TABLE 2.
Population unadjusted and adjusted percent differences and 95% confidence limits in perceived availability of healthy foods by categories of store densities and store variety (all models adjusted for race and categorical household income)
| Number of participants | Population unadjusteda | Population adjustedb | |
|---|---|---|---|
| Supermarket | 5,774 | ||
| Low density | −17.0 (−18.8, −15.1) | −16.8 (−18.6, −15.0) | |
| Medium density | −6.7 (−8.8, −4.6) | −5.7 (−7.8, −3.6) | |
| High density | Referent | Referent | |
| Smaller storesc | 2,044 | ||
| Low density | −6.3 (−10.4, −2.1) | −2.8 (−6.9, 1.4) | |
| Medium density | 4.4 (0.1, 8.9) | 7.7 (3.3, 12.2) | |
| High density | Referent | Referent | |
| Variety of smaller storec,d | 2,044 | ||
| Low variety | −5.0 (−8.9, −0.9) | – | |
| Medium variety | 2.6 (−1.7, 7.1) | – | |
| High variety | Referent | Referent |
aBased on tertiles of densities of stores per square mile pooled across sites. Categories of densities of supermarkets were based on cutoffs of 0.4 and 2.1 stores per square mile; categories of densities for smaller stores were 0.3 and 1.7 stores per square mile; categories for smaller store variety were based on cutoffs of 1.0 and 3.0.
bBased on tertiles of densities of stores per 1,000 population pooled across sites.
cDensities for smaller stores and store variety are restricted to persons in areas without supermarkets (n = 2,044).
dVariety measures were not investigated after population density adjustment because these measures are not densities.