According to Météo-France, the summer of 2003 was the hottest in 53 years in terms of both maximum and, importantly, minimum temperatures. Furthermore, weather conditions contributed to a photochemical pollution episode that was exceptional in terms of its duration and geographic scope. From August 1 to 14, nearly the entire country experienced significant episodes of ozone pollution. Ozone, after particulate matter, is the pollutant for which the consistency and coherence of epidemiological study results support a causal relationship between exposure and adverse health effects. This heat wave was accompanied by a surge in short-term excess mortality of exceptional magnitude. Recommendations were formulated in reports published following the summer of 2003 to identify all factors that increase temperature-related risks and to quantify the respective roles of air pollution and heat in the observed excess mortality. Indeed, understanding these respective impacts would allow for more precise targeting of preventive measures and recommendations regarding these two risk factors. Following the heat wave, it was initially planned to conduct an assessment of the health impact of air pollution for the summer of 2003 independently of temperature effects. However, given the exceptionally high and prolonged levels of temperature and pollution (ozone) during the summer of 2003, assessing the health impact based on existing exposure-risk relationships appeared scientifically invalid and could lead to erroneous results. As part of the Air and Health Monitoring Program – 9 Cities (PSAS-9), the primary objective of this study was to reanalyze, taking into account the specific period of the summer of 2003, the relationship between ozone exposure and the risk of mortality. The purpose of this estimation was to conduct a health impact assessment for the heat wave period that was scientifically more valid than one based on previous results. Furthermore, this analysis sought to estimate, for the heat wave period, the excess mortality risk associated with both temperature and ozone, and the relative contribution of each of these factors to this combined effect. The final objective was to quantify a potential short-term shift in mortality, i.e., a period of lower-than-expected mortality occurring immediately following the heat wave.