Who
Who is most affected?
Every age group reduces its emergency department use on cold days, but not by the same amount and not on the same timetable: accounting for 30 days, the response ranges from 23.0% among people aged 12–20 down to an insignificant 1.4% among those aged 80+. Heat runs the other way and is concentrated in the young: on a hot day, visits rise 4.2 to 5.0% for people under 40, 2.5% for those aged 40–60, and do not rise for those over 60.
By age group
estimate, with the 95% interval beneath
| Group | ≤10 °C | 10–15 °C | 15–20 °C | 20–25 °C | 25–30 °C | >30 °C |
|---|---|---|---|---|---|---|
| Baseline (all) | −9.1%−10.0% to −8.1% | −5.3%−5.8% to −4.9% | −2.7%−3.0% to −2.4% | reference | +2.5%+2.1% to +2.8% | +3.7%+3.1% to +4.4% |
| 0–12 | −12.1%−13.3% to −10.9% | −6.7%−7.4% to −6.0% | −3.2%−3.7% to −2.8% | reference | +3.1%+2.5% to +3.7% | +4.8%+3.5% to +6.1% |
| 12–20 | −8.8%−10.0% to −7.5% | −5.8%−6.3% to −5.2% | −3.2%−3.5% to −2.8% | reference | +2.8%+2.4% to +3.3% | +5.0%+3.9% to +6.0% |
| 20–40 | −7.0%−8.2% to −5.7% | −4.7%−5.2% to −4.2% | −2.5%−2.8% to −2.2% | reference | +2.5%+2.2% to +2.9% | +4.2%+3.4% to +5.0% |
| 40–60 | −9.8%−10.9% to −8.7% | −5.0%−5.7% to −4.3% | −2.5%−2.9% to −2.0% | reference | +2.0%+1.4% to +2.5% | +2.5%+1.6% to +3.4% |
| 60–80 | −9.2%−10.6% to −7.8% | −4.5%−5.3% to −3.7% | −1.8%−2.4% to −1.3% | reference | +1.1%+0.4% to +1.7% | −0.3%−1.4% to +0.7% |
| 80+ | −8.1%−10.0% to −6.2% | −3.8%−5.0% to −2.5% | −1.7%−2.6% to −0.8% | reference | +0.9%−0.1% to +2.0% | −2.0%−4.3% to +0.3% |
estimate, with the 95% interval beneath
| Group | ≤10 °C | 10–15 °C | 15–20 °C | 20–25 °C | 25–30 °C | >30 °C |
|---|---|---|---|---|---|---|
| Baseline (all) | −16.0%−19.7% to −12.4% | −11.8%−13.9% to −9.7% | −5.3%−6.6% to −3.9% | reference | +4.5%+2.8% to +6.1% | +2.7%+0.4% to +5.0% |
| 0–12 | −11.7%−16.6% to −6.9% | −4.7%−7.8% to −1.5% | −1.1%−3.0% to +0.9% | reference | −1.2%−3.2% to +0.8% | −12.0%−15.7% to −8.2% |
| 12–20 | −23.0%−28.4% to −17.5% | −19.2%−21.9% to −16.6% | −9.4%−11.1% to −7.7% | reference | +7.9%+6.0% to +9.9% | +6.7%+3.4% to +10.1% |
| 20–40 | −19.5%−23.6% to −15.3% | −14.8%−17.1% to −12.5% | −6.8%−8.2% to −5.3% | reference | +7.4%+5.5% to +9.4% | +8.6%+5.9% to +11.2% |
| 40–60 | −15.5%−19.8% to −11.1% | −13.1%−15.8% to −10.4% | −6.3%−8.2% to −4.5% | reference | +5.2%+2.7% to +7.6% | +7.2%+4.6% to +9.7% |
| 60–80 | −8.6%−13.8% to −3.4% | −5.6%−8.9% to −2.3% | −2.5%−4.5% to −0.4% | reference | +1.3%−1.3% to +4.0% | +1.9%−1.9% to +5.6% |
| 80+ | −1.4%−10.1% to +7.3% | −5.1%−9.8% to −0.4% | −3.4%−6.8% to +0.1% | reference | +2.3%−2.0% to +6.6% | +5.8%−1.4% to +13.0% |
What to read here
- Cold, same day. Every age group comes less on a cold day. Children aged 0–12 show the largest same-day reduction, 12.1%; the other groups fall between 7.0% and 9.8%.
- Cold, 30 days. The gap opens in the weeks after. The 12–20 group reaches 23.0%, the largest cumulative effect of any age group, while among people aged 80+ it is a statistically insignificant 1.4%. This describes demand for emergency care, not deaths; the mortality literature finds the opposite age gradient (Cohen and Dechezleprêtre, 2022).
- Heat, same day. The increase is concentrated in the young: visits rise 4.2 to 5.0% for people under 40 and 2.5% for those aged 40–60. Over 60 there is no significant increase, and the over-80s show a 2.0% reduction.
- Heat, 30 days. For adults the effect persists: 6.7% at 12–20, 8.6% at 20–40, 7.2% at 40–60. For children aged 0–12 it reverses to a 12.0% reduction, so their same-day increase is followed by fewer visits in the weeks after. Over 60 the cumulative effect is not significant.
By sex
estimate, with the 95% interval beneath
| Group | ≤10 °C | 10–15 °C | 15–20 °C | 20–25 °C | 25–30 °C | >30 °C |
|---|---|---|---|---|---|---|
| Baseline (all) | −9.1%−10.0% to −8.1% | −5.3%−5.8% to −4.9% | −2.7%−3.0% to −2.4% | reference | +2.5%+2.1% to +2.8% | +3.7%+3.1% to +4.4% |
| Women | −8.1%−9.1% to −7.0% | −4.9%−5.4% to −4.5% | −2.6%−2.9% to −2.3% | reference | +2.3%+1.9% to +2.6% | +3.4%+2.6% to +4.1% |
| Men | −11.0%−12.1% to −10.0% | −6.1%−6.7% to −5.5% | −2.9%−3.3% to −2.5% | reference | +2.8%+2.3% to +3.3% | +4.4%+3.6% to +5.2% |
estimate, with the 95% interval beneath
| Group | ≤10 °C | 10–15 °C | 15–20 °C | 20–25 °C | 25–30 °C | >30 °C |
|---|---|---|---|---|---|---|
| Baseline (all) | −16.0%−19.7% to −12.4% | −11.8%−13.9% to −9.7% | −5.3%−6.6% to −3.9% | reference | +4.5%+2.8% to +6.1% | +2.7%+0.4% to +5.0% |
| Women | −16.8%−21.0% to −12.7% | −12.4%−14.6% to −10.2% | −5.8%−7.1% to −4.4% | reference | +4.9%+3.2% to +6.5% | +3.7%+1.4% to +6.0% |
| Men | −14.6%−18.5% to −10.7% | −10.7%−13.2% to −8.2% | −4.3%−5.9% to −2.8% | reference | +3.8%+1.8% to +5.7% | +1.0%−2.2% to +4.1% |
Women and men
The difference by sex is small next to the differences by age. Over 30 days the cold response is 16.8% for women and 14.6% for men, on a scale where the age groups span 1.4% to 23.0%.
What is being estimated
Each group is a separate Poisson distributed lag model fitted to that group alone, with 30 lags, and effects are measured against a day between 20 and 25 °C.
Every group is drawn on one set of axes, so the curves are directly comparable. The vertical scale differs between the same-day and 30-day views because the 30-day confidence intervals are roughly twice as wide.
Percentages are 100 times the estimated coefficient. Bands are 95% confidence intervals, clustered at the municipality level.