[Estimates of Attributable Fraction of occupational cancers in the recent epidemiological literature].

The article reviews the estimates of Attributable Fraction (AF) of occupational cancers. Because of their relevance, it starts with a synthesis and a comment on the estimates of avoidable cancers, and among these of cancers due to occupation, as expressed in 1981 by Doll and Peto. The main studies that have brought back into the epidemiological pathway the exercise of producing FA are quoted and the results of the new studies reviewed. The dimension of occupational cancers is of public health importance, because caused by exposures assumed to be avoidable. However, the estimates of AF are prone to uncertainties and limitations. Occupational cancers represent the major killer among deaths due to occupational diseases and injuries and a relevant fraction, especially among males, of total cancers. When related to developed countries, estimates may underestimate the risks, because of the insufficient data on exposures for workers in small firms, the lack of information on exposures in agriculture, the scanty availability of epidemiological studies on cancer risks among women. The trend towards a reduction of employees in industrial activities, the elimination or control of some exposures should suggest, instead, a reduction for occupational cancer. However, the more recent estimates of AF of occupational cancers are still in line with the estimates expressed more that 20 years ago. In developing countries, the estimates of AF are plenty of assumptions. In addition, some characteristics (i.e. the younger age of starting work, the extension of work at older age) suggest the need of better data, whereas information on exposures and events are dramatically lacking.