Table 2. Multiscale Geographically Weighted Regression Results Using Age-Adjusted Female Breast Cancer Mortality Rates as the Dependent Variable.
| County variable (N = 2174)a | Mean standardized (SD) β estimate | Range | Counties with significant result, %b | Bandwidth |
|---|---|---|---|---|
| Intercept | 22.76 (1.05) | 19.19 to 26.78 | 100 | 89 |
| Lifestyle | ||||
| Smoking (% adults) | −0.75 (0.92) | −2.99 to 1.50 | 16.3 | 259 |
| Obesity (% adults) | 0.72 (0.02) | 0.67 to 0.75 | 100 | 2179 |
| Food environment index | −1.69 (0.70) | −2.85 to 0.36 | 80.3 | 389 |
| Long commute (% workers) | 0.20 (0.04) | 0.06 to 0.23 | 0 | 2179 |
| Exercise opportunities (% population) | −0.59 (0.81) | −3.46 to 1.05 | 13.5 | 280 |
| Population demographics | ||||
| Unemployment (% population aged ≥16 y) | −0.13 (0.06) | −0.31 to −0.10 | 0 | 2179 |
| Segregation (total population:White ratio) | −0.47 (0.41) | −1.81 to 0.02 | 22.6 | 851 |
| Disability (% population) | 0.40 (0.17) | 0.08 to 0.61 | 45.0 | 1672 |
| Income inequality (80th:20th percentile ratio) | −0.30 (0.01) | −0.34 to −0.27 | 0 | 2179 |
| Access to health care | ||||
| Uninsured (% population) | −0.17 (0.02) | −0.23 to −0.12 | 0 | 2179 |
| Mammograms (% adults screened) | −1.07 (0.16) | −1.29 to −0.74 | 100 | 2059 |
| Mental health care physicians (ratio to population) | −0.48 (0.92) | −2.40 to 2.21 | 14.0 | 418 |
| Primary care physicians (ratio to population) | −1.06 (0.57) | −2.07 to −0.01 | 40.6 | 925 |
| Hospital access (No. per capita) | −0.14 (0.03) | −0.18 to −0.04 | 0 | 2179 |
| Environment | ||||
| Mean radiance (Watts × cm−2 × sr−1) | 0.27 (0.04) | 0.22 to 0.37 | 42.4 | 2179 |
| Transit access (No. stops per capita) | −0.31 (0.02) | −0.34 to −0.24 | 83.0 | 2179 |
| Proportion of natural land per county | 0.10 (0.01) | 0.08 to 0.11 | 0 | 2179 |
| Grocery store access (No. per capita) | 0.33 (0.05) | 0.26 to 0.46 | 0 | 2179 |
a
Basis for comparison between states included in parentheses.
b
Threshold for significance P < .05.