Table 4.
Association between principal component scores with NDVI, tree canopy, and leaf area in adjusted models. Beta per IQR of greenness exposure. NDVI IQR = 0.053. Tree canopy IQR = 7.827. Leaf area IQR = 0.278.
| Outcome | Exposure | Adjusted β | 95% CI | p-Value | Adjusted p-Value |
|---|---|---|---|---|---|
| PC 1 | NDVI | 0.161 | −0.008, 0.330 | 0.062 | 0.111 |
| PC 2 | NDVI | 0.088 | −0.085, 0.262 | 0.316 | 0.355 |
| PC 3 | NDVI | 0.265 | 0.092, 0.438 | 0.003 | 0.026 |
| PC 1 | Tree Canopy | 0.194 | −0.027, 0.415 | 0.084 | 0.122 |
| PC 2 | Tree Canopy | 0.300 | 0.078, 0.523 | 0.009 | 0.026 |
| PC 3 | Tree Canopy | 0.296 | 0.068, 0.524 | 0.011 | 0.026 |
| PC 1 | Leaf Area Index | 0.085 | −0.116, 0.286 | 0.404 | 0.404 |
| PC 2 | Leaf Area Index | 0.173 | −0.030, 0.376 | 0.095 | 0.122 |
| PC 3 | Leaf Area Index | 0.280 | 0.075, 0.485 | 0.008 | 0.026 |
Values represent adjusted Betas and 95% confidence intervals per IQR of exposure. Principal Component scores were regressed against greenness metrics, adjusting for age, sex, race, income, education, and BMI. PC 1 = Natural vegetation and spaces. PC 2 = Trees. PC 3 = Playgrounds, parks, and sports fields. NDVI = normalized difference vegetation index. Adjusted p-value represents False Discovery rate adjusted p-value using the Benjamini-Hochberg procedure. The adjusted alpha was set at p ≤ 0.028. Significance was set at Adjusted p ≤ 0.05.