Table 3.
Wave IV health outcomes and air pollution exposure (2002–2007) in adjusted GEE logistic regression models.a
| Ozone |
PM2.5 | |||
|---|---|---|---|---|
| Health outcome (n) | OR (95% CI) |
p-value | OR (95% CI) |
p- value |
| Hypertension (n = 10,883) | 1.015 (1.011, 1.029) | 0.0342 | 1.022 (1.001, 1.045) | 0.0457 |
| Hyperlipidemia (n = 11,259) | 1.009 (0.991, 1.028) | 0.310 | 1.013 (0.986, 1.041) | 0.353 |
| Obese (n = 11,101) | 1.022 (1.004, 1.040) | 0.0192 | 0.991 (0.965, 1.017) | 0.476 |
| Diabetes (n = 11,259) | 1.032 (1.009, 1.054) | 0.00502 | 0.975 (0.938, 1.014) | 0.208 |
| Inflammation (n = 10,514) | 1.012 (0.999, 1.025) | 0.0627 | 1.003 (0.982, 1.025) | 0.759 |
| Metabolic syndrome (n = 9,518) | 1.028 (1.014, 1.041) | 0.0000770 | 0.994 (0.967, 1.022) | 0.692 |
a
GEEs were fit as logistic regression models with a binary health outcome, Primary Sampling Unit School Identifier (PSUSCID) as the cluster ID variable, and an exchangeable correlation matrix. Models adjusted for 2002-07 pollutant concentration, age at Wave IV (years), sex (female or male), and race/ethnicity (non-Hispanic Black, non-Hispanic White, Hispanic, and Other).