Using a Geographical Information System to Assess Physical Environmental Influences on Walking in New Orleans

In New Orleans, only 39% of adults meet physical activity recommendations. In spite of being the simplest form of physical activity, walking is very effective. However, only few studies have explored the impact of physical environment on walking behavior. This study aimed to examine the relationship of crime rates and food store density with walking - both exercise-related and utility-related. Data on exercise/utility walking for 3116 participants in the New Orleans BRFSS and residence by census tract was extracted and analysed. Two crime databases provided information on reported violent crime and police complaint locations. A health department permit database provided information about retail food store locations. These data were geocoded to create census tract-level objective measures of crime rates and food store density. Other variables included in the analyses at the census tract level were socioeconomic status and household vehicle ownership. Additional variables at the individual level were demographics, socioeconomic status, attitudes toward exercise and walking, perceived neighborhood safety, and perceived neighborhood barriers to walking. Hierarchical modeling was used to study the relationships between tract-level crime and food store density and both exercise-related walking and utility-related walking, while controlling for individual covariates, such as gender, race, age and income. Several models were fitted for utility-related walking. After controlling for individual covariates, we found that people living in a neighborhood with higher food store density were more likely to engage in utility walking even when the crime rate was high. Motivation for walking seems to be necessity-driven, despite high crime rates.