This study set out to examine if subjective reports of perceived neighborhood disorder were associated with objectively measured physical activity and obesity status among African American adolescents. For the current study, accelerometer data indicated that the adolescents were engaged in low levels of moderate-to-vigorous physical activity (less than 30 minutes daily) and almost half of the adolescents (43 percent) were either overweight or obese. Using the conceptual model developed by Burdette and Hill , it was hypothesized that living in disordered neighborhoods would affect obesity status through less participation in physical activity. The current research provides partial support for the conceptual model and suggests that while adolescents’ perceptions of neighborhood disorder are significantly related to obesity status, physical activity may not be a significant mediator of this relationship.
The current study findings are supported by the literature evaluating relationships of neighborhood social context on obesity. Evenson and colleagues assessed the relationship between neighborhood factors and BMI for adolescent girls and their findings suggest that neighborhood factors such as low levels of crime, seeing other children playing outside, and the availability of recreational facilities are associated with lower BMI . Although they note that the influence of neighborhood factors is minimal, the neighborhood context may be important and may operate to protect adolescent girls from obesity . In the current study, the associations are also minimal, but statistically significant. While some research studies do not find significant associations of neighborhood context with obesity among children [29, 44], the associations observed during adolescence may suggest that the obesogenic effects of neighborhood disorder emerge over time. For the current study, adolescents who live in disordered neighborhoods are more obese and because obesity tracks into adulthood, they may have increased risks for adverse health outcomes later in life. Future studies should longitudinally assess perceived neighborhood disorder characteristics and childhood adiposity to examine the timing and extent to which perceived neighborhood disorder characteristics begin to develop into increased obesity risk among youth.
Contrary to the expected findings outlined by the conceptual model, we did not find significant associations between perceived neighborhood disorder and objectively measured physical activity. While a few studies report significant associations between neighborhood disorder and physical activity, these studies typically rely upon self-reported physical activity assessments [32, 45–47]. However, the use of accelerometry in the current study provides an objective measurement of physical activity and contributes to the literature that suggests a null relationship. Previous studies using objectively measured physical activity also indicate that neighborhood safety and disorder may not be associated with child and adolescent physical activity levels . Studies that examine area-level SES also do not find significant relationships with physical activity . Cumulatively, these research findings may suggest that neighborhood social contexts are not associated with physical activity when physical activity is measured objectively.
However, there may be specific neighborhood features of social and physical disorder that directly affect physical activity. When we examined bivariate correlations between individual items from the perceived neighborhood disorder scale and physical activity (data not shown), specific elements of neighborhood disorder such as drug use in the neighborhood, was significantly and negatively associated with participation in moderate-to-vigorous physical activity among adolescents. Evenson and colleagues also examined specific social and built environment factors related to physical activity and showed that while perceived neighborhood crime is not associated with physical activity, factors such as street lights, recreational facility access, and the presence of other children playing outside are associated with greater non-school related physical activity participation for adolescent girls . Therefore, while perceived neighborhood social disorder such as crime levels may decrease feelings of safety, it appears that population density on the streets may provide a buffering effect . Future studies that incorporate accelerometry should separate school based physical activity from non-school based physical activity and disaggregate features of neighborhood disorder to determine the extent to which neighborhood factors may affect physical activity among children. This type of research is especially important if children get most of their physical activity at school .
The current study findings provide support for the inverse association of moderate-to-vigorous physical activity with obesity status. Although most children in this study did not meet daily recommendations of at least 60 minutes of moderate-to-vigorous physical activity, their participation in physical activity appeared to be protective against obesity. We also examined other types of physical activity data from the accelerometers including models of participation in light or vigorous physical activity in association with obesity, but found no significant associations (data not shown). As a supplement to the objective accelerometer data, we also examined adolescents’ qualitative reports of the types of physical activities that they participated in and the duration of these activities. Among adolescent girls, the most frequently reported physical activities were walking for exercise, dance, and basketball. Among boys, the most frequently reported activities were basketball, football and walking for exercise. Although adolescents subjectively reported being physically active, the accelerometer data reflected that the activity levels were insufficient to meet the recommended physical activity guidelines. However, research findings from a nationally representative sample of children and adolescents, indicate that children and adolescents are more likely to engage in unstructured sporadic physical activity than structured physical activity . These sporadic bouts of physical activity are inversely associated with overweight and obesity and are independent of total physical activity. This suggests it is important to measure both sporadic physical activity in addition to a measure of moderate-to-vigorous physical activity .
Because physical activity did not appear to mediate the association of perceived neighborhood disorder and obesity, future research should analyze the additional neighborhood disorder pathway of diet, which is a key component of the perceived neighborhood disorder conceptual model outlined by Burdette and Hill . Fast food density and grocery store access, and fruit and vegetable availability are known to differ by neighborhood context [50–55] and these neighborhood features are associated with poorer diet quality and adolescent obesity [56, 57]. An exploration of the relationship to diet in conjunction with the neighborhood pathway through physical activity may provide more evidence for the mechanisms through which perceived neighborhood disorder affects adolescent obesity.
While informative, this study is not without limitation. The sample size is relatively small and was obtained from a limited geographical region. As such, the current findings may not be generalizable to African American adolescents as a whole. Further, the sample data are cross-sectional in nature and therefore causality could not be established. Also, while the use of accelerometer data strengthened the findings, the data presented may not represent regular physical activity patterns of adolescents throughout the year. Additionally, research suggests that since children engage in very short bouts of sporadic vigorous physical activity, the 60 second epoch length of the accelerometery may underestimate the amount of moderate-to-vigorous physical activity patterns of the adolescents . Further, this study did not include subjective or objective measures of stress as suggested by the conceptual model of Burdette and Hill . Future work should incorporate these measures to examine the extent to which neighborhood disorder directly affects psychological and physiologic functioning.
Despite the study limitations, this research has significant strengths and contributes to the neighborhood and health literature by incorporating a conceptual model to examine the mechanisms through which perceived neighborhood disorder may affect obesity risk among African American adolescents, a population who experiences higher prevalence of obesity. Further, the inclusion of traditional measures of body mass index coupled with objective measures to assess physical activity, strengthen the current study findings and suggest that perceived neighborhood disorder and the low levels of physical activity observed among adolescents may contribute to obesity. Future intervention efforts to reduce obesity among African American adolescents should be developed to address strategies to increase physical activity and to modify features of the perceived neighborhood environment contexts that are directly associated with obesity.