• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of medjournmedLink to Publisher's site
Medscape J Med. 2009; 11(1): 15.
Published online Jan 15, 2009.
PMCID: PMC2654697

Food Availability, Neighborhood Socioeconomic Status, and Dietary Patterns Among Blacks With Type 2 Diabetes Mellitus

Abstract

Context

High diabetes prevalence among low-income and urban African American populations.

Objectives & Main Outcome Measures

This study aimed to determine associations between neighborhood-level food sources and socioeconomic status (SES), and dietary patterns and body-mass index (BMI). The hypotheses were that the presence of food stores in neighborhoods would be associated with better dietary habits and BMI, and that the presence of convenience stores, and lower neighborhood SES, would be associated with poorer dietary habits and BMI.

Design, Setting, & Patients

Black adults (n = 132) with type 2 diabetes in Project Sugar 2 (Baltimore, Maryland) underwent the Ammerman dietary assessment: total dietary risk score and subscores for meat, dairy, starches, and added fat. Food source availability (food stores, convenience stores, other food stores, restaurants, and other food service places) and SES data from the 2000 US census at the tract-level were linked to individual-level data. Linear mixed-effects regression models with random intercepts were used to account for neighborhood clustering and for individual-level SES and potential confounders.

Results

The presence of restaurants and other food service places in census tracts were associated with better dietary patterns (adjusted added fat subscore β = −1.1, 95% confidence interval [CI] = −1.8, −0.4, and β = −1.0, 95% CI = −1.7, −0.3, respectively). The presence of convenience stores and lower neighborhood SES was not significantly associated with worse dietary patterns or body-mass index, although trends were in the hypothesized direction.

Conclusions

These findings provide some evidence for structural improvements to food environments in urban and low-income black neighborhoods.

Introduction

Among US adults, diabetes disproportionately affects racial and ethnic minorities and low-income individuals.[13] Low socioeconomic status (SES) may be a major explanatory factor for these health disparities. For example, it is now well-documented that lower individual-level SES is associated with increased risk for cardiovascular disease, and similar trends are emerging in the diabetes literature.[46] Socioeconomic disadvantage may be one of the main contributors to the excess type 2 diabetes prevalence among blacks.[5,7] In the United States, it is difficult to disentangle the effects of low SES from those of race. However, with regard to diabetes mellitus, it does appear that low SES confers risk independent of race.[5,8]

Interest in interpersonal, social, and area-level environmental characteristics that may affect the more established risk factors for development of chronic diseases has grown substantially in recent years.[911] Living in low-income neighborhoods may be an independent risk factor for increased mortality, cardiovascular disease, obesity, insulin resistance, and ultimately type 2 diabetes.[1115] Some explanations for this increased risk among residents of low-income or resource-poor areas include increased stress, low access to medical and preventive care, and adverse built environment. The built environment includes the physical environment (buildings, street-connectivity, walkability of neighborhoods), and locations of different resources, such as food stores.[1116] These factors can affect physical activity and diet, and thus, obesity and diabetes.[12,17]

One such neighborhood-level factor is food availability and access to healthy foods. Research has shown that the local food environment, as defined by the presence of different types of food stores and restaurants, as well as food prices, can play a strong role in people's diets and in preventing obesity and overweight.[9,17,18] Low-income neighborhoods, especially urban neighborhoods composed mainly of blacks, tend to have reduced access to healthy foods and possibly greater access to fast food.[1621] These same communities suffer high rates of obesity, diabetes mellitus, poor blood sugar control, and diabetic complications.[13,2224] it is reasonable to hypothesize that there might be a causal link. Within neighborhoods, the availability of recommended foods for those with diabetes, including whole grains, fruits, vegetables, and diet sodas, may be a strong predictor of meeting dietary recommendations for reducing diabetic complications and maintaining good health.[22]

This is a rapidly evolving field with associations that remain to be clarified, and to-date, the associations between food availability and dietary patterns have not been fully assessed in the Baltimore community, especially with a focus on low-income blacks with diabetes. Given the economic and environmental stressors and disadvantages present, this population represents one which is potentially very vulnerable to negative health effects of the food environment. Thus, the goals of this paper, using Project Sugar 2 (PS2) ancillary participants, are threefold: to (1) characterize the local food environment, (2) determine the association between the different neighborhood food sources and dietary patterns, and (3) determine the association between neighborhood SES and dietary patterns. We hypothesized that the presence of food stores in neighborhoods would be associated with better dietary patterns and body mass index (BMI), and that the presence of convenience stores in neighborhoods and lower neighborhood SES would be associated with poorer dietary patterns and BMI.

Methods

Study Sample

The detailed methods of the PS2 study have been previously published.[25] Briefly, PS2 was a randomized controlled trial comparing two diabetes management interventions among 542 blacks with type 2 diabetes living in Baltimore City, Maryland. Participants were randomly assigned to receive, in addition to their usual medical care, either a minimal intervention (consisting of occasional telephone calls and mailing of standard informational brochures) or an intensive intervention (consisting of ongoing, individually tailored efforts by a nurse case manager/community health worker team aimed at improving diabetes-related health behaviors of patients and preventive health practices by their physicians). Individual-level data were obtained from PS2 participants at baseline. The ancillary participants were a random subsample of 132 (25%) patients (regardless of intervention assignment) who answered an additional questionnaire at baseline, with information on health behaviors and psychosocial factors such as diet, physical activity, depression, diabetes knowledge and attitudes, and problem-solving. Baseline visits were conducted between October 2000 and June 2002, and the intervention period began in November 2002 and ended in 2005.

Study Design

This study used individual baseline data and participants' census tract (neighborhood) SES and food source data. Census tracts are standard geographic subregions that aim to include about 3000–5000 people.[26] For the purpose of this study, census tracts were used to represent neighborhoods, as is common in studies of neighborhood-level associations.[10,16,17] Food source locations were geocoded to the census tracts of residence of PS2 participants using ArcGIS 9.1.[27] As an indicator of census tract-level SES, the percent of the census tract living below the poverty line (% below poverty), data from the national census, was used.[26] At the time of the study, the federal poverty line was defined as earning $8860 per year for an individual, or a combined income of $18,100 per year for a family of 4.[26] This standard was selected because of its ease of interpretation, and because it correlated highly with the Cubbin and Diez Roux scores, widely used comprehensive neighborhood SES indices.[12,15]

Characterization of the Local Food Environment

To characterize local food environment, we purchased food source data from Mapping Analytics.[28] The food sources were classified based on the 2002 North America Industry Classification System (NAICS) codes or 1987 Standard Industrial Classification (SIC) codes.[29] Five food source categories were acquired: (1) food stores were defined as supermarkets and other grocery (except convenience) stores (NAICS code 445110). These stores are those that sell a wide range of foods, including canned and frozen foods, fresh fruits and vegetables, and fresh and prepared meats, fish, and poultry.[29] Supermarkets tend to have more healthy foods and at lower prices.[30] (2) Convenience stores (NAICS code 445120) are those unattached to gas stations, which sell a limited range of foods, usually including milk, bread, snack food, and soda. (3) Other food stores (NAICS codes 445210, 445220, 445230, 445291, 445292, 445299; SIC major group 54, industry groups 542–546, 549) consist of meat and fish markets; fruit and vegetable markets; candy, nut, and confectionary stores; dairy product stores; retail bakeries; and miscellaneous food stores. (4) Restaurants include all full-service restaurants, pizzerias, diners, family restaurants, and fine dining restaurants (NAICS code 722110, SIC code 5812-full service). (5) Other food service includes all limited-service eating places such as carry-out restaurants, fast food restaurants, snack bars, delis, cafes, coffee shops, and grills (NAICS codes 722211, 722212, 722213, SIC code 5812-limited service and all others). The number of each different type of food source was summed for each census tract.

Dependent Variables

Dietary patterns were assessed among the PS2 ancillary questionnaire respondents at baseline using the Ammerman Dietary Risk Assessment tool.[31] This tool was designed specifically for use among low-income populations in the Southern United States, or those who are likely to follow similar traditional dietary patterns. It emphasizes food-based patterns rather than nutrient-based results by assessing the relative difference between recommended and actual intakes of different foods. There were 42 questions about food frequency grouped into 4 food group risk categories, ranked to reflect their atherogenic risk: (1) meat, (2) dairy, (3) starches (side dishes, desserts, and snacks), and (4) added fat (spreads, salad dressings, and oils). For each question on food frequency, responses are scored as either “doing well” (0 points), “needs work” (1 point), or “problem” (2 points). Points are summed for each food category and all 4 categories are summed to obtain the total (continuous) dietary risk assessment score, where higher values indicate higher risk. The minimum score for each category is 0; the maximum scores are meat, 28; dairy, 16; starches, 22; and added fat, 18 – corresponding to a maximum total score of 84. This tool has been successfully validated against other brief dietary assessments.[31]

The fruit and vegetable items were also analyzed separately from the total Ammerman dietary risk score. The fruit item asked, “How many times a day do you eat or drink fruit or fruit juices?” Responses choices were: 0, 1, or 2 or more. The vegetable questions were: “How many times a day do you eat vegetables of any kind?” Responses choices were: 0, 1, or 2 or more. The salad question asked, “How many times a week do you eat tossed salads?” Response choices were 0, 1, 2, 3, or 4 or more. To combine the 2 vegetable items into 1 vegetable score, the salad responses were divided by 7 to determine salad consumption per day. This value was added to the vegetables per day responses to calculate the vegetable sum score.

Height and weight were measured at the baseline assessment in the Johns Hopkins Outpatient Department General Clinical Research Center. BMI was calculated as weight in kilograms divided by height in meters squared and analyzed as a continuous variable.

Statistical Methods

For the first goal of this study, characterizing the local food environment, we calculated the total number of each of the different types of food sources for each census tract of residence and mapping geographic patterns.

For the second goal, determining the associations between different food sources and dietary patterns, multilevel linear mixed effects models with a random intercept for each census tract were used to account for the clustering of individuals within census tracts. Food sources were dichotomized as “any” vs “none” of each type, in keeping with commonly used methods in this field.[10,17] We dichotomized to improve interpretability of results and to reflect the fact that living in an area where there is none of a particular food source is likely very different than living in an area where there is at least 1 location of a particular food source. The adjusted models included individual-level confounders that are commonly used in similar analyses,[10,17] those that were likely to be associated with participants' census tract of residence and dietary patterns: age, gender, annual household income, education, employment status, and marital status.

The third goal examined the same models as for goal 2 using the percent of the neighborhood living below the federal poverty line (% below poverty) as the exposure. All data were analyzed using STATA 9.2.[32]

Results

Baseline Characteristics of the Study Sample

The baseline sociodemographic and clinical characteristics of the 132 PS2 ancillary questionnaire respondents are presented in Table 1. A majority of the participants were female (70.4%), the mean age was 57.5 years, and 25.8% reported an annual household income of over $15,000. Over half of the participants (60.3%) had completed high school or college, one third (33.3%) of the participants were married, just under one third (29.5%) were employed full or part-time, and 59.8% reported trying to lose weight. Just over two thirds of the participants (67.2%) were obese (BMI ≥ 30 kg/m2), over 60% had a hemoglobin A1c above the recommended 7%, and roughly half of the sample had a total cholesterol level below the recommended 200 mg/dL. Almost two thirds of the sample had a systolic blood pressure above the recommended 130 mm Hg, and half had a diastolic blood pressure above the recommended 80 mm Hg. More than one fourth of the participants (27.3%) had visited a nutritionist in the past 12 months and 46.5% reported consuming fast food 0 times per week.

Table 1
Baseline Characteristics of the Study Sample

Characterization of the Local Food Environment and Dependent Variables

Characteristics of the local food environment of PS2 ancillary questionnaire respondents are presented in Table 2, which also summarizes census tract SES characteristics, dietary patterns, and BMI. The mean number of total food sources per census tract was just under 8. The mean number of the individual food sources per census tract were : 2.3 food stores, 0.4 convenience stores, and 2.2 restaurants. The mean percentage of people living below the federal poverty line in the PS2 census tracts was almost 31%, with a range of 5.9% to 58.6%.

Table 2
Neighborhood (Census Tract) Food Source Availability, Socioeconomic Status, and Dependent Variables

The individual-level dietary patterns and BMI are described in Table 2. The Ammerman total dietary risk score had a mean of 24.4 and ranged from 7–49, with higher scores indicating increased risk and the maximum possible score being 84 (see Methods). The meat sub-score had the highest mean score (9.2) and added fat had the lowest (2.9). The means for daily fruit and vegetable consumption were 1.6 and 1.7, respectively. These scores are lower, in general, than those of the original Ammerman assessment population.[31]

Association Between the Food Environment And Dietary Patterns and Body Mass Index

The adjusted results are presented in Table 3. The negative association between the presence of food stores and the added fat risk score remained marginally significant after adjustment (β = −0.7, 95% CI = −1.5, 0.0). The presence of convenience stores was nonsignificantly associated with increased risk scores for the Ammerman total score and the meat and dairy subscores, as well as higher BMI. The presence of other food stores in a census tract was associated with decreased risk for the Ammerman total, meat, dairy, and added fat scores, although none of the associations reached statistical significance. Restaurants were significantly associated with a decreased added fat score (β = −1.1, 95% CI = −1.8, −0.4). The presence of food service places was negatively associated with the added fat score (β = −1.0, 95% CI = −1.7, −0.3). Fruit and vegetable consumption did not appear to be associated with the presence of any of the food sources.

Table 3
Adjusted¥ Associations Between Presence (or Absence) of Food Sources in Census Tracts of Residence and Dependent Variables

Association Between Neighborhood Poverty And Dietary Patterns and Bmi

Table 4 presents the unadjusted and adjusted associations of census tract SES with dietary patterns and BMI. In general, associations were in the expected directions, with higher percent poverty associated with increased dietary risk and BMI. In unadjusted analyses, Ammerman total risk score, the meat, dairy, and added fat subscores, and BMI, appeared to be strongly associated with % below poverty, although none reached statistical significance. After adjusting for the individual-level confounders, the associations in general remained constant or were attenuated. The Ammerman total risk score and the dairy subscore, and to a lesser extent, added fat and BMI, were associated with percentage below poverty. However, none of these associations were statistically significant.

Table 4
Unadjusted and Adjusted¥ Associations Between Percent of People in Census Tract Living Below Federal Poverty Line and Dependent Variables

Discussion

This study examined the associations between food availability, neighborhood SES, dietary patterns, and BMI among this sample of 132 blacks with diabetes living in Baltimore City, Maryland. As there had previously been no published work on this population's food environment, it represents a foundation for future research on this and similar groups and helps to further illuminate food environment associations. Compared to Baltimore's population on average, this study's population was older, lower-income, and comprised of more females, and more blacks.[26] The main findings of this study were that the presence of food stores, and to some extent, restaurants, in census tracts was associated with encouraging trends in dietary risk patterns. Additionally, the presence of convenience stores in census tracts was associated with worse dietary patterns and BMI. Moreover lower census tract SES was associated with higher dietary risk and BMI. However, most of these associations did not reach statistical significance.

Several interesting trends emerged when characterizing the local food environment. While many census tracts where PS2 ancillary participants lived did not have any of the food sources, most of the participants' tracts (n = 106) had at least 1 food store (a grocery store or supermarket). These types of stores generally carry a wide range of food items, including healthy recommended foods for diabetic people, and at reasonable costs.[30,33] While it is promising that many low-income neighborhoods in Baltimore do have food stores, the issue of food availability within grocery stores also arises. It has been found that stores in low-income neighborhoods may carry fewer healthy choices, or that healthy foods are not affordable.[18,22] Thus, even though food stores that may carry healthy items are present in a neighborhood, it does not necessarily mean that residents' diets will be healthier, given the multitude of other influences on food choices, such as cultural norms, pricing, marketing, food quality within stores, and taste preference.

We also found that most participants' census tracts (n = 90) did not have any convenience stores. Convenience stores in this study were the only food source type to correlate significantly with percentage below poverty, but in the opposite direction than expected: more poverty was related to fewer convenience stores. It has been shown that wealthy neighborhoods have fewer convenience stores than poorer ones.[16] In keeping with the theories presented above for food stores, convenience stores are less likely to carry foods recommended for people with diabetes, such as low-fat milk, high-fiber or low-carbohydrate bread, and fresh fruits and vegetables.[22] Thus, while there may be fewer convenience stores in low-income neighborhoods of PS2 participants, shopping at convenience stores may still have negative ramifications for diet quality.

In general, the direction of the association between food source and Ammerman risk scores was in agreement with other studies. For instance, the presence of grocery stores, supermarkets, and full-service restaurants in census tracts of residents has been found to be associated with a healthy diet among blacks.[17,18] Also, the presence of convenience stores in census tracts was generally associated with increased dietary risk scores and BMI. This latter finding makes sense in the context of previous studies showing reduced availability of healthy foods in convenience stores in low-income areas.[10,22] However, the fact that the associations between food stores, restaurants, and other food service places with lower added fat risk scores remained significant after adjusting for individual-level confounders, lends credibility to the notion that neighborhood-level factors account for dietary patterns above that which is contributed by individual behaviors and characteristics.[12,15]

The presence of restaurants and other food service places was inversely associated with the added fat subscale. The Ammerman total risk score and meat subscore were also lower with the presence of any restaurants or other food service places, although not significantly. Given that the added fat score is composed of items such as butter, margarine, oil, shortening, mayonnaise, and deep fried foods, it is a somewhat surprising negative association with restaurants, which often serve rich foods, and other food service places, which includes fast food restaurants. One possible explanation is that several of the questions comprising the added fat score inquire about fat added or eaten in the home or in food preparation.[31] Thus, these practices would not be accounted for in this food frequency score if they occurred when eating at restaurants or outside of the home. Additional food frequency questions that ask specifically about food consumption outside the home would be helpful in future investigations.

The unadjusted models showed positive associations between higher poverty and worse dietary patterns and BMI. After adjusting for individual-level confounders, including annual household income, the direction of the associations was the same, with the exception of the starches score, but the magnitude of the associations were generally smaller and still not significant. These findings, both of the positive association between neighborhood SES and individual diabetic risk factors,[13] as well as the attenuation after adjustment, agrees with prior research.[12,34] While in this study, such associations were not significant, future studies should include more participants and those with a wider range of SES exposures.

There were several limitations within this study, leading to the mainly negative findings. First, the small sample size could have limited power to detect significant associations. The ancillary PS2 questionnaire, which included the Ammerman dietary risk assessment and other measures peripheral to the main PS2 hypotheses, was also time-consuming to answer, and so it was administered only to this smaller group of participants.[25] The fact that several significant associations were seen despite this small sample size, however, lends credibility to these findings. Another potential limitation of this analysis was the fact that, by the nature of the trial's design, the population was relatively homogenous (low-income, obese, older African American Baltimore residents with diabetes). Such homogeneity could make it difficult to detect differences when examining diabetes-related outcomes and confounders such as BMI, age, and gender.

Finally, the classifications of the food sources may have been too broad, in some cases, to make meaningful interpretations. For instance, the other food stores category encompassed markets selling foods from meat and fish to candy and baked goods.[28] Thus, within this category, there are items that are both recommended and discouraged for people with diabetes, making it difficult to interpret the results. Similarly, the other food service category was composed of a wide variety of eating and drinking places, from ice cream stands to fast food restaurants.[28] While most of the establishments in this group would likely not be recommended for people with diabetes, it is still difficult to interpret this category. This limitation comes from the data source, in which there was no way to make finer distinctions among store subtypes. In general, this heterogeneity within food source classifications is a limitation of this field and problematic in other studies as well. However, some studies have distinguished between more specific subgroupings of food stores and restaurants, such as supermarkets, grocery stores, limited service restaurants, and fast food restaurants.[10,16,18] Future work would benefit from using data on more specific and homogenous categories of food sources that are classified in a consistent manner.

This study provides a preliminary examination of Baltimore's local food environment and its association with dietary patterns and BMI. The results presented here help lay the groundwork for future research focusing on structural and environmental influences of diabetes risk factors. What is evident from this study is that environmental factors, beyond individual behaviors, although they are critical as well, affect dietary patterns. While the findings presented here are mainly suggestions of associations, it will be important to continue research, policy, and advocacy efforts focusing on making structural, environmental-level changes for the food environment in poor urban neighborhoods. Such changes could include improving healthy food selections in convenience stores and developing effective social marketing campaigns. In addition, this study can help physicians and other public health officials understand the broader context: that food environment is a critical component of controlling disease. Taken together with previous research, this study can serve to improve health-related quality of life for people living with chronic diseases in urban low-income neighborhoods.

Acknowledgments

We would like to acknowledge the efforts of the Project Sugar 2 research staff and the Johns Hopkins General Clinical Research Center (GCRC). We also acknowledge the Project Sugar 2 participants whose cooperation made this research possible.

Notes

Funding Information

The project was funded by grants from the National Institutes of Health (R01-DK48117 and R00052). Dr. Gary was funded by a grant from the NHLBI (K01-HL084700) and Dr. Brancati was funded by a grant from the NIDDK (K24-DK6222).

The results were presented in part at the 6th Meeting of the International Conference on Urban Health, Baltimore, Maryland, November 2007.

Disclaimer

Researchers identified relationships between dietary patterns, socioeconomic status, and the proximity of different types of food outlets in low-income neighborhoods.

Contributor Information

Rachel A. Millstein, Department of Epidemiology, The Johns Hopkins Medical Institutions, Baltimore, Maryland Author's email address: ude.hpshj@etsllimr.

Hsin-Chieh Yeh, Departments of Epidemiology, and Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland.

Frederick L. Brancati, Departments of Epidemiology, and Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland.

Marian Batts-Turner, Department of Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland.

Tiffany L. Gary, Departments of Epidemiology, and Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland.

References

1. Centers for Disease Control and Prevention. National diabetes fact sheet: General information and national estimates on diabetes in the United States, 2003. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2004.
2. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA. 2002;288:1723–1727. [PubMed]
3. Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS, Koplan JP. The continuing epidemics of obesity and diabetes in the United States. JAMA. 2001;286:1195–1200. [PubMed]
4. Marmot MG, Adelstein AM, Robinson N, Rose GA. Changing social-class distribution of heart disease. BMJ. 1978;2:1109–1112. [PMC free article] [PubMed]
5. Robbins JM, Vaccarino V, Zhang H, Kasl SV. Excess type 2 diabetes in african-american women and men aged 40–74 and socioeconomic status: Evidence from the third national health and nutrition examination survey. J Epidemiol Commun Health. 2000;54:839–845. [PMC free article] [PubMed]
6. Black SA. Diabetes, diversity, and disparity: What do we do with the evidence? Am J Public Health. 2002;92:543–548. [PMC free article] [PubMed]
7. Brancati FL, Kao WH, Folsom AR, Watson RL, Szklo M. Incident type 2 diabetes mellitus in African American and white adults: The atherosclerosis risk in communities study. JAMA. 2000;283:2253–2259. [PubMed]
8. Brancati FL, Whelton PK, Kuller LH, Klag MJ. Diabetes mellitus, race, and socioeconomic status. A population-based study. Ann Epidemiol. 1996;6:67–73. [PubMed]
9. Yen IH, Kaplan GA. Neighborhood social environment and risk of death: Multilevel evidence from the alameda county study. Am J Epidemiol. 1999;149:898–907. [PubMed]
10. Morland K, Diez Roux AV, Wing S. Supermarkets, other food stores, and obesity: The atherosclerosis risk in communities study. Am J Prev Med. 2006;30:333–339. [PubMed]
11. Diez-Roux AV, Nieto FJ, Muntaner C, et al. Neighborhood environments and coronary heart disease: A multilevel analysis. Am J Epidemiol. 1997;146:48–63. [PubMed]
12. Diez Roux AV, Jacobs DR, Kiefe CI, Coronary Artery Risk Developoment in Young Adults (CARDIA) Study. Neighborhood characteristics and components of the insulin resistance syndrome in young adults: The coronary artery risk development in young adults (CARDIA) study. Diabetes Care. 2002;25:1976–1982. [PubMed]
13. Lee RE, Cubbin C. Neighborhood context and youth cardiovascular health behaviors. Am J Public Health. 2002;92:428–436. [PMC free article] [PubMed]
14. Cubbin C, Sundquist K, Ahlen H, Johansson SE, Winkleby MA, Sundquist J. Neighborhood deprivation and cardiovascular disease risk factors: Protective and harmful effects. Scand J Public Health. 2006;34:228–237. [PubMed]
15. Winkleby MA, Cubbin C. Influence of individual and neighbourhood socioeconomic status on mortality among black, Mexican American, and white women and men in the united states. J Epidemiol Community Health. 2003;57:444–452. [PMC free article] [PubMed]
16. Morland K, Wing S, Diez Roux A, Poole C. Neighborhood characteristics associated with the location of food stores and food service places. Am J Prev Med. 2002;22:23–29. [PubMed]
17. Morland K, Wing S, Diez Roux A. The contextual effect of the local food environment on residents' diets: The atherosclerosis risk in communities study. Am J Public Health. 2002;92:1761–1767. [PMC free article] [PubMed]
18. Zenk SN, Schulz AJ, Israel BA, James SA, Bao S, Wilson ML. Fruit and vegetable access differs by community racial composition and socioeconomic position in Detroit, Michigan. Ethn Dis. 2006;16:275–280. [PubMed]
19. Sooman A, Macintyre S, Anderson A. Scotland's health – a more difficult challenge for some? the price and availability of healthy foods in socially contrasting localities in the west of Scotland. Health Bull (Edinb) 1993;51:276–284. [PubMed]
20. Block JP, Scribner RA, DeSalvo KB. Fast food, race/ethnicity, and income: A geographic analysis. Am J Prev Med. 2004;27:211–217. [PubMed]
21. Fitzgibbon ML, Stolley MR. Environmental changes may be needed for prevention of overweight in minority children. Pediatr Ann. 2004;33:45–49. [PubMed]
22. Horowitz CR, Colson KA, Hebert PL, Lancaster K. Barriers to buying healthy foods for people with diabetes: Evidence of environmental disparities. Am J Public Health. 2004;94:1549–1554. [PMC free article] [PubMed]
23. Cowie CC, Eberhardt MS. In: Sociodemographic characteristics of persons with diabetes. 2nd ed. National Diabetes Data Group; Diabetes in America, editor. Bethesda, MD: National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health; 1995. pp. 85–116.
24. Harris MI, Eastman RC, Cowie CC, Flegal KM, Eberhardt MS. Racial and ethnic differences in glycemic control of adults with type 2 diabetes. Diabetes Care. 1999;22:403–408. [PubMed]
25. Gary TL, Batts-Turner M, Bone LR, et al. A randomized controlled trial of the effects of nurse case manager and community health worker team interventions in urban African Americans with type 2 diabetes. Control Clin Trials. 2004;25:53–66. [PubMed]
26. US Census Bureau. Summary file 1 (and file 3) 2000 census of population and housing technical documentation. Washington, DC: US Census Bureau; 2001.
27. Environmental Systems Research Institute (ESRI) Redlands. Vol. 9.1. Calif: ArcGIS; 2004.
28. Mapping Analytics L. Mapping Analytics. Rochester, NY: Mapping Analytics; 2005.
29. Economic Classification Policy Committee (EPCP); US Bureau of the Census. NAICS: New Data for a New Economy. Washington, DC: US Bureau of the Census; 2002.
30. Sallis JF, Nader PR, Rupp JW, Atkins CJ, Wilson WC. San diego surveyed for heart-healthy foods and exercise facilities. Public Health Rep. 1986;101:216–219. [PMC free article] [PubMed]
31. Ammerman AS, Haines PS, DeVellis RF, et al. A brief dietary assessment to guide cholesterol reduction in low-income individuals: Design and validation. J Am Diet Assoc. 1991;91:1385–1390. [PubMed]
32. StataCorp L. Stata statistical software. College Station, Tx: StataCorp; 2006. Release 9.2.
33. Turrell G. Structural, material and economic influences on the food purchasing choices of socioeconomic groups. Aust NZ J Public Health. 1996;20:611–617. [PubMed]
34. Diez-Roux AV, Nieto FJ, Caulfield L, Tyroler HA, Watson RL, Szklo M. Neighbourhood differences in diet: The atherosclerosis risk in communities (ARIC) study. J Epidemiol Community Health. 1999;53:55–63. [PMC free article] [PubMed]

Articles from The Medscape Journal of Medicine are provided here courtesy of WebMD/Medscape Health Network
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • EST
    EST
    Published EST sequences
  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...