Introduction. The eastern coast and mountain ranges of Corsica are among the areas in France that experienced the highest levels of fallout from the Chernobyl accident. In October 2000, in order to assess the health consequences, the Territorial Collectivity of Corsica commissioned the Regional Health Observatory of Corsica (ORS) to conduct studies. In early 2003, due to methodological difficulties, the French Institute for Public Health Surveillance (InVS), and in particular the Southern Interregional Epidemiology Unit (Cire), continued the work in agreement and collaboration with the ORS. Given the scientific knowledge regarding the health effects of the Chernobyl accident in the most exposed countries, it was decided to focus the study on thyroid cancers. Initially, it seemed necessary to compare Corsica with the rest of France by estimating the incidence of thyroid cancer in Corsica. Methods. The study includes patients of all ages residing in Corsica who were diagnosed with primary thyroid cancer between 1998 and 2001, the period for which data from the national database of the Medical Information Systems Program (PMSI) were available at the time the protocol was developed. Data were collected from three sources: 1) medical records from healthcare facilities identified via the national PMSI database (hospital stays of patients residing in Corsica with thyroid cancer as the primary or associated diagnosis and a thyroid surgery procedure or iodine therapy); 2) recognition of thyroid cancer as a long-term condition by local health insurance funds; 3) medical information from pathology laboratories. A database of incident cases of thyroid cancer was created by cross-referencing information from these various data sources. A description of the cases (age, sex, geographic distribution, histology, tumor size, etc.) was performed. Comparisons of incidence rates with departments having a cancer registry were conducted using incidence rates standardized to the world population. Results. In Corsica, the average annual age-standardized incidence rate of thyroid cancer is 12.7 per 100,000 person-years among women, a rate not significantly different from those observed in certain registries (Calvados, Isère, Marne-Ardennes, and Tarn). Among men, this rate is 6.8 per 100,000 person-years, which is statistically higher than all rates in the registries. Thyroid cancer affects women twice as often as men (a ratio of 3 in the Francim network registries). The median age is 46 years among men and 48.5 years among women. In 2000, according to the registries, it was 52 years for men and 51 years for women. Thirteen cases (approximately 10% of recorded cases) involve individuals who were aged 0 to 14 at the time of the Chernobyl accident. The proportion of tumors smaller than 1 cm (51%) is statistically higher than in the Marne-Ardennes registry (approximately 40%). The histology is primarily papillary (79%). Histologies other than papillary and vesicular are statistically less common than in the Francim network registries. Discussion. This study, whose implementation foreshadows the planned national multisource cancer surveillance project, shows a high incidence of thyroid cancer in Corsica, particularly among men. The case characteristics (age, tumor size, histological types) suggest a probable role for screening practices among patients residing in Corsica. However, these results justify the initiation of etiological studies on risk factors for thyroid cancer to better clarify the situation in Corsica. It is also necessary to continue research on the incidence of thyroid cancer in Corsica for the 2002–2005 period to verify whether these initial findings are confirmed.