Table 2.
Associations of LBW and TLBW with quartiles of greenness exposure within a 3000-m buffer
| Quantiles | Median | LBW | TLBW | ||
|---|---|---|---|---|---|
| OR (95% CI) | P for trenda | OR (95% CI) | P for trenda | ||
| NDVI | |||||
| Q1 | 0.111 | 1 (Ref.) | <0.001 | 1 (Ref.) | <0.001 |
| Q2 | 0.242 | 0.984 (0.982–0.985) | 0.976 (0.975–0.978) | ||
| Q3 | 0.338 | 0.970 (0.969–0.972) | 0.963 (0.961–0.965) | ||
| Q4 | 0.512 | 0.961 (0.960–0.963) | 0.956 (0.955–0.958) | ||
| EVI | |||||
| Q1 | 0.074 | 1 (Ref.) | <0.001 | 1 (Ref.) | <0.001 |
| Q2 | 0.176 | 0.984 (0.982–0.985) | 0.976 (0.974–0.977) | ||
| Q3 | 0.242 | 0.972 (0.971–0.974) | 0.966 (0.964–0.968) | ||
| Q4 | 0.382 | 0.963 (0.961–0.964) | 0.958 (0.956–0.960) | ||
These associations are derived from the fully adjusted model, where covariates include maternal demographic characteristics, fetal variables, and environmental factors. The statistical tests are two-sided.
OR odds ratio, CI confidence interval, NDVI normalized difference vegetation index, EVI enhanced vegetation index, LBW low birth weight, TLBW term low birth weight.
aTested by introducing the median value of each quartile as a continuous variable in the logistic regression model.