Estimates for the season-specific percentage change in risk (95% CI) per 10 µg/m3 increase in pollutant and tests for modification by season*: Single pollutant models†
| All MI | STEMI | NSTEMI | ||||
|---|---|---|---|---|---|---|
| Pollutant‡ | Percentage change (95% CI) | p Value* | Percentage change (95% CI) | p Value* | Percentage change (95% CI) | p Value* |
| O3 | ||||||
| Autumn | −0.38 (−0.85 to 0.10) | 0.446 | −1.15 (−1.98 to −0.31) | 0.040 | −0.10 (−0.76 to 0.57) | 0.785 |
| Winter | 0.00 (−0.37 to 0.38) | 0.18 (−0.48 to 0.85) | 0.04 (−0.49 to 0.56) | |||
| Spring | 0.19 (−0.29 to 0.67) | 0.39 (−0.46 to 1.24) | −0.03 (−0.71 to 0.65) | |||
| Summer | −0.07 (−0.52 to 0.38) | −0.24(−1.04 to 0.56) | −0.14 (−0.77 to 0.50) | |||
| NO2 | ||||||
| Autumn | 0.11 (−0.26 to 0.47) | 0.419 | −0.07 (−0.71 to 0.58) | 0.060 | 0.13 (−0.38 to 0.64) | 0.571 |
| Winter | 0.16 (−0.18 to 0.51) | 0.26 (−0.34 to 0.86) | 0.12 (−0.36 to 0.60) | |||
| Spring | −0.05 (−0.39 to 0.28) | −0.55 (−1.13 to 0.03) | 0.38 (−0.09 to 0.86) | |||
| Summer | 0.23 (−0.27 to 0.73) | −0.21 (−1.09 to 0.68) | 0.64 (−0.07 to 1.36) | |||
| PM2.5 | ||||||
| Autumn | 0.98 (0.20 to 1.76) | 0.030 | 0.76 (−0.62 to 2.16) | 0.639 | 1.13 (0.05 to 2.23) | 0.119 |
| Winter | −0.70 (−1.52 to 0.13) | −0.92 (−2.35 to 0.54) | −1.13 (−2.27 to 0.02) | |||
| Spring | −0.39 (−1.09 to 0.31) | −0.41 (−1.64 to 0.84) | −0.46 (−1.45 to 0.54) | |||
| Summer | 0.01 (−1.12 to 1.15) | −1.28 (−3.26 to 0.75) | −0.19 (−1.78 to 1.42) | |||
| PM10 | ||||||
| Autumn | 0.51 (−0.14 to 1.16) | 0.155 | 0.53 (−0.62 to 1.70) | 0.280 | 0.47 (−0.44 to 1.39) | 0.510 |
| Winter | −0.43 (−0.99 to 0.15) | −0.75 (−1.75 to 0.26) | −0.64 (−1.43 to 0.15) | |||
| Spring | −0.61 (−1.20 to −0.01) | −0.26 (−1.30 to 0.80) | −0.79 (−1.62 to 0.04) | |||
| Summer | −0.13 (−0.87 to 0.61) | −1.06 (−2.37 to 0.26) | −0.20 (−1.24 to 0.86) | |||
*The p values in the table relate to likelihood ratio tests for a season interaction. Seasons defined as: Autumn (September–November); winter (December–February); spring (March–May); and summer (June–August).
†The conditional logistic regression model fits the pollutant as unconstrained distributed lags 0–2 and adjusts for, the weekly RCGP influenza-like illness consultation rates per 100 000 England and Wales population, two natural cubic splines (df=5) for temperature (mean lag 0–1 and mean lag 2–6), public holidays and a sine/cosine annual cycle.
‡Pollutant metrics: daily mean PM2.5, daily mean PM10, daily maximum 1-hour NO2, daily maximum 8-hour mean O3.
MI, myocardial infarction; NSTEMI, non-ST-elevation MI; STEMI, ST-elevation MI.