Food Environment Index Research Articles

Social vulnerability and disparities in incidence of diagnosed diabetes in the united states, 2004 to 2019

Abstract Introduction Vulnerable populations experience a higher burden of cardiovascular disease.(1) However, the incidence of cardiovascular risk factors, such as diabetes, within these populations, is incompletely characterized. Recently, the United States Centers for Disease Control and Prevention (US CDC) developed the Social Vulnerability Index (SVI) to categorize counties that may be vulnerable to worse health outcomes. The SVI is composed of 4 themes: socioeconomic status (Theme 1), household composition & disability (Theme 2), minority status & language (Theme 3), and housing type & transportation (Theme 4).(2) We sought to evaluate the association of the 4 SVI themes with county incidence of diagnosed diabetes. Method We used CDC data (2004-2019; reported for 94% of US counties), on county incidence of diagnosed diabetes. The SVI themes were scored from 0 to 1 (low to high vulnerability) based on national percentile ranking. We used weighted least squares regression with restricted cubic splines to model the relationship between SVI themes and incidence of diagnosed diabetes (median data 2004−2019), adjusting for median age, sex ratio, state, percent uninsured, food environment index, healthcare providers, and rurality.(3) The partial effects of SVI themes and diabetes incidence were reported as F-statistics (P-values) and diabetes incidence rate ratios (IRR [95% confidence interval]), which were plotted for all values compared with the 50th percentile (SVI 0.5). For each theme, we also present the IRR for a 0.5-point increase when moving over the lower (SVI score 0.5 vs. 0.0) and the upper (1.0 vs. 0.5) half of the SVI score distribution. Results The 4 SVI theme rank scores were significantly associated with diabetes incidence (all, P<.001). However, Theme 1 and 2 scores showed stronger associations with IRR than Theme 3 and 4 (Figure). For Theme 1, SVI score 0.5 vs. 0.0 was associated with a 15% higher incidence (IRR 1.15 [1.13–1.18]) and score 1.0 vs. 0.5 was associated with a 13% higher incidence (IRR 1.13 [1.11–1.16]). For Theme 2, SVI score 0.5 vs. 0.0 was associated with a 16% higher incidence (IRR 1.16 [1.14–1.18]), but there was no difference between rates for SVI score 1.0 vs. 0.5 (IRR 1.01 [0.99–1.03]). For Theme 3, there was no difference between SVI score 0.5 vs. 0.0 (IRR 0.98 [0.96–1.00]), but rates were 8% lower for SVI score 1.0 vs. 0.5 (IRR 0.92 [0.90–0.94]). For Theme 4, SVI score 0.5 vs. 0.0 was associated with a 3% lower incidence (IRR 0.97 [0.96–0.99]) and score 1.0 vs. 0.5 was associated with a 2% higher incidence (IRR 1.02 [1.01–1.04]). Conclusion In US counties, the strength and directionality of the association between SVI themes and diabetes incidence differed by theme and level of SVI scores. These themes, particularly socioeconomic status and household composition & disability may be leveraged to identify high-risk counties for primary prevention of diabetes, with implications for reduction in cardiovascular risk.

Residential food environment and dietary antioxidant consumption among pregnant women: Modifying effects of socioeconomic factors and ethnic identity

BackgroundResidential food environment influences dietary patterns, however the impact of individuals' perceived social identity on the antioxidant intake, an indicator of healthy dietary patterns, remains underexplored. ObjectiveWe conducted a cross-sectional analysis using data collected in a longitudinal cohort. In the study, we aim to investigate the interactions between the food environment and two indicators of social identity, specifically a subjective ladder measure of socioeconomic status (SES) and the Multigroup Ethnic Identity Measure (MEIM) score, in relation to dietary antioxidant intake (DAI) among n = 512 Boston and NYC-resident pregnant women. MethodsThe modified Retail Food Environment Index (mRFEI) was calculated using the Centers for Disease Control and Prevention's equation, with higher scores indicating a healthier food environment. DAI was estimated by summing standardized data from six micronutrients (magnesium, selenium, zinc, and vitamins A, C, and E) obtained through the Block98 Food Frequency Questionnaires administered during pregnancy, with higher scores indicating increased intake. The mRFEI and DAI were dichotomized based on a median split. Multivariable-adjusted logistic regressions were used to analyze associations, both with and without considering women's subjective SES or MEIM levels as effect modifiers. ResultsWomen were racially/ethnically mixed (19.2 % White, 42.7 % Black, and 33.1 % Hispanic) with 32.2 % reporting more than high school education. In the main effect models, no significant association was observed between mRFEI and DAI. Women with higher MEIM scores exhibited higher DAI [Odds ratio (OR) =1.85, 95 % Confidence interval (CI) = 1.26–2.73]. Exploratory interaction models showed that subjective SES significantly modified the association (p-value for interaction = 0.03), women perceiving themselves to have a lower SES compared to their community (n = 45) exhibited a significantly positive association between mRFEI and DAI. ConclusionThese findings suggest that women perceiving their SES to be lower than their neighborhoods may benefit from better access to healthy food.

Abstract A040: Effects of food environment, income inequality, and racial segregation on obesity- related cancer mortality

Abstract BACKGROUND: Rural areas experience 7-12% higher risk of obesity-related cancer mortality compared to their urban counterparts. Complex interactions between social and structural factors in rural areas, combined with limited access to healthy foods and lower income, warrant further study to understand drivers of this disparity. Methods: County-level rates of age-adjusted obesity-related cancer mortality per 100,000 (2009-2020 were obtained from the Centers for Disease Control and Prevention’s Wide-Ranging ONline Data for Epidemiologic Research database. Based on data from the Food Environment Atlas, we calculated food swamp using the modified Retail Food Environment Index (mRFEI) as the ratio of fast-food restaurants and convenience stores to grocery and specialized food stores, supercenters, and farmers markets. Food swamp scores were then divided into quartiles. Next, we used Global G and local Gi statistics to identify the presence of spatial autocorrelation in food swamp and subsequently calculated a spatial moving average (SMA) across counties. Measures of economic segregation, racial segregation, and racialized economic segregation were calculated using the Index of Concentration at the Extremes (ICE) using population and demographic estimates from the 2005- 2009 American Community Survey. Larger values reflect concentrated extremes of disadvantage. Poisson regression was used to assess the impact of food swamp, urban-rural status, economic segregation, racial segregation, and racialized economic segregation on rates of obesity-related cancer mortality. Results: Among the 3014 counties included, median food swamp score was 3.6 (interquartile range=2.6-4.9), and counties in the 2nd, 3rd, and 4th quartiles for food swamp had 2%, 4%, and 6% higher rates (respectively) of obesity-related cancer mortality than those in the 1st quartile. Incidence rate ratios for the association of food swamp quartile with obesity-related cancer mortality were 1-3% smaller after control for food swamp SMA and for racialized economic Urban-rural differences in obesity-related cancer mortality varied with all ICE measures. Rural counties had 5% higher rates of obesity-related cancer mortality (IRR=1.05, 95%CI=1.04-1.06), even after controlling for food swamp and food swamp SMA, but there was no longer an association after controlling for economic segregation (IRR=0.99, 95%CI=0.98-1.00). Rural counties had 5% higher rates of obesity-related cancer mortality (IRR=1.05, 95%CI=1.04-1.07) when controlling for racial segregation and 2% higher rates when controlling for racialized economic segregation (IRR=1.02, 95%CI=1.01-1.03). Thus, we stratified our models by urban-rural status and found that each ICE measure had larger effects on obesity-related cancer mortality in rural counties than in urban. CONCLUSION: While food swamp is important, the socioeconomic environment of a county has a larger impact on obesity- related cancer mortality. Individuals may have access to healthy food stores, but whether they can afford the healthy food is more important. Citation Format: Elizabeth S. Davis, Sara Myers, Jeffrey A. Franks, Michael R. Poulson, Julie R. Palmer, Kelly M. Kenzik. Effects of food environment, income inequality, and racial segregation on obesity- related cancer mortality [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 A040.

The food environment and hypertension: A cross-sectional analysis in Black breast cancer survivors in Maryland

ObjectivesThe Food Environment Index (FEI) has shown varying positive impacts on health outcomes related to diabetes, obesity, and hypertension. However, a relationship between FEI and hypertension among breast cancer (BC) survivors, particularly Black women survivors, remains underexplored. Black women who are BC survivors have a high prevalence of hypertension and increased risk of mortality compared to White women with BC. Our analysis aims to fill this gap by assessing the FEI's association with hypertension in this population. DesignUtilizing social media recruitment strategies and BC survivor networks, 100 Black female BC survivors completed an online survey, that included sociodemographic and clinical characteristics as well as lifestyle factors. The 2023 FEI County Health Rankings was used to assess the food environment and the index ranges from 0 (worst) to 10 (best). Adjusted prevalence odds ratios (PORs) and 95 % confidence intervals (CI) were calculated for the associations between FEI, sociodemographic and clinical factors, and hypertension status. ResultsAmong the 94 study participants with data on hypertension status, 54.3 % reported a diagnosis of hypertension. Residing in counties with a below-median FEI (<8.8 v. above median: ≥8.8) was significantly associated with hypertension (POR = 4.10, 95 % CI: 1.19–14.13). Age at survey (≥50 years compared to <50 years: POR= 0.29, 95 % CI: 0.10–0.87) and household income ($75,000-$99,999 compared to > $99,999/year: POR = 12.02, 95 % CI: 2.08–69.43) were also significantly associated with hypertension. ConclusionOur study highlights the potential impact of the food environment on hypertension among Black BC survivors living in Maryland. Our findings call attention to the need for targeted interventions to improve food accessibility and quality in underserved communities, especially for special populations such as cancer survivors.

Abstract PO3-27-07: Age at diagnosis and the food environment are associated with postdiagnosis weight gain among Black American breast cancer survivors in Maryland

Abstract Background: Weight management is now included in the American Cancer Society/American Society of Clinical Oncology Breast Cancer Survivorship Care Guidelines due to its clinical impact on breast cancer (BC) survivorship outcomes. While numerous individual-level risk factors for weight gain after BC diagnosis have been identified, such as younger age at diagnosis, cancer stage, and cancer treatments (chemotherapy), there are limited data on the impact of neighborhood-level factors associated with weight change postdiagnosis among Black BC survivors. Methods: Utilizing social media recruitment strategies and survivor networks, we recruited 100 Black female BC survivors to complete an online survey, including demographic, clinical characteristics, and lifestyle factors, between January 5, 2022, and August 18, 2022. To capture the food environment, we utilized the 2023 Food Environment Index County Health Rankings, which accounts for access to healthy foods due to proximity and income (considering the distance an individual lives from a grocery store or supermarket, locations for health food purchases in most communities, and the inability to access healthy food because of cost barriers). The index ranges from 0 (worst) to 10 (best). Our outcome of postdiagnosis weight gain was calculated as percent weight change from time at diagnosis to time of survey, calculated as ([weight at survey – weight at diagnosis]/weight at diagnosis) × 100. Participants were grouped into mutually exclusive categories of stable weight (within ± 3%) or weight loss (≤-3%) compared to weight gain (≥ 3%). Adjusted odds ratios (aORs) and 95% confidence intervals (CI) were calculated for the associations between clinical factors, county-level food environment index, and postdiagnosis weight gain. Results: The average ages at time of survey and at primary BC diagnosis were 58.6 years (standard deviation [SD]= 10.1) and 49.1 years (SD=10.2), respectively. Most women (70%) were diagnosed with invasive BC and only 22% were currently undergoing treatment at the time of the survey. While only 7% reported being obese at the time of BC diagnosis, 54% reported being obese at the time of survey which was on average 9 years post-BC diagnosis; and while 61 women had weight loss or stable weight, 34 women experienced weight gain postdiagnosis. Among the 95 women included in multivariable models who reported weight and height measurements for calculating weight change, women ≥50 years of age at BC diagnosis (versus &amp;lt; 50 years) were 3.43 times more likely to gain weight after diagnosis (95% CI 1.13-11.50). Women living in counties with a Food Environment Index ≥8.8 (median cut-point compared to &amp;lt; 8.8) were significantly less likely to experience weight gain (aOR 0.11; 95% CI 0.01-0.90). Stage and BC treatments were not significantly associated with weight gain. Conclusions: Our study findings demonstrate the importance of evaluating the food environment during BC survivorship and longitudinal monitoring of weight postdiagnosis to mitigate weight gain among Black BC survivors. Our findings will be used to inform a larger prospective study and future interventions among this population. Citation Format: Avonne Connor, Katherine Ho, Kate Dibble, Kala Visvanathan. Age at diagnosis and the food environment are associated with postdiagnosis weight gain among Black American breast cancer survivors in Maryland [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO3-27-07.

Neighborhood physical environments and change in cardiometabolic risk factors over 14 years in the study of Women's health across the nation

BackgroundNeighborhood physical environments may influence cardiometabolic health, but prior studies have been inconsistent, and few included long follow-up periods. MethodsChanges in cardiometabolic risk factors were measured for up to 14 years in 2830 midlife women in the Study of Women's Health Across the Nation, a multi-ethnic/racial cohort of women from seven U.S. sites. Data on neighborhood food retail environments (modified Retail Food Environment Index) and walkability (National Walkability Index) were obtained for each woman's residence at each follow-up. Data on neighborhood access to green space, parks, and supermarkets were available for subsets (32–42%) of women. Models tested whether rates of change in cardiometabolic outcomes differed based on neighborhood characteristics, independent of sociodemographic and health-related covariates. ResultsLiving in more (vs. less) walkable neighborhoods was associated with favorable changes in blood pressure outcomes (SBP: −0.27 mmHg/year, p = 0.002; DBP: −0.22 mmHg/year, p < 0.0001; hypertension status: ratio of ORs = 0.79, p < 0.0001), and small declines in waist circumference (−0.09 cm/year, p = 0.03). Small-magnitude associations were also observed between low park access and greater increases in blood pressure outcomes (SBP: 0.37 mmHg/year, p = 0.003; DBP: 0.15 mmHg/year, p = 0.04; hypertension status: ratio of ORs = 1.16, p = .04), though associations involving DBP and hypertension were only present after adjustment for sociodemographic variables. Other associations were statistically unreliable or contrary to hypotheses. ConclusionNeighborhood walkability may have a meaningful influence on trajectories of blood pressure outcomes in women from midlife to early older adulthood, suggesting the need to better understand how individuals interact with their neighborhood environments in pursuit of cardiometabolic health.

The association between neighborhood obesogenic factors and prostate cancer risk and mortality: the Southern Community Cohort Study.

Prostate cancer is one of the leading causes of cancer-related mortality among men in the United States. We examined the role of neighborhood obesogenic attributes on prostate cancer risk and mortality in the Southern Community Cohort Study (SCCS). From the total of 34,166 SCCS male participants, 28,356 were included in the analysis. We assessed the relationship between neighborhood obesogenic factors [neighborhood socioeconomic status (nSES) and neighborhood obesogenic environment indices including the restaurant environment index, the retail food environment index, parks, recreational facilities, and businesses] and prostate cancer risk and mortality by controlling for individual-level factors using a multivariable Cox proportional hazards model. We further stratified prostate cancer risk analysis by race and body mass index (BMI). Median follow-up time was 133 months [interquartile range (IQR): 103, 152], and the mean age was 51.62 (SD: ± 8.42) years. There were 1,524 (5.37%) prostate cancer diagnoses and 98 (6.43%) prostate cancer deaths during follow-up. Compared to participants residing in the wealthiest quintile, those residing in the poorest quintile had a higher risk of prostate cancer (aHR = 1.32, 95% CI 1.12-1.57, p = 0.001), particularly among non-obese men with a BMI < 30 (aHR = 1.46, 95% CI 1.07-1.98, p = 0.016). The restaurant environment index was associated with a higher prostate cancer risk in overweight (BMI ≥ 25) White men (aHR = 3.37, 95% CI 1.04-10.94, p = 0.043, quintile 1 vs. None). Obese Black individuals without any neighborhood recreational facilities had a 42% higher risk (aHR = 1.42, 95% CI 1.04-1.94, p = 0.026) compared to those with any access. Compared to residents in the wealthiest quintile and most walkable area, those residing within the poorest quintile (aHR = 3.43, 95% CI 1.54-7.64, p = 0.003) or the least walkable area (aHR = 3.45, 95% CI 1.22-9.78, p = 0.020) had a higher risk of prostate cancer death. Living in a lower-nSES area was associated with a higher prostate cancer risk, particularly among Black men. Restaurant and retail food environment indices were also associated with a higher prostate cancer risk, with stronger associations within overweight White individuals. Finally, residing in a low-SES neighborhood or the least walkable areas were associated with a higher risk of prostate cancer mortality.

Weight Loss Maintainers Sustain High Diet Quality in Diverse Residential Retail Food Environments

BackgroundThe relationship between the retail food environment and diet quality has received minimal investigation among weight loss maintainers. ObjectiveThe aim of this study was to investigate the association between the residential retail food environment and diet quality in weight loss maintainers from WeightWatchers in the United States. DesignCross-sectional data were collected between January 2018 and February 2020. The Retail Food Environment Index (RFEI), based on geocoded home addresses, classified the environment as follows: RFEI <1.6 = healthiest; RFEI 1.6 to <2.5 = moderately healthy; RFEI 2.5 to <4.0 = moderately unhealthy; RFEI ≥4.0 = least healthy. Dietary data were obtained using a food frequency questionnaire. Participants/SettingAdult participants (n = 1,159) who had lost weight using WeightWatchers and maintained ≥9.1-kg weight loss for ≥1 year (mean 24.7-kg loss for 3.4 years). Main Outcome MeasuresHealthy Eating Index 2015 (HEI-2015) component and total scores (0-100; higher scores indicate better alignment with the 2015-2020 Dietary Guidelines for Americans). Statistical Analyses PerformedRegression models included RFEI category, the independent variable, and HEI-2015 and component scores (outcomes) controlling for age, sex, race and ethnicity, educational attainment, and household income. ResultsCompared with individuals living in the healthiest food environments (mean HEI-2015 score = 71.5) those in the unhealthiest environments had a mean HEI-2015 score of 70.1 (95% CI 68.8 to 71.3), those in moderately unhealthy environments had a score of 71.3 (95% CI 70.3 to 73.1) and those in moderately healthy environments had a score of 70.3 (95% CI 68.9 to 71.2), indicating a nonlinear relationship. Compared with those in the healthiest environments, those in the least healthy environments had an approximately 0.47 lower added sugar HEI-2015 component score (95% CI –0.86 to –0.08), indicating approximately 5% higher added sugar intake. ConclusionsWeight loss maintainers maintained high diet quality in diverse retail food environments. Compared with those in the healthiest food environments, those in the least healthy had a higher consumption of added sugars.

Influence of community food environment on adult body mass index (BMI): a systematic review

To verify the association between community food environment and Body Mass Index (BMI) of adults. Systematic review conducted in EMBASE, PubMed, and Web of Science databases, considering the period from 2010 to 2022. Out of 10,407 articles, 24 observational studies were eligible according to the inclusion criteria. The methodological approaches were evaluated using STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) and OSQE (Observational Study Quality Evaluation). The protocol was registered at PROSPERO (number 42021260594). Most studies reported that BMI tends to increase with proximity to and a greater number of supermarkets, fast-food establishments, and convenience stores. The prevalence of adults with BMI greater than 25 kg/m2 was higher in locations with lower socioeconomic status. BMI was lower in more financially advantaged neighborhoods near grocery stores and fruit and vegetable markets. The selected studies indicate that a community food environment with higher availability of unhealthy foods is related to high BMI. The socioeconomic level can worsen this association, showing that people in social vulnerability have more difficulty accessing healthy food.

What North American retail food environment indices miss in Guatemala: Cultural considerations for the study of place and health

We evaluated the cross-context validity and equivalence of the US- and Canada-originated Retail Food Environment Index (RFEI) and modified RFEI (mRFEI) against a retail food environment dataset from the indigenous-majority city of Quetzaltenango (Xela), Guatemala. The RFEI/mRFEI failed to identify 77% of retailers and misclassified the healthiness of 42% of the remaining retailers in Xela, inaccurately labeling the city a food swamp. The RFEI/mRFEI are not currently suitable for mapping retail food environments in places like Quetzaltenango. Alternative functional and temporal classifications of retail food environments may provide measures with greater contextual fit, highlighting important cultural considerations for the study of place and dietary health.

Deserts, Swamps and Food Oases: Mapping around the Schools in Bahia, Brazil and Implications for Ensuring Food and Nutritional Security.

Deserts, swamps and food oases terms have been used to characterize the food environment according to the identified food availability. Food swamps are defined as neighborhoods with a predominance of food establishments selling ultra-processed foods compared to establishments selling healthy options. In contrast, food oases are areas with easy access to healthy and nutritious foods, such as fruits, vegetables, and other fresh foods. Food deserts describe densely populated urban areas where residents face difficulty obtaining healthy food. In this context, this work aimed to map deserts, swamps, and food oases around federal schools in Bahia, Brazil, emphasizing the importance of implementing the Brazilian National School Feeding Program (PNAE) in these schools, to guarantee food security and nutrition. An ecological study was carried out in all 35 federal schools in Bahia, Brazil, using an 800 m buffer analysis, with the school as the centroid. The geographic coordinates of schools and food establishments were initially obtained using Google Maps and later confirmed onsite. To evaluate food deserts and swamps, the methodology proposed by the CDC was used and the Modified Retail Food Environment Index (mRFEI) was calculated; when the result was equal to zero, the surrounding area was considered a food desert and values between 0.01 and 20 determined food swamps; for values above 20, the neighborhood was classified as mixed. Food oases were considered regions containing at least one supermarket within the analyzed buffer. Descriptive analyses were carried out with frequency measurements, measures of central tendency (mean and median) and dispersion (standard deviation). The food environment of schools was compared considering the number of students impacted, the area where the school was located (urban or rural) and the size of the municipalities. The average number of food establishments found was 22.39 (±13.03), with the highest averages for snack bars (7.33 ± 4.43), grocery stores (5.83 ± 4.09) and restaurants (2.94 ± 2.19). Food deserts and mixed environments were identified in 40% of the sample, while swamps represented 20% and oases 65%. An association was observed between food deserts and social vulnerability, making it necessary to emphasize the importance of adequate implementation of the PNAE in these schools to reduce food and nutritional insecurity, guaranteeing the human right to adequate and healthy food and providing better nutrition and health perspectives within the school environment and impact on students' lives through food and nutrition education actions, which are also part of the context of PNAE activities.

The Retail Food Environment Index and its association with dietary patterns, body mass index, and socioeconomic position: A multilevel assessment in Mexico.

In Mexico, 75% of the population are affected by overweight or obesity, and the availability and affordability of high-calorie-dense foods and beverages are high. This study tested the association between the retail food environment index (RFEI), dietary patterns, body mass index (BMI), and socioeconomic position (SEP) in Mexico. Cross-sectional diet, health, and sociodemographic population-based secondary data analyses were conducted. The RFEI was calculated by dividing the total number of fast-food outlets and convenience stores by the total number of supermarkets and fruit and vegetable stores per census tract area. Associations between BMI, dietary patterns, SEP and the RFEI were tested using multilevel linear regression, including interactions of the RFEI with SEP, gender, and age. Living in neighbourhoods with a higher RFEI was associated with a 0.01kg/m2 higher BMI (β = 0.01, 95%CI: 0.0005, 0.02, p = 0.04), equivalent to a mean 0.046 weight gain for a 1.60m tall person per 10% higher RFEI. Unhealthy dietary patterns were more likely in neighbourhoods with a higher RFEI (β = 0.100, 95%CI: 0.03, 0.12, p = 0.001). Multilevel linear regression showed that lower SEP households had a higher RFEI compared to higher SEP households (β = 0.020, 95% CI: -0.006 to 0.04, p = 0.10). Generalised structural equation models revealed a graded relationship between RFEI and SEP, showing that lower SEP households were exposed to a higher RFEI (β = 0.060, 95% CI: 0.05 to 0.07, p < 0.001.) The study identified significant associations between higher proportions of fast-food outlets and convenience stores, higher BMI, and unhealthy dietary patterns. It was particularly evident that low-income populations are more likely to be exposed to obesogenic food environments.

Predicting incident cardiovascular disease among African-American adults: A deep learning approach to evaluate social determinants of health in the Jackson heart study.

The present study sought to leverage machine learning approaches to determine whether social determinants of health improve prediction of incident cardiovascular disease (CVD). Participants in the Jackson Heart study with no history of CVD at baseline were followed over a 10-year period to determine first CVD events (i.e., coronary heart disease, stroke, heart failure). Three modeling algorithms (i.e., Deep Neural Network, Random Survival Forest, Penalized Cox Proportional Hazards) were used to evaluate three feature sets (i.e., demographics and standard/biobehavioral CVD risk factors [FS1], FS1 combined with psychosocial and socioeconomic CVD risk factors [FS2], and FS2 combined with environmental features [FS3]) as predictors of 10-year CVD risk. Contrary to hypothesis, overall predictive accuracy did not improve when adding social determinants of health. However, social determinants of health comprised eight of the top 15 predictors of first CVD events. The social determinates of health indicators included four socioeconomic factors (insurance status and types), one psychosocial factor (discrimination burden), and three environmental factors (density of outdoor physical activity resources, including instructional and water activities; modified retail food environment index excluding alcohol; and favorable food stores). Findings suggest that whereas understanding biological determinants may identify who is currently at risk for developing CVD and in need of secondary prevention, understanding upstream social determinants of CVD risk could guide primary prevention efforts by identifying where and how policy and community-level interventions could be targeted to facilitate changes in individual health behaviors.

Food environment index and preterm birth rate in the counties of the United States.

To determine the relationship between Food Environment Index (FEI) and Preterm Birth (PTB) rate at the county level of the United States of America (USA) (primary), while evaluating the interaction of multiple factors within a framework of sociodemographic, maternal health, maternal behavioral, and environmental factors. This is a population-based retrospective cohort ecological study from 2015-2018. The study compares the characteristics of the population of the counties of the USA. All counties with complete data on their PTB rate and the independent variables were included in the study. Independent variables with greater than 20% missing data were excluded from the study. Purposive sampling technique was applied. A total of 2983/3142 counties were included in the study. The median PTB rate of all counties was 9.90%. The highest PTB rate (23.3%) was in Tallapoosa County, Alabama and the lowest (3.4%) in San Juan County, Washington State. After adjusting for variables, PTB rate had a significant association with FEI (coefficient of correlation - 0.36, p < 0.01, 95% CI - 0.19 to - 0.04). Increase in the rate of unemployment, African American race, adult smoking, obesity, uninsured rate, sexually transmitted diseases (STD), high school education and air pollution was associated with an increase in PTB rate, while an increase in FEI and alcohol abuse rates was associated with a decrease in PTB rate. FEI can predict the PTB rate in USA counties after adjusting for sociodemographic, health, behavioral and environmental factors. Future studies are needed to confirm these associations and consider them when making policies to reduce PTBs.

Refining large-scale spatial metrics of food access

Large scale metrics of food access are frequently used to identify spatial inequities and for targeting local government intervention. The Food Access Research Atlas by the USDA and the Modified Retail Food Environment Index by the CDC are two such measures frequently used at the census-tract level. This paper uses spatial analyses to identify ways in which they can be refined for better actionability. Outcomes from this paper are threefold: (i) Upon examination of the average food environment, it finds that we might be undercounting rural census tracts with low food access relative to urban tracts; (ii) Disaggregating the often-conflated goals of physical access for food equity and nutritional access for public health might help targeting interventions; (iii) Data on food environments, and underlying assumptions on food consumption behavior should be investigated with greater caution.

SynthEco - A multi-layered digital ecosystem for analysing complex human behaviour in context

Introduction &amp; BackgroundHuman behaviour is multi-faceted and complex, with different dimensions interacting and impacting each other and individuals operating in an environmental context. In order to understand this behaviour better, the combination of data from different sources is useful to uncover some of those interactions and complexities. We present a multi-layered digital ecosystem based on a data platform providing statistically representative synthetic population derived from census data at different geo-spatial granularity, which we call SynthEco. This platform is enriched with individual data stemming from cohorts and cross-sectional surveys and geo-scanning of different layers of socio-environmental actors and conditions to create a complex digital ecosystem.&#x0D; Objectives &amp; ApproachThe objective of SynthEco is to allow for the analysis of behaviour, as well as health and wellbeing outcomes, through the integration of cohort and cross-sectional data into a geospatially anchored synthetic population embedded into environmental data which is forming the backdrop. We demonstrate the use of this platform on the example of Montreal, Canada. The synthetic population is first generated from census data using iterative proportional fitting, which allows for the creation of a population data set that is artificial yet statistically representative for a given geospatial granularity, such as a city. Each individual household is assigned a geospatial location, which allows for the consideration of their surrounding environment including enterprises or institutions such as schools, hospitals and the local food environment. Through fuzzy matching and statistical extrapolation, different cohort and cross-sectional survey data are then merged to individual records, in order to describe them in more detail. This includes health, as well as financial wellbeing or social environment descriptors.&#x0D; Relevance to Digital FootprintsThere are two important points made through the presented work in relation to Digital Footprints data: the first is the technical approach to merging multiple datasets describing different dimensions of interacting human characteristics and behaviour by anchoring them into a synthetic population through fuzzy record matching. The second is the consideration of a spatial dimension when describing human behaviour. This is especially important when describing behaviour within local environments, such as the interaction with local food outlets.&#x0D; ResultsRecent work in this context includes an analysis of the food environment in Montreal, Canada. It introduces a way of utilising the synthetic population to predict the healthfulness of their local environment in terms of healthy food outlets, as well as providing a platform for the analysis of food environment surveillance and intervention simulations. For this purpose, the healthfulness of different census tract regions in Montreal is calculated to identify food deserts, food swamps, as well as healthy areas as defined through the Modified Retail Food Environment Index. We test different machine learning approaches to then predict these healthfulness scores using census variables from the synthetic population in their respective census tract, achieving accuracy scores of around 0.53 to 0.60. This demonstrates that census data has some limited predictive power in explaining the healthiness of the local food environment, which could be especially relevant for situations in which no information on the retailers is available to local policy makers. Future work can extend this approach to also include further data describing the population, stemming from the integrated cohorts and survey data, which could improve the prediction accuracy or help in identifying areas of concern.&#x0D; Conclusions &amp; ImplicationsThe presented SynthEco platform views individuals as agents nested within modular systems of systems, trying to capture both internal systems and processes as well as environmental ones within which individuals are operating. The platform thus enables the application of computational systems modelling for the analysis of individual human behaviour in contexts. As demonstrated through the example of using SynthEco in the context of healthier food environments, the approach is especially relevant to practitioners and policy makers interested in local intervention strategies and identifying areas for targeted policy related to different dimensions of health and wellbeing.

Food Environment Quality and Cardiovascular Disease Mortality in the United States: a County-Level Analysis from 2017 to 2019.

Upstream socioeconomic circumstances including food insecurity and food desert are important drivers of community-level health disparities in cardiovascular mortality let alone traditional risk factors. The studyassessed the association between differences in food environment quality and cardiovascular mortality in US adults. Retrospective analysis of the association between cardiovascular mortality among US adults aged 45 and above and food environment quality, measured as the food environment index (FEI), in 2615 US counties. FEI was measured by equal weights of food insecurity (limited access to a reliable food source) and food desert (limited access to healthy food), ranging from 0 (worst) to 10 (best).Age-adjusted cardiovascular mortality rates per 100,000 adults aged 45 and above in the calendar year 2017-2019. County-level association between CVD mortality rate and FEI was modeled using generalized linear regression. Data were weighted using county population. Median CVD deaths per 100,000 population were 645.4 (IQR 561.5, 747.0) among adults aged 45years and above across US counties in 2017-2019. About 12.8% (IQR 10.7%, 15.1%) of residents were food insecure and 6.3% (IQR 3.6%, 9.9%) were living in food desert areas. Comparing counties by FEI quartiles, the CVD mortality rate was higher in the least healthy FE counties (704.3 vs 598.6 deaths per 100,000 population) compared to the healthiest FE counties. One unit increase in FEI was associated with - 12.95 CVD deaths/100,000 population. In the subgroup analysis of counties with higher income inequality, the healthiest food environment was associated with 46.4 lower CVD deaths/100,000 population than the least healthy food environment. One unit increase in FEI in counties with higher income inequality was associated with a fivefold decrease in CVD mortality difference in African American counties (- 18.4 deaths/100,000 population) when compared to non-African American counties (- 3.63 deaths/100,000 population). In this retrospective multi-county study in the USA, a higher food environment index was significantly associated with lower cardiovascular mortality.

Impact of Food Environments on Obesity Rates: A State-Level Analysis.

Limited access to healthy food in areas that are predominantly food deserts or food swamps may be associated with obesity. Other unhealthy behaviors may also be associated with obesity and poor food environments. We calculated Modified Retail Food Environment Index (mRFEI) to assess food retailers. Using data collected from the Behavioral Risk Factor Surveillance System (BRFSS) survey, the NJ Department of Health (NJDOH), and the US Census Bureau, we conducted a cross-sectional analysis of the interaction of obesity with the food environment and assessed smoking, leisure-time physical activity (LPA), and poor sleep. There were 17.9% food deserts and 9.3% food swamps in NJ. There was a statistically significant negative correlation between mRFEI and obesity rate (Pearson's r -0.13, p < 0.001), suggesting that lack of access to healthy food is associated with obesity. Regression analysis was significantly and independently associated with increased obesity prevalence (adjusted R square 0.74 and p=0.008). Obesity correlated positively with unhealthy behaviors. Each unhealthy behavior was negatively correlated with mRFEI. The mean prevalence for smoking, LPA, and sleep <7 hours was 15.4 (12.5-18.6), 26.5 (22.5-32.3), and 37.3 (34.9-40.4), respectively. Obesity tracks with food deserts and especially food swamps. It is also correlated with other unhealthy behaviors (smoking, LPA, and poor sleep).

A longitudinal assessment of racial and ethnic inequities in food environment exposure and retail market concentration.

This paper assesses trends in food environment and market concentration and racial and ethnic inequities in food environment exposure and food retail market concentration at the US census tract level from 2000 to 2019. Establishment-level data from the National Establishment Time Series were used to measure food environment exposure and food retail market concentration. We linked that dataset to race, ethnicity and social vulnerability information from the American Community Survey and the Agency for Toxic Substances and Disease Registry. A geospatial hot-spot analysis was conducted to identify relatively low and high healthy food access clusters based on the modified Retail Food Environment Index (mRFEI). The associations were assessed using two-way fixed effects regression models. Census tracts spanning all US states. 69 904 US census tracts. The geospatial analysis revealed clear patterns of areas with high and low mRFEI values. Our empirical findings point to disparities in food environment exposure and market concentration by race. The analysis shows that Asian Americans are likelier to live in neighbourhoods with a low food environment exposure and low retail market concentration. These adverse effects are more pronounced in metro areas. The robustness analysis for the social vulnerability index confirms these results. US food policies must address disparities in neighbourhood food environments and foster a healthy, profitable, equitable and sustainable food system. Our findings may inform equity-oriented neighbourhood, land use and food systems planning. Identifying priority areas for investment and policy interventions is essential for equity-oriented neighbourhood planning.

The Impact of Fast-Food Density on Obesity during the COVID-19 Lockdown in the UK: A Multi-Timepoint Study on British Cohort Data

Poor food environments are considered to trigger obesity and related health complications by restricting the local food options to predominantly low quality, energy-dense foods. This study investigated the impact of the food environment on obesity with a focus on any changes that might have occurred around the COVID lockdown period in the UK when majority of the population relied on food delivery and the local food environments. The proportion of fast-food retailers in the area and the Retail Food Environment Index (RFEI) were calculated for participants of the 1970 British Cohort Study (BCS70) at three timepoints: pre-COVID (2016), the first UK nation-wide lockdown (April–May 2020) and post lockdown (September–October 2020). The association of the food environment and the odds of obesity was estimated through multivariable logistic regression, with adjustments being made for selected socioeconomic variables. A model using the fast-food proportion as the sole predictor estimated that higher fast-food proportion increased the odds of obesity by 2.41 in 2016, 2.89 during the lockdown and 1.34 post lockdown, compared with 1.87, 2.23, and 0.73, respectively, for the same three periods with adjustments being made for select socioeconomic variables. On the other hand, RFEI increased the odds of obesity only slightly at 1.01, 1.02 and 1.03, respectively, with the model with adjustments yielding respective similar values. The fast-food proportion model indicates that proximity to a poor food environment is linked to obesity, especially during the COVID lockdown period, but the impact of a poor-food environment is limited if the RFEI is used as its indicator. The findings will add much needed insights on the UK data and will inform public health planning and policy.