The prevalence of obesity among children and adolescents has tripled since the 1970s.^1–3 In the United States alone, 32% of children and adolescents ages 2 to 19 years are overweight.⁴ Although obesity has increased significantly throughout the population, sharply higher rates of obesity and being overweight occur among children and adolescents of urban lower socioeconomic status^5,6 and minority groups.⁴ Weight increases among children have prompted nationwide programs to prevent obesity and obesity-related illnesses among children, including First Lady Michelle Obama’s “Let’s Move” initiative aimed at improving food quality in schools, increasing healthy food accessibility, and enhancing parents’ nutrition and exercise knowledge. The significant impact that obesity has on the physical and psychosocial health of children^7,8 calls for a change in policy. There is an increased prevalence of insulin resistance, glucose intolerance, advanced growth/early maturity, hypertension, sleep apnea, and weight stigmatization^9–11 associated with obesity and being overweight during childhood. Further, these children are at greater risk for ill health persisting into adulthood.¹² Current estimates confirm that obesity is one of the nation’s most serious childhood health problems.^2,3,13 As such, there is a need to examine obesity-related risk^14–17 and protective factors (e.g., behavioral, environmental, socioeconomic, and cultural) among urban children.

Pioneering work of urban planners like Jane Jacobs observed that the design of physical spaces can create a repelling or an inviting effect.¹⁸ Using those findings, social scientists undertook empirical studies on neighborhood geography and its contribution to obesity and obesity-related disparities. For example, Franco et al.¹⁹ examined associations between neighborhood income and differential availability of healthy foods. Using a “healthy food availability index” and 159 census tracts, it was found that predominantly African-American, lower-income neighborhoods have lower access to healthy, less energy-dense food options (e.g., fruits, vegetables, low-fat foods) because of the types and locations of stores compared with predominantly white, higher-income neighborhoods.¹⁹ In a similar study by Morland et al.,²⁰ the authors provided additional evidence that making healthy dietary choices was difficult for people in lower-income communities because of the differential distribution of quality supermarkets and convenience stores by neighborhood wealth and racial segregation.

Other studies, using primarily adult samples, examined neighborhood physical and social surroundings and obesity outcomes. Research using the socioecologic model with adults noted associations between obesity and residential density, land use patterns,²¹ poor terrain/sidewalk conditions, fewer community recreational outlets,^22,23 proximity to fast food restaurants,^24,25 and reduced access to supermarkets.²⁶ Further, indicators of neighborhood quality, safety, disorder, and perceptions of neighborhood conditions as stressful contributed to decreased healthy nutrition seeking and physical activity engagement^27,28 among adults,^29–36 affecting obesity. Yet, there are few analyses of the association between aspects of the built environment and childhood obesity in published peer-reviewed journals. This is especially true for studies including ethnic minority youth.

Existing studies using youth samples have found correlations between the availability of safe places to play,^37,38 carbohydrate-rich dietary intake,^39–41 increased caloric intake, hours spent watching television,^42,43 physical activity,^44,45 and childhood obesity. Molnar et al.⁴⁶ showed that an environment (including social disorder and neighborhood safety) can affect children’s recreational involvement/physical activity and that unsafe neighborhoods may deprive children of time spent in outdoor play and increase their susceptibility for obesity.⁴⁶ Moreover, disadvantaged African-American and Hispanic youth fare worse than their majority white counterparts regarding their neighborhood’s features (e.g., abandoned buildings, poor lighting, poorly maintained sidewalks, and fewer green spaces),⁴⁷ and this pattern of risk also correlates with gender, with the highest obesity rates found among girls.⁴⁸

Despite these early findings, results do not consistently demonstrate an association between neighborhood safety and disorder and childhood obesity.^49–51 For example, in a study of the effects of neighborhood deprivation on strenuous exercise, Ross⁵² found no association between neighborhood disadvantage and walking; however, high levels of walking in low-income neighborhoods are likely due to greater reliance on public transportation.⁵³ Further, schoolchildren in such neighborhoods are more likely to walk to school.⁵⁴ The conflicting results regarding neighborhood effects on childhood obesity prompted literature reviews to identify conceptual gaps and methodologic limitations. The use of cross-sectional study designs and compositional measures derived from U.S. Census data (e.g., aggregated data used to approximate certain neighborhood characteristics) limits conclusions derived from studies using these designs.¹⁷ Additionally, the lack of theoretical foundations and understanding of interdependency among various built and social environmental variables was frequently identified as a limitation of existing studies.^55,56

In more recent years, a new method called the Neighborhood Inventory for Environmental Typology (NIfETy) was devised to objectively measure physical and social disorder in a neighborhood. The NIfETy evaluates both positive and negative features of the social environment and neighborhood structures, with an emphasis on those that are malleable.^57,58 This study uses the NIfETy method to examine associations between 11 physical and social hazards and being overweight in a sample of urban, primarily African-American, elementary schoolchildren enrolled in an ongoing study.⁵⁹ This study also examined the interaction of gender and environmental determinants of being overweight, because evidence shows that these determinants affect girls differently.^3,4

The conceptual model for the current study (Figure) is guided by the socioecologic model of behavior.^60,61 The socioecologic model of behavior holds that multiple contexts and elements in the environment are interrelated. Given that individuals and their behavior are nested within socioenvironmental influences, the socioecologic model of behavior in this conceptual model illustrates that neighborhood physical and social disorder hazards may restrict child residents’ range of behavior (e.g., by limiting safe opportunities for physical activity), which in turn increases the risk of obesity and being overweight.

This study investigated the association between urban neighborhood disorder and the risk of being overweight in childhood. Among urban schoolchildren, gender was associated with an increased likelihood of being overweight. An interaction term comparing girls with boys living in the same level of disorder was also associated with increases in overweight status in childhood, indicating that although urban male and female youth are affected by unhealthy weight status, urban neighborhood hazards may impact school-aged girls more than boys. Lastly, neither race nor comparative time spent on a sport was significantly associated with being overweight, after adjusting for demographic characteristics. The results of this study on children are similar to those in the extant adult literature, suggesting that neighborhood characteristics¹⁴ and physical surroundings encourage obesity, obesity-generating behaviors,^15,16 such as sedentary lifestyle and unhealthy eating habits,¹⁷ and poorer physical health status.^{23,27,69, 70}

Evidence from this and other studies tends to support the notion that disorder at the neighborhood level may affect obesity by keeping residents indoors. The lack of significant race effects is not surprising given the work of Powell et al.³⁷ Evidence from their work suggests that the reduced availability of physical activity settings in ethnic minority and socioeconomically disadvantaged communities is the likely culprit for higher weight and lower physical activity among this population. Similarly, the finding of an association between gender and being overweight is consistent with the literature.¹ Between 1999 and 2004, African-American and Hispanic female children and adolescents were significantly more likely to be overweight compared with their white counterparts.¹

The results of this study should be interpreted in view of its limitations. There are exogenous variables potentially related to obesity and neighborhood disorder that were not available for study, including a dietary intake measure. In addition, we did not quantitatively measure physical activity because of a lack of available metrics. Future studies should use a more robust and empirically validated measure for physical activity. Also, data on parental or familial history of obesity were not available and may have strengthened the study. Lastly, this is a cross-sectional study. Longitudinal prospective studies are needed to evaluate the association of neighborhood characteristics, gender, and weight over time.

A major criticism of prior research was the degree to which aggregated individual compositional measures limited the investigation of causal links between specific neighborhood features and health outcomes.⁷¹ The current study used an objective measure of neighborhood features, the NIfETy, which enabled a more direct exploration of environmental triggers for being overweight in childhood. Additionally, this work suggests measures for intervention. Intervention designs may benefit from the inclusion of family, school, and/or community involvement in efforts to increase active-recreational orientation for urban youth, particularly those that are gender specific, given that girls’ and boys’ obesity risk profiles (defined as the social, cultural, and environmental factors influencing obesity for children) differ. Further, strategies might also include increasing the development of safe neighborhood structures that support gender-specific physical activity. In addition, addressing neighborhood-based obesogenic risks in urban communities (e.g., revitalization projects to improve neighborhood quality and reduce crime) may lead to a more health-promoting environment. Finally, the Child Nutrition Act and campaigns such as the “Let’s Move” Initiative must include the importance of neighborhood factors associated with the development and remediation of obesity.

Although this study represents a first exploration, further research is warranted. Obesity and overweight status among urban, ethnic minority children is likely related to neighborhood socioeconomic status, neighborhood disorder, and the availability of safe physical activity settings. Accordingly, efforts to avert, suspend, and reverse this alarming trend may be more effective if culturally and ecologically relevant, long-term, and gender-specific interventions were employed at the individual, family, school, and/or community levels.