Table 3.
RRs of HA-AKI by a 1 MAD increase in the ambient level of air pollutants
| Pollutant | RR (95% CI) from single-pollutant modelsa |
RR (95% CI) from six-pollutant model with 1-day lag | |||
|---|---|---|---|---|---|
| No lag | 1-day lag | 2-day lag | 3-day lag | ||
| PM2.5 | 1.000 (0.981, 1.020) | 1.008 (0.987, 1.028) | 1.006 (0.986, 1.026) | 1.006 (0.986, 1.026) | 0.963 (0.922, 1.007) |
| PM10 | 1.009 (0.986, 1.032) | 1.022 (0.999, 1.047) | 1.018 (0.994, 1.042) | 1.023 (0.999, 1.047) | 1.029 (0.980, 1.082) |
| NO2 | 1.042 (1.016, 1.067) | 1.051 (1.025, 1.077) | 1.046 (1.020, 1.072) | 1.036 (1.011, 1.062) | 1.063 (1.026, 1.101) |
| SO2 | 1.021 (1.001,1.041) | 1.022 (1.002, 1.043) | 1.030 (1.010, 1.051) | 1.017 (0.997, 1.038) | 1.000 (0.975, 1.026) |
| CO | 1.009 (0.987, 1.031) | 1.014 (0.992, 1.036) | 1.009 (0.987, 1.032) | 1.008 (0.986, 1.030) | 0.996 (0.964, 1.029) |
| O3 | 1.003 (0.971, 1.036) | 0.997 (0.965, 1.030) | 1.031 (0.998, 1.065) | 1.012 (0.979, 1.046) | 0.988 (0.955, 1.022) |
a
RRs (95% CIs) were estimated from the conditional Poisson regression models with adjustment for ambient temperature and scaled by 1 MAD of the daily concentration of the air pollutant. Since a 1-day lag produced the strongest association in the single-pollutant models, the six-pollutant model was analyzed with a 1-day lag only.