Headlines - Cold Weather Kills Three Times More People Than Heat in France

Santé publique France analyzed (1) the impacts of heat and cold on overall mortality between 2000 and 2010 in 18 areas of metropolitan France (including Dijon) with a combined population of 15.5 million. The comparison between the different regions was based on the percentiles of observed average temperatures: in Dijon, 1% of days between 2000 and 2010 had a temperature below -4.4°C (1st percentile, or P1), 99% of days between 2000 and 2010 had an average temperature below 25.7°C (99th percentile, or P99), meaning 1% were above 25.7°C.

Across all 18 areas studied between 2000 and 2010, cold weather accounted for 3.9% (95% CI: [3.2–4.6]) of mortality (impact over 0–21 days), and heat accounted for 1.2% [1.1–1.2] (impact over 0–3 days). Temperatures observed on cold and hot days caused more than 60,000 deaths. Cold has a weak influence (38% for P0.1), but this effect is observed even at mild temperatures (P25), with a gradual increase persisting up to 21 days after exposure. Its overall impact is significant: 46,256 deaths. Heat strongly influences the risk of death, but at temperatures that remain rare (P99, i.e., 25.7°C in Dijon) with a very rapid increase in mortality (a 3% increase for P99, but 96% for P99.9). Heat is responsible for 13,855 deaths over 10 years. The net impact of heat 21 days after exposure is 5,804 deaths, which is three times less than that of cold.

Heat and cold require different public health responses. The high risk of death at very hot temperatures (P99.9) is not observed at very cold temperatures (P0.1) and suggests that protective mechanisms against extreme cold (home heating, clothing, behaviors) are effective for a large portion of the population. However, this protection is not accessible to a segment of the population, which must be identified.

Finally, deaths linked to moderate heat may be premature by a few days, as the initial excess mortality is partially offset by lower mortality in the following days (harvest effect). It can be assumed that these deaths affect people in very poor health, who are poorly acclimated to heat, and, for some, likely overexposed. Consideration of these high-risk individuals is necessary because these deaths occur at temperatures well below the alert thresholds of the National Heat Wave Plan.

This study highlights the strong non-linearity of the relationship between temperature and mortality: the asymmetry between cold and heat must be incorporated into prevention efforts, which should not be limited to cold or heat wave alerts.