In 2006, the French Institute for Public Health Surveillance was commissioned by the Directorate General for Health to assess the impact of cold weather on health, identify measures likely to reduce this impact, and evaluate the feasibility of an alert system to support the implementation of all or some of these measures. To this end, eight criteria were defined to test the relevance and feasibility of a cold-weather and health alert system. Some of these criteria were verified, notably the existence of a documented and preventable impact. We also demonstrate, based on a time-series study applied to Paris and Marseille, that it is possible to develop biometeorological indicators predictive of mortality and to propose thresholds for a defined excess mortality. Finally, health indicators are available and allow for monitoring the impact of cold weather. However, the main preventive measures are structural measures concerning housing improvements and access to clean, affordable energy. Short-term measures, on the other hand, are limited to communication (dissemination of protective advice) and can be implemented based on weather alerts such as those already available via Météo-France’s vigilance map. This study therefore concluded that it was not appropriate to implement a complex monitoring and alert system based on meteorological and health data, but that it was important to advance research on the relationship between cold weather and health and to improve long-term cold-weather prevention measures. (R.A.)