Table 2.
Neighborhood violence environment's association with BMI and WC (N = 90).
| Beta coefficient (β) | Standard Error (SE) | p-Value | |
|---|---|---|---|
| Model 1. Effect of neighborhood violent crime on BMI | |||
| Intercept | 8.709 | 3.080 | 0.007 |
| Number of violent crimes within 500 m | 1.241 | 0.489 | 0.016 |
| Child sex (female vs. male) | −0.125 | 1.113 | 0.911 |
| Child age (ordinal) | 2.268 | 0.858 | 0.012 |
| Maternal education (ordinal) | 1.467 | 0.674 | 0.036 |
| Lower daily fruit and vegetable consumption (yes) | 0.191 | 1.156 | 0.869 |
| Lower daily physical activity (yes) | 2.787 | 1.394 | 0.053 |
| R-square | 27% | ||
| c (PCV)a | 70% | ||
| Model 2. Effect of neighborhood violent crime on waist circumference | |||
| Intercept | 35.292 | 9.503 | 0.001 |
| Number of violent crimes within 500 m | 0.723 | 0.493 | 0.151 |
| Child sex (female vs. male) | 1.510 | 3.138 | 0.634 |
| Child age (ordinal) | 9.882 | 2.417 | 0.0002 |
| Maternal education (ordinal) | 0.915 | 2.317 | 0.695 |
| Lower daily fruit and vegetable consumption (yes) | 0.256 | 3.467 | 0.942 |
| Lower daily physical activity (yes) | 5.510 | 4.271 | 0.206 |
| R-square | 15% | ||
| Proportional Change in Variance (PCV)a | 29% | ||
| Model 3. Indirect effect of telomere length on BMI | |||
| Indirect Effect = 0.318 (95% Bootstrapped CI = 0.048, 0.806) = the proportion of the effect of neighborhood violent crime on BMI that acts through telomere length | |||
| Model 4. Indirect effect of telomere length on waist circumference | |||
| Indirect Effect = 0.201 (95% Bootstrapped CI = 0.038, 0.520) = the proportion of the effect of neighborhood violent crime on WC that acts through telomere length | |||
a
PCVneighborhood(n) = [(Vn−1 − Vn−2) / Vn−1], where Vn−1 = neighborhood variance in the empty model and Vn−2 is neighborhood variance in the model including violent crime rate.