Neighborhood-level Interventions Research Articles

Abstract A058: Investigating the role of race/ethnicity on the association between neighborhood deprivation and breast cancer outcomes among Kentucky breast cancer patients years 2010-2022

Abstract Background: Despite advances in medical knowledge and treatment, disparities in breast cancer outcomes persist along demographic and socioeconomic lines. To quantify these disparities, the Area Deprivation Index (ADI) emerges as a powerful metric. It offers insights into the interplay between neighborhood socioeconomic status and breast cancer outcomes by encompassing factors pertaining to income, education, employment, and housing quality. There have been limited studies to examine the role of race and neighborhood disadvantage on specific breast cancer outcomes. Kentucky also faces a unique burden as it has the third highest breast cancer mortality rate in the United States. This study aims to examine the association between neighborhood socioeconomic disadvantage, and breast cancer outcomes including receipt of surgical treatment, radiation therapy, chemotherapy, and breast cancer-specific survival. While also examining how any associations are modified by race and ethnicity among Kentucky breast cancer patients. Methods: We conducted a retrospective cohort analysis using data from the Kentucky Cancer Registry (KCR) among breast cancer patients diagnosed between 2010 and 2017 with follow-up data through December 31, 2022. The KCR dataset was then linked with census tract data to examine ADI. We conducted consecutive multilevel logistic regression for binary outcomes and Cox Proportional hazard models using SAS (version 9.4, SAS Institute, Inc., Cary, North Carolina, USA) for time-to-event of breast cancer survival. Results: Women in the most disadvantaged quartile (4th quartile) were more likely to be Non-Hispanic Black (NH-Black), Appalachian residents, current tobacco users, divorced and using government insurance. Compared to women in the least disadvantaged neighborhoods (1st quartile) women living in the most disadvantaged neighborhoods (4th quartile) were 50% more likely to die from breast cancer (HR: 1.50, 95 CI: 1.28-1.76). Women in the most disadvantaged neighborhoods (4th quartile) were also more likely to be diagnosed at later stages (aOR:1.26, 95 CI:1.13-1.42). These associations were all observed after adjusting for age, race, tobacco use, marital status, insurance status, family history, and residence in Appalachia. Conclusions: Women living in areas of greater disadvantage had significantly increased odds of late-stage diagnosis and risk of breast cancer death regardless of race suggesting that neighborhood-level factors contribute to breast cancer disparities. Our findings underscore the need for neighborhood-level interventions to improve cancer outcomes. Citation Format: Breyanna E. Walker, Elinita Pollard, Justin Moore, Sydney Howard, Pamela Hull, Kathleen O'Connor, Samantha Jones, Lovoria Williams, Adebola Adegboyega, Veronica Jones, Eric Durban, Xiaoqin Wang, Wendi Owen. Investigating the role of race/ethnicity on the association between neighborhood deprivation and breast cancer outcomes among Kentucky breast cancer patients years 2010-2022 [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr A058.

Correlations between Hotspots of Child Maltreatment and Neighborhood-Level Interventions

Correlations between Hotspots of Child Maltreatment and Neighborhood-Level Interventions

Histories of neighborhood socioeconomic status contribute to race differences in later-life cognition.

Neighborhood characteristics are increasingly implicated in cognitive health disparities, but no research has investigated how the historical context of neighborhoods shapes these disparities. Four hundred sixty-four Black (55%) and White older adults (Mage=63.6) were drawn from the Michigan Cognitive Aging Project, a community-based, prospective study of older adults. Participants' addresses at baseline (2017-2020) were geocoded and linked to 2000-2017 measures of neighborhood socioeconomic status (NSES): disadvantage [NDis] and affluence [NAff]. Latent class growth analysis (LCGA) characterized 18 interpolated year trajectories of NSES across 1344 census tracts. Path analysis examined whether NSES trajectory classes mediated the association between race and a global cognition composite. LCGA identified three NDis and two NAff trajectory classes, which were associated with participant race. Only one NDis class was associated with cognition, and it mediated the association between the Black race and cognition. Disinvestment in neighborhoods may be particularly salient in race disparities in cognitive function. Race is implicated in the likelihood of living in more disadvantaged neighborhoods. Historical trends in neighborhood disadvantage are associated with cognitive function in older adulthood. Identifying patterns of neighborhood change may inform neighborhood-level interventions.

Abstract HUP14: Racial And Ethnic Disparities In Neighborhood Socioeconomic Status Associated With Cardiovascular Disease And Stroke: The Women’s Health Initiative

Introduction: The role of socioeconomic status (SES) across cardiovascular diseases (CVD) is well researched at the individual-level but fails to explain the greater CVD rates among Black and American Indian and Alaska Native (AIAN) groups at every level of SES. This research evaluated whether lower Neighborhood SES (NSES), a structural determinant of health, was associated with racial and ethnic disparities in CVD or stroke among aging women. Methods: The data came from the Women’s Health Initiative (WHI) cohort, consisting of 161,808 postmenopausal women with 30 years of data on stroke and CVD incidence. The NSES data were obtained from the US Census based on participant’s census tract of residence. We evaluated racial and ethnic disparities in the exposure and outcome prevalence using ANOVA. Associations between NSES and time to stroke and CVD were evaluated using multivariate adjusted Cox Regression models, and models were stratified by race and ethnicity to evaluate disparities. Results: NSES differed by race and ethnicity. Black, AIAN, and Latina groups were significantly (P<0.01) more likely to live in neighborhoods with NSES Z-scores below the population mean (Z-score=0) compared with White and Asian groups. The mean NSES Z-score among Black participants was -5.21 standard deviations below the population mean. Thirty-year CVD and stroke incidence followed the same trajectory of racial disparities. Preliminary analysis suggests results from Cox regression models will identify lower NSES to be longitudinally associated with greater stroke and CVD incidence overall, and greater effect sizes among groups experiencing structural racism. Conclusion: This research evaluates racial disparities in NSES as a structural determinant of stroke and CVD. These findings provide evidence emphasizing the need for neighborhood-level interventions to improve CVD disparities that coalesce in marginalized communities.

DO SLEEP QUALITY AND SOCIAL COHESION MEDIATE ASSOCIATION BETWEEN NEIGHBORHOOD DISORDER AND COGNITIVE FUNCTIONING?

Abstract Daily exposure to neighborhoods with high levels of disorder, such as public drinking, graffiti, and vandalism, has been linked to worse cognitive functioning and a higher risk for dementia in older adults. The present study examined the association between neighborhood disorder and cognitive functioning. This study also examined sleep quality and social cohesion as potential mediators of this association using Hayes’ (2013) Macro Process via bootstrapping method. The sample included cross-sectional data for 2,529 older adults (1,115 men and 1,482 women, age 70+) using the 2021 National Health and Aging Trends Study. Neighborhood disorder was measured as a 3-item scale capturing physical neighborhood characteristics completed by the interviewer (e.g., trash on sidewalks). The social cohesion scale measured participant agreement with 3 questions about their community (e.g., “People can be trusted”). The sleep quality scale measured participant responses to 3 questions about their sleep in the last month (e.g., taking 30+ minutes to fall asleep). Cognitive functioning was assessed using performance on tasks evaluating memory, orientation, and executive functioning. Results showed greater neighborhood disorder was associated with worse performance on cognitive tasks (B=-0.417; p<0.001). Results of mediation testing indicated that social cohesion was a significant mediator (B=0.224; p<0.0001). Though sleep quality was not a mediator, better sleep quality was associated with better cognitive functioning (B=0.053; p<0.05). Understanding the contribution of neighborhood disorder, sleep quality, and social cohesion to cognitive functioning will inform neighborhood-level interventions that can buffer cognitive decline in older adults aging in place.

Trusted residents and housing assistance to decrease violence exposure in New Haven (TRUE HAVEN): a strengths-based and community-driven stepped-wedge intervention to reduce gun violence

BackgroundWe describe the rationale and study design for “TRUsted rEsidents and Housing Assistance to decrease Violence Exposure in New Haven (TRUE HAVEN),” a prospective type 1 hybrid effectiveness/implementation study of a multi-level intervention using a stepped wedge design. TRUE HAVEN aims to lower rates of community gun violence by fostering the stability, wealth, and well-being of individuals and families directly impacted by incarceration through the provision of stable housing and by breaking the cycle of trauma.DesignTRUE HAVEN is an ongoing, multi-level intervention with three primary components: financial education paired with housing support (individual level), trauma-informed counseling (neighborhood level), and policy changes to address structural racism (city/state level). Six neighborhoods with among the highest rates of gun violence in New Haven, Connecticut, will receive the individual and neighborhood level intervention components sequentially beginning at staggered 6-month steps. Residents of these neighborhoods will be eligible to participate in the housing stability and financial education component if they were recently incarcerated or are family members of currently incarcerated people; participants will receive intense financial education and follow-up for six months and be eligible for special down payment and rental assistance programs. In addition, trusted community members and organization leaders within each target neighborhood will participate in trauma-informed care training sessions to then be able to recognize when their peers are suffering from trauma symptoms, to support these affected peers, and to destigmatize accessing professional mental health services and connect them to these services when needed. Finally, a multi-stakeholder coalition will be convened to address policies that act as barriers to housing stability or accessing mental healthcare. Interventions will be delivered through existing partnerships with community-based organizations and networks. The primary outcome is neighborhood rate of incident gun violence. To inform future implementation and optimize the intervention package as the study progresses, we will use the Learn As You Go approach to optimize and assess the effectiveness of the intervention package on the primary study outcome.DiscussionResults from this protocol will yield novel evidence for whether and how addressing structural racism citywide leads to a reduction in gun violence.Trial registrationClinicalTrials.gov Identifier: NCT05723614. Registration date: February 01, 2023. Please refer to https://clinicaltrials.gov/ct2/show/NCT05723614 for public and scientific inquiries.

Socioeconomic Disadvantage Is Associated With Risk and Severity of Anemia [ID: 1377641

INTRODUCTION: Anemia on admission for delivery is a significant contributor to peripartum transfusion. Risk of maternal morbidity including transfusion has been linked to neighborhood deprivation. We sought to explore the relationship between a composite measure of neighborhood-level deprivation and frequency and severity of anemia on admission for delivery. METHODS: We performed an IRB-approved retrospective cohort analysis of delivery admissions in a single health care network from 2015 to 2020. Area deprivation index (ADI) was used to represent neighborhood socioeconomic disadvantage and is a composite index of neighborhood that spans income, education, household characteristics, and housing. The index ranges from 1 to 100 with higher values indicating higher disadvantage. Anemia was defined as hemoglobin less than 11.0 mg/dL and severe anemia less than 9.0 mg/dL. Multivariable binomial regression models assessed the relationship between ADI and anemia. RESULTS: Eighty-five thousand five hundred fifty-three delivery admissions were included. A monotonic linear positive relationship was observed between ADI and anemia. Individuals who reside in the most disadvantaged neighborhoods (ADI>95) had 2.2 times the odds (95% CI 2.1–2.3) of anemia on admission for delivery. For example, individuals who live in a less disadvantaged neighborhood (ADI of 5) had a 11% predicted probability of anemia compared to 23% in individuals who lived in a more disadvantaged neighborhood (ADI of 95). Individuals who lived in the most disadvantaged neighborhoods had a higher rate of severe anemia (1% versus 3%, P<.001). CONCLUSION: Anemia is more likely and more severe in individuals living in disadvantaged neighborhoods. Neighborhood-level interventions could be used to target populations at high risk for anemia and potentially affect maternal outcomes.

NEIGHBORHOOD CHARACTERISTICS INFLUENCE OLDER ADULTS' HEALTH AND MENTAL HEALTH OUTCOMES

Abstract Both social and physical aspects of neighborhood characteristics are related to a wide range of health and mental health outcomes. There has been increasing evidence pointing to the link between neighborhood-level factors and health among older adults. Specifically, older adults living in disadvantaged neighborhoods with under-resourced infrastructure may experience more daily activity limitations, mental health symptoms, and increased morbidity and mortality. Positive aspects of the neighborhood, such as social cohesion, may serve as a social capital resource and protect against adverse health outcomes. On the contrary, negative aspects of the neighborhood, such as physical disorder, can be a substantial stressor leading to poor health. The neighborhood environment also disproportionally affects racial and ethnic minorities in the US. This symposium session will present four studies exploring important topics related to neighborhood factors of health among older adults. Collectively, the findings will inform neighborhood-level interventions to promote health and well-being among community-dwelling older adults. This session will start with a talk by Dr. Chan on the link between neighborhood and disability across six ethnic groups of older Asian Americans residing in New York City. Dr. Perry will present a qualitative study to explore the environmental and infrastructure challenges in the neighborhood from the perspective of older adults in Detroit. This will be followed by Dr. Jiang’s talk on the relationship between neighborhood cohesion and mortality among a sample of older Chinese in Chicago. Finally, the session will conclude with Dr. Qin’s presentation on how neighborhood characteristics affect older adults’ mental health trajectories.

Health outcomes in redlined versus non-redlined neighborhoods: A systematic review and meta-analysis

BACKGROUND AND AIM: Redlining was a racialized zoning practice in the U.S. that blocked fair access to home loans during the 1930s, and recent research is illuminating health problems in the current residents of these historically redlined areas. However, this work has not yet been holistically summarized. Here, we present the first systematic review and meta-analysis comparing health outcomes in redlined versus non-redlined neighborhoods in U.S. cities. METHODS: We extracted relevant articles in PubMed, Web of Science, Cochrane and Science Direct databases published from January 2010 to September 2021. RESULTS: The search revealed 12 studies on preterm births (n=3), gunshot-related injuries (n=2), cancer (n=1), asthma (n=1), self-rated health (n=1), multiple health outcomes (n = 2), heat-related outcomes (n=1) and COVID-19 incidence and mortality (n=1). A meta-analysis of three studies found the odds of having preterm birth was significantly higher (OR=1.41, 95% CI: 1.05-1.88; p=0.02) among women living in redlined areas compared to those in non-redlined areas. Review of other outcomes revealed that gunshot-related injuries, asthma, heat-related outcomes, and multiple chronic conditions were worse in redlined areas, while associations with cancer varied by cancer type. In terms of cause-specific mortality, one study revealed no link between residential redlining and infant mortality rate, while one study on COVID-19 outcomes was inconclusive. CONCLUSIONS: Overall, this review presents evidence that living in historically redlined areas is associated with increased risk of multiple serious adverse health outcomes. Further research on mechanisms, remediation, and neighborhood-level interventions is needed to strengthen the understanding of the impacts of redlining on health. KEYWORDS: Redlining; health inequities; structural racism; environmental justice; racial disparities; place-based disparities

Neighborhood-level out-of-hospital cardiac arrest risk and the impact of local CPR interventions

IntroductionIt is unclear how best to identify “high-risk” areas for out-of-hospital cardiac arrest (OHCA) and if neighborhood-level interventions improve bystander cardiopulmonary resuscitation (BCPR). Our objectives were to 1) identify and compare community characteristics between high and low-risk neighborhoods; and 2) examine change in BCPR after a targeted hands-only CPR intervention. MethodsThis was a cross-sectional analysis of OHCA events in Franklin County, Ohio between 1/1/2010–12/31/2017. Adult (≥18 years) OHCAs in a non-healthcare setting with emergency medical services resuscitation attempted were included. High-risk neighborhoods based on OHCA incidence and BCPR rates were identified using global Empirical Bayes, Local Moran’s I, and spatial scan statistic. We compared characteristics of high and low-risk neighborhoods and examined change in BCPR. ResultsFrom the 3,841 included OHCAs, the mean adjusted OHCA incidence per census tract was 0.81 per 1,000, BCPR rate was 37.2%, and survival to hospital discharge was 11.5%. Of the 35 census tracts identified as high-risk, ten persisted from previous work. OHCA incidence was higher in high-risk neighborhoods (1.30 per 1,000 vs. 0.73, p < 0.001) and BCPR rates were lower (30.2% vs. 38.5%, p < 0.001). There were significant differences in characteristics between high and low-risk neighborhoods (e.g., Black population: 45.3% vs. 25.7%, p < 0.001). The neighborhoods targeted for the community education intervention had similar pre- and post-intervention BCPR rates. ConclusionsDemographic and socioeconomic characteristics differed between high- and low-risk neighborhoods. BCPR rates were lower in high-risk neighborhoods despite a targeted BCPR intervention. Educational interventions may be necessary, but not sufficient, to improve OHCA outcomes.

Associations between neighborhood characteristics and dating violence: does spatial scale matter?

BackgroundDating violence (DV) is a public health problem that could have serious repercussions for the health and well-being of a large number of adolescents. Several neighborhood characteristics could influence these behaviors, but knowledge on such influences is still limited. This study aims at (1) evaluating the associations between neighborhood characteristics and DV, and (2) assessing how spatial scale influences the estimations of the latter associations.MethodsThe Québec Health Survey of High School Students (2016–2017) was used to describe DV. Neighborhoods were operationalized with polygon-based network buffers of varying sizes (ranging from 250 to 1000 m). Multiple data sources were used to describe neighborhood characteristics: crime rate, alcohol outlet density (on-premises and off-premises), walkability, greenness, green spaces density, and youth organizations density. Gendered-stratified logistic regressions were used for assessing the association between neighborhood characteristics and DV.ResultsFor boys, off-premises alcohol outlet density (500 m) is associated with an increase in perpetrating psychological DV. Crime rate (500 m) is positively associated with physical or sexual DV perpetration, and crime rate (250 m) is positively associated with physical or sexual DV victimization. Greenness (1000 m) has a protective effect on psychological DV victimization. For girls, walkability (500 m to 1000 m) is associated with a decrease in perpetrating and experiencing psychological DV, and walkability (250 m) is negatively associated with physical or sexual DV victimization.ConclusionsSeveral neighborhood characteristics are likely to influence DV, and their effects depend on the form of DV, gender, and spatial scale. Public policies should develop neighborhood-level interventions by improving neighborhood living conditions.

Identifying Geographic Areas of Washington, DC, With Increased Potential for Sexual HIV Transmission Among People With HIV With STIs and Concurrent Elevated HIV RNA: Data From the DC Cohort.

BackgroundThe Undetectable = Untransmittable (U = U) campaign advances the goal of ending the HIV epidemic by promoting durable viral suppression and therefore reducing sexual transmission. We used geospatial analysis to assess the potential for sexual HIV transmission by ZIP code of residence in the District of Columbia (DC) using data from the DC Cohort Longitudinal HIV Study (DC Cohort), a city-wide cohort of persons with HIV (PWH).MethodsDC Cohort participants aged ≥13 years were included in the study period between April 1, 2016, and March 31, 2018. Potential for sexual HIV transmission was defined as the proportion of participants with incident sexually transmitted infection (STI; gonorrhea, chlamydia, syphilis) and with HIV RNA ≥200 copies/mL from 9 months before to 3 months after STI diagnosis. We performed geographic information system (GIS) analysis to determine the ZIP codes with the highest potential for sexual HIV transmission.ResultsOf 3467 participants, 367 (10.6%) had at least 1 incident STI, with 89.4% residing in 11 of the 20 residential ZIP codes in DC. Of the 367 participants with an incident STI, at least 1 HIV RNA was available for 348 (94.8%). Ninety-seven (27.9%) individuals with an incident STI had HIV RNA ≥200 copies/mL in the defined time window. Of these 97, 66 (68.0%) resided in 5 of the 20 DC ZIP codes.ConclusionsIn DC, 5 ZIP codes of residence accounted for the majority of the estimated potential for HIV transmission among participants in the DC Cohort. These results support focused neighborhood-level interventions to help end the HIV epidemic.

Neighborhood Poverty and Incident Heart Failure: an Analysis of Electronic Health Records from 2005 to 2018.

Neighborhood-level characteristics, such as poverty, have been associated with risk factors for heart failure (HF), including hypertension and diabetes mellitus. However, the independent association between neighborhood poverty and incident HF remains understudied. To evaluate the association between neighborhood poverty and incident HF using a "real-world" clinical cohort. Retrospective cohort study of electronic health records from a large healthcare network. Individuals' residential addresses were geocoded at the census-tract level and categorized by poverty tertiles based on American Community Survey data (2007-2011). Patients from Northwestern Medicine who were 30-80 years, free of cardiovascular disease at index visit (January 1, 2005-December 1, 2013), and followed for at least 5 years. The association of neighborhood-level poverty tertile (low, intermediate, and high) and incident HF was analyzed using generalized linear mixed effect models adjusting for demographics (age, sex, race/ethnicity) and HF risk factors (body mass index, diabetes mellitus, hypertension, smoking status). Of 28,858 patients included, 75% were non-Hispanic (NH) White, 43% were men, 15% lived in a high-poverty neighborhood, and 522 (1.8%) were diagnosed with incident HF. High-poverty neighborhoods were associated with a 1.80 (1.35, 2.39) times higher risk of incident HF compared with low-poverty neighborhoods after adjustment for demographics and HF risk factors. In a large healthcare network, incident HF was associated with neighborhood poverty independent of demographic and clinical risk factors. Neighborhood-level interventions may be needed to complement individual-level strategies to prevent and curb the growing burden of HF.

Neighbourhood flOURISHing (NOURISH): A new short and inclusive interpersonal measure of subjective wellbeing

BackgroundShort individual Subjective Wellbeing (SWB) survey instruments are increasingly used to measure the impact of policy and community health initiatives. Neighbourhood-level interventions should affect social SWB, where intervention impact relates to a geographically bounded area. Existing social SWB measures lack spatial focus and have not been tested with low literacy users. The objective of the study was to develop and validate a short and inclusive neighbourhood flourishing tool. MethodsFive key stages, started with conceptualisation built on recent reviews and a community survey. Second, a set of survey items were operationalised. Third, cognitive interviews with six low education readers refined and reduced the pool. Fourth, these items plus new and established measures of SWB and predictors of both, were tested with 292 participants and subject to psychometric analyses. Fifth, the nine strongest items were validated in two deprived UK neighbourhoods with a further 221 participants and subjected to readability formulas. ResultsInitial work revealed 12 relevant concepts and 101 survey items assessing these were developed. Approximately half the new short items were acceptable in ‘think aloud’ interviews (n = 6). Two rounds of completion of items revealed a 2-factor solution consisting of neighbourhood feelings and functionings. The final solution consisted of three high loading items assessing feelings (safety, trust and respect) and three items assessing functionings (celebration, autonomy and participation), that did not cross-load, and demonstrated predictive and discriminant validity. Based on eight reading score formulas, the final measure was found to suit a reading age of 12–14 years old. ConclusionThe new valid six-item measure was tested with 519 participants, is acceptable and usable with people from deprived neighbourhoods, and with evidence of validity. This is a robust practical and inclusive tool for the measurement of Neighbourhood Flourishing, relevant to urban regeneration, inequalities and climate change policy and research.

Abstract 256: Outcome Prediction of Out of Hospital Cardiac Arrest with Environmental Information Using Machine Learning

Background: The lethality of out-of-hospital cardiac arrest (OHCA) among minority and low income patients in large urban centers has been extensively described. Neighborhood disparities in OHCA survival outcomes cannot be fully explained by clinical variables alone. Machine Learning (ML) is a sub-field of Artificial Intelligence where algorithms can learn and improve themselves by studying high volumes of data and evaluating interactions that affect OHCA. We sought to determine whether incorporating neighborhood variables into a ML model of OHCA can increase its predictive accuracy of survival outcomes. Methods: Using OHCA data from the Cardiac Arrest Registry to Enhance Survival (CARES) database, we geocoded Chicago incidents that occurred outside of healthcare facilities and nursing homes from 2014 - 2019 into the 77 Chicago community areas. Neighborhood characteristics (e.g., access to healthcare, educational attainment, crime rates) were assigned to OHCA cases based on community area using publicly available data from the Chicago Health Atlas (CHA). The merged CARES and CHA data were used to train, validate, test, and analyze ML models for OHCA survival outcomes, defined as a binary outcome based on Cerebral Performance Category or CPC (Good = CPC 1 or 2; Poor=CPC 3, 4, or 5). ML models including deep neural networks were developed and compared using average class sensitivity to models developed using only CARES data. Results: ML algorithms were tested on 9,595 cases of OHCA. Baseline results using only CARES data on the Embedded Fully Convolutional Network (EFCN) show an average class sensitivity of 84%. Including neighborhood variables from the CHA dataset increased the average class sensitivity of the EFCN model to 86.5-86.9%. Conclusion: The use of neighborhood variables in combination with CARES data can predict neurologic outcomes with a higher level of accuracy than using CARES registry data alone. Future studies will use this newly developed model to perform simulation experiments to identify neighborhood level interventions to improve OHCA survival outcomes.

Neighborhood-level stroke and its neighborhood-level determinants: a Bayesian machine learning approach

The substantial health and economic burden of stroke has prompted numerous studies examining the risk factors of stroke using individual patient level data. Increasingly, place-based evidence is recognized as a critical part of stroke management, but descriptions of risk factors and their associations with stroke at the same neighborhood level smaller than counties are lacking.We applied a novel Bayesian machine learning approach to an integrated data set consisted of 24 potential variables to identify factors with high predictive power for stroke and quantify the associations between predictors and stroke prevalence. Both the stroke outcome and its predictors were measures at neighborhood level, defined as census tracts (n=26697) from 500 large cities in the United States. Variable selection was conducted using Bayesian Additive Regression Trees (BART)-Machine, and compared with conventional models: step-wise regression and regression with all predictors. The exposure-outcome associations were further assessed using Bayesian linear regression. We identified six high importance tract-level predictors for tract-level stroke prevalence. They were prevalence of no leisure-time physical activity (%LPA), proportions of population who were aged 65 years and above (%OlderAdults) and who were non-Hispanic black (%NHB), median household income, and ozone level, as well as the interaction term between %LPA and %OlderAdults. Among these factors, %OlderAdults, %NHB, %LPA, and ozone were positively associated with stroke prevalence, while median household income was inversely associated with stroke prevalence. The interaction term showed an exacerbated adverse effect of aging population structure and low physical activity on tract-level stroke prevalence.High-performance machine learning identified the most important determinants of neighborhood-level cardiovascular health from a wide-ranging variables in an agnostic, data-driven and reproducible way. The identified neighborhood-level predictors were consistent with known patient-level risk factors. The results can be used to prioritize and allocate resources to develop targeted neighborhood-level interventions for stroke prevention and control.

Risk Factors for Physical Inactivity Among Children With and Without Asthma Living in Peri-Urban Communities of Lima, Peru.

The authors sought to examine physical activity patterns among children with and without asthma in 2 peri-urban communities in Lima, Peru, to identify socioeconomic and demographic risk factors for physical inactivity and examine the relationship between asthma and physical activity. The authors measured mean steps per day in 114 children (49 with asthma and 65 without) using pedometers worn over a 1-week period. They also used the 3-day physical activity recall to determine the most common activities carried out by children. The authors found that 84.2% of the children did not meet the daily international physical activity recommendations. Girls took significantly fewer mean steps per day as compared with boys (2258 fewer steps, 95% confidence interval, 1042-3474), but no other factors, including asthma status, showed significant differences in the mean daily steps. Mean daily steps were positively associated with higher socioeconomic status among girls, and current asthma had a larger inverse effect on daily steps in boys when compared with girls. Physical activity levels were below recommended guidelines in all children. There is a need for policy and neighborhood-level interventions to address low physical activity levels among Peruvian youth. Special focus should be given to increasing the physical activity levels in girls.

A Latent Class Analysis of Perceived Neighborhood Conditions Associated with Mental Disorders Among Children in the United States.

The current study examined the association between perceived neighborhood conditions and common childhood mental disorders in a nationally representative sample of children in the U.S. The datawere derived from the 2017 National Survey of Children's Health, including American children aged 6-17years (N = 15,438). Latent class analysis was used to identify subtypes of perceived neighborhood conditions regarding neighborhood physical environment, social capital, and violence. Three classes were identified: Ideal Neighborhood (55.99%); Insufficient Assets (27.38%), and Broken and Unsafe Neighborhood (16.63%). The effects of latent classes on psychiatric outcomes (i.e. attention deficit hyperactivity disorder, depression, anxiety, conduct problem, and any of these four disorders) were examined. Class membership was differentially associated with the mental disorders after adjustment for demographic variables, food insufficiency, and guardian's mental health. The Broken and Unsafe Neighborhood class was associated with greater odds of all childhood psychiatric disorders than the Ideal Neighborhood and Insufficient Assets class. Insufficient Assets class was associated with greater odds of all childhood psychiatric disorders than the Ideal Neighborhood class. The findings suggest that neighborhood-level interventions to decrease children's mental health burdens are critically needed.

Associations among Early Exposure to Neighborhood Disorder, Fathers’ Early Involvement, and Children’s Internalizing and Externalizing Problems

ABSTRACT Purpose This study aimed to examine (1) the effects of early exposure to neighborhood disorder and fathers’ early involvement on children’s long-term internalizing and externalizing problems, and (2) whether fathers’ early involvement buffered effects of early exposure to neighborhood disorder on children’s internalizing and externalizing problems. Method We used five waves of Fragile Family and Child Wellbeing study data and conducted multi-level longitudinal mixed-effects models to examine relationships among early exposure to neighborhood disorder, fathers’ early involvement, and children’s internalizing and externalizing problems. Results Results indicated that early exposure to neighborhood disorder was associated with increased children’s internalizing and externalizing problems, while fathers’ early involvement was associated with decreased children’s internalizing and externalizing problems. However, fathers’ early involvement did not buffer the negative effects of early exposure to neighborhood disorder on children’s internalizing and externalizing problems. Discussion The findings suggest the importance of neighborhood order and fathers’ early involvement in decreasing children’s internalizing and externalizing problems. Developing neighborhood-level interventions and improving fathers’ involvement in early childhood are potential strategies to prevent children’s behavioral problems in the long term.

Preferences Regarding a Web-Based, Neighborhood-Level Intervention Program to Promote Household Energy Conservation

ABSTRACT The aim of this study is to investigate people’s willingness to participate in and their preferences regarding a web-based, neighborhood-level energy conservation program. We created a hypothetical program that incorporates several intervention strategies and presented it to 715 respondents in the form of a stated choice experiment. A strong majority of participants indicated that they would be willing to join such a program. Their preferences as to how the program should be constructed reflected interest in personalized feedback, privacy protection, monetary incentives for reducing energy use, and a relatively short-term initial commitment. However, there was variation in people’s taste for such attributes. We conclude with reflections on how to design and implement a web-based intervention program and on future research possibilities.