Neighborhood Data Research Articles

A high-performance voting-based ensemble model of graph convolutional extreme learning machines for identifying geochemical anomalies related to mineralization

How to effectively identify mineralization anomalies from geochemical survey data is one of the key problems that must be properly solved in geochemical exploration. Although supervised machine learning and deep learning techniques have been adopted to solve this problem, the performance of the established mineralization anomaly identification models is not satisfactory because the mutual relationship between neighborhood data points is not taken into account in the modeling of geochemical exploration data. The graph convolutional extreme learning machines (GCELMs) provide a new technique for modeling the graph data transformed from geochemical exploration data. To improve the performance of mineralization anomaly identification algorithms, the GCELM techniques were adopted to establish the supervised classification models for identifying mineralization anomalies from geochemical exploration data. To improve the robustness of the GCELM models in mineralization anomaly identification, an ensemble classification model called voting-based GCELM (V-GCELM) model was established by voting the outputs of multiple GCELM models. The Baishan area of Jilin Province (China) was taken as the study area for carrying out an experiment of mineralization anomaly identification from geochemical exploration data. The synthetic minority over-sampling technique (SMOTE) was used to alleviate the class-imbalance of the training data. The GCELM model, V-GCELM model and extreme learning machine (ELM) model were established for identifying mineralization anomalies, and their performances were compared in terms of the receiver operating characteristic (ROC) curves, area under the curves (AUCs) and lift indices. The results show that the V-GCELM model has the best performance, and its AUC and lift index are 0.9989 and 28.6, respectively. The GCELM model has the second-best performance, and its AUC and lift index are 0.9977 and 20.1, respectively. The ELM model has the worst but still high performance, its AUC and lift index are 0.9402 and 9.2, respectively. The mineralization anomalies identified by the three models have close spatial correlation with the polymetallic deposits found in the study area, and spatially controlled by the main controlling factors in the study area. Therefore, GCELM and V-GCELM are promising techniques for building the high-performance classification models for identifying mineralization anomalies from geochemical exploration data.

Local neighbor propagation on graphs for mismatch removal

Recently, despite the promising progress having been achieved, most mismatch removal methods only consider the connection relationships of feature matches in a single neighborhood, and ignore their relationships between different neighborhoods, which will lead to the unavailability of local topological structure for feature matching. In this paper, we propose a novel robust Local Neighbor Propagation on Graphs based mismatch removal (LNPG) method for robust feature matching. LNPG starts from a novel neighborhood graph construction strategy, which leverages both the spatial and the residual information to preserve the local neighborhood structures of potential inliers. Subsequently, LNPG incorporates local neighbor propagation into the graph to enhance connection relationships of the data in different neighborhoods, by using the path-based similarity measurement and the adaptive graph partition. In addition, LNPG includes a novel consistency-filtering-based clustering algorithm, which introduces a reliable neighborhood consistency measure function and an effective cluster merging criterion for robust clustering. Overall, LNPG not only effectively distinguishes inliers from outliers, but also reliably classifies inliers into different transformation models between pairs of images. Extensive experiments on publicly available datasets show the superiority of our LNPG in comparison with other state-of-the-art methods.

POINT: Pipeline for Offline Conversion and Integration of Geocodes and Neighborhood Data.

Geocoding, the process of converting addresses into precise geographic coordinates, allows researchers and health systems to obtain neighborhood-level estimates of social determinants of health. This information supports opportunities to personalize care and interventions for individual patients based on the environments where they live. We developed an integrated offline geocoding pipeline to streamline the process of obtaining address-based variables, which can be integrated into existing data processing pipelines. POINT is a web-based, containerized, application for geocoding addresses that can be deployed offline and made available to multiple users across an organization. Our application supports use through both a graphical user interface and application programming interface to query geographic variables, by census tract, without exposing sensitive patient data. We evaluated our application's performance using two datasets: one consisting of 1 million nationally representative addresses sampled from Open Addresses, and the other consisting of 3,096 previously geocoded patient addresses. A total of 99.4 and 99.8% of addresses in the Open Addresses and patient addresses datasets, respectively, were geocoded successfully. Census tract assignment was concordant with reference in greater than 90% of addresses for both datasets. Among successful geocodes, median (interquartile range) distances from reference coordinates were 52.5 (26.5-119.4) and 14.5 (10.9-24.6) m for the two datasets. POINT successfully geocodes more addresses and yields similar accuracy to existing solutions, including the U.S. Census Bureau's official geocoder. Addresses are considered protected health information and cannot be shared with common online geocoding services. POINT is an offline solution that enables scalability to multiple users and integrates downstream mapping to neighborhood-level variables with a pipeline that allows users to incorporate additional datasets as they become available. As health systems and researchers continue to explore and improve health equity, it is essential to quickly and accurately obtain neighborhood variables in a Health Insurance Portability and Accountability Act (HIPAA)-compliant way.

Association of Provider Zip Code Sociodemographic Characteristics with Radiation Therapy Modality Use in the Medicare Population

Variations in healthcare delivery by provider and patient-level characteristics have been well studied in the US health system as a whole, though these patterns are not as well defined in radiation oncology. We evaluated the association of provider zip-code level sociodemographic characteristics with variations in radiation therapy (RT) modality use. We used the Centers for Medicare & Medicaid Services (CMS) Physician and Other Supplier Public Use File, CMS Physician Compare, and National Neighborhood Data Archive databases to link non-facility based radiation oncologists and their submitted Medicare charges in 2020 with sociodemographic information of the provider's zip code. We selected radiation oncologists who submitted for any of the following RT treatments: proton RT (PRT), stereotactic body RT (SBRT), intensity modulated RT (IMRT), 3D conformal RT (3DCRT), and brachytherapy (BT). Sociodemographic covariates included predominant race/ethnicity of area (Black, White, or Hispanic), percentage of population earning above $75,000 annually, provider rurality (metropolitan vs non-metropolitan), and predominant education level attained of area (Bachelor's degree vs no Bachelor's degree). We measured the association of these covariates with the use of different RT modalities with multivariable logistic regression. We identified 1,126 radiation oncologists for analysis. PRT was more likely to be used by providers in areas with a predominantly Black population compared with providers in areas with a predominantly White or Hispanic population (Black as reference, White odds ratio (OR) 0.11; 95% CI 0.05-0.24; Hispanic OR 0.23; 95% CI 0.08-0.56). IMRT was more likely to be used by providers in areas with a predominantly White or Hispanic population compared with providers in areas with a predominantly Black population (White OR 3.40; 95% CI 1.63-6.78; Hispanic OR 2.46; 95% CI 1.04-5.97). SBRT was less likely to be used in non-metro areas compared with metro areas (OR 0.28, 95% CI 0.12-0.56). The annual income measure was not associated with any RT modality. Providers in areas with predominantly Bachelor's degree attained were more likely to use PRT (Bachelor's degree as reference, OR 0.34; 95% CI 0.17-0.64) and less likely to use SBRT (OR 1.53; 95% CI 1.04-2.26), IMRT (OR 2.04; 95% CI 1.20-3.48), and 3DCRT (OR 1.52; 95% CI 1.04-2.22). No covariates were associated with BT use. We identified notable variations in RT delivery with multiple sociodemographic factors. A predominantly Black population was associated with a higher likelihood of PRT use, and BT use was not associated with any sociodemographic covariate. These findings might reflect reduced barriers to care afforded by the universal healthcare setting of Medicare or ongoing improvements in RT delivery with respect to previously identified health disparities. Additional work is needed to further elucidate these patterns of RT use and determine if they represent effective care.

Public transit stop density is associated with walking for exercise among a national sample of older adults

BackgroundWalking is the primary and preferred mode of exercise for older adults. Walking to and from public transit stops may support older adults in achieving exercise goals. This study examined whether density of neighborhood public transit stops was associated with walking for exercise among older adults.Methods2018 National Health and Aging Trends Study (NHATS) data were linked with the 2018 National Neighborhood Data Archive, which reported density of public transit stops (stops/mile2) within participants’ neighborhood, defined using census tract boundaries. Walking for exercise in the last month was self-reported. The extent to which self-reported public transit use mediated the relationship between density of neighborhood public transit stops and walking for exercise was examined. Covariates included sociodemographic characteristics, economic status, disability status, and neighborhood attributes. National estimates were calculated using NHATS analytic survey weights.ResultsAmong 4,836 respondents with complete data, 39.7% lived in a census tract with at least one neighborhood public transit stop and 8.5% were public transit users. The odds of walking for exercise were 32% higher (OR = 1.32; 95% confidence interval: 1.08, 1.61) among respondents living in a neighborhood with > 10 transit stops per mile compared to living in a neighborhood without any public transit stops documented. Self-reported public transit use mediated 24% of the association between density of neighborhood public transit stops and walking for exercise.ConclusionsDensity of neighborhood public transit stops was associated with walking for exercise, with a substantial portion of the association mediated by self-reported public transit use. Increasing public transit stop availability within neighborhoods may contribute to active aging among older adults.

Depth-defying OOF-GNN: Sailing smoothly amidst GNN waves

When it comes to machine learning on graphs, Graph Neural Networks(GNNs) is a potent tool. By iteratively propagating neural messages along the edges of the input graph, GNNs integrate node feature information with graph structure. Nonetheless, not all neighborhood data is valuable. Some of the neighbor data may not be necessary for the downstream task. For a controlled and relevant information aggregation, we propose a flexible method OOF-GNN (Over-smoothing and Over-squashing Free Graph Neural Network). The main notion of this technique is to limit the impact of irrelevant and excessive information propagation during message passing. An extensive discussion of the theoretical and experimental justifications, associated with this method is presented. This technique works on all baseline models (e.g. GCN, GraphSAGE, and GAT). It allows them to gather information from the K-hop neighborhood without node information being over-smoothed and over-squashed. The efficacy of OFF-GNN in addressing the challenges of under-reaching, over-smoothing, and over-squashing, as well as improving the performance of GNN models in node classification tasks, has been demonstrated through comprehensive experimentation on both academic benchmarks and real-world datasets.

Crimes committed by recent immigrants: Characteristics and community patterns

The notion persists that recent immigrants commit substantively more serious crimes than citizens, but prior research has only used aggregate-level data or youth-focused samples. We address this gap using individual-level 2017 crime arrest data from the Houston Police Department (HPD) that include citizenship, supplemented by Houston Super Neighborhoods data and American Community Survey estimates. We conduct bivariate and multilevel multinomial analyses to compare crime characteristics and neighborhood-level influences on offending by citizenship, finding that non-U.S. citizen arrestees were less likely to have been arrested for felonies, drug crimes, and Part I property crimes than U.S. citizen arrestees. An exploration of neighborhood contextual influences and interaction effects reveals further nuances. Directions for future research and implications for evidence-based policy are discussed.

Spatial Variation in COVID-19 Mortality in New York City and Its Association with Neighborhood Race, Ethnicity, and Nativity Status.

We examined the association between variation in COVID-19 deaths and spatial differences in the racial, ethnic, and nativity-status composition of New York City neighborhoods, which has received little scholarly attention. Using COVID-19 mortality data (through 31 May 2021) and socioeconomic and demographic data from the American Community Survey at the Zip Code Tabulation Area level as well as United-Hospital-Fund-level neighborhood data from the Community Health Survey of the New York City Department of Health and Mental Hygiene, we employed multivariable Poisson generalized estimating equation models and assessed the association between COVID-19 mortality, racial/ethnic/nativity-status composition, and other ecological factors. Our results showed an association between neighborhood-level racial and ethnic composition and COVID-19 mortality rates that is contingent upon the neighborhood-level nativity-status composition. After multivariable adjustment, ZCTAs with large shares of native-born Blacks and foreign-born Hispanics and Asians were more likely to have higher COVID-19 mortality rates than areas with large shares of native-born Whites. Areas with more older adults and essential workers, higher levels of household crowding, and population with diabetes were also at high risk. Small-area analyses of COVID-19 mortality can inform health policy responses to neighborhood inequalities on the basis of race, ethnicity, and immigration status.

An examination of the relationships between the neighborhood social environment, adiposity, and cardiometabolic disease risk in adolescence: a cross-sectional study

BackgroundDisadvantaged neighborhood environments are a source of chronic stress which undermines optimal adolescent health. This study investigated relationships between the neighborhood social environment, specifically, chronic stress exposures, adiposity, and cardiometabolic disease risk factors among 288 Louisiana adolescents aged 10 to 16 years.MethodsThis cross-sectional study utilized baseline data from the Translational Investigation of Growth and Everyday Routines in Kids (TIGER Kids) study. Adolescent data were obtained using self-reported questionnaires (demographics and perceived neighborhood disorder), anthropometry, body imaging, and a blood draw while objective neighborhood data for the concentrated disadvantage index were acquired from the 2016 American Community Survey five-year block group estimates, 2012–2016. Multilevel linear regression models were used to examine whether neighborhood concentrated disadvantage index and perceived neighborhood disorder were associated with body mass index, waist circumference, body fat, adipose tissue, blood pressure, and lipids. We performed multilevel logistic regression to determine the odds of elevated adiposity and cardiometabolic disease risk for adolescents living in neighborhoods with varying levels of neighborhood concentrated disadvantage and disorder.ResultsAdolescents living in neighborhoods with higher disadvantage or disorder had greater waist circumference and total percent body fat compared to those in less disadvantaged and disordered neighborhoods (p for trend < 0.05). Neighborhood disadvantage was also positively associated with percentage of the 95th Body Mass Index percentile and visceral abdominal adipose tissue mass while greater perceived neighborhood disorder was related to higher trunk fat mass and diastolic blood pressure (p for trend < 0.05). Living in the most disadvantaged was associated with greater odds of obesity (OR: 2.9, 95% CI:1.3, 6.5) and being in the top tertile of body fat mass (OR: 3.0, 95% CI: 1.4, 6.6). Similar results were found with neighborhood disorder for odds of obesity (OR: 2.1, 95% CI:1.1, 4.2) and top tertile of body fat mass (OR: 2.1, 95% CI:1.04, 4.1).ConclusionsNeighborhood social environment measures of chronic stress exposure were associated with excess adiposity during adolescence, and relationships were most consistently identified among adolescents living in the most disadvantaged and disordered neighborhoods. Future studies should account for the influences of the neighborhood environment to stimulate equitable improvements in adolescent health.Clinical Trials Registration# NCT02784509.

Fear of a Black Neighborhood: Anti-Black Racism and the Health of White Americans.

Does anti-Black racism harm White Americans? We advance hypotheses that address this question within the neighborhood context. Hypotheses are tested with neighborhood and survey data from a probability sample of White residents of Nashville, Tennessee. We find that regardless of neighborhood crime rates or socioeconomic compositions, Whites report heightened perceptions of crime and danger in their neighborhoods as the proportion of Black residents increases. Perceived neighborhood danger, in turn, predicts increased symptoms of psychophysiological distress. When stratified by socioeconomic status (SES), however, low-SES Whites also report perceptions of higher status when living near more Black neighbors, which entirely offsets their distress. We conclude that although anti-Black racism can ironically harm the health of White Americans, compensatory racist ideologies can also offset these harms, particularly for lower-status Whites. We situate our findings within broader discussions of anti-Black racism, residential segregation, and psychiatric disorders commonly observed among White Americans.

Realization of a database‐driven PID control system using a CMAC memory

AbstractAs a nonlinear control approach, a database‐driven PID (DD‐PID) controller has been proposed that uses distance calculations of an initial database to tune PID parameters. However, sorting operations are required in extracting neighborhood data from the database, which imposes a large computational load. Therefore, it is impossible to finish this computation within a restricted sampling period in control systems with fast response time and insufficient memory capacity. On the other hand, the cerebellar model articulation controller (CMAC) has advantages for these problems and has similar constructs to the DD approach. Therefore, this paper combines advantages of these two approaches, and proposes a CMAC‐PID controller design based on off‐line optimization of a database. Finally, a comparison with conventional off‐line CMAC‐PID controller and DD‐PID controller through simulations are considered.

Differential health outcomes of the COVID‐19 pandemic among minority populations: An analysis based on Chicago's neighborhoods

AbstractBackgroundThe study examines the effects of the COVID‐19 pandemic on different ethnic and racial groups. It aims to investigate the existence or nonexistence of significant variations in COVID‐19 health outcomes among two ethnic and racial minorities that resided in Chicago neighborhoods during the onslaught of the pandemic. Researchers have traditionally studied health disparities by comparing the health of minorities representing “underserved” populations and those with adequate healthcare. This study focuses on the heterogeneity of health outcomes between different minority populations, mainly Black and Hispanic, traditionally considered underserved populations.MethodsThis cross‐sectional study uses secondary data from a public reporting site. The unit of analysis is neighborhood units based on US postal zip codes that are cross‐referenced with the US Census Bureau's zip code tabulation area codes. We used Chicago neighborhood data and applied geographic analyses to describe the patterns of similarities and differences in the outcomes of the COVID‐19 pandemic among neighborhoods with different ethnic and racial minorities residing in them. Using the one‐way analysis of variance technique, we also tested research hypotheses about the COVID‐19 outcome differences and/or similarities among the neighborhoods.ResultsOur findings show that although Hispanic neighborhoods disproportionately carried a higher burden of infection by the disease, the mortality due to the illness or the case fatality rate was not much higher than in the other neighborhoods. In contrast, African American neighborhoods did experience significantly higher case fatality rates—although their infection rate was not statistically significantly higher than the average infection rates of the other Chicago neighborhoods.ConclusionsMinority status creates distinct adverse effects on different minority groups. The patterns of distinct outcomes need to be well understood through further studied and considered by policymakers when health policies are designed to address the impact of health disparities.

Generalizable prediction of childhood ADHD symptoms from neurocognitive testing and youth characteristics

Childhood attention-deficit/hyperactivity disorder (ADHD) symptoms are believed to result from disrupted neurocognitive development. However, evidence for the clinical and predictive value of neurocognitive assessments in this context has been mixed, and there have been no large-scale efforts to quantify their potential for use in generalizable models that predict individuals’ ADHD symptoms in new data. Using data drawn from the Adolescent Brain Cognitive Development Study (ABCD), a consortium that recruited a diverse sample of over 10,000 youth (ages 9–10 at baseline) across 21 U.S. sites, we develop and test cross-validated machine learning models for predicting youths’ ADHD symptoms using neurocognitive abilities, demographics, and child and family characteristics. Models used baseline demographic and biometric measures, geocoded neighborhood data, youth reports of child and family characteristics, and neurocognitive tests to predict parent- and teacher-reported ADHD symptoms at the 1-year and 2-year follow-up time points. Predictive models explained 15–20% of the variance in 1-year ADHD symptoms for ABCD Study sites that were left out of the model-fitting process and 12–13% of the variance in 2-year ADHD symptoms. Models displayed high generalizability across study sites and trivial loss of predictive power when transferred from training data to left-out data. Features from multiple domains contributed meaningfully to prediction, including neurocognition, sex, self-reported impulsivity, parental monitoring, and screen time. This work quantifies the information value of neurocognitive abilities and other child characteristics for predicting ADHD symptoms and provides a foundational method for predicting individual youths’ symptoms in new data across contexts.

Living Alone, Environmental Hazards, and Falls Among U.S. Older Adults.

Physical conditions of living environments can affect the incidence of falls; however, prior work has focused typically on 1 domain at a time-either neighborhood or home, capturing limited environmental boundaries of older adults. We examined how neighborhood together with the home environment affect the incidence of falls over time and whether living arrangement modifies the influence of the environmental risks on falls. Using the 2012-2020 waves of the Health and Retirement Study (HRS; N = 1,893), we fitted logistic regression to estimate the incidence of falls over an 8-year study period. We used the neighborhood and housing data that are collected systematically by trained observers in the HRS to assess environmental hazards. Sidewalk quality, neighborhood disorder, and the presence of green space were measured to capture outdoor environmental hazards. Indoor environmental hazards included the presence of housing decay and poorly maintained stairways. All models were stratified by living arrangement. Neighborhood and housing environment were independently associated with the odds of falls net of demographic characteristics and preexisting health conditions, and effects were significant for people living alone only. The presence of green space and poorly maintained stairways were associated with greater odds of falling, net of covariates during 8 years of follow-up (odds ratios = 2.10 and 2.65, p < .05, respectively). None of the environmental risk factors were significant for those living with others. Falls in old age may be determined in part by a combination of outdoor and indoor risk factors. More research is needed to understand pathways that lead to greater vulnerability among older adults living alone to environmental hazards.

Representativeness of patients with lung cancer in an integrated health care delivery system.

e13600 Background: Lung cancer is the leading cause of cancer death in the US, and research is critical to reducing its morbidity and mortality. Minoritized groups (e.g., low income) that face higher lung cancer burden are underrepresented in clinical trials, which are often conducted at NCI-designated cancer centers (NCICCs). Large integrated health systems such as Kaiser Permanente Northern California (KPNC) may be well-positioned to produce research findings that are more representative of patient populations in important clinical and demographic characteristics. In this study, we evaluated the representativeness of lung cancer patients from KPNC compared to those from NCICCs and non-NCICCs in Northern California. Methods: We extracted individual and neighborhood data from the California Cancer Registry, California Neighborhoods Data System, and CalEnviroScreen 4.0 for patients with invasive lung cancer diagnosed from 2015 through 2019 within the KPNC catchment area. We compared these characteristics among patients from KPNC, NCICCs (Stanford, UC Davis, UCSF), and all other non-NCICCs. Results: Of 27,288 patients, 31% are from KPNC, 8% from NCICCs, and 61% from other non-NCICCs (Table). KPNC patients are more similar to other non-NCICC patients and the total lung cancer population than to NCICC patients on most characteristics. KPNC patients are slightly older, more likely to be female, and less likely to be Asian American/Pacific Islander (AAPI), compared to other non-NCICC patients. In contrast, NCICC patients are younger, less likely to be female or non-AAPI minoritized racial and ethnic groups, and present with less stage IV disease and more adenocarcinoma at diagnosis. Both KPNC and NCICC patients are more likely to reside in neighborhoods with the highest socioeconomic status (SES) and lowest pollution burden than other non-NCICC patients. Given large numbers, all differences were statistically significant (p&lt;0.001). Conclusions: These analyses demonstrate that lung cancer patients in KPNC are more broadly representative of the overall lung cancer population than those in NCICCs, and can produce generalizable research findings to improve outcomes for all lung cancer patients. [Table: see text]

A Review of the Role of Intermediate Partners in the Community Participatory Mapping Project at Slum Upgrading Settlement

Surakarta has put a lot of effort into promoting civic engagement. Numerous projects have been taken out since 2017 to help the local government with the upgrading of the slum community on the Kali Pepe and Bengawan Solo River riverbanks. The initiation comes from grass root programme. Clarifying the function of intermediary actors in the administration of participatory mapping on slum improvement projects is the goal of this study. The initiatives were based on the writers’ experiences while working on several community engagement programs for project upgrades between 2018 and 2022 at Surakarta City (Solo). The information is gathered through field research that is done concurrently with project execution. Selected participatory mapping projects included four case studies. The Kampung Layak Anak Project was run by Urban Rural Design and Conservation (URDC Labo) and ARKOM Solo in 2017, The Green Kampung Project was run by URDC Labo UNS and SIBAT Sewu in 2018, Bangkit Berbarengan was run by URDC Labo and ARKOM Solo in 2020, and Mojo’s participatory mapping of a children-friendly environment (2022) are just a few examples. Several participants were involved in the neighborhood data collection process, and participatory mapping technologies were applied. It will explore these conflicts to look at the role intermediate partners play in empowering communities. Since participatory mapping pertains to both top-down and bottom-up methodologies, the findings primarily focus on the ambiguity of participation in the context of Solo. Another conclusion highlights the role of intermediary actors, such as academics, ARKOM, KOTAKU, and Sibat Sewu, in efforts to restructure urban areas.

Using electronic medical records to develop a predictive model of 30‐day hospital readmission for people with ADRD

AbstractBackgroundHospitals are insufficiently equipped for Alzheimer’s disease and related dementia (ADRD) patients. Thus, readmission incidence is much higher and costlier among ADRD patients than the general population. Hospital discharges often occur without adequate preparation for the complex care management needs of ADRD patients and their caregivers. This study’s objective was to develop a risk‐assessment tool for hospitalized patients with ADRD. By supporting timely and better identification of who is at risk and why, already‐scarce resources can be allocated more efficiently to reduce readmissions.MethodsWe used 2016‐2019 EMR data from the University of Michigan health system (Michigan Medicine) and applied machine learning techniques (Random Forest, XGBoost, and Logistic LASSO) to develop a readmission risk‐assessment tool. We identified 2,899 individuals with ADRD who had at least one index hospital admission. All data features available in EMR – demographics, lab results, prior counts of healthcare use, and characteristics of index hospitalization – were included in our predictive models. Additionally, we geocoded the street address of place of residence using the National Neighborhood Data Archive (NaNDA) using the U.S. Census tract‐level information to include two composite measures of socioeconomic status: disadvantage and affluence.ResultsThe readmission rate for ADRD patients was 22% versus 17% for the general population. The best predictive model was the Random Forest (area under the receiver operating characteristic curve = 0.66; sensitivity = 0.64; specificity = 0.61). The accuracy of our model (0.61) was 42% higher than the LACE score (0.43), which is currently used by the hospital for all patients. The top 5 predictors of 30‐day readmission among people with ADRD included length of hospital stay, frailty index, living in a disadvantaged neighborhood, and total prior‐year healthcare charges.ConclusionADRD patients are highly vulnerable and require many resources, with substantially greater readmission risk and elevated rates of other adverse health events. Leveraging EMR data in a readmission risk‐assessment tool can help inform appropriate and efficient coordination and transitions of care for ADRD patients. Our risk‐assessment tool can identify ADRD patients at high risk for readmission and why they are at higher risk. This can enable better decision‐making upon discharge.

The Built Environment and Social and Emotional Support among Rural Older Adults: The Case for Social Infrastructure and Attention to Ethnoracial Differences.

Social and emotional support (SaES) is essential for older adult mental health and is shaped by individual-level factors and the built environment. However, much of the focus on the built environment, and specifically social infrastructure - the physical places that facilitate social interaction and social tie formation - relies heavily on urban settings or samples with limited diversity. Consequently, there is little understanding if social infrastructure matters for the SaES of older adults in rural America, and across race and ethnicity. Therefore, we use social cohesion as a conceptual lens and the community gerontology framework to determine if availability of social infrastructure is associated with SaES among older adults in rural America and if this relationship varies across race and ethnicity. Using data from 110,850 rural older adults from the Behavioral Risk Factors Surveillance System and data from the National Neighborhood Data Archive, we show that among rural ethnoracial minority older adults, higher densities of social infrastructure are associated with higher SaES. This is not true for rural non-Hispanic White older adults. Results highlight the importance of accounting for both social infrastructure as part of the built environment and heterogeneity across race and ethnicity in studies that examine older adult mental and emotional health.

Spatiotemporal Distribution and Influencing Factors of Theft during the Pre-COVID-19 and COVID-19 Periods: A Case Study of Haining City, Zhejiang, China

Theft is an inevitable problem in the context of urbanization and poses a challenge to people’s lives and social stability. The study of theft and criminal behavior using spatiotemporal, big, demographic, and neighborhood data is important for guiding security prevention and control. In this study, we analyzed the theft frequency and location characteristics of the study area through mathematical statistics and hot spot analysis methods to discover the spatiotemporal divergence characteristics of theft in the study area during the pre-COVID-19 and COVID-19 periods. We detected the spatial variation pattern of the regression coefficients of the local areas of thefts in Haining City by modeling the influencing factors using the geographically weighted regression (GWR) analysis method. The results explained the relationship between theft and the influencing factors and showed that the regression coefficients had both positive and negative values in the pre-COVID-19 and COVID-19 periods, indicating that the spatial distribution of theft in urban areas of Haining City was not smooth. Factors related to life and work indicated densely populated areas had increased theft, and theft was negatively correlated with factors related to COVID-19. The other influencing factors were different in terms of their spatial distributions. Therefore, in terms of police prevention and control, video surveillance and police patrols need to be deployed in a focused manner to increase their inhibiting effect on theft according to the different effects of influencing factors during the pre-COVID-19 and COVID-19 periods.

Population density: What does it really mean in geographical health studies?

Population density is an indicator in many studies, but often with only a cursory explanation of why. Unfortunately, elected officials and the media draw misleading conclusions about population density and public health. After providing three reasons why population density is linked to human health outcomes, using state, county, municipal and neighborhood scale data, we show that population density serves as a surrogate for explaining the geographical distribution of life expectancy and broadband access. However, population density loses its unique contribution when other factors influencing health are included. We urge authors to explain why they include population density in their studies.