The objective of this study is to estimate the number of lung cancer deaths attributable to radon in Corsica. The characterization of exposure is based on a residential measurement campaign conducted in 1995–1996. The risk model used made it possible to establish a lifetime-averaged relative risk for the entire population. Two hypotheses regarding the type of interaction between the effects of tobacco and radon were tested (submultiplicative and multiplicative interactions). The average of the measurements used is 197 Bq/m³, while the average level observed in France is 91 Bq/m³. After adjusting for geology, measurement season, and housing type, the average is 134 Bq/m³. The risk assessment indicates that 21.5 to 28.0% of lung cancer deaths are attributable to radon across the entire population (between 33 and 44 deaths per year). The proportion attributable to concentrations exceeding 400 Bq/m³ is estimated to be between 30% and 48%. To better estimate radon levels and thus its impact, a study on seasonal effects in Corsica and additional measurements in multi-unit housing are necessary. Due to the strong interaction between the effects of tobacco and radon, an accurate assessment of smoking habits in Corsica appears essential. (R.A.)