Table 3

Association between body mass index and zip code-level characteristics of the food environment† for the study sample and stratified by poverty status‡ (estimate (95% CI))

Neighbourhood characteristic§Total study sample (N=37 584)¶Low poverty‡ (N=16 351)¶High poverty‡ (N=21 233)¶
Proportion poverty0.270 (0.084 to 0.457)**0.060 (−0.576 to 0.697)0.371 (0.058 to 0.685)*
Proportion African American0.115 (−0.002 to 0.233)0.261 (0.092 to 0.431)**0.031 (−0.135 to 0.197)
Neighbourhood walkability index0.029 (−0.113 to 0.171)0.007 (−0.190 to 0.204)0.006 (−0.224 to 0.237)
Food diversity−0.071 (−0.172 to 0.030)−0.079 (−0.226 to 0.068)−0.054 (−0.203 to 0.094)
Proportion of BMI-unhealthy food outlets0.261 (0.093 to 0.429)**0.507 (0.284 to 0.731)***−0.024 (−0.288 to 0.239)
Store density−0.324 (−0.446 to −0.202)***−0.273 (−0.444 to −0.102)**−0.305 (−0.496 to −0.113)**
  • *p<0.05.

  • **p<0.01.

  • ***p<0.001.

  • †All models adjust for gender, age, race/ethnicity, income-to-poverty ratio, nativity, marital status, self-reported health and employment and the β-coefficients presented in the table are mutually adjusted for the other variables listed in the table.

  • ‡Low-poverty zip codes are defined as zip codes below the median per cent of the population that is below the poverty level in New York City (median=17.1%).

  • §β-Coefficient estimated for a difference equivalent to a 1 IQR change.

  • ¶Number of individual level-1 units available for analysis after accounting for missing data from individual-level covariates.