Built Environment and Health Behaviors Among African Americans A Systematic Review

Transcription

1 Built Environment and Health Behaviors Among African Americans A Systematic Review Sarah Stark Casagrande, MHS, Melicia C. Whitt-Glover, PhD, Kristie J. Lancaster, PhD, RD, Angela M. Odoms-Young, PhD, RD, Tiffany L. Gary, PhD Context: Evidence acquisition: Evidence synthesis: Conclusions: An overall understanding of environmental factors that affect weight-related behaviors and outcomes in African American adults is limited. This article presents a summarization of the literature on the built environment and its association with physical activity, diet, and obesity among African Americans. A systematic review was conducted by searching the PubMed electronic database from inception to July 31, 2007, reviewing bibliographies of eligible articles, and searching authors personal databases using various search terms for the built environment, physical activity, diet, and obesity. Eligible articles were observational studies that included a study population 90% African American (or subgroup analysis), adults ( 18 yrs), and were published in English; final article data abstraction occurred from October 2007 through February A total of 2797 titles were identified from the initial search, and 90 were deemed eligible for abstract review. Of these, 17 articles were eligible for full review and ten met all eligibility criteria. The median sample size was 761 (234 to 10,623), and half of the articles included only African Americans. Light traffic, the presence of sidewalks, and safety from crime were more often positively associated with physical activity, although associations were not consistent (OR range ). Additionally, perceived barriers to physical activity were associated with obesity. The presence of supermarkets and specialty stores was consistently positively associated with meeting fruit and vegetable guidelines. With relatively few studies in the literature focused on African Americans, more research is needed to draw conclusions on features of the built environment that are associated with physical activity, diet, and obesity. (Am J Prev Med 2009;36(2): ) 2009 American Journal of Preventive Medicine Introduction Obesity is a major concern among U.S. adults, and the prevalence has roughly doubled in the past 25 years with no signs of slowing. 1 Further, African Americans share a disproportionate burden of the epidemic. Data from the National Health and Nutrition Examination Survey (NHANES) indicate that 45.0% of African-American adults are From the Johns Hopkins Bloomberg School of Public Health, Department of Epidemiology (Casagrande, Gary), Baltimore, Maryland; Wake Forest University School of Medicine, Department of Epidemiology and Prevention, Division of Public Health Sciences (Whitt- Glover), Winston-Salem, North Carolina; New York University, Department of Nutrition, Food Studies and Public Health (Lancaster), New York, New York; and the Northern Illinois University, School of Nursing and Health Studies (Odoms-Young), DeKalb, Illinois Address corresponding and reprint requests to: Tiffany L. Gary, PhD, 615 N Wolfe Street, Room E6531, Baltimore MD tgary@jhsph.edu. The full text of this article is available via AJPM Online at 1 unit of Category-1 CME credit is also available, with details on the website. obese compared with 30.6% of whites. 2 Consistent with the high prevalence of obesity, several studies have also reported disparities between physical activity and healthy eating in African Americans and whites. 3 7 Although physical activity helps to prevent weight gain, 26% of U.S. adults do not engage in any physical activity; African Americans have an even lower prevalence. 5 According to the 2000 Behavioral Risk Factor Surveillance System (BRFSS), 33.2% of African Americans did not participate in any physical activity in the previous month compared with 22.0% of whites. 5 Likewise, African Americans are less likely to meet the Dietary Guidelines for Americans, which provide dietary guidance. For example, data from NHANES indicated that while most U.S. adults (89%) do not meet fruit and vegetable recommendations ( 5 servings per day), African Americans were 38% less likely to meet fruit and vegetable recommendations compared with whites. 3 Similar racial disparities have been found in intakes of saturated fat Am J Prev Med 2009;36(2) /09/$ see front matter 2009 American Journal of Preventive Medicine Published by Elsevier Inc. doi: /j.amepre

2 Although previous studies have identified multiple contributors to overweight, an emerging body of literature has emphasized the importance of the built environment in shaping weight-related behaviors, 8 20 specifically dietary intake and physical activity, and weight outcomes. With individual-level behavioral interventions generally showing unsuccessful results in producing long-term behavior modification, 8 there has been an increase in nutrition, physical activity, and obesity research focused on population-level approaches. 9,10 Whereas several recent reviews have examined the impact of environmental factors on individual physical activity, dietary behaviors, and/or weight status, few studies have focused specifically on minority or vulnerable populations. In a 2002 review with no minority focus, accessibility, opportunities, and aesthetic attributes of neighborhoods had significant associations with physical activity. 11 Other attributes that have been associated with physical activity include residential density, 12,13 land-use mix, intersection density, walkability, 13,14 and park availability. 15 There is evidence that dietary patterns differ across neighborhoods and that these differences are not fully explained by individuallevel socioeconomic characteristics. 16,17 For example, the presence of supermarkets, compared to convenience stores that have fewer healthy options, has been associated with a higher intake of fruits and vegetables in adults. 18 A review published in 2005 on the built environment found that area of residence, resources, television viewing, walkability, land use, sprawl, and level of deprivation were associated with obesity. 19 Similarly, a review published in 2007 found that the majority of studies identified some aspect of access, availability of physical activity resources, and food sources as being significantly associated with obesity. 20 Since measurement and analysis methods differ significantly across studies, more research is needed to understand the mechanisms underlying these results. With the growing number of studies focused primarily on African Americans, the purpose of this review is to (1) quantify the existing literature, (2) acknowledge gaps in the literature that could affect future research, and to (3) surmise any salient environmental characteristics that are associated with diet, physical activity, and obesity among African Americans that may be important targets for environmental interventions. Evidence Acquisition Eligibility Criteria Eligible articles evaluated the objective or perceived physical built environment and the association with physical activity, dietary behaviors, and obesity or BMI in study populations of 90% African Americans (or studies that included a subgroup analysis of African Americans). The physical built environment was defined as features or structures of the environment that may affect physical activity and healthy dietary practices. In addition, eligible articles were observational studies, including population and community surveys, cohort and cross-sectional studies, that were published in English through July 2007, and that studied African American adults ( 18 years). Articles were excluded that evaluated neighborhood socioeconomic status or racial segregation as environmental measures, evaluated only health status or mortality as outcomes, provided no subgroup analyses for African Americans, or included 90% African Americans in populations where analyses for race/ethnic groups were combined. Intervention studies, design manuscripts, and qualitative studies were also excluded. Search Strategy Published studies were identified by searching the PubMed database from inception through July 31, The following keywords were used to search for relevant articles: African American AND built environment, urban environment, neighborhood, neighbourhood, physical environment, environmental justice, walk* (truncated term; includes all terms beginning with walk), neighborhood destinations, food store availability, food availability, food environment, environment resources, grocery store, communities, physical activity, exercise, diet, obesity, BMI, food, and nutrition. Additional strategies included searching the bibliographies of eligible studies and searching the authors personal databases for relevant articles on the built environment among African Americans. These additional searches were reviewed using the same processes outlined below. Data Abstraction Article titles were independently reviewed by two investigators and included if at least one reviewer selected the article for further review. Abstracts were then independently reviewed by two investigators and were selected for full article review when both investigators agreed that the article should move forward to full review. If both investigators could not resolve differences (i.e., eligibility criteria, methods) in abstract selection, conflicts were adjudicated by a third investigator. Using a standard review form, full articles were independently serially reviewed by two investigators. Data abstraction occurred between October 2007 and February Data Synthesis Since there were a small number of articles and a variety of methods were used in eligible articles, a meta-analysis was not performed. Results of the review are summarized qualitatively in evidence tables that include major findings and conclusions. Odds ratios shown in the evidence tables were extracted directly from the tables of reviewed articles. For comparison purposes, odds ratios were presented for similar exposures across studies. Any significant findings were also presented. Evidence Synthesis Search Strategy Figure 1 depicts the search outcomes. Due to the small number of articles fitting the eligibility criteria, articles in which the study population did not meet the eligibility criteria ( 90% African Americans or sub-analyses February 2009 Am J Prev Med 2009;36(2) 175

3 presented) but included some proportion of African Americans were additionally summarized. Study Characteristics Table 1 summarizes the ten eligible articles. 18,21 29 Publication dates ranged from 2000 to The median sample size was 761 (range ,623) for all included articles (n 10), and 483 (range ) for articles consisting of only African Americans (n 5). Of the four articles reporting a mean age, it was 48.2 years (range ) 18,22,28,29 ; overall age ranged from 18 to 96 years. The percentage of female participants ranged from 56.0% to 100.0% with seven articles including only female participants ,27,29 Five articles had a study population composed of only African Americans (mean percentage of African Americans, 73.0%; range 22.5% 100.0%). Five articles were published from the same multi-site parent study The two studies had populations who resided exclusively in rural areas 24,26 and five studies had populations who resided in urban or suburban areas. 21,23,25,28,29 Recruitment methods included randomdigit dialing in specific ZIP codes, recruiting through Figure 1. Results of search strategy Table 1. Characteristics of articles on the built environment and health behaviors/obesity among African Americans % African American Setting Location Recruitment and methods Sample size Age (years) % female Study Sumter, Orangeburg counties, South Carolina Recruitment: RDD Face-to-face interview Ainsworth (2003) (range) Metropolitan/ nonmetropolitan Eyler (2003) 22 4, (M) Mixed Multi-site; Universities of AL, SC, IL, MD, NC Recruitment: RDD Face-to-face interview Rohm Young (2003) (range) Urban Baltimore MD Recruitment: community organization Face-to-face interview Sanderson (2003) (range) Rural Greene, Lowndes, Wilcox counties, AL Recruitment: RDD Wilbur (2003) (range) Urban Chicago IL Recruitment through community Face-to-face interview Hooker (2005) 26 1, (range) Rural One county SC Recruitment: RDD Questionnaire King (2000) 27 2, Mixed National sample Telephone survey by ZIP code Morland (2002) 18 10, (M) Mixed Multi-site; Washington county MD; Forsyth Existing study Objective environment measures county NC; Jackson City MS; Minneapolis MN Wanko (2004) (M) Urban Atlanta GA Consecutive patients for first visit at diabetes clinic Self-administered questionnaire Zenk (2005) (M) Urban Detroit MI Community survey RDD, random-digit dialing 176 American Journal of Preventive Medicine, Volume 36, Number 2

4 community organizations and clinics, and the use of an existing study. Table 2 shows the measures and outcomes used in the eligible articles. One study used objective measures of the built environment 18 while the remaining articles used questionnaires to assess participants perceptions of the built environment. The majority of articles focused on built environment features related to physical activity; only two articles were focused on food-related aspects of the built environment 18,29 ; one article focused on built environment features related to obesity. 22 Built Environment Features Related to Physical Activity and Obesity Of the seven studies focused on environmental features associated with physical activity, all used questionnaires examining participants perceptions. Five articles used the Women and Physical Activity Survey 31 to assess the perception of neighborhood traffic, presence of sidewalks, street lighting at night, unattended dogs, safety from crime, places within walking distance, and places to exercise. The other articles assessed similar environmental features and also included safety of public facilities, feeling safe walking or jogging, hills, and enjoyable scenery Environmental Features Related to Dietary Intake Studies examining the built environment and diet focused primarily on associations between retail food environments, specifically the types of food stores and services available, and fruit and vegetable intake. Only two studies were identified that specifically targeted African Americans. Morland and colleagues, 18 assessed the availability of food stores and food service places in participants census tracts using data collected from local health departments and state departments of agriculture. In contrast, in a study by Zenk and colleagues 29 retail food environments were defined based on the primary store locations where participants shopped. Description of Health Behaviors/Weight Status and Measures Health behaviors included physical activity and dietary intake, and weight status included obesity. Physical activity was self-reported in all studies using the BRFSS survey or the BRFSS survey in combination with the National Health Interview Survey (NHIS). The BRFSS survey includes questions on the average frequency and intensity of overall physical activity. Dietary intake was assessed using either a semi-quantitative food frequency questionnaire (FFQ) or the BRFSS Survey. Two studies assessing dietary intake focused on the frequency of fruit and vegetable consumption. 18,29 The study assessing obesity objectively measured height and weight to determine BMI. 28 Associations Between the Built Environment and Physical Activity In general, associations between the built environment and physical activity among African Americans were not consistent. Light traffic (OR 1.53, 95% CI ) and the presence of sidewalks (OR 1.57, 95% CI ) were significantly associated with meeting physical activity recommendations among participants living in both metropolitan and nonmetropolitan areas 21 but were not significantly associated with physical activity in other studies. 23,24 Among urban participants, safety from crime was associated with physical activity (OR 2.43, 95% CI ) 25 but not in other studies. 23,24,26 In a national sample that evaluated many environmental features, the presence of unattended dogs was the only environmental attribute associated with physical activity (OR 1.51, 95% CI ). 27 Associations Between the Built Environment and Dietary Intake The presence of supermarkets increased the likelihood of meeting fruit and vegetable (OR 1.54, 95% CI ), total fat (OR 1.22, 95% CI ), and saturated fat recommendations (OR 1.30, 95% CI ) in one study. 18 In the other study, women shopping at supermarkets (beta 1.22, p 0.001) and specialty stores (beta 2.37, p 0.05) consumed fruit and vegetables more often than those shopping at independent grocers. 29 Association Between the Built Environment and Obesity One article assessed the association between the perceived barriers to physical activity and obesity. 28 Individuals with a higher BMI were more likely to report a barrier to exercise compared with individuals with a lower BMI (OR 1.04, 95% CI ). For every 4% increase in reported barriers, BMI increased by one unit. Other Articles Since few articles included only African Americans or presented sub-analyses, the 12 articles that had 90% African Americans in the study population and had no sub-analyses were additionally summarized (Table 3). 30,32 42 The median percentage of African Americans in these studies was 27.0% (range 9.5% 41.0%) and the median percentage of females was 58.2% (range 52.1% 100.0%). Six articles found positive significant associations between perceived environmental supports and physical February 2009 Am J Prev Med 2009;36(2) 177

5 178 American Journal of Preventive Medicine, Volume 36, Number 2 Table 2. Measures and outcomes of articles on the built environment and health behaviors/obesity among African Americans (AA) Study Built environment features Measure Behavior/weight status Measure Variable of interest Point est (95% CI) Conclusions Ainsworth Traffic a (2003) 21 Presence of sidewalks Street lighting at night Unattended dogs Eyler (2003) 22 Safety from crime Places within walking distance Places to exercise Women and PA Survey Physical activity BRFSS Traffic, light Sidewalks Urban Safety from crime Mixed Sidewalks Rohm Young Traffic, light (2003) 23 Sidewalks Safety from crime Sanderson Traffic, light (2003) 24 Sidewalks Safety from crime Wilbur (2003) 25 Hooker (2005) 26 King (2000) 27 Morland (2002) 18 Traffic Streetlight quality Unattended dogs Safe from crime Public facilities are safe Feel safe walking/jogging Sidewalks Heavy traffic Hills Street lights Unattended dogs Enjoyable scenery High crime Food stores and food service places Adapted from Brownson and focus groups Women s determinants study Geocoded addresses from local health & state departments Traffic, light Safety from crime Physical activity BRFSS, 2001 Traffic, light Safe from crime Physical activity BRFSS & NHIS Unattended dogs Sidewalks High crime Fruit and vegetable, total and saturated fat intake Wanko (2004) 28 Perceived barriers to PA Survey Obesity Objective height & weight Zenk (2005) 29 Food stores Self-report Fruit and vegetable BRFSS, 2001 intake FFQ Fruits & vegetables Total fat guideline Saturated fat guideline OR 1.53 (1.00,2.34) OR 1.57 (1.14,2.17) OR 1.27 (0.51,3.15) OR 1.07 (0.70,1.65) OR 0.59 (0.20,1.78) OR 0.82 (0.22,3.09) OR 0.93 (0.41,2.11) OR 0.53 (0.22,1.32) OR 0.95 (0.52,1.76) OR 0.66 (0.36,1.21) OR 1.05 (0.34,3.27) OR 2.43 (1.19,4.99) OR 0.75 (0.36,1.55) OR 1.32 (0.73,2.38) OR 1.51 (1.06,2.15) OR 1.51 (0.98,2.32) OR 0.91 (0.58,1.42) OR 1.54 (1.11,2.12) OR 1.22 (1.03,1.44) OR 1.30 (1.07,1.56) a Rohm Young, Sanderson, Ainsworth, Eyler, and Wilbur all used the Women and PA Survey for environmental measures and the BRFSS for assessment of physical activity BRFSS, Behavioral Risk Factor Surveillance System; FFQ, food-frequency questionnaire; NHIS, National Health Interview Survey; PA, physical activity Lighter traffic and the presence of sidewalks were associated meeting PA recommendations; social factors also associated Physical environmental factors were not consistently related to PA For any active vs inactive, physical environment factors not associated with PA level; social factors (few social roles) associated with PA For any active vs inactive, physical environment factors not associated with PA level; social factors (seeing people exercise, attending religious services) associated with PA Safety from crime was associated with any PA vs no activity Among AA, no association between perceived neighborhood environment supports and PA Among AA, unattended dogs was the only attribute associated with any PA vs no activity The presence of supermarkets and the local food environment increased the likelihood of meeting dietary recommendations BMI OR 1.04 (1.01,1.06) A higher BMI increased patients likelihood of reporting an exercise barrier Supermarket beta 1.22 (0.33) Shopping at supermarket or specialty store Specialty food beta 2.37 (1.06) increased likelihood to meet fruit and stores vegetable recommendations

6 Table 3. Main characteristics and results for articles with 90% African Americans or no subgroup analyses Associations of interest Built environment/physical activity Perceived support for physical activity Study % African American % female Main association a Relative point estimate b Addy (2004) Moderate strong Brownson (2001) Weak Huston (2003) Moderate strong McNeill (2006) Weak moderate Sharpe (2004) Weak moderate Wilcox (2003) Weak Environmental barriers of Greenberg (2005) Weak physical activity Parks (2003) Moderate strong Urban form Berrigan (2002) Weak King (2005) Moderate Wilcox (2000) Weak moderate Built environment/diet Food stores Rose (2004) Unknown Weak a ( ) indicates significance and positive association; ( ) indicates significance and negative association; NA indicates no association b weak, OR ; moderate, OR ; strong, OR 2.0 activity with weak to moderate relative point estimates. 30,32 36 One article found a significant positive association between physical activity and neighborhood safety and a significant negative association with sidewalks and traffic. 30 A number of studies found significant positive associations of physical activity with environmental barriers and living in neighborhoods built pre-1950, which represents urban form conducive for walking Store proximity had a weak but statistically significant association with dietary intake in one study. 42 Discussion The results from this review summarize the growing body of literature on the built environment and its association with health behaviors and obesity among African Americans. To our knowledge this is the first review to summarize the literature examining the association between the built environment and physical activity, dietary intake, and obesity among African Americans. Few environmental attributes were consistently associated with physical activity and diet; the presence of light traffic, sidewalks, and safety from crime were more often positively associated with physical activity. Safety from crime was shown to have the strongest association with physical activity among urban dwellers. Among the two studies assessing the built environment and dietary intake, the presence of supermarkets and specialty food stores were consistently positively associated with meeting guidelines for fruit, vegetable, total fat, and saturated fat intake. Results from the study on obesity indicated that a higher BMI increased the likelihood of reporting a barrier to physical activity. Similar environments may have different health impacts on African-American populations compared to other populations. 26 Behavioral research has indicated that physical activity preferences, such as group versus individual exercises, tend to differ between African Americans and whites. Thus, environments that support a particular type of physical activity, such as safe streets to run or recreation centers, may not be conducive if African Americans do not prefer that activity. Additionally, traditional dietary patterns may present different challenges for African Americans compared with other populations. 43,44 Indeed, there is evidence to suggest that African Americans make food choices different from whites Some supermarkets, although providing healthy foods generally, may not provide the types of healthy foods that are preferred by African Americans. Understanding cultural and ethnic differences is important for successful environmental interventions to improve health and health behaviors. 48 For this review, 22 articles were identified that measured the built environment using either objective or perceived measures and assessed the association with physical activity, dietary behaviors, or obesity in African Americans. A previous review 11 assessed objective or perceived measures of the built environment and the association with physical activity. While accessibility, opportunities, and aesthetic attributes had significant associations with physical activity in their review, safety had a weaker association. Although safety from crime was shown to have one of the strongest relationships with physical activity in the current review, it was not a consistent predictor. In a previous nonsystematic review on the built environment and obesity among children and adults, 19 the authors noted that the investigative methods used to assess relationships were dissimilar across studies and varied between objective and perceived. Although a variety of methods were used to assess the built envi- February 2009 Am J Prev Med 2009;36(2) 179

7 ronment in the current review, there were a number of articles that used the same physical activity survey or an adaptation of the survey among different study populations of African Americans. Even though some of the articles in the current review are directly comparable by assessment method, direct comparisons are still difficult due to differences in study location. A recent systematic review 20 on the built environment and obesity identified 20 articles that had a direct measure of weight status and at least one objective measure of the built environment. Only one study examined differences in the built environment and obesity by race/ethnicity. Although most studies demonstrated a statistically significant association between at least one aspect of the built environment and obesity, the authors noted that methodological differences made comparisons across studies difficult. Similar to the current findings, the previously published review found that most investigations were cross-sectional study designs and did not focus on both physical activity and diet. However, few of the studies in the current review used objective measures of the built environment or assessed obesity as the major outcome. Future research on the built environment should focus on consistency in measurement, methodology, longitudinal study designs, and racial/ethnic differences. The majority of studies eligible for review focused on participants residing in urban areas. Features of the built environment may vary considerably among rural, urban, and suburban locations. Thus, individuals are likely to interact differently depending on the type of environment. These geographic differences are important to understand when conceptualizing and assessing the ways in which the built environment affects health and health behaviors. Therefore, more investigation of the rural environment is warranted. This review has several strengths and limitations. Even though a systematic review was conducted, a meta-analysis was not feasible due to the small number of eligible articles and the varied methods used to measure environment and outcomes of interest. As previously noted in the literature, researchers interested in studying the built environment would benefit from consistent measures so comparisons and conclusions could be more readily drawn across studies. Compared to other reviews on the built environment that focused only on obesity as an outcome, this review was broadened to include physical activity and dietary practices as outcomes since they are major health behaviors influencing obesity that African Americans disproportionately face. 11,19,20 Although varied aspects of the built environment have been associated with health outcomes and behaviors, more research is needed to draw definitive conclusions and make policy recommendations. First, more studies using objective measures of the built environment, such as sidewalk or food-availability audits, are needed to develop new policy initiatives. Since objective measures can be directly changed in the environment, they provide the greatest support for policy change and implementation. It is through objective changes in combination with individually-targeted health promotion efforts that perceived barriers or perception of an individual s surrounding environment may change and, consequently, affect individual health behavior. On a cautionary note, a disconnect between observational studies and interventions has been noted. 49 Associations between factors in the built environment and health behaviors that are not significant in observational studies were significant when implemented in an intervention. Thus, the focus should be on the theory and methods of observational studies before large intervention studies are conducted. Second, longitudinal study designs are necessary to identify causal associations, and interventions are needed to determine the health impact of environmental changes. Third, researchers should devote time to constructing consistent methodologies that would enhance the understanding of measured constructs and underlying mechanisms. Standard measurements would allow more accurate comparisons across studies. Finally, given the growing number of studies indicating the importance of physical contexts on individual health and the unsuccessful results of individual-level interventions, future research assessing the impact of the built environment on health is needed. However, with relatively few studies in the literature focused on African Americans, the findings for features of the built environment that are associated with physical activity, diet, and obesity are inconclusive. Future research goals should focus on the inclusion of African Americans to clarify the relationships in this subgroup. The project was funded by a grant from the National Institutes of Health. Dr. Gary was funded by a grant from NHLBI (K01-HL084700). This project was also supported by the African American Collaborative Obesity Research Network (AACORN). No financial disclosures were reported by the authors of this paper. References 1. CDC. Trends in overweight and obesity. Atlanta GA: USDHHS, Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, JAMA 2006;295: Casagrande SS, Wang Y, Anderson C, Gary TL. Have Americans increased their fruit and vegetable intake? The trends between 1988 and Am J Prev Med 2007;32: Kurian AK, Cardarelli KM. Racial and ethnic differences in cardiovascular disease risk factors: a systematic review. Ethn Dis 2007;17: CDC. Prevalence of physical activity, including lifestyle activities among adults United States MMWR Morb Mortal Wkly Rep American Journal of Preventive Medicine, Volume 36, Number 2

8 6. Patterson BH, Block G, Rosenberger WF, Pee D, Kahle LL. Fruit and vegetables in the American diet: data from the NHANES II survey. Am J Public Health 1990;80: Patterson BH, Harlan LC, Block G, Kahle L. Food choices of whites, blacks, and Hispanics: data from the 1987 National Health Interview Survey. Nutr Cancer 1995;23: Taylor WC, Baranowski T, Young DR. Physical activity interventions in low-income, ethnic minority, and populations with disability. Am J Prev Med 1998;15: Susser M. The tribulations of trials intervention in communities. Am J Public Health 1995;85: Svetkey LP, Stevens VJ, Brantley PJ, et al. Comparison of strategies for sustaining weight loss: the weight loss maintenance randomized controlled trial. JAMA 2008;299: Humpel N, Owen N, Leslie E. Environmental factors associated with adults participation in physical activity: a review. Am J Prev Med 2002;22: Atkinson JL, Sallis JF, Saelens BE, Cain KL, Black JB. The association of neighborhood design and recreational environments with physical activity. Am J Health Promot 2005;19: Frank LD, Schmid TL, Sallis JF, Chapman J, Saelens BE. Linking objectively measured physical activity with objectively measured urban form: findings from SMARTRAQ. Am J Prev Med 2005;28: Berke EM, Koepsell TD, Moudon AV, Hoskins RE, Larson EB. Association of the built environment with physical activity and obesity in older persons. Am J Public Health 2007;97: Cohen DA, McKenzie TL, Sehgal A, Williamson S, Golinelli D, Lurie N. Contribution of public parks to physical activity. Am J Public Health 2007;97: Diez Roux AV, Nieto FJ, Caulfield L, Tyroler HA, Watson RL, Szklo M. Neighborhood differences in diet: the Atherosclerosis Risk in Communities (ARIC) Study. J Epidemiol Community Health 1999;53: Ellaway A, Macintyre S. Does where you live predict health related behaviours?: a case study in Glasgow. Health Bull (Edinb) 1996;54: Morland K, Wing S, Diez Roux AV. The contextual effect of the local food environment on residents diets: the atherosclerosis risk in communities study. Am J Public Health 2002;92: Booth KM, Pinkston MM, Poston WS. Obesity and the built environment. J Am Diet Assoc 2005;105:S110 S Papas MA, Alberg AJ, Ewing R, Helzlsouer KJ, Gary TL, Klassen AC. The built environment and obesity. Epidemiol Rev 2007;29: Ainsworth BE, Wilcox S, Thompson WW, Richter DL, Henderson KA. Personal, social, and physical environmental correlates of physical activity in African-American women in South Carolina. Am J Prev Med 2003;25: Eyler AA, Matson-Koffman D, Young DR, et al. Quantitative study of correlates of physical activity in women from diverse racial/ethnic groups: the Women s Cardiovascular Health Network Project summary and conclusions. Am J Prev Med 2003;25: Rohm YD, Voorhees CC. Personal, social, and environmental correlates of physical activity in urban African-American women. Am J Prev Med 2003; 25: Sanderson BK, Foushee HR, Bittner V, et al. Personal, social, and physical environmental correlates of physical activity in rural African-American women in Alabama. Am J Prev Med 2003;25: Wilbur J, Chandler PJ, Dancy B, Lee H. Correlates of physical activity in urban Midwestern African-American women. Am J Prev Med 2003;25: Hooker SP, Wilson DK, Griffin SF, Ainsworth BE. Perceptions of environmental supports for physical activity in African American and white adults in a rural county in South Carolina. Prev Chronic Dis 2005;2:A King AC, Castro C, Wilcox S, Eyler AA, Sallis JF, Brownson RC. Personal and environmental factors associated with physical inactivity among different racial ethnic groups of U.S. middle-aged and older-aged women. Health Psychol 2000;19: Wanko NS, Brazier CW, Young-Rogers D, et al. Exercise preferences and barriers in urban African Americans with type 2 diabetes. Diabetes Educ 2004;30: Zenk SN, Schulz AJ, Hollis-Neely T, et al. Fruit and vegetable intake in African Americans income and store characteristics. Am J Prev Med 2005;29: Wilcox S, Bopp M, Oberrecht L, Kammermann SK, McElmurray CT. Psychosocial and perceived environmental correlates of physical activity in rural and older African American and white women. J Gerontol B Psychol Sci Soc Sci 2003;58: Eyler AA, Matson-Koffman D, Rohm YD, et al. Quantitative study of correlates of physical activity in women from diverse racial/ethnic groups: Women s Cardiovascular Health Network Project introduction and methodology. Am J Prev Med 2003;25: Addy CL, Wilson DK, Kirtland KA, Ainsworth BE, Sharpe P, Kimsey D. Associations of perceived social and physical environmental supports with physical activity and walking behavior. Am J Public Health 2004;94: Brownson RC, Baker EA, Housemann RA, Brennan LK, Bacak SJ. Environmental and policy determinants of physical activity in the United States. Am J Public Health 2001;91: Huston SL, Evenson KR, Bors P, Gizlice Z. Neighborhood environment, access to places for activity, and leisure-time physical activity in a diverse North Carolina population. Am J Health Promot 2003;18: McNeill LH, Wyrwich KW, Brownson RC, Clark EM, Kreuter MW. Individual, social environmental, and physical environmental influences on physical activity among black and white adults: a structural equation analysis. Ann Behav Med 2006;31: Sharpe PA, Granner ML, Hutto B, Ainsworth BE. Association of environmental factors to meeting physical activity recommendations in two South Carolina counties. Am J Health Promot 2004;18: Greenberg MR, Renne J. Where does walkability matter the most? An environmental justice interpretation of New Jersey data. J Urban Health 2005;82: Parks SE, Housemann RA, Brownson RC. Differential correlates of physical activity in urban and rural adults of various socioeconomic backgrounds in the United States. J Epidemiol Community Health 2003;57: Berrigan D, Troiano RP. The association between urban form and physical activity in U.S. adults. Am J Prev Med 2002;23: King WC, Belle SH, Brach JS, Simkin-Silverman LR, Soska T, Kriska AM. Objective measures of neighborhood environment and physical activity in older women. Am J Prev Med 2005;28: Wilcox S, Castro C, King AC, Housemann R, Brownson RC. Determinants of leisure time physical activity in rural compared with urban older and ethnically diverse women in the United States. J Epidemiol Community Health 2000;54: Rose D, Richards R. Food store access and household fruit and vegetable use among participants in the U.S. Food Stamp Program. Public Health Nutr 2004;7: Airhihenbuwa CO, Kumanyika S, Agurs TD, Lowe A, Saunders D, Morssink CB. Cultural aspects of African American eating patterns. Ethn Health 1996;1: James DC. Factors influencing food choices, dietary intake, and nutritionrelated attitudes among African Americans: application of a culturally sensitive model. Ethn Health 2004;9: Arab L, Carriquiry A, Steck-Scott S, Gaudet MM. Ethnic differences in the nutrient intake adequacy of premenopausal U.S. women: results from the Third National Health Examination Survey. J Am Diet Assoc 2003;103: Cabe-Sellers BJ, Bowman S, Stuff JE, Champagne CM, Simpson PM, Bogle ML. Assessment of the diet quality of U.S. adults in the Lower Mississippi Delta. Am J Clin Nutr 2007;86: Hoerr SL, Tsuei E, Liu Y, Franklin FA, Nicklas TA. Diet quality varies by race/ethnicity of Head Start mothers. J Am Diet Assoc 2008;108: Kumanyika SK, Whitt-Glover MC, Gary TL, et al. Expanding the obesity research paradigm to reach African American communities. Prev Chronic Dis 2007;4:A Brug J, van Lenthe FJ, Kremers SP. Revisiting Kurt Lewin: how to gain insight into environmental correlates of obesogenic behaviors. Am J Prev Med 2006;31: February 2009 Am J Prev Med 2009;36(2) 181