124 records met eligibility criteria for MaxEnt models. The training Area Under the Curve (AUC) ranged from 0.771 to 0.837 and the test AUC ranged from 0.769 to 0.804. Population density had the highest contribution in predicting CO-MRSA and CO-MSSA locations, which was confirmed by jack-knife tests. Conclusion By applying MaxEnt to pediatric CO-S. aureus infections in the Atlanta MSA, it was found that higher risks of CO-S. aureus infections may exist in more densely populated areas. MaxEnt can be utilized to identify potential future areas of CO-MRSA and CO-MSSA infections based on estimated or predicted changes to the place-based factors used to build these models. Disclosures Lilly Immergluck, MD, MS, GSK: Clinical Trial- PI|Merck: Vaccine Trial Site- serve as PI|Moderna: Board Member|Novavax: Part of CoVID-19 Phase 3 Trial through US Covid Prevention Network.

Preface: 2022 International Conference on Management Engineering and Economic Analysis (MEEA 2022)

The 2022 International Conference on Management Engineering and Economic Analysis (MEEA 2022) was held in London, UK during September 24-25, 2022. This conference is to bring together innovative academics and experts in the fields of management engineering, business statistics, marketing strategies, financial economics and economic analysis to a common forum. The primary goal of the conference is to promote research and developmental activities and another goal is to promote information interchange between researchers, developers, engineers, students, and practitioners working all around the world.
 The conference will be held every year to make it an ideal platform for people to share views and experiences in related areas. The conference model of MEEA 2022 was divided into two sessions, including oral presentations and keynote speeches. In the first part, some scholars, whose submissions were selected as the excellent papers, were given 10-15 minutes to perform their oral presentations one by one. Then in the second part, keynote speakers were each allocated 35-45 minutes to hold their speeches. More than 280 participants attended the meeting. There were over 16 experts and scholars in the area of management engineering, business statistics, and economic analysis representing different famous universities and institutes around the globe.
 We are glad to share with you that we received lots of submissions from the conference and we selected a bunch of high-quality papers and compiled them into the proceedings after rigorous reviewing. All the papers have been through rigorous review and process to meet the requirements of international publication standard.
 We are really grateful to the keynote speakers, session chairs, organizing committee members, student volunteers and publication house. Also, we are thankful to all the authors for contributing a large number of papers in the conference. It was the quality of their presentations and their passion to communicate with the other participants that really make this conference series a great success.
 MEEA 2022 Organizing Committees
 London, UK

Do ridesharing services cause traffic congestion?

AbstractThis paper uses a difference‐in‐differences strategy and an event‐study analysis to evaluate the effect of Uber entries on vehicle miles travelled in 346 US Metropolitan statistical areas between 2011 and 2017. Empirical results demonstrate that Uber entries have no significant effect on the total vehicle miles travelled in a city. However, Uber is found to decrease the vehicle miles travelled on highways but increase them on collector roads. All these effects come mainly from large metropolitan statistical areas. A simulation exercise suggests that Uber entries cause longer travel times if the share of travel time spent on collector roads is high.

Notes on Generalized Perron Complement of Nonnegative Matrices

In the areas of statistics, matrix analysis, and several applications of applied mathematics, the Perron complement is fundamental. For a matrix with nonnegative elements, some inequalities and equalities involving generalized Perron complement and Perron complement of the matrices are derived.

Clinical biostatistics in the 2020s

Clinical biostatistics in the 2020s

MT24 Oxidized Zirconium Can Positively Impact Outcomes in Revision Arthroplasty: A 90-Day Episode of Care Analysis in a Large US Metropolitan Statistical Area (MSA) Following Revision Knee (RKA) and Hip Arthroplasty (RHA)

MT24 Oxidized Zirconium Can Positively Impact Outcomes in Revision Arthroplasty: A 90-Day Episode of Care Analysis in a Large US Metropolitan Statistical Area (MSA) Following Revision Knee (RKA) and Hip Arthroplasty (RHA)

MT42 Oxidized Zirconium Use in Primary Joint Arthroplasty: A 90-Day Episode of Care Analysis in a Large US Metropolitan Statistical Area (MSA) for Elective Total Knee Arthroplasty (TKA) and Total Hip Arthroplasty (THA)

Centers for Medicare and Medicaid Services has made hospitals accountable for better clinical and economical outcomes of total joint arthroplasty by introducing the Comprehensive Care for Joint Replacement Model. The objective of this study was to evaluate 90-day outcomes and costs of using Oxidized Zirconium (OxZr)1 knee and hip implants compared to non-OxZr from the New York-Newark-Jersey MSA.

Industrial Structure and a Tradeoff Between Productivity and Economic Resilience

Abstract The structures of regional economies play a critical role in determining both a region’s productivity and its resilience to shocks. We extend previous work on the regional occupation and skills structure by analyzing the effect of a region’s industry structure. We operationalize the concept of economic structure by constructing a network of interdependent economic components, employing ecological techniques of co-occurrence analysis to infer interactions between industries. For each U.S. metropolitan statistical area, we create an aggregate measure of economic tightness that captures the degree of interconnectedness among a region’s industries. We find that industry tightness, which we find is partly driven by rare industry pairs, is positively correlated with a region’s economic productivity, negatively correlated with a region’s change in productivity following the Great Recession. This study contributes to an understanding of the tradeoff between productivity and resilience, which is intended to help policy makers that face similar real-world tradeoffs.

Socio-economic and ethnocultural influences on geographical disparities in breast cancer screening participation in Victoria, Australia.

To examine the socio-economic and ethnocultural characteristics of geographical areas that may influence variation in breast cancer screening participation. In a cross-sectional analysis breast cancer screening participation for statistical areas in Victoria, Australia (2015-2017) was linked with data from the 2016 Australian Census. We selected four commonly used area-level measures of socio-economic status from the Australian Census (i) income (ii) educational level (iii) occupational status and (iv) employment profile. To assess the ethnocultural characteristics of statistical areas we used the Census measures (i) country of birth (ii) language spoken at home (iii) fluency in English (iv) religion and (v) the proportion of immigrants in an area, together with their recency of migration. All the selected measures were related to screening participation. There was a high degree of association both within and between socio-economic and ethnocultural characteristics of areas as they relate to screening. Ethnocultural characteristics alone accounted for most of the explained geographical disparity in screening participation. Geographical disparities in breast cancer screening participation may be due to ethnocultural factors that are confounded with socio-economic factors.

Inequalities in Access to Mental Health Treatment by Australian Youths During the COVID-19 Pandemic.

The authors aimed to evaluate changes in use of government-subsidized primary mental health services, through the Medicare Benefits Schedule (MBS), by young people during the first year of the COVID-19 pandemic in Australia and whether changes were associated with age, sex, socioeconomic status, and residence in particular geographical areas. Interrupted time-series analyses were conducted by using quarterly mental health MBS service data (all young people ages 12-25 years, 2015-2020) for individual Statistical Area Level 3 areas across Australia. The data captured >22.4 million service records. Meta-analysis and meta-regression models estimated the pandemic interruption effect at the national level and delineated factors influencing these estimates. Compared with expected prepandemic trends, a 6.2% (95% CI=5.3%-7.2%) increase was noted for all young people in use of MBS mental health services in 2020. Substantial differences were found between age and sex subgroups, with a higher increase among females and young people ages 18-25. A decreasing trend was observed for males ages 18-25 (3.5% reduction, 95% CI=2.5%-4.5%). The interruption effect was strongly associated with socioeconomic status. Service uptake increased in areas of high socioeconomic status, with smaller or limited uptake in areas of low socioeconomic status. During 2020, young people's use of primary mental health services increased overall. However, increases were inequitably distributed and relatively low, compared with increases in population-level mental health burden. Policy makers should address barriers to primary care access for young people, particularly for young males and those from socioeconomically disadvantaged backgrounds.

Simulating and Forecasting the COVID-19 Spread in a U.S. Metropolitan Region with a Spatial SEIR Model

The global COVID-19 pandemic has taken a heavy toll on health, social, and economic costs since the end of 2019. Predicting the spread of a pandemic is essential to developing effective intervention policies. Since the beginning of this pandemic, many models have been developed to predict its pathways. However, the majority of these models assume homogeneous dynamics over the geographic space, while the pandemic exhibits substantial spatial heterogeneity. In addition, spatial interaction among territorial entities and variations in their magnitude impact the pandemic dynamics. In this study, we used a spatial extension of the SEIR-type epidemiological model to simulate and predict the 4-week number of COVID-19 cases in the Charlotte-Concord-Gastonia Metropolitan Statistical Area (MSA), USA. We incorporated a variety of covariates, including mobility, pharmaceutical, and non-pharmaceutical interventions, demographics, and weather data to improve the model's predictive performance. We predicted the number of COVID-19 cases for up to four weeks in the 10 counties of the studied MSA simultaneously over the time period 29 March 2020 to 13 March 2021, and compared the results with the reported number of cases using the root-mean-squared error (RMSE) metric. Our results highlight the importance of spatial heterogeneity and spatial interactions among locations in COVID-19 pandemic modeling.

Electricity consumption variation versus economic structure during COVID-19 on metropolitan statistical areas in the US

The outbreak of novel coronavirus disease (COVID-19) has resulted in changes in productivity and daily life patterns, and as a result electricity consumption (EC) has also shifted. In this paper, we construct estimates of EC changes at the metropolitan level across the continental U.S., including total EC and residential EC during the initial two months of the pandemic. The total and residential data on the state level were broken down into the county level, and then metropolitan level EC estimates were aggregated from the counties included in each metropolitan statistical area (MSA). This work shows that the reduction in total EC is related to the shares of certain industries in an MSA, whereas regardless of the incidence level or economic structure, the residential sector shows a trend of increasing EC across the continental U.S. Since the MSAs account for 86% of the total population and 87% of the total EC of the continental U.S., the analytical result in this paper can provide important guidelines for future social-economic crises.

DOES MACHINE LEARNING PREDICTION DAMPEN THE INFORMATION ASYMMETRY FOR NON-LOCAL INVESTORS?

In this study, we examine the prediction accuracy of machine learning methods to estimate commercial real estate transaction prices. Using machine learning methods, including Random Forest (RF), Gradient Boosting Machine (GBM), Support Vector Machine (SVM), and Deep Neural Networks (DNN), we estimate the commercial real estate transaction price by comparing relative prediction accuracy. Data consist of 19,640 transaction-based office properties provided by Costar corresponding to the 2004–2017 period for 10 major U.S. CMSA (Consolidated Metropolitan Statistical Area). We conduct each machine learning method and compare the performance to identify a critical determinant model for each office market. Furthermore, we depict a partial dependence plot (PD) to verify the impact of research variables on predicted commercial office property value. In general, we expect that results from machine learning will provide a set of critical determinants to commercial office price with more predictive power overcoming the limitation of the traditional valuation model. The result for 10 CMSA will provide critical implications for the out-of-state investors to understand regional commercial real estate market.

The impact of the COVID-19 pandemic on racial disparities in patients undergoing total shoulder arthroplasty in the United States

The purpose of this study was to assess racial disparities in total shoulder arthroplasty (TSA) in the United States and to determine whether these disparities were affected by the COVID-19 pandemic. Centers for Medicare and Medicaid Services (CMS) 100% sample was used to examine primary TSA volume from April to December from 2019 to 2020. Utilization was assessed for White, Black, Hispanic, and Asian populations to determine if COVID-19 affected these groups differently. A regression model adjusted for age, sex, CMS-hierarchical condition categories (HCC) score, dual enrollment (proxy for socioeconomic status), time-fixed effects, and core-based statistical area fixed effects was used to study difference across groups. In 2019, the TSA volume per 1000 beneficiaries was 1.51 for White and 0.57 for non-White, with a 2.6-fold difference. In 2020, the rate of TSA in White patients (1.30/1000) was 2.9 times higher than non-White (0.45/1000) during the COVID-19 pandemic (P<.01). There was an overall 14% decrease in TSA volume per 1000 Medicare beneficiaries in 2020; non-White patients had a larger percentage decrease in TSA volume than White (21% vs. 14%, estimated difference; 8.7%, P=.02). Black patients experienced the most pronounced disparity with estimated difference of 10.1%, P=.05, compared with White patients. Similar disparities were observed when categorizing procedures into anatomic and reverse TSA, but not proximal humerus fracture. During the COVID-19 pandemic, overall TSA utilization decreased by 14% with White patients experiencing a decrease of 14%, and non-White patients experiencing a decrease of 21%. This trend was observed for elective TSA, while disparities were less apparent for proximal humerus fracture.

Delineating and comparing local labor market geographies of Millennials, Generation Xers, and Baby Boomers

There has been an increasing share of older workers in America’s workforce as Baby Boomers continue to age into this group. This raises questions about what their local labor markets look like in space and how they differ from Millennials and Generation Xers who are also active in today’s workforce. However, even though significant generational (or age) differences in commuting distance have been well documented, questions about generational disparities in local labor market geographies have not been investigated. Although several definitions of market areas exist, they were delineated using all commuting trips made by all workers in a region and thus may not be representative of subgroups of workers. The aim of this research is to delineate generation-specific local labor market areas and compare how they differ from one another. Using flow visualizations and three employment self-containment indices applied to a Longitudinal Employer-Household Dynamics dataset of Florida, this research finds that the core-based statistical areas, a popular definition of local labor market areas, understate the true market geographies across generational cohorts. It then delineates generation-specific market areas using a popular community detection method and compares their geographies using the self-containment indices. Results demonstrate largely consistent market geographies across the three generational cohorts in most regions in Florida. Most of these consistent market areas encompass two or more metros, indicating the emergence of functional megaregions centering on major metropolitan hubs. Substantial generational differences in market geographies do exist and concentrate in regions appealing to seniors along the Gulf of Mexico.

Trends in Prescription Opioid Use in Motor Vehicle Crash Injuries in the United States: 2014-2018.

Motor vehicle crashes (MVC) cause over three million people to be nonfatally injured each year in the United States alone. We investigated trends and patterns in prescription opioid usage among nonfatal MVC injuries in 50 states in the US and the District of Columbia from 2014 to 2018. All emergency department visits for an MVC event (N = 142,204) were identified from the IBM® MarketScan® Databases. Using log-binomial regression models, we investigated whether the prevalence of prescription opioids in MVC injuries varied temporally, spatially, or by enrollees' characteristics. Adjusting for age, relationship to the primary beneficiary, employment status, geographic region, and residence in metropolitan statistical area, the prevalence decreased by 5% (95% CI: 2-8%) in 2015, 18% (95% CI: 15-20%) in 2016, 31% (95% CI: 28-33%) in 2017, and 49% (95% CI: 46-51%) in 2018, compared to 2014. Moreover, the prevalence decreased by 28% (95% CI: 26-29%) after the publication of the CDC Guidelines for Prescribing Opioids for Chronic Pain. Spatial variations were observed in the prevalence and temporal trend of prevalence. The decreasing trend in the prevalence of prescription opioids in MVC is consistent with the decrease in the dispensing rate of opioids and the percentage of high-dosage opioids in the study population.