Homogeneity versus Heterogeneity in Household Structure: The Recent Experience of Twenty US Cities

P Gober

doi:10.1068/a180715

Abstract P873: Structural Racism in Columbus, Ohio Associated With Stroke Prevalence

Stroke ◽

10.1161/str.52.suppl_1.p873 ◽

2021 ◽

Vol 52 (Suppl_1) ◽

Author(s):

Jeffrey J Wing ◽

Emily E Lynch ◽

Sarah E Laurent ◽

Bruce C Mitchell ◽

Jason Richardson ◽

...

Keyword(s):

Census Tract ◽

Sociodemographic Factors ◽

Structural Racism ◽

Stroke Outcomes ◽

Census Tracts ◽

Census Tract Level ◽

Stroke Prevalence ◽

Level Information ◽

Fundamental Cause ◽

Loan Corporation

Introduction: Racial disparities exist in stroke and stroke outcomes. However, the fundamental cause for these disparities are not biological differences, but structural racism. Using the Home Owners’ Loan Corporation (HOLC) ‘redlining’ scores, as indicator of structural lending practices from middle of the last century, we hypothesize that census tracts with high historic redlining are associated with higher stroke prevalence. Methods: Weighted historic redlining scores (HRS) were calculated using the proportion of 1930s HOLC residential security grades contained within 2010 census tract boundaries of Columbus, Ohio. Stroke prevalence (adults >=18) was obtained at the census tract-level from the CDC’s 500 Cities Project. Sociodemographic factors, as measured by census tract level information (American Community Survey 2014-2018), were considered mediators in the causal association between historic redlining (measured in 1936) and stroke prevalence (measured in 2017) and were not controlled for in regression analysis. The functional form of the association was non-linear, so stroke prevalence within quartiles of the HRS were compared using linear regression instead of a continuous score. Results: Higher HRS, representing greater redlining, were associated with greater prevalence of stroke when comparing the highest to the lowest quartile of HRS (Figure). Census tracts in the highest quartile of HRS had 1.48% higher stroke prevalence compared to those in the lowest quartile (95% CI: 0.23-2.74). No other interquartile differences were observed. Conclusions: Historic redlining practices are a form of structural racism that established geographic systems of disadvantage and consequently, poor health outcomes. Our findings demonstrate disparate stroke prevalence by degree of historic redlining in census tracts across Columbus, Ohio. While ecologic, this study demonstrates the need to acknowledge that racism, not race, drive stroke disparities.

Download Full-text

2325. Relationship Between Neighborhood Census-Tract-Level Poverty and Respiratory Syncytial Virus Infection in hospitalized Adults in the San Francisco Bay area, CA 2015–2017

Open Forum Infectious Diseases ◽

10.1093/ofid/ofz360.2003 ◽

2019 ◽

Vol 6 (Supplement_2) ◽

pp. S798-S798

Author(s):

Jenna Holmen ◽

Bryanna Cikesh ◽

Lindsay Kim ◽

Art Reingold

Keyword(s):

Respiratory Syncytial Virus ◽

San Francisco ◽

Census Tract ◽

San Francisco Bay ◽

San Francisco Bay Area ◽

Poverty Level ◽

Census Tracts ◽

Bay Area ◽

Census Tract Level ◽

Syncytial Virus

Abstract Background In the United States, respiratory syncytial virus (RSV) is a leading cause of admission for adults with respiratory illness. In adults > 50 years of age, it accounts for up to 12% of medically-attended acute respiratory illnesses and has a case fatality proportion of ~6–8%. Poverty can have an important influence on health. Few studies have evaluated the relationship of RSV incidence and poverty level, and no identified studies have evaluated this relationship among adults. We evaluated the incidence of RSV-associated hospitalizations in adults in the San Francisco Bay Area, CA by census-tract-level poverty. Methods Medical record data abstraction was conducted for all adults with a laboratory-confirmed RSV infection who were admitted to a hospital within the 3 counties comprising the catchment area (Alameda, Contra Costa, and San Francisco counties) during the 2015–2016 and 2016–2017 RSV seasons. Patient addresses were geocoded to their corresponding census-tract (CT). Census tracts were divided into four levels of poverty based on American Community Survey data of percentage of people living below the poverty level: 0–4.9%, 5–9.9%, 10-–9.9%, and ³20%. Incidence rates were calculated by dividing the number of RSV cases in each CT poverty-level (numerator) by the number of adults living in each CT poverty level (denominator), as determined from the 2010 US census, and standardized for age. Results There were 526 RSV case-patients with demographic characteristics as outlined in Table 1. The highest incidence of RSV-associated hospitalization was in CTs associated with the highest levels of poverty (>20%). However, the second highest incidence of RSV-associated hospitalization occurred among adults living in CTs with <5% poverty (Figure 1 and Table 2). Conclusion The incidence rate of RSV-associated hospitalization in adults appears to be positively correlated with highest census-tract level of poverty; however, there is a high incidence among adults living in the lowest poverty census-tracts. Disclosures All authors: No reported disclosures.

Download Full-text

Using the Poisson Distribution for Estimating Households by Household Size

10.31219/osf.io/v67k9 ◽

2017 ◽

Author(s):

Beth Jarosz

Keyword(s):

Poisson Distribution ◽

Census Tract ◽

Census Data ◽

Statistical Technique ◽

Household Size ◽

Size Distributions ◽

Census Tracts ◽

Reliable Tool ◽

Census Tract Level ◽

Simplified Method

Infrastructure planners often require detail about the number of households by household size at very small levels of geography (census tract or smaller) to calibrate their models. In addition, these data must also be projected into the future in order to support planning efforts.This paper documents a statistical technique for estimating the distribution of households by household size using a modified application of the Poisson distribution. This technique is valuable to demographers as it provides a simple and reliable tool for estimating the distribution of household sizes at nearly any level of geography for a given point in time.There are a wide variety of applications of the Poisson distribution in biology and engineering. However, there are only few documented applications in demographics. This article puts forth two key advancements over prior published work:(1) an entirely new, and greatly simplified method for applying the distribution,(2) evidence of the reliability of the technique for estimating household size distributions in small geographic areas (e.g. counties and census tracts).Tests on U.S. Census data (1990-2010) suggest that the model is suitable for use in estimating the distribution of households by household size at the state, county, and census tract level.

Download Full-text

Understanding Pediatric Surgery Cancellation: Geospatial Analysis (Preprint)

10.2196/preprints.26231 ◽

2020 ◽

Author(s):

Lei Liu ◽

Yizhao Ni ◽

Andrew F Beck ◽

Cole Brokamp ◽

Ryan C Ramphul ◽

...

Keyword(s):

Linear Regression ◽

Regression Model ◽

Linear Regression Model ◽

Census Tract ◽

Geospatial Analysis ◽

Community Level ◽

Minority Status ◽

Data Set ◽

Census Tracts ◽

Census Tract Level

BACKGROUND Day-of-surgery cancellation (DoSC) represents a substantial wastage of hospital resources and can cause significant inconvenience to patients and families. Cancellation is reported to impact between 2% and 20% of the 50 million procedures performed annually in American hospitals. Up to 85% of cancellations may be amenable to the modification of patients’ and families’ behaviors. However, the factors underlying DoSC and the barriers experienced by families are not well understood. OBJECTIVE This study aims to conduct a geospatial analysis of patient-specific variables from electronic health records (EHRs) of Cincinnati Children’s Hospital Medical Center (CCHMC) and of Texas Children’s Hospital (TCH), as well as linked socioeconomic factors measured at the census tract level, to understand potential underlying contributors to disparities in DoSC rates across neighborhoods. METHODS The study population included pediatric patients who underwent scheduled surgeries at CCHMC and TCH. A 5-year data set was extracted from the CCHMC EHR, and addresses were geocoded. An equivalent set of data >5.7 years was extracted from the TCH EHR. Case-based data related to patients’ health care use were aggregated at the census tract level. Community-level variables were extracted from the American Community Survey as surrogates for patients’ socioeconomic and minority status as well as markers of the surrounding context. Leveraging the selected variables, we built spatial models to understand the variation in DoSC rates across census tracts. The findings were compared to those of the nonspatial regression and deep learning models. Model performance was evaluated from the root mean squared error (RMSE) using nested 10-fold cross-validation. Feature importance was evaluated by computing the increment of the RMSE when a single variable was shuffled within the data set. RESULTS Data collection yielded sets of 463 census tracts at CCHMC (DoSC rates 1.2%-12.5%) and 1024 census tracts at TCH (DoSC rates 3%-12.2%). For CCHMC, an L2-normalized generalized linear regression model achieved the best performance in predicting all-cause DoSC rate (RMSE 1.299%, 95% CI 1.21%-1.387%); however, its improvement over others was marginal. For TCH, an L2-normalized generalized linear regression model also performed best (RMSE 1.305%, 95% CI 1.257%-1.352%). All-cause DoSC rate at CCHMC was predicted most strongly by <i>previous no show</i>. As for community-level data, the proportion of African American inhabitants per census tract was consistently an important predictor. In the Texas area, the proportion of overcrowded households was salient to DoSC rate. CONCLUSIONS Our findings suggest that geospatial analysis offers potential for use in targeting interventions for census tracts at a higher risk of cancellation. Our study also demonstrates the importance of home location, socioeconomic disadvantage, and racial minority status on the DoSC of children’s surgery. The success of future efforts to reduce cancellation may benefit from taking social, economic, and cultural issues into account.

Download Full-text

Association of perceived neighborhood problems and census tract income with poor self-rated health in adults: a multilevel approach

Cadernos de Saúde Pública ◽

10.1590/0102-311x00210913 ◽

2015 ◽

Vol 31 (suppl 1) ◽

pp. 79-91 ◽

Cited By ~ 7

Author(s):

Doroteia Aparecida Höfelmann ◽

Ana V. Diez Roux ◽

José Leopoldo Ferreira Antunes ◽

Marco Aurélio Peres

Keyword(s):

Mental Health ◽

Census Tract ◽

Multilevel Models ◽

Physical Problem ◽

Social Disorder ◽

Census Tracts ◽

Neighborhood Problems ◽

Self Rated Health ◽

Census Tract Level ◽

Health Related

Abstract Neighborhood problems constitute sources of chronic stress that may increase the risk of poor self-rated health. The associations of census tract level income and perceived neighborhood problems with self-rated health were examined in Florianópolis, Santa Catarina State, Brazil (1,720 adults). Odds ratios (OR) and their 95% confidence intervals (95%CI) of poor self-rated health were estimated through multilevel models. Residents in census tracts in the lower and intermediate tertiles of income reported poorer health than those in the highest tertile. OR of reporting poorer health was 2.44 (95%CI: 2.35- 2.54) in the higher tertile of social disorder (adjusting for mental health). The chances of reporting the poorer health with neighborhood problems ranged from 1.07 (95%CI: 1.03-1.11) to 2.02 (95%CI: 1.95-2.10) for the higher tertile of social disorder (physical health) and physical problem (health-related variables). Perceived neighborhood problems were independently associated with poor health. The perception of a neighborhood among its residents should be considered by health policymakers.

Download Full-text

Using safety inspection data to estimate shaking intensity for the 1994 Northridge earthquake

Bulletin of the Seismological Society of America ◽

10.1785/bssa0880051243 ◽

1998 ◽

Vol 88 (5) ◽

pp. 1243-1253

Author(s):

Katharina Thywissen ◽

John Boatwright

Keyword(s):

Los Angeles ◽

Census Tract ◽

Residential Buildings ◽

Intensity Level ◽

Northridge Earthquake ◽

Single Family ◽

Census Tracts ◽

Wood Frame ◽

Inspection Data ◽

1994 Northridge Earthquake

Abstract We map the shaking intensity suffered in Los Angeles County during the 17 January 1994, Northridge earthquake using municipal safety inspection data. The intensity is estimated from the number of buildings given red, yellow, or green tags, aggregated by census tract. Census tracts contain from 200 to 4000 residential buildings and have an average area of 6 km2 but are as small as 2 and 1 km2 in the most densely populated areas of the San Fernando Valley and downtown Los Angeles, respectively. In comparison, the zip code areas on which standard MMI intensity estimates are based are six times larger, on average, than the census tracts. We group the buildings by age (before and after 1940 and 1976), by number of housing units (one, two to four, and five or more), and by construction type, and we normalize the tags by the total number of similar buildings in each census tract. We analyze the seven most abundant building categories. The fragilities (the fraction of buildings in each category tagged within each intensity level) for these seven building categories are adjusted so that the intensity estimates agree. We calibrate the shaking intensity to correspond with the modified Mercalli intensities (MMI) estimated and compiled by Dewey et al. (1995); the shapes of the resulting isoseismals are similar, although we underestimate the extent of the MMI = 6 and 7 areas. The fragility varies significantly between different building categories (by factors of 10 to 20) and building ages (by factors of 2 to 6). The post-1940 wood-frame multi-family (≧5 units) dwellings make up the most fragile building category, and the post-1940 woodframe single-family dwellings make up the most resistant building category.

Download Full-text

Abstract 005: Violent Crime, Physical Inactivity, And Obesity: Examining Differences In Spatial Relationships By Racial/ethnic Composition Of Community Residents

Circulation ◽

10.1161/circ.143.suppl_1.005 ◽

2021 ◽

Vol 143 (Suppl_1) ◽

Author(s):

Chelsea R Singleton ◽

Fikriyah Winata ◽

Oluwafikayo S Adeyemi ◽

Kaustubh V Parab ◽

Susan Aguiñaga

Keyword(s):

Violent Crime ◽

Census Tract ◽

Crime Rate ◽

Physical Inactivity ◽

Ethnic Composition ◽

Census Tracts ◽

Census Tract Level ◽

Violent Crime Rate ◽

Community Residents ◽

Racial Ethnic

Introduction: Violent crime (e.g., homicide, aggravated assault) is a major public health issue that disproportionately affects communities of color in large urban centers. Studies have reported that residents in high crime communities are less likely to engage in physical activity. There is limited understanding of how violent crime influences physical inactivity and obesity at the community level. We aimed to address this gap by examining differences in spatial relationships between violent crime rate, physical inactivity, and obesity by racial/ethnic composition of community residents in Chicago, IL. Hypothesis: We assessed the hypothesis that violent crime rate is associated with the prevalence of physical inactivity and obesity at the census tract level in Chicago, IL. Methods: We conducted an ecological assessment of 2018 census tract data obtained from various sources. We used data from the City of Chicago to calculate per capita violent crime rate (number of incidents per 1,000 residents) for all census tracts (N = 801). Data on physical inactivity and obesity prevalence (%) were acquired from the CDC. Socio-demographic data (i.e., % Non-Hispanic (NH) White, % NH Black, % Hispanic, median household income) were obtained from the census bureau. We examined spatial lag and error models to determine if violent crime rate is associated with % physical inactivity and % obesity after controlling for socio-demographic characteristics and amenity availability (i.e., per capita outdoor parks and grocery stores). Stratified models were examined to identify differences in associations among majority NH White, NH Black, and Hispanic census tracts (defined as ≥ 50% representation). Results: NH Black census tracts (n = 278) had significantly higher rates of violent crime, physical inactivity, and obesity than Hispanic (n = 169) and NH White tracts (n = 240). Overall, violent crime rate was positivity associated with % physical inactivity (p<0.001) but not % obesity (p=0.77) in Chicago after controlling for covariates. Stratified models revealed that violent crime rate was positively associated with % physical inactivity (p<0.001) and % obesity (p=0.01) among NH Black tracts. Violent crime rate was not associated with % physical inactivity or % obesity among Hispanic and NH White census tracts. Conclusions: Racial/ethnic composition of residents appears to influence census-tract level associations between violent crime rate, physical inactivity, and obesity. Violent crime appears to be more relevant to physical inactivity and obesity in Chicago’s NH Black communities compared to Hispanic and NH White communities.

Download Full-text

An Examination of Geographic Differences in Social Determinants of Health Among Women with Diagnosed HIV in the United States and Puerto Rico, 2017

The Open AIDS Journal ◽

10.2174/1874613602115010010 ◽

2021 ◽

Vol 15 (1) ◽

pp. 10-20

Author(s):

Ndidi Nwangwu-Ike ◽

Chan Jin ◽

Zanetta Gant ◽

Shacara Johnson ◽

Alexandra B. Balaji

Keyword(s):

United States ◽

Puerto Rico ◽

Health Insurance ◽

Social Determinants ◽

Census Tract ◽

The United States ◽

Adult Women ◽

Hiv Diagnosis ◽

Census Tracts ◽

Census Tract Level

Objective: To examine differences, at the census tract level, in the distribution of human immunodeficiency virus (HIV) diagnoses and social determinants of health (SDH) among women with diagnosed HIV in 2017 in the United States and Puerto Rico. Background: In the United States, HIV continues to disproportionately affect women, especially minority women and women in the South. Methods: Data reported in the National HIV Surveillance System (NHSS) of the Centers for Disease Control and Prevention were used to determine census tract-level HIV diagnosis rates and percentages among adult women (aged ≥18 years) in 2017. Data from the American Community Survey were combined with NHSS data to examine regional differences in federal poverty status, education level, income level, employment status, and health insurance coverage among adult women with diagnosed HIV infection in the United States and Puerto Rico. Results: In the United States and Puerto Rico, among 6,054 women who received an HIV diagnosis in 2017, the highest rates of HIV diagnoses generally were among those who lived in census tracts where the median household income was less than $40,000; at least 19% lived below the federal poverty level, at least 18% had less than a high school diploma, and at least 16% were without health insurance. Conclusion: This study is the first of its kind and gives insight into how subpopulations of women are affected differently by the likelihood of an HIV diagnosis. The findings show that rates of HIV diagnosis were highest among women who lived in census tracts having the lowest income and least health coverage.

Download Full-text

Opioid Overdose in Ohio: Comprehensive Analysis of Associated Socioeconomic Factors

10.1101/19005140 ◽

2019 ◽

Author(s):

Chihyun Park ◽

Sara Crawford ◽

Rocio Lopez ◽

Anna Seballos ◽

Jean R. Clemenceau ◽

...

Keyword(s):

Socioeconomic Factors ◽

Census Tract ◽

Rural Areas ◽

Urban Areas ◽

Death Rate ◽

Opioid Overdose ◽

Death Rates ◽

Census Tracts ◽

Socioeconomic Characteristics ◽

Census Tract Level

ABSTRACTObjectiveOur study focused on identifying socioeconomic factors associated with death by opioid overdose in Ohio communities at the census tract level.Materials and MethodsA large-scale vital statistic dataset from Ohio Department of Health (ODH) and U.S. Census datasets were used to obtain opioid-related death rate and socioeconomic characteristics for all census tracts in Ohio. Regression analysis was performed to identify the relationships between socioeconomic factors of census tracts and the opioid-related death rate for both urban and rural tracts.ResultsIn Ohio from 2010-2016, whites, males, and people aged 25-44 had the highest opioid-related death rates. At the census tract level, higher death rates were associated with certain socioeconomic characteristics (e.g. percentage of the census tract population living in urban areas, percentage divorced/separated, percentage of vacant housing units). Predominately rural areas had a different population composition than urban areas, and death rates in rural areas exhibited fewer associations with socioeconomic characteristics.DiscussionPredictive models of opioid-related death rates based on census tract-level characteristics held for urban areas more than rural ones, reflecting the recently observed rural-to-urban geographic shift in opioid-related deaths. Future research is needed to examine the geographic distribution of opioid abuse throughout Ohio and in other states.ConclusionRegression analysis identified associations between population characteristics and opioid-related death rates of Ohio census tracts. These analyses can help government officials and law official workers prevent, predict and combat opioid abuse at the community level.

Download Full-text

Neighborhood disadvantage and firearm injury: does shooting location matter?

Injury Epidemiology ◽

10.1186/s40621-021-00304-2 ◽

2021 ◽

Vol 8 (1) ◽

Author(s):

Kimberly Dalve ◽

Emma Gause ◽

Brianna Mills ◽

Anthony S. Floyd ◽

Frederick P. Rivara ◽

...

Keyword(s):

Census Tract ◽

Social Deprivation ◽

Public Health Problem ◽

Credible Interval ◽

Neighborhood Disadvantage ◽

Future Research ◽

Census Tracts ◽

Firearm Injury ◽

Firearm Injuries ◽

Census Tract Level

Abstract Background Firearm violence is a public health problem that disparately impacts areas of economic and social deprivation. Despite a growing literature on neighborhood characteristics and injury, few studies have examined the association between neighborhood disadvantage and fatal and nonfatal firearm assault using data on injury location. We conducted an ecological Bayesian spatial analysis examining neighborhood disadvantage as a social determinant of firearm injury in Seattle, Washington. Methods Neighborhood disadvantage was measured using the National Neighborhood Data Archive disadvantage index. The index includes proportion of female-headed households with children, proportion of households with public assistance income, proportion of people with income below poverty in the past 12 months, and proportion of the civilian labor force aged 16 and older that are unemployed at the census tract level. Firearm injury counts included individuals with a documented assault-related gunshot wound identified from medical records and supplemented with the Gun Violence Archive between March 20, 2016 and December 31, 2018. Available addresses were geocoded to identify their point locations and then aggregated to the census tract level. Besag-York-Mollie (BYM2) Bayesian Poisson models were fit to the data to estimate the association between the index of neighborhood disadvantage and firearm injury count with a population offset within each census tract. Results Neighborhood disadvantage was significantly associated with the count of firearm injury in both non-spatial and spatial models. For two census tracts that differed by 1 decile of neighborhood disadvantage, the number of firearm injuries was higher by 21.0% (95% credible interval: 10.5, 32.8%) in the group with higher neighborhood disadvantage. After accounting for spatial structure, there was still considerable residual spatial dependence with 53.3% (95% credible interval: 17.0, 87.3%) of the model variance being spatial. Additionally, we observed census tracts with higher disadvantage and lower count of firearm injury in communities with proximity to employment opportunities and targeted redevelopment, suggesting other contextual protective factors. Conclusions Even after adjusting for socioeconomic factors, firearm injury research should investigate spatial clustering as independence cannot be able to be assumed. Future research should continue to examine potential contextual and environmental neighborhood determinants that could impact firearm injuries in urban communities.

Download Full-text