Article Text
Abstract
Background In the UK, obesity is more common in some ethnic minority groups than in Whites but little is known about the extent to which ethnic minorities are more exposed to obesity promoting environments. We examine whether area deprivation and ethnic density are associated with access to fast food outlets, supermarkets and physical activity structures.
Design Population sizes of Indians, Pakistanis, Bangladeshis, Black Caribbeans and Black Africans (2001 Census), income deprivation (Index of Multiple Deprivation), and number of fast food outlets, supermarkets, indoor (eg, sports clubs) and outdoor physical activity (eg, football grounds) structures were obtained for lower super output areas (LSOA). Ethnic density was measured using index of dissimilarity (evenness in distribution of a group relative to the White group), isolation index (extent to which ethnic minority group members are exposed to each other), cluster size (proportion in LSOA) and concentration (proportion of local authority district's ethnic population in an LSOA).
Setting England
Main Outcome Measures Rate ratios (RR), derived from multilevel Poisson models, using the rate of structures in low ethnic density areas as baseline rate.
Results Ethnic densities were generally higher in the most than least deprived areas, least consistent for Indians. Fast food outlets and supermarkets were also more likely to be found in the most than least deprived areas. In contrast, outdoor PA structures were more likely to found in least deprived areas. Adjusted for area deprivation, the index of dissimilarity and concentration measures reflected a pattern of more fast food outlets in high than low ethnic density areas. For example, RRs for fast food outlets in the highest ethnic density areas using the concentration measure were: Indians 1.91 (95% CI 1.22 to 3.00), Pakistanis 1.41 (1.05 to 1.89), Bangladeshis 1.80 (1.43 to 2.26), Black Caribbeans 1.29 (0.85 to 1.96), Black Africans 1.42 (1.01 to 1.98). Supermarkets were more likely to be in higher than lower-density Pakistani and Bangladeshi areas using these two ethnic density measures. Across all ethnic groups the concentration measure reflected a positive association with occurrence of indoor PA structures, while all ethnic density measures reflected a pattern of inverse association with occurrence of outdoor PA structures.
Conclusion These findings indicate that ethnic minorities might be more exposed to fast food outlets and less exposed to outdoor PA structures in high than low ethnic density areas. These issues might contribute to ethnic differences in food choices and engagement in physical activity.