Table 3.
Cumulative percent change (95% confidence interval)a in number of deaths associated with air pollutant levels for different cumulative lag structures.b
| Deaths | Percent change (95% confidence interval) | ||
|---|---|---|---|
| Lag 0 to 10 | Lag 0 to 20 | Lag 0 to 30 | |
| PM10 | |||
| Nonaccidental | 1.22 (0.84, 1.60) | 0.88 (0.33, 1.44) | –0.10 (–0.82, 0.62) |
| Circulatory | 0.72 (0.11, 1.33) | 0.04 (–0.85, 0.93) | –1.11 (–2.21, 0.01) |
| Cerebrovascular | 2.33 (1.14, 3.54) | 1.05 (–0.58, 2.71) | 0.72 (–1.39, 2.87) |
| Respiratory | 3.40 (2.45, 4.36) | 3.45 (2.05, 4.87) | 2.81 (0.99, 4.66) |
| Chronic respiratory | 2.95 (1.34, 4.57) | 1.58 (–0.75, 3.96) | 2.41 (–0.66, 5.57) |
| Cancer | 1.55 (0.75, 2.43) | 1.59 (0.50, 2.68) | 1.98 (0.58, 3.40) |
| NO2 | |||
| Nonaccidental | 1.47 (1.05, 1.89) | 1.15 (0.52, 1.78) | 0.22 (–0.60, 1.05) |
| Circulatory | 1.34 (0.66, 2.02) | 0.55 (–0.46, 1.57) | –1.45 (–2.77, –0.12) |
| Cerebrovascular | 1.99 (0.69, 3.30) | 0.62 (–1.20, 2.47) | –0.14 (–2.47, 2.24) |
| Respiratory | 3.18 (2.13, 4.25) | 3.15 (1.58, 4.76) | 2.98 (0.94, 5.06) |
| Chronic respiratory | 2.83 (1.03, 4.66) | 0.83 (–1.81, 3.54) | 1.34 (–2.09, 4.90) |
| Cancer | 1.84 (0.93, 2.76) | 2.38 (1.05, 3.72) | 2.35 (0.67, 4.06) |
| CO | |||
| Nonaccidental | 2.01 (1.05, 2.97) | 0.41 (–0.99, 1.83) | –1.47 (–3.29, 0.39) |
| Circulatory | 1.18 (–0.30, 2.68) | –0.11 (–3.29, 1.11) | –4.22 (–7.05, –1.30) |
| Cerebrovascular | 3.54 (0.70, 6.45) | 0.88 (–3.09, 5.01) | 0.34 (–4.78, 5.74) |
| Respiratory | 6.41 (3.96, 8.92) | 5.96 (3.36, 9.69) | 3.21 (–1.32, 7.94) |
| Chronic respiratory | 5.39 (1.37, 9.57) | 1.81 (–3.91, 7.86) | 2.82 (–4.59, 10.80) |
| Cancer | 1.36 (–0.64, 3.41) | 3.50 (0.50, 6.58) | 2.85 (–0.96, 6.81) |
| Notes: CO, carbon monoxide; NO2, nitrogen dioxide; PM10, particulate matter smaller than 10 μm. aAssociated with a 10-μg/m3 increase in PM10 and NO2 and with a 1-ppm increase in CO. bResults from a Poisson generalized additive distributed lag model, constrained with a second degree polynomial, using cumulative lag structures of lags 0–10, 0–20, and 0–30 days for PM10, NO2, and CO, adjusted by trend, seasonality, temperature, relative humidity, weekdays, and holidays. | |||