Neighborhood-level Variables Research Articles

A data pipeline for secure extraction and sharing of social determinants of health.

Linking neighborhood- and patient-level data provides valuable information about the influence of upstream social determinants of health (SDOH). However, sharing of these data across health systems presents challenges. We set out to develop a pipeline to acquire, deidentify, and share neighborhood-level SDOH data across multiple health systems. We created a pipeline centered around Decentralized Geomarker Assessment for Multi-Site Studies (DeGAUSS) that utilizes containerization to geocode patient addresses and obtain neighborhood-level SDOH variables. We compared DeGAUSS to a third-party vendor geocoding tool available at Duke Health using a cohort of adult patients referred for abdominal transplant from January 1, 2016, to December 31, 2022. We calculated Cohen's Kappa and percent disagreement at census block group and tract levels, and by Area Deprivation Index, urbanicity, and year. The pipeline successfully generated SDOH data for 97.8% of addresses. There was high concordance between DeGAUSS and the vendor tool at the census block group (0.93) and tract levels (0.95). At the block group level, disagreement proportion differed by year and urbanicity, with larger disagreement in the rural category than in micropolitan and metropolitan categories (13%, 7%, 6.2%, respectively). We describe a novel pipeline that can facilitate the secure acquisition and sharing of neighborhood-level SDOH without sharing PHI. The pipeline can be scaled to include additional social, climate, and environmental variables, and can be extended to an unlimited number of health systems.

Predicting blood lead in Uruguayan children: Individual- vs neighborhood-level ensemble learners.

Predicting childhood blood lead levels (BLLs) has had mixed success, and it is unclear if individual- or neighborhood-level variables are most predictive. An ensemble machine learning (ML) approach to identify the most relevant predictors of BLL ≥2μg/dL in urban children was implemented. A cross-sectional sample of 603 children (~7 years of age) recruited between 2009-2019 from Montevideo, Uruguay participated in the study. 77 individual- and 32 neighborhood-level variables were used to predict BLLs ≥2μg/dL. Three ensemble learners were created: one with individual-level predictors (Ensemble-I), one with neighborhood-level predictors (Ensemble-N), and one with both (Ensemble-All). Each ensemble learner comprised four base classifiers with 50% training, 25% validation, and 25% test datasets. Predictive performance of the three ensemble models was compared using area under the curve (AUC) for the receiver operating characteristic (ROC), precision, sensitivity, and specificity on the test dataset. Ensemble-I (AUC: 0.75, precision: 0.56, sensitivity: 0.79, specificity: 0.65) performed similarly to Ensemble-All (AUC: 0.75, precision: 0.63, sensitivity: 0.79, specificity: 0.69). Ensemble-N (AUC: 0.51, precision: 0.0, sensitivity: 0.0, specificity: 0.50) severely underperformed. Year of enrollment was most important in Ensemble-I and Ensemble-All, followed by household water Pb. Three neighborhood-level variables were among the top 10 important predictors in Ensemble-All (density of bus routes, dwellings with stream/other water source and distance to nearest river). The individual-level only model performed best, although precision was improved when both neighborhood and individual-level variables were included. Future predictive models of lead exposure should consider proximal predictors (i.e., household characteristics).

The TBI Model Systems Neighborhood Socioeconomic Disadvantage Index (TBIMS-NSDI): Development and Comparison to Individual Socioeconomic Characteristics.

To create a census-based composite neighborhood socioeconomic deprivation index (NSDI) from geocoded residential addresses and to quantify how NSDI aligns with individual-level socioeconomic factors among people with traumatic brain injury (TBI). Community. People enrolled in the TBI Model Systems National Database (TBIMS NDB). Secondary analysis of a longitudinal cohort study. The TBIMS-NSDI was calculated at the census tract level for the United States population based on a principal components analysis of eight census tract-level variables from the American Community Survey. Individual socioeconomic characteristics from the TBIMS NDB were personal household income, education (years), and unemployment status. Neighborhood:Individual NSDI residuals represent the difference between predicted neighborhood disadvantage based on individual socioeconomic characteristics versus observed neighborhood disadvantage based on the TBIMS-NSDI. A single principal component was found to encompass the eight socioeconomic neighborhood-level variables. It was normally distributed across follow-up years 2, 5, and 10 post-injury in the TBIMS NDB. In all years, the TBIMS-NDSI was significantly associated with individual-level measures of household income and education but not unemployment status. Males, persons of Black and Hispanic background, Medicaid recipients, persons with TBI caused by violence, and those living in urban areas, as well as in the Northeast or Southern regions of the United States, were more likely to have greater neighborhood disadvantage than predicted based on their individual socioeconomic characteristics. The TBIMS-NSDI provides a neighborhood-level indicator of socioeconomic disadvantage, an important social determinant of outcomes from TBI. The Neighborhood:Individual NSDI residual adds another dimension to the TBIMS-NSDI by summarizing how a person's socioeconomic status aligns with their neighborhood socioeconomics. Future studies should evaluate how both measures affect TBI recovery and life quality. Research studying neighborhood socioeconomic disadvantage may improve our understanding of how systemic adversity influences outcomes after TBI.

Characteristics of Youth Presenting for Gender Care Compared to Background Populations: Examination of Social Determinants of Health.

Purpose: Transgender and gender diverse (TGD) youth in North American clinical reports are predominantly White with relatively high socioeconomic status suggesting that access to gender-affirming care is inequitable. This study examined whether socioeconomic and social determinant of health discrepancies exist between a clinical population of TGD youth and surrounding communities. Methods: Patient postal codes were used to link the Ontario Marginalization Index (ON-MARG) to a clinic-based TGD youth cohort (n = 298). Using ON-MARG, each patient was assigned a quintile score from 1 (least marginalized) to 5 (most marginalized) on four marginalization measures. Mean quintile scores were compared to background populations. Census-based Toronto neighborhood-level data on ethnic diversity and educational status were also examined. Neighborhoods were categorized as highly represented, less represented, or unrepresented based on representation in the clinic cohort. One-way analysis of covariance was used to determine associations between neighborhood-level variables and the degree of neighborhood representation. Results: ON-MARG data demonstrated that clinic patients hailed from areas with more individuals having paid employment. Patients from Toronto and surrounding areas came, in general, from communities with fewer recent immigrants and visible minorities. Highly represented Toronto neighborhoods had smaller proportions of visible minorities and immigrants compared with less and unrepresented neighborhoods. Educational status, represented by adults with bachelor's degrees, was lower among unrepresented neighborhoods. Conclusion: TGD youth seen in clinic, particularly those from Toronto, are disproportionally White and socioeconomically advantaged. Further research is needed to better understand the underrepresentation of racialized and low-socioeconomic status youth and to inform strategies to improve access to care.

Socioeconomic Status, Length of Stay, and Postoperative Complications in Oral Cavity Squamous Cell Carcinoma.

Background: Despite universal healthcare in Canada, low socioeconomic status (SES) has been associated with worse survival in oral cavity squamous cell carcinoma (OCSCC) patients. However, the relationship between SES and outcomes during the acute postoperative period is poorly defined. Hamilton, Ontario, presents a unique population with widely varying SES within the same geography. The objective of this study was to examine the relationship between SES, length of hospital stay (LOHS), and postoperative complications in OCSCC. Methods: Newly diagnosed OCSCC patients receiving primary surgical treatment from 2010 to 2014 were identified within a prospectively collected database. Inclusion criteria included age >18 years old, pathological diagnosis of oral cavity cancer, and primary surgical treatment with curative intent. Patients were excluded if they were undergoing palliative treatment or had previous head and neck surgery/radiotherapy. Postal codes were used to identify neighborhood-level socioeconomic variables via 2011 Canada Census data. Income quartiles were defined from groups of neighboring municipalities based on Canada Census definitions. Demographic, social, pathological, staging, and treatment data were collected through chart review. Results: One hundred and seventy-four patients were included in the final analysis. OCSCC patients with lower SES were more likely to be younger (P = .041), male (P = .040), have significant tobacco and alcohol use (P = .001), higher Charlson Comorbidity Index (CCI; P = .014), lower levels of education (P = .001), and have lower employment levels (P = .001). Lower SES patients had higher clinical tumor (P = .006) and clinical nodal (P = .004) staging and were more likely to receive adjuvant therapy (P = .001) and G-tubes (P = .001). Multivariable regression analysis showed that low SES was a statistically significant predictor of postoperative complications [β 2.50 (95% confidence interval (CI) 0.200, 3.17); P = .014] and LOHS [β 2.03 (95% CI 1.06, 2.99); P = .0001]. Tobacco and alcohol use, clinical tumor, and nodal stage, CCI, and planned adjuvant therapy were also statistically significant predictors of postoperative complications and LOHS (P < .05). Conclusion: Patients with lower SES have more advanced OCSCC disease with increased comorbidities that owes itself to more acute postoperative complications and LOHS within this study population. Patients with low SES should be identified as patients that require more support during their cancer treatment.

Neighborhood-level socioeconomic disparities in Radon testing in North Carolina from 2010 to 2020

Radon is a naturally occurring radioactive gas that poses significant health risks to humans, including increased risk of lung cancer. This study investigates the association of neighborhood-level socioeconomic variables with radon testing and radon exposure levels in North Carolina between 2010 and 2020. Our analysis of the two largest commercial household radon tests reveals that 67% of census tracts had testing rates below 10 tests per 1000 population, indicating low testing prevalence. Low radon levels (<2 pCi/L) were detected in 74.1% of the tracts (n = 1626), while medium levels of 2–4 pCi/L and ≥4 pCi/L were observed in 17.2% (n = 378) and 1.6% (n = 36) of the tracts. A generalized spatial regression model was employed to analyze the association between neighborhood-level socioeconomic variables and radon testing rates (per 1000 households), controlling for median radon testing results. The results show a positive correlation (P-value <0.001) of testing rate with various indicators of neighborhood affluence including education level, income, and occupation. In contrast, neighborhood disadvantage, including poverty, unemployment, and public assistance, was associated with a lower radon-testing rate (P-value <0.001). These findings highlight the need for targeted interventions to address socioeconomic disparities in radon testing and promote awareness and access to testing resources in lower socio-economic neighborhoods. Improving testing rates can effectively address radon-related health risks in North Carolina and across the U.S.

Socioeconomic inequality in prostate cancer diagnostics, primary treatment, rehabilitation, and mortality in Sweden.

We designed a nationwide study to investigate the association between socioeconomic factors (household income and education) and different aspects of prostate cancer care, considering both individual- and neighbourhood-level variables. Data were obtained from Prostate Cancer data Base Sweden (PCBaSe), a research database with data from several national health care registers including clinical characteristics and treatments for nearly all men diagnosed with prostate cancer in Sweden. Four outcomes were analysed: use of pre-biopsy magnetic resonance imaging (MRI) in 2018-2020 (n = 11,843), primary treatment of high-risk non-metastatic disease in 2016-2020 (n = 6633), rehabilitation (≥2 dispensed prescriptions for erectile dysfunction within 1 year from surgery in 2016-2020, n = 6505), and prostate cancer death in 7770 men with high-risk non-metastatic disease diagnosed in 2010-2016. Unadjusted and adjusted odds and hazard ratios (OR/HRs) with 95% confidence intervals (CIs) were calculated. Adjusted odds ratio (ORs) comparing low versus high individual education were 0.74 (95% CI 0.66-0.83) for pre-biopsy MRI, 0.66 (0.54-0.81) for primary treatment, and 0.82 (0.69-0.97) for rehabilitation. HR gradients for prostate cancer death were significant on unadjusted analysis only (low vs. high individual education HR 1.41, 95% CI 1.17-1.70); co-variate adjustments markedly attenuated the gradients (low vs. high individual education HR 1.10, 95% CI 0.90-1.35). Generally, neighbourhood-level analyses showed weaker gradients over the socioeconomic strata, except for pre-biopsy MRI. Socioeconomic factors influenced how men were diagnosed with prostate cancer in Sweden but had less influence on subsequent specialist care. Neighbourhood-level socioeconomic data are more useful for evaluating inequality in diagnostics than in later specialist care.

Examining Treatment Attendance Dynamics: Role of Immigrant Status and Neighborhood Factors

In the United States, where one in every seven residents identifies as an immigrant, this study explores the impact of both individual and neighborhood-level factors on treatment attendance, with a particular emphasis on immigrant status. The aim is to identify the drivers and obstacles contributing to disparities in treatment attendance. Hierarchical Linear Modeling (HLM) was employed to integrate individual and neighborhood-level variables and to assess interactions between immigrant status and neighborhood characteristics. Neighborhood factors like Neighborhood Immigrant Density and Neighborhood Concentrated Disadvantage did not show a significant association with treatment attendance. However, immigrant status was significantly associated with higher attendance rates, a trend that persisted even after controlling for other individual and neighborhood factors. Age, identification as Black, and preference for the English language also emerged as significant influences on attendance. This study utilized secondary data merging publicly available census information with electronic medical records (EMRs) from an integrated care clinic. Data integration was achieved using ArcGIS software. The findings underscore the predictive role of immigrant status in treatment attendance, shedding light on the potential challenges immigrants face. This research improves our understanding of the relationship between treatment attendance and both individual and neighborhood factors, guiding healthcare providers in creating targeted interventions for equal treatment access. Although neighborhood attributes yielded no significant associations, the study underscores the need for further exploration. Ultimately, these findings equip healthcare providers with insights into diverse influences on attendance, contributing to efforts aimed at addressing health disparities.

Environmental justice burden and Black-White disparities in spontaneous preterm birth in Harris County, Texas.

Given limited evidence of previous studies, we evaluated the role of environmental justice (EJ) burden (i.e., a neighborhood characterized by both increased environmental burden and socioeconomic deprivation) in Black-White disparities in spontaneous preterm birth (sPTB) in Harris County, Texas and compared results that evaluated neighborhood-level socioeconomic deprivation alone. We conducted a retrospective analysis using PeriBank, a database and biospecimen repository of gravidae giving birth at two hospitals in the Texas Medical Center. We included 3,703 non-Hispanic Black and 5,475 non-Hispanic white gravidae who were U.S.-born, delivered from August 2011-December 2020, and resided in Harris County, TX. We used data from the U.S. EPA EJScreen to characterize the EJ burden of participant's zip code of residence from fine particulate matter (PM2.5), ozone, and proximity to National Priorities List (NPL) sites and calculated zip-code level Area Deprivation Index (ADI). We assessed the contribution of neighborhood-level variables to the Black-White disparity in sPTB by evaluating attenuation of the odds ratio (OR) representing the effect of race in multivariable logistic regression models, controlling for individual-level characteristics. We also conducted race-stratified analyses between each neighborhood variable and sPTB. Exposure indices were treated as continuous variables; in stratified models, ORs and 95% Confidence Intervals (CIs) are presented per 10-unit increase in the neighborhood variable. Accounting for individual-level variables, Black gravidae had 79% higher odds of sPTB than white gravidae (OR = 1.79, 95%CI = 1.32, 2.44); the disparity was moderately attenuated when accounting for EJ burden or ADI (ORs ranged from 1.58 to 1.69). Though we observed no association between any of the EJ burden indices and sPTB among white gravidae, we found increased risks among Black gravidae, with ORs of similar magnitude for each EJ variable. For example, Black gravidae experienced 17% increased odds of sPTB associated with a 10-unit increase in the EJ burden index for PM2.5 (OR = 1.17, 95%CI = 0.97, 1.40). No racial differences were observed in the association of ADI with sPTB. Though we observed limited evidence of the contribution of living in EJ neighborhoods to the Black-White disparity in sPTB, our study suggests living in an EJ neighborhood may differentially impact Black and white gravidae.

1378. Seroprevalence and risk factors for cysticercosis in Mexican-Americans in Starr County, Texas

Abstract Background Neurocysticercosis (NCC) is a parasitic infection caused by the tapeworm Taenia solium, and is a leading cause of epilepsy and chronic headaches in endemic areas including Latin America, Africa, and Asia. The epidemiology of NCC has not been well-studied in the United States to date, but small epidemiological studies suggest evidence of this disease in populations living in poverty and who have immigrated from an endemic region. There is a population living in Starr County, Texas along the Texas-Mexico border in whom the presence of other neglected tropical diseases has been demonstrated due to the intersection of poverty, climate, and high rate of immigration. Methods Our study was a serologic survey and risk factor analysis of cysticercosis in a pre-existing cohort of Mexican-Americans in Starr County, Texas. We used a triplex enzyme-linked immunoassay (ELISA) against cysticercosis-specific antigens and self-reported survey data and neighborhood-level variables to conduct a risk-factor analysis of seropositive participants. Results We identified an overall seropositivity to T. solium in 7.4% (45/605) of the cohort which likely represents a combination of parasite exposure, T. solium disease outside of the central nervous system, old calcified NCC, and viable or degenerating NCC. Female gender, specific occupation and indoor versus outdoor occupation were found to be significantly associated with cysticercosis seropositivity. Thirty-nine of the 45 seropositive participants were female (p 0.021), 26 were employed in healthcare, caregiving, or social service (p 0.009), and 42 were employed in an indoor occupation (p &amp;lt; 0.001). These occupations could pose a risk for T. solium transmission. A geospatial analysis of cases at the census tract level showed increased percentage of seropositivity centered around urban cities Rio Grande City and Roma, as well as smaller city La Grulla and census-designated place Alto Bonito. Conclusion There is a critical need to investigate these cysticercosis-seropositive cases with brain imaging to evaluate for neurocysticercosis, as well as conduct an epidemiological investigation to determine disease transmission in Starr County. Disclosures All Authors: No reported disclosures

Influence of Childhood Adversity on Students’ Delinquent Activities: Interplay with Neighborhood Context and Delinquent Peer Association

Juvenile delinquency is often linked with various proximal family and environmental factors during a child’s upbringing. Richard Jessor’s problem behavior theory (PBT) emphasizes that a combined interplay of these factors may explain this phenomenon appropriately. This study employed the PBT framework to investigate the impact of family on students’ delinquency, considering the influence of neighborhood and delinquent peer association. A model was developed for analyzing the variables by structural equation modeling (SEM). Data were collected through interviews with 1026 students aged between 12 and 18 years from a child development center and eight educational institutes in Bangladesh. The findings revealed that family-level factors (adverse childhood experiences (ACEs) and family deprivation) can significantly predict students’ delinquency directly and through the moderation effect of neighborhood-level variables (neighborhood social capital and neighborhood disorganization). Delinquent peer association exhibited a significant mediating role in the model. It could directly predict delinquency as well. The research has notable theoretical and practical implications for understanding the complex dynamics of family, neighborhood, and delinquent peer association in explaining students’ delinquency.

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.

Prostate Cancer in New York City: Impact of Neighborhood Level Social Determinants of Care

Social determinants of health may impact stage and aggressiveness of cancer presentation. As neighborhoods may influence adjacent neighborhoods across often porous and arbitrary borders, an analysis that measures the spatial impact of these social determinants provides a more holistic evaluation than an analysis that treats each neighborhood as independent and unrelated units. In this paper, we study the impact of sociodemographic and environmental characteristics of neighborhoods on cancer presentation in New York City, using both patient-level analysis and spatial autoregression on the New York State Public Access Cancer Epidemiology Data (NYSPACED) and the NYC Open neighborhood-level dataset. Two outcomes of interest were chosen: Presentation with high grade prostate cancer and presentation with distant disease. We performed logistic regression on individual patient level analysis. We created generalized cross sectional autoregressive spatial regression models to assess direct and indirect (via adjacent neighborhoods) impact of neighborhood level independent variables. We analyzed the following neighborhood level variables: 1) Percent population that was Black, 2) Percent population living in poverty, 3) Percent population obese based on self-reported height and weight, 4) Percent who report being current smokers, 5) Percent who report eating one or more servings of fruits and/or vegetables in the last day, and 6) annual average micrograms of fine particulate matter per cubic meter of air. We identified 95,749 patients diagnosed with prostate adenocarcinoma in New York City from 2001-2018. With reported Black race as the referent value, White patients were less likely to present with high grade disease (OR 0.92 [95% CI 0.89-0.95]). Similarly, self-identified White patients were also less likely to present with distant disease (vs. Black, OR 0.76 [95% CI 0.72-0.81]). Spatial autoregression models revealed no neighborhood level variables associated with direct or indirect impacts on presentation with high grade disease. However, there were several variables initially associated with distant disease presentation, including race, poverty, obesity, fruit and vegetable consumption, and air pollution. However, only neighborhood poverty and obesity levels continued to have significant direct impact on presentation with distant disease when included in the regression model with percent Black residents. In conclusion, we discovered a strong association between Race and poverty with the likelihood of presenting with advanced prostate cancer, at both the individual and neighborhood levels. Geographically targeted outreach to improve prostate cancer screening and access to care should be focused on Black men and neighborhoods with high levels of poverty.

Gray space and default mode network-amygdala connectivity.

Aspects of the built environment relate to health factors and equity in living conditions, and may contribute to racial, ethnic, or economic health disparities. For example, urbanicity is linked with negative factors including exposure to gray space (e.g., impervious surfaces such as concrete, streets, or rooftops). While there is existing research on access to green space and urbanicity on some mental health and cognitive outcomes, there is limited research on the presence of gray space linked with cognitive functioning in youth. The goal of this study was to investigate the link between gray space and amygdala-default mode network (DMN) connectivity. This study used data from the ABCD Study. Participants (n = 10,144; age M = 119.11 months, female = 47.62%) underwent resting-state fMRI acquisition at baseline. Impervious surfaces (gray space) were measured via the Child Opportunity Index (COI). To examine the relationship between presence of gray space and -amygdala-DMN (left/right) connectivity, we employed linear mixed effects models. Correlations were run between amygdala-DMN connectivity and internalizing and externalizing symptoms. Finally, post hoc sensitivity analyses were run to assess the impact of race. More gray space, adjusting for age, sex, and neighborhood-level variables, was significantly associated with increased left amygdala-DMN connectivity (p = 0.0001). This association remained significant after sensitivity analyses for race were completed (p = 0.01). No significant correlations were observed between amygdala-DMN and internalizing or externalizing symptoms. Findings suggest gray space was linked with increased left amygdala-DMN connectivity, circuits that have been implicated in affective processing, emotion regulation, and psychopathology. Thus gray space may be related to alterations in connectivity that may enhance risk for emotion dysregulation. Future investigation of these relationships is needed, as neuroimaging findings may represent early dysregulation not yet observed in the behavioral analyses at this age (i.e., the present study did not find significant relationships with parent-reported behavioral outcomes). These findings can help to inform future public policy on improving lived and built environments.

Small area variations in non-affective first-episode psychosis: the role of socioeconomic and environmental factors.

There is strong evidence supporting the association between environmental factors and increased risk of non-affective psychotic disorders. However, the use of sound statistical methods to account for spatial variations associated with environmental risk factors, such as urbanicity, migration, or deprivation, is scarce in the literature. We studied the geographical distribution of non-affective first-episode psychosis (NA-FEP) in a northern region of Spain (Navarra) during a 54-month period considering area-level socioeconomic indicators as putative explanatory variables. We used several Bayesian hierarchical Poisson models to smooth the standardized incidence ratios (SIR). We included neighborhood-level variables in the spatial models as covariates. We identified 430 NA-FEP cases over a 54-month period for a population at risk of 365,213 inhabitants per year. NA-FEP incidence risks showed spatial patterning and a significant ecological association with the migrant population, unemployment, and consumption of anxiolytics and antidepressants. The high-risk areas corresponded mostly to peripheral urban regions; very few basic health sectors of rural areas emerged as high-risk areas in the spatial models with covariates. Increased rates of unemployment, the migrant population, and consumption of anxiolytics and antidepressants showed significant associations linked to the spatial-geographic incidence of NA-FEP. These results may allow targeting geographical areas to provide preventive interventions that potentially address modifiable environmental risk factors for NA-FEP. Further investigation is needed to understand the mechanisms underlying the associations between environmental risk factors and the incidence of NA-FEP.

Neighborhood-Level Risk Factors for Severe Hyperglycemia among Emergency Department Patients without a Prior Diabetes Diagnosis.

A person's place of residence is a strong risk factor for important diagnosed chronic diseases such as diabetes. It is unclear whether neighborhood-level risk factors also predict the probability of undiagnosed disease. The objective of this study was to identify neighborhood-level variables associated with severe hyperglycemia among emergency department (ED) patients without a history of diabetes. We analyzed patients without previously diagnosed diabetes for whom a random serum glucose value was obtained in the ED. We defined random glucose values ≥ 200mg/dL as severe hyperglycemia, indicating probable undiagnosed diabetes. Patient addresses were geocoded and matched with neighborhood-level socioeconomic measures from the American Community Survey and claims-based surveillance estimates of diabetes prevalence. Neighborhood-level exposure variables were standardized based on z-scores, and a series of logistic regression models were used to assess the association of selected exposures and hyperglycemia adjusting for biological and social individual-level risk factors for diabetes. Of 77,882 ED patients without a history of diabetes presenting in 2021, 1,715 (2.2%) had severe hyperglycemia. Many geospatial exposures were associated with uncontrolled hyperglycemia, even after controlling for individual-level risk factors. The most strongly associated neighborhood-level variables included lower markers of educational attainment, higher percentage of households where limited English is spoken, lower rates of white-collar employment, and higher rates of Medicaid insurance. Including these geospatial factors in risk assessment models may help identify important subgroups of patients with undiagnosed disease.

Identifying Young Adults at High Risk for Weight Gain Using Machine Learning

IntroductionWeight gain among young adults continues to increase. Identifying adults at high risk for weight gain and intervening before they gain weight could have a major public health impact. Our objective was to develop and test electronic health record-based machine learning models to predict weight gain in young adults with overweight/class 1 obesity. MethodsSeven machine learning models were assessed, including three regression models, random forest, single-layer neural network, gradient-boosted decision trees, and support vector machine (SVM) models. Four categories of predictors were included: 1) demographics; 2) obesity-related health conditions; 3) laboratory data and vital signs; and 4) neighborhood-level variables. The cohort was split 60:40 for model training and validation. Area under the receiver operating characteristic curves (AUC) were calculated to determine model accuracy at predicting high-risk individuals, defined by ≥ 10% total body weight gain within 2 y. Variable importance was measured via generalized analysis of variance procedures. ResultsOf the 24,183 patients (mean [SD] age, 32.0 [6.3] y; 55.1% females) in the study, 14.2% gained ≥10% total body weight. Area under the receiver operating characteristic curves varied from 0.557 (SVM) to 0.675 (gradient-boosted decision trees). Age, sex, and baseline body mass index were the most important predictors among the models except SVM and neural network. ConclusionsOur machine learning models performed similarly and had modest accuracy for identifying young adults at risk of weight gain. Future models may need to incorporate behavioral and/or genetic information to enhance model accuracy.

Neighborhood Food Environment and Birth Weight Outcomes in New York City

Infants born with unhealthy birth weight are at greater risk for long-term health complications, but little is known about how neighborhood characteristics (eg, walkability, food environment) may affect birth weight outcomes. To assess whether neighborhood-level characteristics (poverty rate, food environment, and walkability) are associated with risk of unhealthy birth weight outcomes and to evaluate whether gestational weight gain mediated these associations. The population-based cross-sectional study included births in the 2015 vital statistics records from the New York City Department of Health and Mental Hygiene. Only singleton births and observations with complete birth weight and covariate data were included. Analyses were performed from November 2021 to March 2022. Residential neighborhood-level characteristics, including poverty, food environment (healthy and unhealthy food retail establishments), and walkability (measured by both walkable destinations and a neighborhood walkability index combining walkability measures like street intersection and transit stop density). Neighborhood-level variables categorized into quartiles. The main outcomes were birth certificate birth weight measures including small for gestational age (SGA), large for gestational age (LGA), and sex-specific birth weight for gestational age z-score. Generalized linear mixed-effects models and hierarchical linear models estimated risk ratios for associations between density of neighborhood-level characteristics within a 1-km buffer of residential census block centroid and birth weight outcomes. The study included 106 194 births in New York City. The mean (SD) age of pregnant individuals in the sample was 29.9 (6.1) years. Prevalence of SGA and LGA were 12.9% and 8.4%, respectively. Residence in the highest density quartile of healthy food retail establishments compared with the lowest quartile was associated with lower adjusted risk of SGA (with adjustment for individual covariates including gestational weight gain z-score: risk ratio [RR], 0.89; 95% CI 0.83-0.97). Higher neighborhood density of unhealthy food retail establishments was associated with higher adjusted risk of delivering an infant classified as SGA (fourth vs first quartile: RR, 1.12; 95% CI, 1.01-1.24). The RR for the association between density of unhealthy food retail establishments and risk of LGA was higher after adjustment for all covariates in each quartile compared with quartile 1 (second: RR, 1.12 [95% CI, 1.04-1.20]; third: RR, 1.18 [95% CI, 1.08-1.29]; fourth: RR, 1.16; [95% CI, 1.04-1.29]). There were no associations between neighborhood walkability and birth weight outcomes (SGA for fourth vs first quartile: RR, 1.01 [95% CI, 0.94-1.08]; LGA for fourth vs first quartile: RR, 1.06 [95% CI, 0.98-1.14]). In this population-based cross-sectional study, healthfulness of neighborhood food environments was associated with risk of SGA and LGA. The findings support use of urban design and planning guidelines to improve food environments to support healthy pregnancies and birth weight.

Socioeconomic disparities in fertility treatments and associated likelihood of livebirth following in vitro fertilization

To determine whether neighborhood-level socioeconomic characteristics are associated with the likelihood of livebirth (LB) following in vitro fertilization (IVF). Specifically, we evaluated neighborhood-level household income, unemployment rate, and educational attainment. A retrospective cross-sectional study was conducted for patients undergoing autologous IVF cycles. Large academic health system. For each patient, ZIP code of residence was used as a proxy for neighborhood. Neighborhood characteristics were compared between patients with and without LB. Generalized estimating model was used to adjust the association between SES factors and likelihood of a live birth with respect to relevant clinical factors. A total of 4942 autologous IVF cycles from 2768 patients were included: 1717 (62.0%) had at least one associated LB. Patients who achieved LB from IVF were younger, had higher anti-Mullerian hormone (AMH) levels, lower body mass index (BMI), and differed by ethnic background, primary language, and neighborhood socioeconomic characteristics. In a multivariable model, language, age, AMH, and BMI were associated with a live birth from IVF. None of the neighborhood-level socioeconomic variables were associated with the total number of IVF cycles or cycles required to achieve first LB. Patients living in neighborhoods with lower annual household income have lower odds of livebirth after IVF compared to those living in more affluent areas, despite undergoing the same number of IVF stimulation cycles.

Socio-ecologic Correlates of HIV/AIDS-Related Sexual Risk Behavior of African American Emerging Adults.

There is growing evidence showing that patterns of individual sexual risk behaviors are insufficient in explaining the disproportionate HIV/AIDS burden borne by African Americans. Instead, dynamic features of social, economic, political, and geographic contexts play a more determining role. However, not enough studies have examined the impact of multi-level factors including neighborhood-level influences on HIV/AIDS sexual risk among African American emerging adults using a socio-ecologic perspective. Anchored on the socio-ecologic framework, this study examines the collective role of relevant socio-ecologic determinants of sexual risk-taking among African American emerging adults. Results from both bivariate and multivariate analyses revealed that individual and neighborhood-level variables were significantly associated with sexual risk in our study population partially confirming the hypothesis of the study. Male gender, educational attainment, and neighborhood social disorder were the strongest predictors of sexual risk. Our findings contribute to the vast literature on sexual risk behavior patterns of young adults, and increasing evidence demonstrating the role of contextual factors as stronger predictors of sexual risk and HIV infection among at-risk youth. Our findings, however, underscore the need for further research on the pathways of HIV socio-behavioral vulnerability in this demographic group.