In a rural area near Florence a population-based screening program for breast cancer was started in 1970, offering a mammography test every 2.5 years to all women between 40 and 70 years of age. In order to evaluate the efficacy of this program a case-control study has been carried out comparing the screening history of all women who died from breast cancer in this area in the years 1977-1984, and who had been diagnosed after the start of the program, with that of a matched group of living controls. The case-control study showed that the overall O.R. of dying from breast cancer in the study period for "screened" versus "never-screened" women was 0.53 (95% C.I. = 0.29-0.95). After evaluation of other variables as potential confounders the adjusted O.R. estimates were respectively 0.57 (95% C.I. = 0.35-0.92) and 0.32 (95% C.I. = 0.20-0.52) for women screened only once or at least twice. No significant protective effect was shown for women below the age of 50 years.