[Risk factors of cancer of the pancreas from analytic epidemiologic studies].

The main case control studies and over 30 cohort studies of pancreatic cancer, performed during the period 1973-1988 are reviewed. For each type of risk factor a table is given which summarizes the conclusions of the various studies. The multifactorial etiology of this cancer is reflected by the number of different exposures studied. In our opinion, the effects of tobacco are indisputable even though they have not been systematically described in all studies. Cigarette smoking is associated with two-fold increase in risk. The effect of different forms of smoking, such as filter utilization, pipe, colour of tobacco, etc. on the risk are unclear. The effect of alcohol is less important although this needs to be confirmed. Tobacco use was not controlled in many studies where a positive results was obtained for alcohol. The same confounding effect has perhaps biased the results of several studies carried out for coffee drinkers. While not definitively confirmed, the elevated risk associated with coffee drinking appears to be plausible. Results on other dietary factors have not been widely reported. However, a positive association has been found for consumption of meat and fat and a negative one for vegetables and fruits. In some cases, these results are amplified by a dose-response effect. Further nutritional studies are required in order to resolve this question. Previous medical conditions have also been considered, in particular diseases of the pancreas, i.e. diabetes mellitus and pancreatitis. The results are not consistent, as causes and effects are not easy to detect. However, positive results for diabetes mellitus are more convincing than those for other conditions. Potential effects of work-related exposure have been analysed by many researchers although no firm conclusions can be drawn upon review of those results. Among the suspected risks, one has to consider the chemical industry and industrial exposure in oil refinery, aluminium reduction, and coke transformation. Occupations involving contact with dyes, paint and thinners could also be in the risk group.