Transforming geographic scale: a comparison of combined population and areal weighting to other interpolation methods

A Comparison of Hospice Care Utilization Between Rural and Urban Children in Appalachia: A Geographic Information Systems Analysis

Long driving times from hospice providers to patients lead to poor quality of care, which may exacerbate in rural and highly isolated areas of Appalachia. This study aimed to investigate geographic patterns of pediatric hospice care across Appalachia. Using person-level Medicaid claims of 1,788 pediatric hospice enrollees who resided in the Appalachian Region between 2011 and 2013. A database of boundaries of Appalachian counties, postal addresses of hospices, and population-weighted county centroids of residences of hospice enrollees driving times from the nearest hospices were calculated. A choropleth map was created to visualize rural/urban differences in receiving hospice care. The average driving time from hospice to child residence was 28 minutes (SD = 26). The longest driving time was in Eastern Kentucky-126 minutes (SD = 32), and the shortest was in South Carolina-11 min (SD = 9.1). The most significant differences in driving times between rural and urban counties were found in Virginia 28 (SD = 7.5) and 5 minutes (SD = 0), respectively, Tennessee-43 (SD = 28) and 8 minutes (SD = 7), respectively; and West Virginia-49 (SD = 30) and 12 minutes (SD = 4), respectively. Many pediatric hospice patients reside in isolated counties with long driving times from the nearest hospices. State-level policies should be developed to reduce driving times from hospice providers.

Using POI and multisource satellite datasets for mainland China's population spatialization and spatiotemporal changes based on regional heterogeneity

Geospatial big data and remote sensing data are widely used in population spatialization studies. However, the relationship between them and population distribution has regional heterogeneity in different geographic contexts. It is necessary to improve data processing methods and spatialization models in areas with large geographical differences. We used land cover data to extract human activity, nighttime light and point-of-interest (POI) data to represent human activity intensity, and considered differences in geographical context to divide mainland China into northern, southern and western regions. We constructed random forest models to generate gridded population distribution datasets with a resolution of 500 m, and quantitatively evaluated the importance of auxiliary data in different geographical contexts. The street-level accuracy assessment showed that our population dataset is more accurate than WorldPop, with a higher R2 and smaller deviation. The improved datasets provided broad potential for exploring the spatial-temporal changes in grid-level population distribution in China from 2010 to 2020. The results indicated that the population density and settlement area have increased, and the overall pattern of population distribution has remained highly stable, but there are significant differences in population change patterns among cities with different urbanization processes. The importance of the ancillary data to the models varied significantly, with POI contributing the most to the southern region and the least to the western region. Moreover, POI had relatively less influence on model improvement in undeveloped areas. Our study could provide a reference for predicting social and economic spatialized data in different geographical context areas using POI and multisource satellite data.

Identifying the Environmental Determinants of Lung Cancer: A Case Study of Henan, China

Lung cancer has become one of the most prevalent cancers in the last several decades. Studies have documented that most cases of lung cancer are caused by inhaling environmental carcinogens while how external environmental factors lead to individual lung cancer is still an open issue as the pathogenesis may come from the combined action of multiple environmental factors, and such pathogenic mechanism may vary from region to region. Based on the data of lung cancer cases from hospitals at the county level in Henan from 2016 to 2020, we analyzed the response relationship between lung cancer incidence and physical ambient factors (air quality, meteorological conditions, soil vegetation) and socioeconomic factors (occupational environment, medical level, heating mode, smoking behavior). We used a Bayesian spatio-temporal interaction model to evaluate the relative risk of disease in different regions. The results showed that smoking was still the primary determinant of lung cancer, but the influence of air quality was increasing year by year, with meteorological conditions and occupational environment playing a synergistic role in this process. The high-risk areas were concentrated in the plains of East and Central Henan and the basin of South Henan, while the low-risk areas were concentrated in the hilly areas of North and West Henan, which were related to the topography of Henan. Our study provides a better understanding of the environmental determinants of lung cancer which will help refine existing prevention strategies and recognize the areas where actions are required to prevent environment and occupation related lung cancer.

Detection of Suicide Clusters using Small-Area Geographic Data from the Virginia Violent Death Reporting System, 2010 - 2015

Introduction

From 1999 to 2020, the suicide rate in Virginia increased from 13.1 to 15.9 per 100,000 persons aged 10 years and older. Few studies have examined spatial patterns of suicide geographies smaller than the county level.

Methods

We analyzed data from suicide decedents aged ≥10 years from 2010 through 2015 in the Virginia Violent Death Reporting System. We identified spatial clusters of high suicide rates using spatially adaptive filtering with standardized mortality ratio (SMR) significantly higher than the state SMR (p < 0.001). We compared demographic characteristics, method of injury, and suicide circumstances of decedents within each cluster to decedents outside any cluster.

Results

We identified 13 high-risk suicide clusters (SMR between 1.7 and 2.0). Suicide decedents in the clusters were more likely to be older (40+ years), non-Hispanic white, widowed/divorced/separated, and less likely to have certain precipitating suicide circumstances than decedents outside the clusters. Suicide by firearm was more common in four clusters, and suicide by poisoning was more common in two clusters compared to the rest of the state.

Conclusions

There are important differences between geographic clusters of suicide in Virginia. These results suggest that place-specific risk factors for suicide may be relevant for targeted suicide prevention.

Predicting COVID-19 cases in diverse population groups using SARS-CoV-2 wastewater monitoring across Oklahoma City

SARS-CoV-2 was discovered among humans in late 2019 and rapidly spread across the world. Although the virus is transmitted by respiratory droplets, most infected persons also excrete viral particles in their feces. This fact prompted a range of studies assessing the usefulness of wastewater surveillance to determine levels of infection and transmission and produce early warnings of outbreaks in local communities, independently of human testing. In this study, we collected samples of wastewater from 13 locations across Oklahoma City, representing different population types, twice per week from November 2020 to end of March 2021. Wastewater samples were collected and analyzed for the presence and concentration of SARS-CoV-2 RNA using RT-qPCR. The concentration of SARS-CoV-2 in the wastewater showed notable peaks, preceding the number of reported COVID-19 cases by an average of one week (ranging between 4 and 10 days). The early warning lead-time for an outbreak or increase in cases was significantly higher in areas with larger Hispanic populations and lower in areas with a higher household income or higher proportion of persons aged 65 years or older. Using this relationship, we predicted the number of cases with an accuracy of 81-92% compared to reported cases. These results confirm the validity and timeliness of using wastewater surveillance for monitoring local disease transmission and highlight the importance of differences in population structures when interpreting surveillance outputs and planning preventive action.

A Comprehensive Analysis of Hurricane Damage across the U.S. Gulf and Atlantic Coasts Using Geospatial Big Data

(1) Background: Hurricane events are expected to increase as a consequence of climate change, increasing their intensity and severity. Destructive hurricane activities pose the greatest threat to coastal communities along the U.S. Gulf of Mexico and Atlantic Coasts in the conterminous United States. This study investigated the historical extent of hurricane-related damage, identifying the most at-risk areas of hurricanes using geospatial big data. As a supplement to analysis, this study further examined the overall population trend within the hurricane at-risk zones. (2) Methods: The Sea, Lake, and Overland Surges from Hurricanes (SLOSH) model and the HURRECON model were used to estimate the geographical extent of the storm surge inundation and wind damage of historical hurricanes from 1950 to 2018. The modeled results from every hurricane were then aggregated to a single unified spatial surface to examine the generalized hurricane patterns across the affected coastal counties. Based on this singular spatial boundary coupled with demographic datasets, zonal analysis was applied to explore the historical population at risk. (3) Results: A total of 775 counties were found to comprise the “hurricane-prone coastal counties” that have experienced at least one instance of hurricane damage over the study period. The overall demographic trends within the hurricane-prone coastal counties revealed that the coastal populations are growing at a faster pace than the national average, and this growth puts more people at greater risk of hurricane hazards. (4) Conclusions: This study is the first comprehensive investigation of hurricane vulnerability encompassing the Atlantic and Gulf Coasts stretching from Texas to Maine over a long span of time. The findings from this study can serve as a basis for understanding the exposure of at-risk populations to hurricane-related damage within the coastal counties at a national scale.

Geographic Associations Between Social Factors and SARS-CoV-2 Testing Early in the COVID-19 Pandemic, February-June 2020, Massachusetts

OBJECTIVES

Widespread SARS-CoV-2 testing is critical to identify infected people and implement public health action to interrupt transmission. With SARS-CoV-2 testing supplies and laboratory capacity now widely available in the United States, understanding the spatial heterogeneity of associations between social determinants and the use of SARS-CoV-2 testing is essential to improve testing availability in populations disproportionately affected by SARS-CoV-2.

METHODS

We assessed positive and negative results of SARS-CoV-2 molecular tests conducted from February 1 through June 17, 2020, from the Massachusetts Virtual Epidemiologic Network, an integrated web-based surveillance and case management system in Massachusetts. Using geographically weighted regression and Moran's I spatial autocorrelation tests, we quantified the associations between SARS-CoV-2 testing rates and 11 metrics of the Social Vulnerability Index in all 351 towns in Massachusetts.

RESULTS

Median SARS-CoV-2 testing rates decreased with increasing percentages of residents with limited English proficiency (median relative risk [interquartile range] = 0.96 [0.95-0.99]), residents aged ≥65 (0.97 [0.87-0.98]), residents without health insurance (0.96 [0.95-1.04], and people residing in crowded housing conditions (0.89 [0.80-0.94]). These associations differed spatially across Massachusetts, and localized models improved the explainable variation in SARS-CoV-2 testing rates by 8% to 12%.

CONCLUSION

Indicators of social vulnerability are associated with variations in SARS-CoV-2 testing rates. Accounting for the spatial heterogeneity in these associations may improve the ability to explain and address the SARS-CoV-2 pandemic at substate levels.

COVID-19 Community Incidence and Associated Neighborhood-Level Characteristics in Houston, Texas, USA

Central to developing effective control measures for the COVID-19 pandemic is understanding the epidemiology of transmission in the community. Geospatial analysis of neighborhood-level data could provide insight into drivers of infection. In the current analysis of Harris County, Texas, we used custom interpolation tools in GIS to disaggregate COVID-19 incidence estimates from the zip code to census tract estimates—a better representation of neighborhood-level estimates. We assessed the associations between 29 neighborhood-level characteristics and COVID-19 incidence using a series of aspatial and spatial models. The variables that maintained significant and positive associations with COVID-19 incidence in our final aspatial model and later represented in a geographically weighted regression model were the percentage of the Black/African American population, percentage of the foreign-born population, area derivation index (ADI), percentage of households with no vehicle, and percentage of people over 65 years old inside each census tract. Conversely, we observed negative and significant association with the percentage employed in education. Notably, the spatial models indicated that the impact of ADI was homogeneous across the study area, but other risk factors varied by neighborhood. The current findings could enhance decision making by local public health officials in responding to the COVID-19 pandemic. By understanding factors that drive community transmission, we can better target disease control measures.

Environmental Characteristics Associated with Older Adults' Social Participation: The Contribution of Sociodemography and Transportation in Metropolitan, Urban, and Rural Areas

Although social participation fosters older adults' health, little is known about which environmental characteristics are related to greater participation in social activities. The Canadian Community Health Survey (n = 2737), a transportation survey, and multiple secondary data sources were used to identify the environmental characteristics associated with older Quebecers' social participation according to living area. Greater social participation was associated with: (1) a higher concentration of older adults (IRR = 2.172 (95% CI 1.600, 2.948); p < 0.001), more kilometers traveled by paratransit (IRR = 1.714 (95% CI 1.286, 2.285); p < 0.01), a lack of medical clinics (IRR = 0.730 (95% CI 0.574, 0.930); p = 0.01), and more funded home adaptations (IRR = 1.170 (95% CI 1.036, 1.320); p = 0.01) in large metropolitan areas; (2) larger paratransit fleets (IRR = 1.368 (95% CI 1.044, 1.791); p = 0.02) and a lower density of road intersections (IRR = 0.862 (95% CI 0.756, 0.982); p = 0.03) in regular metropolitan areas; (3) less social deprivation (IRR = 1.162 (95% CI 1.025, 1.318); p = 0.02) in urban areas; and (4) a higher concentration of older populations (IRR = 2.386 (95% CI 1.817, 3.133); p < 0.001) in rural areas. According to these findings, social participation interventions should target the local environment-for example, by providing more social interaction opportunities for older adults living in younger neighborhoods and by improving access to public transportation, especially paratransit.

Spatial Effects of Livestock Farming on Human Infections With Shiga Toxin-Producing Escherichia coli O157 in Small but Densely Populated Regions: The Case of the Netherlands

The role of environmental transmission of typically foodborne pathogens like Shiga toxin-producing Escherichia coli (STEC) O157 is increasingly recognized. To gain more insights into spatially restricted risk factors that play a role in this transmission, we assessed the spatial association between sporadic STEC O157 human infections and the exposure to livestock (i.e. small ruminants, cattle, poultry, and pigs) in a densely populated country: the Netherlands. This was done for the years 2007-2016, using a state-of-the-art spatial analysis method in which hexagonal areas with different sizes (90, 50, 25 and 10 km2) were used in combination with a novel probability of exposure metric: the population-weighted number of animals per hexagon. To identify risk factors for STEC O157 infections and their population attributable fraction (PAF), a spatial regression model was fitted using integrated nested Laplace approximation (INLA). Living in hexagonal areas of 25, 50 and 90 km2 with twice as much population-weighted small ruminants was associated with an increase of the incidence rate of human STEC O157 infections in summer (RR of 1.09 [95%CI;1.01-1.17], RR of 1.17 [95%CI;1.07-1.28] and RR of 1.13 [95%CI;1.01-1.26]), with a PAF of 49% (95%CI;8-72%). Results suggest exposure to small ruminants to be a risk factor, although no evidence on the mode of transmission is provided. Therefore, the underlying mechanisms warrant further investigation and could offer new targets for control. The newly proposed exposure metric has potential to improve existing spatial modeling studies on infectious diseases related to livestock exposure, especially in densely populated countries like the Netherlands.

Differences in Travel Time to Cancer Surgery for Colon versus Rectal Cancer in a Rural State: A New Method for Analyzing Time-to-Place Data Using Survival Analysis

PURPOSE

Rectal cancer is rarer than colon cancer and is a technically more difficult tumor for surgeons to remove, thus rectal cancer patients may travel longer for specialized treatment compared to colon cancer patients. The purpose of this study was to evaluate whether travel time for surgery was different for colon versus rectal cancer patients.

METHODS

A secondary data analysis of colorectal cancer (CRC) incidence data from the Iowa Cancer Registry data was conducted. Travel times along a street network from all residential ZIP Codes to all cancer surgery facilities were calculated using a geographic information system. A new method for analyzing "time-to-place" data using the same type of survival analysis method commonly used to analyze "time-to-event" data is introduced. Cox proportional hazard model was used to analyze travel time differences for colon versus rectal cancer patients.

RESULTS

A total of 5,844 CRC patients met inclusion criteria. Median travel time to the nearest surgical facility was 9 minutes, median travel time to the actual cancer surgery facilities was 22 minutes, and the median number of facilities bypassed was 3. Although travel times to the nearest surgery facilities were not significantly different for colon versus rectal cancer patients, rectal cancer patients on average traveled 15 minutes longer to their actual surgery facility and bypassed 2 more facilities to obtain surgery.

DISCUSSION

In general, the survival analysis method used to analyze the time-to-place data as described here could be applied to a wide variety of health services and used to compare travel patterns among different groups.

Impact of geo-imputation on epidemiologic associations in a study of outdoor air pollution and respiratory hospitalization

Imputation of missing spatial attributes in health records may facilitate linkages to geo-referenced environmental exposures, but few studies have assessed geo-imputation impacts on epidemiologic inference. We imputed patient Census tracts in a case-crossover analysis of fine particulate matter (PM_2.5) and respiratory hospitalizations in New York State (2000-2005). We observed non-significantly higher PM_2.5 exposures, high accuracy of binary exposure assignment (89 to 99%), and marginally different hazard ratios (HRs) (-0.2 to 0.7%). HR differences were greater in urban versus rural areas. Given its efficiency and nominal influence on accuracy of exposure classification and measures of association, geo-imputation is a candidate method to address missing spatial attributes for health studies.

Using Geosocial Networking Apps to Understand the Spatial Distribution of Gay and Bisexual Men: Pilot Study

BACKGROUND

While services tailored for gay, bisexual, and other men who have sex with men (gbMSM) may provide support for this vulnerable population, planning access to these services can be difficult due to the unknown spatial distribution of gbMSM outside of gay-centered neighborhoods. This is particularly true since the emergence of geosocial networking apps, which have become a widely used venue for meeting sexual partners.

OBJECTIVE

The goal of our research was to estimate the spatial density of app users across Metro Vancouver and identify the independent and adjusted neighborhood-level factors that predict app user density.

METHODS

This pilot study used a popular geosocial networking app to estimate the spatial density of app users across rural and urban Metro Vancouver. Multiple Poisson regression models were then constructed to model the relationship between app user density and areal population-weighted neighbourhood-level factors from the 2016 Canadian Census and National Household Survey.

RESULTS

A total of 2021 app user profiles were counted within 1 mile of 263 sampling locations. In a multivariate model controlling for time of day, app user density was associated with several dissemination area-level characteristics, including population density (per 100; incidence rate ratio [IRR] 1.03, 95% CI 1.02-1.04), average household size (IRR 0.26, 95% CI 0.11-0.62), average age of males (IRR 0.93, 95% CI 0.88-0.98), median income of males (IRR 0.96, 95% CI 0.92-0.99), proportion of males who were not married (IRR 1.08, 95% CI 1.02-1.13), proportion of males with a postsecondary education (IRR 1.06, 95% CI 1.03-1.10), proportion of males who are immigrants (IRR 1.04, 95% CI 1.004-1.07), and proportion of males living below the low-income cutoff level (IRR 0.93, 95% CI 0.89-0.98).

CONCLUSIONS

This pilot study demonstrates how the combination of geosocial networking apps and administrative datasets might help care providers, planners, and community leaders target online and offline interventions for gbMSM who use apps.