O3 concentrations | Model 1a
| Model 2b
| Model 3c
|
---|
Full year | Full year | August–September | October–July |
---|
Same day | 1.023 (1.004–1.043) | 1.023 (1.004–1.043) | 1.036 (1.007–1.066) | 1.017 (0.995–1.039) |
Pinteraction = 0.24 |
Lag 1 day | 1.022 (1.004–1.040) | 1.021 (1.003–1.040) | 1.025 (0.997–1.054) | 1.020 (0.999–1.041) |
Pinteraction = 0.75 |
Lag 2 day | 1.016 (0.998–1.034) | 1.016 (0.998–1.034) | 1.040 (1.012–1.069) | 1.006 (0.986–1.026) |
Pinteraction = 0.03 |
Lag 0–3 days | 1.031 (1.005–1.058) | 1.031 (1.005–1.057) | 1.053 (1.015–1.093) | 1.020 (0.991–1.050) |
Pinteraction = 0.12 |
-
HA hospital admission, ppb parts per billion, SD standard deviation, O
3
ozone
-
aModel 1 adjusted for cubic splines of calendar time (8 d.f. per year), cubic splines of same-day temperature (3 d.f.), cubic splines of the average of lag 1 through 3 day temperature (3 d.f.), cubic splines of same-day dew point (3 d.f.), cubic splines of the average of lag 1 through 3 day dew point (3 d.f.), day of the week, public holidays, total pollen counts (lag 0–2 days, categorical), and total respiratory infection HAs (lag 0–2 days, cubic splines with 3 d.f.)
-
bModel 2 was Model 1 with additional adjustment for the start of school year (indicator variable for August/September)
-
cModel 3 was Model 1 with additional adjustment for the start of school year (indicator variable for August/September) and the interaction between the start of school year and ambient O3 concentrations