INTRODUCTION:the potential health impacts due to the decommissioned Nuclear power plants (NPP) located in Borgo Sabotino and Garigliano in Central Italy (active from the early 1960s to the late 1980s) have raised several concerns. Brain, thyroid, breast and lung cancer and leukaemia have been associated with exposure to ionizing radiations, but the health effects of nuclear plants on the resident populations are controversial. OBJECTIVE:to evaluate whether living close to NPPs is associated with an increased risk of cancer incidence and mortality. METHODS:we defined a cohort of residents within 7 km from the NPPs during the period 1996-2002. Individual follow-up for vital status at 01.01.2007 was conducted using municipality data. Gender specific Standardized Incidence and Mortality Ratios, adjusted for age, were calculated (SIR and SMR) using the regional population as reference. Each participant's address was assigned to a distance from the NPP on the basis of a GIS. A relative risk (RR, CI95%), adjusted for age and socioeconomic status, was calculated in 3 bands of increasing radius from the plants: 0-2, 2-4, and 4-7 km (reference group), using a Poisson regression model. RESULTS:the cohort was of 39,775 people, 32%of whom lived near (0-4 km) the NPP. No differences in mortality was found when comparing the cohort with the regional population; among women living within 7 km from the NPP, we found thyroid cancer incidence higher than expected (SIR 1.53 CI95% 1.18-1.95). However, when the analysis was conducted on the basis of the distance from the NPP, we found a statistically significant increase in male mortality only for causes unrelated to radiation exposure (all causes, stomach cancer, and cardiovascular diseases). No mortality excess was observed among women living close to the NPPs. No statistically significant distance-related gradient was observed for cancer incidence both in men and women. CONCLUSIONS:living close to the NPP was not associated with mortality for causes related to radiation exposure. However, the results suggest to continue the epidemiological surveillance of the population.