Estimated effects from models with threshold and indicator variables
| Model with indicator | Estimated effect for temperature below threshold % per °C | Estimated effect for temperature above threshold % per °C | Effect modification for high winter mortality, below threshold | Effect modification for high winter mortality, above threshold | Estimated effect for RH % per °C |
| TOT | 0.65 (0.17–1.14) | 0.71 (0.28–1.15) | -0.10 (-0.20–0.04) | 0.19 (-0.10–0.49) | 0.17 (0.06–0.27) |
| RES | 0.62 (0.16–1.08) | 0.84 (0.43–1.25) | -0.19 (-0.32– -0.05) | -0.12 (-0.42–0.17) | 0.15 (0.04–0.25) |
| INF | 0.49 (0.03–0.96) | 0.67 (0.23–1.10) | -0.14 (-0.27–0.00) | 0.03 (-0.26–0.32) | 0.15 (0.04–0.26) |
| CVD | 0.74 (0.27–1.21) | 0.88 (0.32–1.16) | -0.29 (-0.48– -0.10) | -0.04 (-0.35–0.26) | 0.16 (0.05–0.27) |
| CVD+RES | 0.79 (0.32–1.27) | 0.88 (0.46–1.31) | -0.29 (-0.49– -0.10) | -0.04 (-0.34–0.26) | 0.16 (0.05–0.26) |
Data are presented as relative risk point estimate (95% confidence interval). RH: relative humidity; TOT: total mortality; RES: respiratory mortality; INF: influenza mortality; CVD: cardiovascular mortality. The threshold corresponded to the 90th percentile of the 2-day mean summer temperature, 21.3°C.